Compare commits

..

362 Commits

Author SHA1 Message Date
Sarah Hoffmann
489653b6ed prepare 3.5.2 release 2020-09-24 11:54:17 +02:00
Sarah Hoffmann
bb0c42e638 update osm2pgsql to same version as master 2020-09-24 11:53:23 +02:00
Sarah Hoffmann
2d226be156 remove ST_Covers check when also testing for ST_Intersects
Using both is slightly problematic because they have different
ways to use the index. Newer versions of Postgis exhibit a
query planner issue when both functions appear together.
As ST_Intersects includes ST_Covers, simply remove the latter.
2020-09-24 11:53:23 +02:00
Sarah Hoffmann
61fe274c6e make sure that all postcodes have an entry in word
It may happen that two different postcodes normalize to exactly
the same token. In that case we still need two different entries
in the word table. Token lookup will then make sure that the correct
one is choosen.

Fixes #1953.
2020-09-24 11:53:23 +02:00
marc tobias
0ac99bc2a9 starting PHP 5.4 get_magic_quotes_gpc() returns false, no need to check 2020-09-24 11:53:23 +02:00
Sarah Hoffmann
76ddace267 tests: use larger grid to avoid rouding errors 2020-09-24 11:53:13 +02:00
Sarah Hoffmann
777c70926a increase splitting for large geometries
When computing the address parts for a geometry, we need to do
a ST_Relates lookup in the location_area_large_* tables. This is
potentially very expensive for geometries with many vertices.
There is already a funtion for splitting large areas to reduce the
impact. This commit reduces the minimum area of a split, effectively
increasing the number of splits.

The effect on database size is minimal (around 3% increase), while
the indexing speed for streets increases by a good 60%.
2020-09-24 11:52:17 +02:00
Sarah Hoffmann
b2886426b7 indexer: allow batch processing of places
Request and process multiple place_ids at once so that
Postgres can make better use of caching and there are less
transactions running.
2020-09-24 10:17:31 +02:00
Sarah Hoffmann
a836ca8991 indexer: move progress tracker into separate class 2020-09-24 10:17:21 +02:00
Sarah Hoffmann
30016b98b7 indexer: get rid of special handling of few places
Given that we do not distiribute geometry sectors to threads anymore,
there is no point in this kind of special handling.
2020-09-24 10:17:12 +02:00
Sarah Hoffmann
0f5fc10e31 make house number reappear in display name on named POIs
After 6cc6cf950c names and house numbers
of POIS got mingled into a single item when creating the display name.
Add the house number as extra information without place_id to avoid
later mangling.
2020-09-24 10:16:03 +02:00
Sarah Hoffmann
72335fb631 make indexing during updates less quiet
Adjust verbosity behaviour to that of indexing during setup.
2020-09-24 10:14:49 +02:00
Sarah Hoffmann
a863392938 add wiki tags to all styles
wikipedia and wikidata tags are needed to compute the importance
so we need to put them into extra tags for all styles.

Fixes #1885.
2020-09-24 10:13:17 +02:00
Sarah Hoffmann
168c2e222e prepare 3.5.1 release 2020-06-29 20:54:50 +02:00
Sarah Hoffmann
770f8e31a8 update libosmium to 2.15.6
Fixes an issue where osm2pgsql hangs on a particularly
complicated multipolygon.
2020-06-29 20:54:50 +02:00
Sarah Hoffmann
dd55a76d6d make phpcs happy 2020-06-28 23:19:49 +02:00
Sarah Hoffmann
670cff0d09 disable JIT and parallel processing for osm2pgsql in updates
This is known to cause issues because of bad indexing
statistics.
2020-06-28 23:19:41 +02:00
Sarah Hoffmann
a0e7d80daf prepare 3.5.0 release 2020-06-06 20:30:29 +02:00
Sarah Hoffmann
d7e2f61e13 Merge pull request #1822 from lonvia/document-address-labels
document which labels may appear in the address info
2020-06-06 18:02:40 +02:00
Sarah Hoffmann
96ed4b02d7 document which labels may appear in the address info
The list is manually generated and only valid for the default
configuration as used on openstreetmap.org.

Fixes #1808.
2020-06-06 17:32:30 +02:00
Sarah Hoffmann
cffc7c0121 parents for large POIs must be address features
There are a couple of places with a search rank < 25 which are
not addressable like waterways and islands. We don't want them
to function as parents for POI-level objects. So use the
address rank for finding parents, not the search rank.

See #1815.
2020-06-03 11:30:51 +02:00
Sarah Hoffmann
d89000cc3d Merge pull request #1813 from lonvia/revert-concurrent-indexing
revert building indexes concurrently
2020-06-01 22:13:33 +02:00
Sarah Hoffmann
3661c75b39 Merge pull request #1814 from lonvia/disable-jit
Disable JIT and parallel workers when indexing
2020-06-01 22:13:07 +02:00
Sarah Hoffmann
3b20b11a9f remove warnings about postgres 12 and postgis 3 2020-05-30 11:25:00 +02:00
Sarah Hoffmann
cca366196d Disable JIT and parallel workers when indexing
Locally disable jit and parallel workers in the connection that
do indexing. The query planner tends to be overenthusiatic about
using JIT. But with the rather less complex queries we have, the
overhead tends to be larger than the performance gain.

Fixes #1677.
2020-05-30 11:20:16 +02:00
Sarah Hoffmann
e09d444068 revert building indexes concurrently
This does not really solve the issue with blocking autovacuum
requests and can lead to incomplete indexes and bogus
out-of-disk messages.

Fixes #1549.
2020-05-30 11:00:05 +02:00
Sarah Hoffmann
4956f5e710 Merge pull request #1809 from lonvia/fix-display-names
Fix some glitches in choice of address tags
2020-05-27 21:28:17 +02:00
Sarah Hoffmann
5bebdfa434 Merge pull request #1804 from lonvia/ranking-improvement-germany
Localized ranking adaptions for Germany, Sweden and Norway
2020-05-27 11:58:21 +02:00
Sarah Hoffmann
aea915aa8d prefer linked place type over own place type
For state cities, tagging might prefer the place=state on
the admin boundary. The linked place is a more reliable indicator.
2020-05-27 11:31:50 +02:00
Sarah Hoffmann
e0d29f398e each address line must fill at most one geocodejson field
This fixes an issue where a postcode with rank_address 5
would also appear in the state field.
2020-05-27 11:16:27 +02:00
Frederik Ramm
c43b39bd88 Fix script names in README (#1805) 2020-05-25 12:45:35 +02:00
Sarah Hoffmann
8218da27b3 adapt tests to new ranks 2020-05-23 19:40:41 +02:00
Sarah Hoffmann
aa4bd00631 Adapt boundary labels for Sweden and Norway
This also gives us the correct labels for address output in
json and xml.
2020-05-23 16:19:27 +02:00
Sarah Hoffmann
af6b9fdb39 fix admin levels for Norway and Sweden
Admin levels 3 and 4 are used for region and county respectively,
so downgrade the ranking.
2020-05-23 15:48:40 +02:00
Sarah Hoffmann
c1b6493373 adapt municipality and region for Germany 2020-05-23 15:20:15 +02:00
Sarah Hoffmann
c386cca73f Merge pull request #1801 from lonvia/rework-classtypes
Rework ClassTypes helper functions
2020-05-20 08:22:56 +02:00
Sarah Hoffmann
cadbdaff18 fix style 2020-05-18 22:20:36 +02:00
Sarah Hoffmann
57510f517a adapt tests to modified address types 2020-05-17 16:53:33 +02:00
Sarah Hoffmann
3a2ddbe2e0 encapsulate icon URL in a function 2020-05-17 16:46:45 +02:00
Sarah Hoffmann
859347523f also adapt uses of ClassTypes in website/ 2020-05-17 16:46:45 +02:00
Sarah Hoffmann
528fe6553f adapt php tests
Also fixes some errors found by the tests.
2020-05-17 16:46:45 +02:00
Sarah Hoffmann
1faa0f4d41 reorganise class/type information
Add a separate function for each property which saves necessary
information independently. Simplify computation of labels and
simple labels to not explicitly save the labels.
2020-05-17 16:46:45 +02:00
Sarah Hoffmann
82a11cae2d first draft 2020-05-17 16:46:45 +02:00
Sarah Hoffmann
431948d768 nominatim: always use deadlock-protected wait
Fixes #1785.
2020-05-15 18:49:27 +02:00
Sarah Hoffmann
f69c3d2b66 Merge pull request #1793 from lonvia/remove-struct-params-in-gui
search UI: hide unused query parameters
2020-05-15 16:50:54 +02:00
mmd
08b05964fa Update travis to bionic=Ubuntu18 (#1800) 2020-05-14 22:52:04 +02:00
Sarah Hoffmann
bd7f597682 Merge pull request #1797 from mtmail/jquery-3-5-1
update jquery dependency 3.5.0 => 3.5.1
2020-05-14 20:54:17 +02:00
marc tobias
6d4fbc9d32 update jquery dependency 3.5.0 => 3.5.1 2020-05-14 15:53:05 +02:00
Sarah Hoffmann
124410a17b improve syntax highlighting for apache and nginx examples 2020-05-13 10:13:15 +02:00
Sarah Hoffmann
a543d57cbd switch to php-fpm 7.3
Also fixes indent.
2020-05-13 10:04:31 +02:00
Sarah Hoffmann
8c3a0efe8b Merge branch 'patch-1' of https://github.com/ganeshkrishnan1/Nominatim into ganeshkrishnan1-patch-1 2020-05-13 09:55:48 +02:00
Sarah Hoffmann
9e2841ad44 search UI: hide unused query parameters
Only send query parameters relevant for the current query
type (simple/structured), hide the other input fields.

This is quite a bit of CSS state changing, so move the intial
setup of the input field states into Javascript.
2020-05-11 00:19:33 +02:00
Sarah Hoffmann
233e5f7c0e show simple query field when no parameters are given 2020-05-10 23:52:53 +02:00
Sarah Hoffmann
d5d9445cfd Fix PHP errors in structured HTML output
Correctly handle missing parameters.
2020-05-10 23:41:04 +02:00
Sarah Hoffmann
7be7417b5b Merge pull request #1792 from lonvia/remove-from-location-area
remove linked places also from the location_area_large tables
2020-05-10 15:49:20 +02:00
Sarah Hoffmann
0a14142156 remove linked places also from the location_area_large tables
We don't want linked places to show up in addresses either,
so remove them from the address lookup table.
2020-05-10 13:59:47 +02:00
galewis2
a5e3785843 Add simple/structured query selector to HTML search page (#1722) 2020-05-08 01:29:44 +02:00
Sarah Hoffmann
fc19ebb218 Merge pull request #1786 from lonvia/remove-ubuntu-1604
remove Ubuntu 16 installation instructions
2020-05-07 22:42:31 +02:00
Sarah Hoffmann
b45411f988 Merge pull request #1782 from Simon-Will/1781-make-tests-work-with-phpunit-8
Make tests work with phpunit 8
2020-05-07 22:01:35 +02:00
Sarah Hoffmann
42f6371e47 remove Ubuntu 16 installation instructions
Also fixes up CentOS 8 links in documentation.
2020-05-07 21:55:04 +02:00
Simon Will
be2aa6ab3a Use Ubuntu’s packaged composer, not the custom installation 2020-05-07 21:44:45 +02:00
Sarah Hoffmann
6e39ed9573 Merge pull request #1780 from Simon-Will/1768-vagrant-installation-for-ubuntu-20
Add vagrant machine for Ubuntu 20.04
2020-05-07 20:46:44 +02:00
Simon Will
daf45a2993 Integrate Ubuntu 20 instructions into documentation 2020-05-07 00:36:13 +02:00
Simon Will
d351b10fde Document minimum phpunit version 2020-05-06 23:47:16 +02:00
Simon Will
0b21050904 Install phpunit 8 on Ubuntu 18 with composer 2020-05-06 23:46:53 +02:00
Sarah Hoffmann
644a7f524c Merge pull request #1784 from krahulreddy/patch-1
Removed redundant question
2020-05-06 21:37:29 +02:00
K Rahul Reddy
53949ace36 Removed redundant question 2020-05-06 21:26:32 +05:30
Simon Will
14dba39157 Use assertEqualsWithDelta for float comparisons
PHPUnit 7.3 introduced the functions assertEqualsWithDelta for comparing
floats with a delta. The old four-argument version of assertEquals is
deprecated in PHPUnit 8 and removed in PHPUnit 9.

This commit means that the tests will fail with PHPUnit < 7.3 because
assertEqualsWithDelta is not defined there.
2020-05-05 23:43:09 +02:00
Simon Will
43fd2a7423 Declare return type of testcase setUp method
PHPUnit 7 changed the signature of the TestCase methods to include the
return type.
2020-05-05 23:40:18 +02:00
Simon Will
4b0ac5356e Add vagrant machine for Ubuntu 20.04
The instructions in
[`VAGRANT.md`](42c80893cb/VAGRANT.md)
still work as before. The names of the Vagrant machines are updated so
that Ubuntu 18.04 (previously called `ubuntu`) is now called `ubuntu18`
and Ubuntu 20.04 is now called `ubuntu20`.

The version changes from Ubuntu 18.04 to Ubuntu 20.04 are:

  - Python: 3.6 to 3.8
  - Postgres: 10 to 12
  - PHP: 7.2 to 7.4

In the `apt-get`, I changed `--force` to `--allow-downgrades --allow-remove-essential --allow-change-held-packages`, because the former is deprecated. Cf. the [manpage of apt-get](http://manpages.ubuntu.com/manpages/focal/man8/apt-get.8.html)

The php module `codesniffer` was previously installed via Composer, but it is available via the Ubuntu repository, so I installed it via `apt-get` now.
2020-05-05 23:10:35 +02:00
Sarah Hoffmann
c2f0d8e5ba docs: add link to new status page 2020-05-04 21:11:57 +02:00
marc tobias
0fb93b1e8a documenation for /status endpoint 2020-05-04 17:06:06 +02:00
Sarah Hoffmann
f94828c3f4 properly escape class parameter
The class parameter was used as is, allowing for potential
SQL injection via the API.

Thanks to @bladeswords for finding this.
2020-05-02 21:54:14 +02:00
Sarah Hoffmann
0e1e7c7df2 Merge pull request #1770 from lonvia/eyusupov-separate-compilation
Separate compilation
2020-04-26 21:48:43 +02:00
Sarah Hoffmann
06110ba358 Merge pull request #1769 from lonvia/display-name-order
Ensure that result object name is always first in display_name
2020-04-26 16:18:56 +02:00
Sarah Hoffmann
bae69f0102 cmake: reintroduce check script 2020-04-26 16:17:43 +02:00
Sarah Hoffmann
77e7f4696b fix docs typos 2020-04-26 15:00:28 +02:00
Sarah Hoffmann
47fb2c9126 cmake: restructure splitting between modules
Make a clear distinction between parts used for the importer
and parts used for the API.
2020-04-26 14:17:21 +02:00
Sarah Hoffmann
2ab9e4acd3 Merge branch 'separate-compilation' of https://github.com/eyusupov/Nominatim into eyusupov-separate-compilation 2020-04-26 10:47:41 +02:00
Sarah Hoffmann
65ee7a8002 Merge pull request #1754 from mtmail/nominatim-db-tests-against-postgres
Nominatim::DB tests against separate postgresql database
2020-04-26 10:20:30 +02:00
marc tobias
a5d0657d9b lonvia PR feedback 2020-04-26 03:33:15 +02:00
Sarah Hoffmann
b8f01f91ca simplify display_name computation 2020-04-26 00:18:29 +02:00
Sarah Hoffmann
6cc6cf950c ensure that result object name is always first in display_name
The display name might be mixed up if the result object has a lower
rank_address than its address members.
2020-04-26 00:14:55 +02:00
Sarah Hoffmann
0b0349f746 Merge pull request #1752 from mtmail/new-oo-shell-class
new PHP Nominatim\Shell class to wrap shell escaping
2020-04-25 16:48:04 +02:00
Sarah Hoffmann
2740974a13 Merge pull request #1758 from krahulreddy/advanced-installations
Advanced installations
2020-04-22 09:59:44 +02:00
Sarah Hoffmann
97a9a262bb Merge pull request #1764 from mtmail/docs-countrycodes-based-on-adminlevel-2
API docs: countrycode assignment happens using admin_level=2 tags
2020-04-22 09:57:59 +02:00
Sarah Hoffmann
207efe700f highway:construction should appear as 'road' in the address list
Fixes #1763.
2020-04-22 09:08:33 +02:00
marc tobias
e33315eaa6 API documentation: clarification countrycode assignment happens using admin_level=2 tags 2020-04-21 17:42:12 +02:00
Sarah Hoffmann
5469d02d03 nominatim.py: fix wrong use of assert
Fixes #1762.
2020-04-19 17:59:49 +02:00
K Rahul Reddy
42c80893cb Fix documentation links (#1760)
Update installation documentation link in VAGRANT.md, update.php
2020-04-19 00:42:24 +02:00
K Rahul Reddy
5c56ea3198 Adjustments made to documentation 2020-04-17 21:53:50 +05:30
K Rahul Reddy
42f86329a9 Added Advanced Installations documentation 2020-04-17 21:53:41 +05:30
K Rahul Reddy
08e273c0c7 Added scripts for multiple country setup and updates 2020-04-17 21:50:59 +05:30
Sarah Hoffmann
5f8f98fa03 Merge pull request #1756 from lonvia/downgrade-waterways
downgrade waterways
2020-04-17 08:46:40 +02:00
Sarah Hoffmann
08c53ae27d downgrade waterways
A lot of streams in OSM are of minor importance, they certainly
should show up lower in the list of results than villages. Those
rivers/streams that are well known have a wikipedia page and get
a higher importance from that.

The disadvantage with downgrading is that the address gets even
more useless but that's something that needs to be solved outside
the rank search.
2020-04-14 17:14:20 +02:00
Sarah Hoffmann
f4f369895c Merge pull request #1753 from mtmail/fix-travis-ci-badge
fix Travis-CI badge in README output
2020-04-13 18:36:02 +02:00
marc tobias
38c21de0ee Nominatim::DB tests against separate postgresql database 2020-04-13 18:01:37 +02:00
marc tobias
43cf36e0c7 fix Travis-CI badge in README output 2020-04-13 17:55:17 +02:00
Sarah Hoffmann
9a9ff95989 fix logging of lookup calls
Log start was called but the actual writing was missing.
2020-04-13 11:55:24 +02:00
marc tobias
fc40939775 new PHP Nominatim\Shell class to wrap shell escaping 2020-04-12 03:50:40 +02:00
Sarah Hoffmann
553d8a828c Merge pull request #1751 from lonvia/respect-admin-hierarchy
Address ranks must not invert admin_level hierarchy
2020-04-11 23:41:19 +02:00
Sarah Hoffmann
80f7392fb1 address ranks must not invert admin_level hierarchy
When inheriting an address rank from a linked place we
must be careful not to destroy the hierarchy established
through boundary admin_level. Therefore, before assigning
an address rank from a linked place, find the next higher
boundary in the admin_level hierarchy, look up its address
rank and then only use the address rank from the linked
place if it is higher.
2020-04-11 20:56:30 +02:00
Sarah Hoffmann
61535c9972 Merge branch 'update-jquery-leaflet' of https://github.com/mtmail/Nominatim into mtmail-update-jquery-leaflet 2020-04-11 20:53:14 +02:00
Sarah Hoffmann
b443c92a7a Merge pull request #1749 from lonvia/ranking-during-updates
Reset search and address ranks on update
2020-04-11 20:52:05 +02:00
marc tobias
22da6c541d website dependencies: jQuery v2.1 => 3.5, leaflet 1.3 => 1.6 2020-04-11 18:18:57 +02:00
Sarah Hoffmann
cd96354bc7 reset address and search ranks on update
With ranks being dynamically changed through linking of places,
it is important to reset the ranks on update, so that changes
of the rank due to changes in linking are correctly taken into
account.
2020-04-11 09:20:13 +02:00
Sarah Hoffmann
c6d859a08a factor out computation of address and search rank 2020-04-10 23:18:31 +02:00
Sarah Hoffmann
ef47515420 make admin levels 3 and 7 distinct ones in addresses
There really is no need to conflate these two levels as they
are in use in various countries.

Also adds province as a distinct place.

Fixes #1736.
2020-04-10 22:58:11 +02:00
Sarah Hoffmann
a471a3d1b0 Merge pull request #1745 from lonvia/shuffle-sql-functions
Some more SQL function reorganisation
2020-04-10 17:23:09 +02:00
Sarah Hoffmann
79a68fc2db Merge branch 'deletable-and-polygons-as-json' of https://github.com/mtmail/Nominatim into mtmail-deletable-and-polygons-as-json 2020-04-10 17:20:51 +02:00
marc tobias
93ddd46231 Add JSON output for /deletable.php and /polygons.php 2020-04-10 15:34:56 +02:00
Sarah Hoffmann
f5f0c197be move ranks-related functions in separate sql file
Also adds a common function for computing the update radius
around place nodes.
2020-04-10 11:34:14 +02:00
Sarah Hoffmann
4a30ec28b9 move helper functions from placex_triggers into utils
Also adds documentation for these functions.
2020-04-10 11:05:11 +02:00
Sarah Hoffmann
37ef9bb3d3 Merge pull request #1742 from mtmail/travis-ci-add-os
Travis-Ci configuration: remove -sudo-, add -os-
2020-04-10 08:46:50 +02:00
marc tobias
3e1d4a87fa travis-ci configuration: remove -sudo-, add -os- 2020-04-10 01:20:51 +02:00
Sarah Hoffmann
320d46cc96 Merge pull request #1741 from filimongeorge/patch-1
Updated Import and Update .md file
2020-04-09 23:00:34 +02:00
Sarah Hoffmann
a06ceeef4c Merge pull request #1740 from mtmail/setupclass-index-outputfile-not-used
SetupClass.php: remove unused variable
2020-04-09 22:57:51 +02:00
Sarah Hoffmann
def573d7b4 Merge pull request #1739 from lonvia/remove-self-from-geojson
Further tweaks to geocodejson output
2020-04-09 22:51:20 +02:00
filimongeorge
7f7d29fdd1 Updated Import and Update .md file 2020-04-09 20:51:38 +03:00
marc tobias
c611d49941 SetupClass.php: remove unused variable 2020-04-08 14:16:06 +02:00
Sarah Hoffmann
11cd648699 remove name from geocodejson when not set 2020-04-08 11:19:43 +02:00
Sarah Hoffmann
98be5bf637 adapt tests to geocodejson format adaptions 2020-04-08 11:19:43 +02:00
Sarah Hoffmann
29df9771bb further tweaks to geocodejson address output
Removes the place itself from the address details and use
the lowest ranking element in the rank range for the output.
2020-04-08 11:11:33 +02:00
Sarah Hoffmann
e68c1132da ignore isaddress in details output when it is not present 2020-04-08 10:28:28 +02:00
Sarah Hoffmann
1047b1c191 Merge pull request #1737 from mtmail/expose-isaddress-in-details-json
details JSON: also print isaddress addressline field
2020-04-07 20:49:31 +02:00
marc tobias
9431e80eb4 details JSON: also print isaddress addressline field 2020-04-07 14:50:41 +02:00
Sarah Hoffmann
178501de61 Merge pull request #1734 from krahulreddy/fixed-parselatlon
Added whitespace support for parseLatLon
2020-04-05 23:25:50 +02:00
Sarah Hoffmann
81c7f618fb avoid deletes on search_name in reverse-only mode 2020-04-04 18:26:27 +02:00
Rahul
eb2d816f2a Added test cases for whitespaces in LatLon 2020-04-04 00:53:40 +05:30
Rahul
244cb0e98c Added whitespace characters support in LatLon parsing 2020-04-04 00:53:40 +05:30
Sarah Hoffmann
300ac4b77b fix phpcs issues 2020-04-03 20:08:08 +02:00
Sarah Hoffmann
0d189ac5df Merge pull request #1733 from krahulreddy/whitespaces-considered-as-single-space
Support whitespace characters(x09-x0d) as single space
2020-04-03 18:01:47 +02:00
Sarah Hoffmann
fed2c307a7 Merge pull request #1732 from lonvia/improve-geocodejson-output
Improve geocodejson output
2020-04-02 21:21:04 +02:00
K Rahul Reddy
7aa2df5389 Support whitespace characters(x09-x0d) as single space 2020-04-02 05:04:40 +05:30
Sarah Hoffmann
975ef0b305 re-add district to geocodejson 2020-04-01 21:24:42 +02:00
Sarah Hoffmann
e59146a733 update documentation for geocodejson
Address parts should be usable now.
2020-04-01 11:17:25 +02:00
Sarah Hoffmann
8150c3602b add tests for geocodejson address fields 2020-04-01 11:14:48 +02:00
Sarah Hoffmann
ca8d776724 determine geocodejson address by rank instead of type
Using the address rank to set the address parts catches
a much wider variety of types like 'town' and 'suburb'.
With recent address ranking changes the rank ranges
are relatively reliable.
2020-04-01 11:12:52 +02:00
Sarah Hoffmann
fdc40d5169 factor out geocodejson address generation
Unifies the two implementations currently used for search and address.
2020-04-01 10:27:17 +02:00
Sarah Hoffmann
d0a97056c4 Merge pull request #1731 from lonvia/remove-polygon-from-docs
docs: remove example with polygon parameter
2020-04-01 10:21:45 +02:00
Sarah Hoffmann
e98619f801 docs: remove example with polygon parameter
This parameter was undocumented, long deprecated and is gone now.
2020-03-31 20:10:03 +02:00
Sarah Hoffmann
86eebc4305 fix typo
Fixes #1730
2020-03-31 19:53:55 +02:00
Sarah Hoffmann
4930f776fe fix handling of postcode areas in addresses
The order of preference is now:
1. a post code on the place itself
2. a post code area in the address
3. the computed postcode from the place

Fixes #1723.
2020-03-30 23:27:48 +02:00
Sarah Hoffmann
19948c378a adapt tests to new borough ranking 2020-03-30 23:04:20 +02:00
Sarah Hoffmann
b3215b802d downgrade borough and remove unincorporated area 2020-03-30 18:37:23 +02:00
Sarah Hoffmann
ed16d5b6aa Merge pull request #1729 from lonvia/fix-details-link-for-boundaries
Fix details link for boundaries
2020-03-29 23:12:16 +02:00
marc tobias
7a94872413 remove polygon=1 (polypoints) feature 2020-03-29 21:58:11 +02:00
Sarah Hoffmann
98750922eb also emit place_type in json version of details 2020-03-29 21:06:39 +02:00
Sarah Hoffmann
60c4c9ef2c rather use new place_type in getAddressNames()
If for a boundary the place_type is defined, handle the address
part like a place node. This is the same behaviour as before
when class/type where patched earlier.
2020-03-29 20:49:35 +02:00
Sarah Hoffmann
101f04bbf2 Fix address link for boundaries in details
Removes the special casing for boundaries with a place
type in get_addressdata(). Instead the place type is now
returned as an extra field, so that the caller has to
handle the situation.

This fixes the details link next to the address in the details
view, which previously would go to a place class instead of the
original boundary class.
2020-03-29 17:40:56 +02:00
Sarah Hoffmann
4c593fa859 Merge pull request #1720 from lonvia/better-linking-of-places
Use wikidata tags for improving linking of places with boundaries
2020-03-27 21:12:39 +01:00
Aakankasha Sharma
6603ad4006 Updated TIGER database link in documentation (#1725)
Updated TIGER database link in documentation
2020-03-27 15:50:05 +01:00
Sarah Hoffmann
d56c69dd01 adapt API tests to place linkage changes
The missing district is due to a data error for wikidata tags.
2020-03-25 11:38:31 +01:00
Sarah Hoffmann
e26a300c2f use wikidata tag for linking places
Having the same wikidata is a strong indicator that the same place
is meant. There are some assignment errors where the wikidata does
not link to the object itself but to something that mentions the
place. To reduce errors there, prefer same name.
2020-03-21 22:46:54 +01:00
Sarah Hoffmann
405482ede4 remove linking via admin_centre role
The admin_centre role is for the seat of government which is not
the same as the administrative entity. This is mostly used
correctly these days, so avoid matching by that role.
2020-03-21 21:59:11 +01:00
Sarah Hoffmann
3db2b05069 linking: better name matching for address-less places
Administrative boundaries that do not figure in the address
should still be able to take part in the name matching.
Use the rank_search for comparison in this case.
2020-03-21 21:57:04 +01:00
Sarah Hoffmann
ce5870223a Merge pull request #1706 from mtmail/warn-if-no-tiger-files-found
print warning if no Tiger files found
2020-03-06 22:55:37 +01:00
Sarah Hoffmann
9c1bb87493 Merge pull request #1707 from lonvia/regression-address-in-area
place node address parts must be in lower rank area
2020-03-06 22:55:24 +01:00
Sarah Hoffmann
1f7394dd54 place node address parts must be in lower rank area
This fixes a regression where the area of the lower ranking
area was not computed correctly.

Also excludes postcodes areas now as they have their own
hierarchy.
2020-03-06 21:51:38 +01:00
marc tobias
bb569aa484 print warning if no Tiger files found 2020-03-06 17:52:46 +01:00
Sarah Hoffmann
b0a350db37 Merge pull request #1705 from lonvia/delete-linkee-from-search-name
Remove linkees from search_name
2020-03-04 11:55:05 +01:00
Sarah Hoffmann
78526a33b4 Remove linkees from search_name
Fixes #722
2020-03-04 11:36:39 +01:00
Sarah Hoffmann
ab997b7fb1 Merge pull request #1704 from lonvia/centroid-within-geometry
linked centroids must always be within geometry
2020-03-04 10:18:57 +01:00
Sarah Hoffmann
6d431aebb7 linked centroids must always be within geometry
When using a linked place as centroid for a boundary, check
first that it is really within the area. If it is outside,
just keep the computed centroid because a centroid outside the
area just causes havok.

Fixes #1352.
2020-03-04 09:59:57 +01:00
Sarah Hoffmann
a00ea23847 Merge pull request #1702 from lonvia/rename-derived-place
Make admin boundaries inherit address rank from place nodes
2020-03-04 08:08:39 +01:00
Sarah Hoffmann
53ca751a02 fix style 2020-03-01 22:24:32 +01:00
Sarah Hoffmann
8c444378bc better grouping 2020-02-28 22:10:35 +01:00
Sarah Hoffmann
55fdf0abda output linked place into address details 2020-02-28 22:07:06 +01:00
Sarah Hoffmann
acd8ca2ebd add testing for rank adaption while linking 2020-02-28 15:22:48 +01:00
Sarah Hoffmann
06fdfad89e link against place nodes by place type
If a boundary relation has no label member preferably
link against a place node with the same place type.

Also inherit the rank_address from the place node (only
has an effect when linking via lable member or place type).
2020-02-28 15:22:48 +01:00
Sarah Hoffmann
00ca493f33 move linked place type into linked_place extratags
Using linked_place means that we don't overwrite any
place tags on the boundary. This is important when we
wanto to use the information for linking.
2020-02-28 15:22:48 +01:00
Sarah Hoffmann
b00d16fd7d Merge pull request #1698 from lonvia/cleanup-partition-functions
Cleanup partition functions
2020-02-26 20:21:10 +01:00
Sarah Hoffmann
03c373a4b3 make all query partition functions stable 2020-02-26 11:41:49 +01:00
Sarah Hoffmann
bdaa39573f remove unused nearfeature types
Also move the remaining nearfeaturecentr type close to the
function that is using it.
2020-02-26 11:01:29 +01:00
Sarah Hoffmann
8a4c7f6e2b simplify getNearestParallelRoadFeature function
The function only ever returns one result of which only the
place_id is used. So simplify it to return a single place_id
only (or NULL if none is found).

Also fix typo in function name.
2020-02-26 11:01:04 +01:00
Sarah Hoffmann
84ea0753d8 simplify getNearestRoadFeature function
The function only ever returns one result of which only the
place_id is used. So simplify it to return a single place_id
only (or NULL if none is found). Rename funciton to avoid
conflicts when updating an existing database.
2020-02-26 10:58:55 +01:00
Sarah Hoffmann
c1ef56c870 advise against using Postgresql 12 and Postgis 3
See also #1677
2020-02-25 09:44:32 +01:00
Sarah Hoffmann
0e3252f045 revert using stricter uniqueness constraint on place
Multiple objects with the same (osm_type, osm_id, class) may
exist when we hold back deleting an area because it is so
large.

Fixes #1695.
2020-02-24 22:55:03 +01:00
Sarah Hoffmann
65df218f91 Merge pull request #1693 from lonvia/reorganize-addressline-computation
Reorganize addressline computation
2020-02-24 22:39:51 +01:00
Sarah Hoffmann
5220a92be4 adapt API tests 2020-02-22 16:46:03 +01:00
Sarah Hoffmann
d643ca8dee move address line computation in its own function 2020-02-21 16:38:14 +01:00
Sarah Hoffmann
de45152028 Merge pull request #1692 from mtmail/tests-for-HasSetAny
unit tests for ParameterParser::hasSetAny
2020-02-20 21:16:39 +01:00
marc tobias
7fd9d0eeef unit tests for ParameterParser::hasSetAny 2020-02-19 16:55:17 +01:00
Sarah Hoffmann
d35a0b392e Merge pull request #1691 from lonvia/structured-query-via-cmdline
add structured search to command-line query tool
2020-02-19 11:12:37 +01:00
Sarah Hoffmann
cbddfcde5b add structured search to command-line query tool 2020-02-19 11:04:07 +01:00
Sarah Hoffmann
02ffa752ea Merge pull request #1690 from lonvia/parenting-large-rank-30-areas
improve parenting for large areas with rank 30
2020-02-19 09:20:12 +01:00
Sarah Hoffmann
6189e0c79b improve parenting for large areas with rank 30
Instead of unconditionally parenting them to a street, the
larger areas get a parent area that contains them. To keep
things computationally light-weight, only use the centroid and
bbox to determine if an area is contained.

Requires renaming of parenting functions because renaming
a parameter of the function causes issues when updating the
function (it requires a manual delete, which I'd like to
avoid).
2020-02-19 08:43:53 +01:00
Sarah Hoffmann
6ed6a0d447 Merge pull request #1689 from mtmail/travis-postgres-stopped-working
Travis: documentation suggests we need to add postgresql-client package
2020-02-19 08:00:08 +01:00
marc tobias
484892ae97 Travis: documentation suggests we need to add postgresql-client package pre-startup 2020-02-18 23:45:21 +01:00
Sarah Hoffmann
027d9e938c Merge pull request #1688 from mtmail/snippet-noun-vs-snipped-verb
documentation wording: snipped->snippet
2020-02-18 22:55:20 +01:00
marc tobias
e171f90d81 documentation wording: snipped->snippet 2020-02-18 22:48:27 +01:00
Sarah Hoffmann
92c5d3b720 make sure that linked places are within a boundary
This is a regression from previous code refactoring.

Fixes #1684.
2020-02-18 22:46:32 +01:00
Sarah Hoffmann
2a6e8ad68e add bbox whereclause to make postgis 3.0 happy
Normally ST_Covers() should include a bbox index use,
so adding a bbox where clause is not really necessary.
However, the query planner messes up and uses a parallel
index search with a second index instead of exclusively
running on the geometry index, when the bbox part is
missing.
2020-02-16 14:10:22 +01:00
Sarah Hoffmann
55c8a0ac08 Merge pull request #1678 from lonvia/early-drop
Clean up intermediate tables earlier with --drop
2020-02-13 22:50:41 +01:00
Sarah Hoffmann
6073d948e6 fix duplicate keys in tests
The tests suddenly failed because the unique key constraint
is more strict and does no longer include the type.
2020-02-12 11:29:33 +01:00
Sarah Hoffmann
b9171dd10b clean up intermediate tables earlier with --drop
When --drop is given, we can remove all node geometry information
already after the import with osm2pgsql. Also drop all unnecessary
tables before creating the final indices.
2020-02-12 11:03:20 +01:00
Sarah Hoffmann
97b892fac2 Merge pull request #1675 from lonvia/refresh-connection-while-indexing
Fix a couple of issues with the new Python nominatim script
2020-02-12 08:18:09 +01:00
Sarah Hoffmann
b3fdf19b85 Merge pull request #1674 from mtmail/testdb-how-to-select-tiger-data
document how to extract subset of TIGER data needed for API tests
2020-02-11 22:57:39 +01:00
Sarah Hoffmann
8c89e16ad2 Merge pull request #1673 from mtmail/wikidata-wget-incomplete
wikipedia: wget didnt download, skip index generation
2020-02-11 22:56:45 +01:00
Sarah Hoffmann
960409c701 psycopg 2.6 is now usable on ubuntu 16 2020-02-11 22:49:03 +01:00
Sarah Hoffmann
d1eeaa59a6 nominatim.py: use async in connect() function
The _async parameter name is only supported since psycopg 2.7.
However, async is a keyword in Python >= 3.7, so using this
gives us a syntax error. Working around this by defining the
parameters in a dict and handing that into the connect function.
2020-02-11 22:16:17 +01:00
Sarah Hoffmann
882f496e0a nominatim.py: also catch deadlocks on final wait 2020-02-11 22:16:17 +01:00
Sarah Hoffmann
8b8aa1b4e6 regularly close connection while indexing
Postgres sooner or later runs out of memory when the connection
is used for too long.
2020-02-11 22:16:17 +01:00
marc tobias
932ac23f18 document how to extract subset of TIGER data needed for API tests 2020-02-11 18:50:27 +01:00
marc tobias
6c6560ca53 wikipedia: wget didnt download, skip index generation 2020-02-10 17:20:11 +01:00
Sarah Hoffmann
0698757e6e Merge pull request #1670 from lonvia/permalinks-for-tiger-and-interpolation
Enable Permalinks to dtails for tiger and interpolation
2020-02-09 21:07:19 +01:00
Sarah Hoffmann
3a3f9b3496 fix formatting 2020-02-09 16:57:55 +01:00
Sarah Hoffmann
97d87895bf details: also look for interpolations when way id is given 2020-02-09 16:50:04 +01:00
Sarah Hoffmann
c36fd72f99 use detailsPermaLink function on main website as well 2020-02-09 16:05:22 +01:00
Sarah Hoffmann
57ae3d03a1 return place_id link to details when not an OSM object
Stop-gap solution to find the right object for Tiger and
interpolation objects.
2020-02-09 15:45:38 +01:00
Sarah Hoffmann
3737712044 Merge pull request #1667 from mtmail/setup-delete-invalid-indices
setup: delete invalid indices in create-search-indices step
2020-02-09 14:21:17 +01:00
Sarah Hoffmann
8531339b4e Remove hack that changes the class/type of cities
This interferes badly with the details view.

Fixes #1668.
2020-02-09 12:14:32 +01:00
marc tobias
540b12537a setup: delete invalid indices in create-search-indices step 2020-02-08 15:16:20 +01:00
Sarah Hoffmann
9e2fb45783 Merge pull request #1665 from mtmail/centos7-php7
CentOS7: update from PHP 5.4 to 7.2, add psycopg2
2020-02-07 20:43:49 +01:00
Sarah Hoffmann
e7c128b973 Merge pull request #1666 from mtmail/ubuntu-installs-add-psycopg2
Vagrant Ubuntu: psycopg2 is required
2020-02-07 20:42:38 +01:00
marc tobias
d4a3470c9e Vagrant Ubuntu: psycopg2 is required 2020-02-07 15:26:09 +01:00
marc tobias
4165b8c011 CentOS7: update from PHP 5.4 to 7.2 2020-02-07 15:18:46 +01:00
Sarah Hoffmann
357ba2f64d Merge pull request #1663 from mtmail/vagrant-centos-8
Vagrant setup for CentOS 8
2020-02-06 21:12:28 +01:00
Sarah Hoffmann
3ce8818045 Merge pull request #1664 from mtmail/check-import-finished-for-reverse-only
check_import_finished.php - reverse_only mode has less indices
2020-02-06 21:07:04 +01:00
marc tobias
76082ac7cb check_import_finished.php - reverse_only mode has less indices 2020-02-06 16:48:06 +01:00
marc tobias
4a451671d3 Vagrant setup for CentOS 8 2020-02-06 00:43:30 +01:00
marc tobias
a3728b7188 details html page: no longer use place_id in URLs 2020-02-02 01:16:31 +01:00
Sarah Hoffmann
a8711ab013 fix verboseness of nominiatim script during updates 2020-01-31 18:18:50 +01:00
Sarah Hoffmann
3e4754febd Merge pull request #1648 from lonvia/nominatim-as-python-script
Replace C Nominatim indexer with Python script
2020-01-31 17:53:49 +01:00
Sarah Hoffmann
da1d661fa0 remove libxml dependency for travis as well 2020-01-31 15:15:35 +01:00
Sarah Hoffmann
1801db523b fix typo 2020-01-29 11:50:30 +01:00
Sarah Hoffmann
2979c39628 also adapt indexing command in update script 2020-01-29 11:36:12 +01:00
Sarah Hoffmann
bb9bb40287 update cMake build documentation
Remove the dependency on libxml, no longer needed.
2020-01-24 22:53:26 +01:00
Sarah Hoffmann
8f6fdfeb0b forgot to index last rank 2020-01-24 22:06:30 +01:00
Sarah Hoffmann
b4e6d72fde replace nominatim C program 2020-01-24 22:06:30 +01:00
Sarah Hoffmann
a338ebfce0 fix log levels 2020-01-24 22:06:30 +01:00
Sarah Hoffmann
4144364a15 add time display for nominatim.py 2020-01-24 22:06:30 +01:00
Sarah Hoffmann
11c0dd235b clean up and document script 2020-01-24 22:06:30 +01:00
Sarah Hoffmann
4a9502bf88 fix SQL and some other stuff 2020-01-24 22:06:30 +01:00
Sarah Hoffmann
6c0d6d3178 Revert "switch to threading"
This reverts commit 8b1c2181be5aa5335c68d36a49cab9c4e2cd8bef.
2020-01-24 22:06:30 +01:00
Sarah Hoffmann
0a26ca7104 switch to threading 2020-01-24 22:06:30 +01:00
Sarah Hoffmann
2a15b2522f use generator for thread choice 2020-01-24 22:06:30 +01:00
Sarah Hoffmann
c11d1d78e9 add prepared statement 2020-01-24 22:06:30 +01:00
Sarah Hoffmann
7e51aa4cef simple implementation 2020-01-24 22:06:30 +01:00
Sarah Hoffmann
9abb96fa6b Merge pull request #1647 from lonvia/split-out-linking
Split up placex update trigger code
2020-01-24 21:55:19 +01:00
Sarah Hoffmann
879aafc916 fix indent 2020-01-24 21:16:26 +01:00
Sarah Hoffmann
5ec25122f6 rename functions where return parameter changed
Postgresql cannot cleanly reimport these functions when
upgrading, so simply rename to avoid errors.
2020-01-23 22:28:43 +01:00
Sarah Hoffmann
9371b1aeb9 forgot new trigger sql 2020-01-23 22:28:43 +01:00
Sarah Hoffmann
6f6d116451 adapt index for changes name lookup 2020-01-23 22:28:43 +01:00
Sarah Hoffmann
3ff6eccfd7 move trigger creation later in setup 2020-01-23 22:28:43 +01:00
Sarah Hoffmann
5d1fa597ea clean up get_word_id function
Replaced by addr_ids_from_name() which also normalises the
string.
2020-01-23 22:28:43 +01:00
Sarah Hoffmann
3b6c2c9155 getNearestNamed*Feature functions better return values 2020-01-23 22:28:43 +01:00
Sarah Hoffmann
f863040b38 factor out parent search from addr:street/addr:place 2020-01-23 22:28:43 +01:00
Sarah Hoffmann
1033f8bce7 factor out searching for parent road for pois 2020-01-23 22:28:43 +01:00
Sarah Hoffmann
cf4dbbd681 remove unused function 2020-01-23 22:28:43 +01:00
Sarah Hoffmann
6dccc693d0 factor out computation of default names
Also moves the computation down the line so that we never
have to do it twice.
2020-01-23 22:28:43 +01:00
Sarah Hoffmann
c3dc66ce9c factor out place linking sql 2020-01-23 22:28:43 +01:00
Sarah Hoffmann
4856f56d61 adapt test to change in hamlet classification 2020-01-23 22:26:47 +01:00
Sarah Hoffmann
2edc15dfb8 doc: clarify the influence of autovacuum on memory 2020-01-22 12:02:38 +01:00
Sarah Hoffmann
69e31baf68 docs: add a note that the faltnode file is needed for updates
Fixes #1644.
2020-01-22 11:44:05 +01:00
Sarah Hoffmann
586ff0c364 Merge pull request #1638 from mtmail/check-for-invalid-indices
check_import_finished: check for invalid indices
2020-01-14 21:34:35 +01:00
Sarah Hoffmann
f3ba358d50 hint that invalid indices must be manually deleted 2020-01-14 21:33:09 +01:00
Sarah Hoffmann
54bf4c3339 Merge pull request #1637 from mtmail/fix-warnings-in-verbose-warm
warm.php verbose mode was printing errors
2020-01-14 21:30:46 +01:00
Sarah Hoffmann
acda4344de Merge pull request #1636 from lonvia/update-introduction
Update help text on website
2020-01-14 21:29:25 +01:00
marc tobias
6b0afd5d9b check_import_finished: check for invalid indices 2020-01-14 19:36:40 +01:00
marc tobias
850910ed9e warm.php verbose mode was printing errors 2020-01-14 18:24:49 +01:00
Sarah Hoffmann
4ffa11a26c update report a problem page 2020-01-13 22:42:49 +01:00
Sarah Hoffmann
f5e60f8c40 update help links
All links should go to nominatim.org now. Also add links
to Help OSM and the Github repository.
2020-01-13 22:20:53 +01:00
Sarah Hoffmann
ddaf1b79d4 remove special handling of rail
Skip railway=rail in the style, so that installations can remove
it if they wish.
2020-01-08 23:53:23 +01:00
Sarah Hoffmann
d732dc3bb2 update place address levels
Adds province and allotments and downgrades hamlet.
2020-01-08 23:53:03 +01:00
Sarah Hoffmann
f0af5c5643 Merge pull request #1628 from lonvia/split-sql-functions
Split monstrous functions.sql
2020-01-08 21:36:40 +01:00
Sarah Hoffmann
7a194789bc remove remaining sql functions into function/ directory 2020-01-08 11:45:51 +01:00
Sarah Hoffmann
827d7a9a62 move postcode table triggers to own file 2020-01-08 11:22:23 +01:00
Sarah Hoffmann
dae2761137 move placex triggers into own file 2020-01-08 11:18:42 +01:00
Sarah Hoffmann
4304c1a7bb move place triggers into own file 2020-01-07 23:55:38 +01:00
Sarah Hoffmann
c537ea18a4 move functions for interpolation table in own file 2020-01-05 16:36:46 +01:00
Sarah Hoffmann
28fa7be75a move functions for address lookup into own file 2020-01-05 16:16:21 +01:00
Sarah Hoffmann
f1a5862f3d move creation function for aux data into own file
This function is currently unused, so don't even load it.
2020-01-05 16:04:04 +01:00
Sarah Hoffmann
4088e4e371 move importance/wikipedia functions into separate file 2020-01-05 15:55:39 +01:00
Sarah Hoffmann
0ef6425847 move SQL functions for normalisation in separate file 2020-01-05 15:38:20 +01:00
Sarah Hoffmann
7489deb1b7 Merge pull request #1627 from lonvia/cleanup-setup-scripts
Minor cleanup of setup scripts
2020-01-05 14:17:39 +01:00
Sarah Hoffmann
2059e18e8b setup: factor out parameter replacement in SQL scripts
Put all into a single function and use for all SQL
templates.
2020-01-04 23:48:49 +01:00
Sarah Hoffmann
d11ee4c6d9 fix osm link in polygon error view 2020-01-04 21:51:14 +01:00
Sarah Hoffmann
c74cbde329 Merge pull request #1626 from lonvia/move-tests-to-osm2pgsql
Move tests to osm2pgsql
2020-01-04 20:08:35 +01:00
Sarah Hoffmann
20d541af06 remove osm2pgsql tag tests
These tests are now part of the osm2pgsql test suite.
2020-01-04 16:23:29 +01:00
Sarah Hoffmann
2c163b3959 update osm2pgsql (gazetteer output tests) 2020-01-04 16:20:52 +01:00
Sarah Hoffmann
c6a7ef5574 Merge pull request #1624 from lonvia/add-extratags-style
Add new extratags style
2020-01-03 10:06:47 +01:00
Sarah Hoffmann
7005c6af12 add new extratags style
This is the same as the full style but also adds all unused tags
except for a couple of internal tags to the extratags column.
2020-01-02 14:24:51 +01:00
Sarah Hoffmann
256986f01f Merge pull request #1622 from mtmail/clang-faq-entry
Documentation: add FAQ entry about clang not found
2020-01-02 14:14:00 +01:00
Sarah Hoffmann
33d322df9d update osm2pgsql (exclude country and postcode from address tags) 2019-12-28 22:36:02 +01:00
marc tobias
631013be02 Documentation: add FAQ entry about clang not found 2019-12-28 21:21:39 +01:00
Sarah Hoffmann
89a990e000 Merge pull request #1612 from mtmail/cleanup-wiki-scripts
refactor import_wiki* script for readability
2019-12-23 22:05:39 +01:00
Sarah Hoffmann
ccddd9d1de Merge pull request #1619 from mtmail/document-check-import-script
Use check_import_finished in test suite, document
2019-12-23 22:04:25 +01:00
marc tobias
9587fc9909 refactor import_wiki* script for readability 2019-12-23 21:38:33 +01:00
marc tobias
22b7aed769 Use check_import_finished in test suite, document 2019-12-23 21:25:06 +01:00
marc tobias
7db0da40ad new script utils/check_import_finished.php 2019-12-23 15:13:18 +01:00
Sarah Hoffmann
2be70b2c36 Merge pull request #1596 from mtmail/remove-obsolete-wikidata-scripts
remove old wikidata script. See data-sources/wikipedia-wikidata/ for new process
2019-12-18 21:55:17 +01:00
Sarah Hoffmann
0e03668cf2 Merge pull request #1601 from mtmail/spheric-distance-on-details-page
details page: differentiate between spheric distance and distance in meters
2019-12-18 21:54:06 +01:00
Sarah Hoffmann
28b89daa22 remove duplicate addr:country from style 2019-12-17 20:48:27 +01:00
Sarah Hoffmann
2bfa2f4292 Merge pull request #1609 from txtor/master
typo - fixes openstreetmap#1606
2019-12-17 20:46:20 +01:00
Francesc Hervada-Sala
3b22b9911b typo - fixes openstreetmap#1606 2019-12-17 17:24:29 +01:00
Sarah Hoffmann
bc68ff1e43 update osm2pgsql (deletion and address updates) 2019-12-16 21:27:41 +01:00
Sarah Hoffmann
f59af7483b Merge pull request #1604 from mtmail/wiki-create-pagelinkcount-earlier
wikipedia: create all language pagelinkcount tables before querying them
2019-12-14 21:43:16 +01:00
marc tobias
394f85a96b wikipedia: create all language pagelinkcount tables before querying them 2019-12-14 20:36:54 +01:00
marc tobias
626e3238f2 details page: differentiate between spheric distance and distance in meters 2019-12-11 00:49:32 +01:00
marc tobias
2051a84a09 remove old wikidata script. See data-sources/wikipedia-wikidata/ for new process 2019-12-03 19:27:32 +01:00
Sarah Hoffmann
f8bd4f5133 add test for finding housenumber 0 2019-12-01 20:36:59 +01:00
Sarah Hoffmann
a4e514033d Merge branch 'housenumber-zero' of https://github.com/mtmail/Nominatim into mtmail-housenumber-zero 2019-12-01 20:17:27 +01:00
Sarah Hoffmann
95f20ed7ab Merge pull request #1585 from mmd-osm/patch-1
Mention FAQ / troubleshooting page in README
2019-12-01 20:13:35 +01:00
Sarah Hoffmann
bfe92ea191 bdd tests: enforce use of full import style 2019-12-01 16:25:39 +01:00
Sarah Hoffmann
d3bacf475a Revert "update osm2pgsql (reduce memory usage)"
This reverts commit 3474464894.
2019-12-01 16:25:02 +01:00
Sarah Hoffmann
3474464894 update osm2pgsql (reduce memory usage) 2019-11-29 19:31:48 +01:00
mmd
5b25bff2d8 Mention FAQ / troubleshooting page in README 2019-11-26 22:33:00 +01:00
Sarah Hoffmann
c36896c524 Merge pull request #1578 from lonvia/docs-installation-support
Update installation documentation wrt memory usage
2019-11-26 22:21:45 +01:00
Sarah Hoffmann
d9fe25ac2e Merge pull request #1581 from mtmail/wrap-leaflet-map
Allow leaflet map to wrap-around, still longitude should be -180..+180
2019-11-26 21:54:26 +01:00
Sarah Hoffmann
0bd006eef8 fix typo 2019-11-26 21:52:37 +01:00
Sarah Hoffmann
081d1f9779 Merge pull request #1582 from mtmail/documentation-direct-link-osm2pgsql
add wiki link to osm2psql README
2019-11-26 21:51:30 +01:00
Sarah Hoffmann
546c975e28 Merge pull request #1583 from mtmail/documentation-pear-db-no-longer-prerequisite
PHP PEAR::DB is not longer a prerequisite
2019-11-26 21:51:01 +01:00
marc tobias
05fb037edb PHP PEAR::DB is not longer a prerequisite 2019-11-25 19:11:08 +01:00
marc tobias
5cdc196df1 add wiki link to osm2psql README 2019-11-25 19:08:34 +01:00
marc tobias
0896c07972 Allow map to wrap-around, still longitude should be -180..+180 2019-11-25 18:56:46 +01:00
Sarah Hoffmann
be9f54d0a9 set default osm2pgsql to 0 when using flatnode file 2019-11-24 14:36:36 +01:00
Sarah Hoffmann
88fab44006 update minimum required RAM to 64GB
Also adds more background explanation on time and RAM use,
as well as a hint that github issues are not good for
installation support.
2019-11-24 10:31:34 +01:00
Sarah Hoffmann
000fe3ddff remove reference to nominatim.xml OpenSearch description
The file was meant for the openstreetmap instance only.

Fixes #1554.
2019-11-20 11:53:04 +01:00
Sarah Hoffmann
f180f99a95 Merge pull request #1570 from lonvia/wikipedia-importance-updates
Wikipedia importance updates
2019-11-20 11:25:23 +01:00
Sarah Hoffmann
8d9aa9bf33 formatting: avoid long lines 2019-11-20 10:57:36 +01:00
Sarah Hoffmann
9fed91a47f adapt tests for new wikipedia tables 2019-11-20 09:57:40 +01:00
Sarah Hoffmann
12f830fbbb fix loading of data without wikipedia files
Also removes unused place_boundingbox table.
2019-11-17 23:58:43 +01:00
Sarah Hoffmann
6d764a05b0 prefer English wikipedia for wikidata matches 2019-11-17 19:53:42 +01:00
Sarah Hoffmann
cd3ddec746 Switch to sql.gz format for wikipedia data
The dump import is no longer needed.
2019-11-17 10:09:31 +01:00
Sarah Hoffmann
e4555a208d fix wikidata import
The loop was not skipping empty results of get_wikipedia_match().
2019-11-17 10:06:34 +01:00
Sarah Hoffmann
d53af96aa4 update documentation for new wikipedia data 2019-11-16 16:52:23 +01:00
Sarah Hoffmann
d1a9dc0f24 use wikidata links for importance as well 2019-11-16 16:52:23 +01:00
Sarah Hoffmann
a1bcb28cea also update wikipedia from linked places 2019-11-16 16:52:23 +01:00
Sarah Hoffmann
3fbba8b9db add command for recomputing importance 2019-11-16 16:52:23 +01:00
Sarah Hoffmann
5fb850982a move importance computation into its own function 2019-11-16 16:52:23 +01:00
Sarah Hoffmann
6f2e767c77 Merge pull request #1557 from mtmail/document-boundingbox
documentation: add note what bounding box can be used for
2019-11-13 10:30:22 +01:00
Sarah Hoffmann
1ead5b0f3f Merge pull request #1563 from mtmail/remove-pear-db-faq-entries
documentation: remove FAQ entries related to PEAR::DB
2019-11-13 10:29:52 +01:00
Sarah Hoffmann
eb6681d486 Merge pull request #1564 from mtmail/typo-sql-in-postcodes-md
typo in SQL in Postcodes.md
2019-11-13 10:28:57 +01:00
marc tobias
7503987630 typo in SQL in Postcodes.md 2019-11-11 23:34:44 +01:00
marc tobias
1d337e8a76 documentation: remove FAQ entries related to PEAR::DB, we no longer use that 2019-11-11 16:54:12 +01:00
Sarah Hoffmann
f42e40712e Merge pull request #1555 from mtmail/setup-escape-shell-args
setup: escape arguments when executing shell commands (psql, createdb)
2019-11-06 22:47:00 +01:00
Sarah Hoffmann
b5fb8608ba some reformatting of documentation changes and typo fixes
The newest mkdocs is more demanding when it comes to code block
formatting.
2019-11-06 22:34:43 +01:00
marc tobias
dea1d67d03 documentation: new page explaining calculated postcodes 2019-11-06 22:15:44 +01:00
Sarah Hoffmann
9e6fc8f073 Merge pull request #1548 from mtmail/centos7-postgresql-11
Update CentOS7 instruction to postgresql 11 (default was 9.2)
2019-11-06 21:52:45 +01:00
marc tobias
db6da75683 documentation: add note what bounding box can be used for 2019-11-06 17:33:20 +01:00
marc tobias
9cfd891fb9 setup: escape arguments when executing shell commands (psql, createdb) 2019-11-05 23:50:46 +01:00
marc tobias
f985680d2c Update CentOS instruction to postgresql 11 (default was 9.2) 2019-11-02 14:17:53 +01:00
Sarah Hoffmann
8ae317e002 Merge pull request #1545 from mtmail/details-index-html-page
HTML page with search form when /details.php called without params
2019-10-31 18:30:35 +01:00
marc tobias
36e99c43ce HTML page with search form when /details.php called without params 2019-10-31 17:13:43 +01:00
marc tobias
eeb26aaa6f Addresses with housenumber 0 are found 2019-10-31 13:52:10 +01:00
Sarah Hoffmann
260dbe302a Merge pull request #1542 from mtmail/tiger-import-more-verbose
Tiger data import: display which path is searched for files
2019-10-31 10:24:02 +01:00
marc tobias
c297584726 Tiger data import: display which path is searched for files 2019-10-30 15:50:29 +01:00
Sarah Hoffmann
5930383404 Merge branch 'fix-getcol-zero' of https://github.com/eyusupov/Nominatim into eyusupov-fix-getcol-zero 2019-10-28 23:07:28 +01:00
Sarah Hoffmann
7395eb9791 Merge pull request #1532 from eyusupov/use-extradir
Use ExtraDataPath for country grid
2019-10-28 23:01:42 +01:00
Sarah Hoffmann
e0de838b13 adapt tests to short_name demotion 2019-10-28 22:53:41 +01:00
Sarah Hoffmann
5292239714 demote short_name in the list of names to display
The usage of short_name in OSM has changed with time. It is no
longer suitable as a display name.

Fixes #783.
2019-10-28 22:12:21 +01:00
Eldar Yusupov
b54fca4f7e Handle zero in getCol() 2019-10-26 20:19:36 +03:00
Sarah Hoffmann
26f47d2eb7 switch to pygments for mkdocs hilighting 2019-10-25 23:57:23 +02:00
Eldar Yusupov
72c0898409 Add optional compilation of osm2pgsl 2019-10-25 23:01:19 +03:00
Eldar Yusupov
92d1f5122b Separate Nominatim build 2019-10-25 23:01:19 +03:00
Eldar Yusupov
314de3c3c0 Add options to compile only PG module/frontend
It makes it easier to install Nominatim server and PostgresSQL on
separate servers or use separate docker images for them.
2019-10-25 23:01:19 +03:00
Eldar Yusupov
544db43026 Separate more dependencies 2019-10-25 23:01:19 +03:00
Eldar Yusupov
4aca3700b2 Move nominatim lib deps to its dir 2019-10-25 23:01:19 +03:00
Eldar Yusupov
af833ff042 Include CheckSymbolExists CMake module 2019-10-25 23:01:19 +03:00
Eldar Yusupov
96c1a0a101 Use ExtraDataPath for country grid 2019-10-23 06:19:04 +03:00
Ganesh Krishnan
043f9d8298 allow nginx to serve files without php extensions
The apache config allows api calls without extension for eg /search?q=query string.
This does not work on nginx and we need to enable this via this patch
2018-08-29 12:59:29 -04:00
172 changed files with 8383 additions and 10892 deletions

View File

@@ -1,11 +1,15 @@
---
sudo: required
dist: xenial
os: linux
dist: bionic
language: python
python:
- "3.6"
addons:
postgresql: "9.6"
apt:
packages:
postgresql-server-dev-9.6
postgresql-client-9.6
git:
depth: 3
env:
@@ -30,5 +34,6 @@ script:
- if [[ $TEST_SUITE == "monaco" ]]; then wget --no-verbose --output-document=../data/monaco.osm.pbf http://download.geofabrik.de/europe/monaco-latest.osm.pbf; fi
- if [[ $TEST_SUITE == "monaco" ]]; then /usr/bin/env php ./utils/setup.php --osm-file ../data/monaco.osm.pbf --osm2pgsql-cache 1000 --all 2>&1 | grep -v 'ETA (seconds)'; fi
- if [[ $TEST_SUITE == "monaco" ]]; then /usr/bin/env php ./utils/specialphrases.php --wiki-import | psql -d test_api_nominatim >/dev/null; fi
- if [[ $TEST_SUITE == "monaco" ]]; then /usr/bin/env php ./utils/check_import_finished.php; fi
notifications:
email: false

View File

@@ -19,7 +19,7 @@ list(APPEND CMAKE_MODULE_PATH "${CMAKE_SOURCE_DIR}/cmake")
project(nominatim)
set(NOMINATIM_VERSION_MAJOR 3)
set(NOMINATIM_VERSION_MINOR 4)
set(NOMINATIM_VERSION_MINOR 5)
set(NOMINATIM_VERSION_PATCH 2)
set(NOMINATIM_VERSION "${NOMINATIM_VERSION_MAJOR}.${NOMINATIM_VERSION_MINOR}.${NOMINATIM_VERSION_PATCH}")
@@ -28,30 +28,40 @@ add_definitions(-DNOMINATIM_VERSION="${NOMINATIM_VERSION}")
#-----------------------------------------------------------------------------
#
# Find external dependencies
#
# Configuration
#-----------------------------------------------------------------------------
set(BUILD_TESTS off CACHE BOOL "Build test suite" FORCE)
set(WITH_LUA off CACHE BOOL "Build with lua support" FORCE)
set(ONLY_DOCS off CACHE BOOL "Build documentation only")
set(BUILD_IMPORTER on CACHE BOOL "Build everything for importing/updating the database")
set(BUILD_API on CACHE BOOL "Build everything for the API server")
set(BUILD_MODULE on CACHE BOOL "Build PostgreSQL module")
set(BUILD_TESTS on CACHE BOOL "Build test suite")
set(BUILD_DOCS on CACHE BOOL "Build documentation")
set(BUILD_OSM2PGSQL on CACHE BOOL "Build osm2pgsql (expert only)")
if (NOT ONLY_DOCS)
#-----------------------------------------------------------------------------
# osm2pgsql (imports/updates only)
#-----------------------------------------------------------------------------
if (BUILD_IMPORTER AND BUILD_OSM2PGSQL)
if (NOT EXISTS "${CMAKE_SOURCE_DIR}/osm2pgsql/CMakeLists.txt")
message(FATAL_ERROR "The osm2pgsql directory is empty.\
Did you forget to check out Nominatim recursively?\
\nTry updating submodules with: git submodule update --init")
endif()
set(BUILD_TESTS_SAVED "${BUILD_TESTS}")
set(BUILD_TESTS off)
set(WITH_LUA off CACHE BOOL "")
add_subdirectory(osm2pgsql)
set(BUILD_TESTS ${BUILD_TESTS_SAVED})
endif()
find_package(Threads REQUIRED)
unset(PostgreSQL_TYPE_INCLUDE_DIR CACHE)
set(PostgreSQL_TYPE_INCLUDE_DIR "/usr/include/")
find_package(PostgreSQL REQUIRED)
include_directories(${PostgreSQL_INCLUDE_DIRS})
link_directories(${PostgreSQL_LIBRARY_DIRS})
#-----------------------------------------------------------------------------
# python and pyosmium (imports/updates only)
#-----------------------------------------------------------------------------
if (BUILD_IMPORTER)
find_package(PythonInterp 3)
find_program(PYOSMIUM pyosmium-get-changes)
if (NOT EXISTS "${PYOSMIUM}")
@@ -61,115 +71,121 @@ if (NOT ONLY_DOCS)
set(PYOSMIUM_PATH "${PYOSMIUM}")
message(STATUS "Using pyosmium-get-changes at ${PYOSMIUM_PATH}")
endif()
find_program(PG_CONFIG pg_config)
execute_process(COMMAND ${PG_CONFIG} --pgxs
OUTPUT_VARIABLE PGXS
OUTPUT_STRIP_TRAILING_WHITESPACE)
if (NOT EXISTS "${PGXS}")
message(FATAL_ERROR "Postgresql server package not found.")
endif()
find_package(ZLIB REQUIRED)
find_package(BZip2 REQUIRED)
find_package(LibXml2 REQUIRED)
include_directories(${LIBXML2_INCLUDE_DIR})
# Setting PHP binary variable as to command line (prevailing) or auto detect
if (NOT PHP_BIN)
find_program (PHP_BIN php)
endif()
# sanity check if PHP binary exists
if (NOT EXISTS ${PHP_BIN})
message(FATAL_ERROR "PHP binary not found. Install php or provide location with -DPHP_BIN=/path/php ")
endif()
message (STATUS "Using PHP binary " ${PHP_BIN})
endif()
#-----------------------------------------------------------------------------
#
# Setup settings and paths
#
# PHP
#-----------------------------------------------------------------------------
set(WEBSITESCRIPTS
website/deletable.php
website/details.php
website/hierarchy.php
website/lookup.php
website/polygons.php
website/reverse.php
website/search.php
website/status.php
)
# Setting PHP binary variable as to command line (prevailing) or auto detect
set(CUSTOMSCRIPTS
utils/country_languages.php
utils/importWikipedia.php
utils/export.php
utils/query.php
utils/setup.php
utils/specialphrases.php
utils/update.php
utils/warm.php
if (NOT PHP_BIN)
find_program (PHP_BIN php)
endif()
# sanity check if PHP binary exists
if (NOT EXISTS ${PHP_BIN})
message(FATAL_ERROR "PHP binary not found. Install php or provide location with -DPHP_BIN=/path/php ")
endif()
message (STATUS "Using PHP binary " ${PHP_BIN})
#-----------------------------------------------------------------------------
# import scripts and utilities (importer only)
#-----------------------------------------------------------------------------
if (BUILD_IMPORTER)
set(CUSTOMSCRIPTS
utils/check_import_finished.php
utils/country_languages.php
utils/importWikipedia.php
utils/export.php
utils/query.php
utils/setup.php
utils/specialphrases.php
utils/update.php
utils/warm.php
)
foreach (script_source ${CUSTOMSCRIPTS})
configure_file(${PROJECT_SOURCE_DIR}/cmake/script.tmpl
${PROJECT_BINARY_DIR}/${script_source})
endforeach()
endif()
#-----------------------------------------------------------------------------
# webserver scripts (API only)
#-----------------------------------------------------------------------------
if (BUILD_API)
set(WEBSITESCRIPTS
website/deletable.php
website/details.php
website/hierarchy.php
website/lookup.php
website/polygons.php
website/reverse.php
website/search.php
website/status.php
)
foreach (script_source ${CUSTOMSCRIPTS})
configure_file(${PROJECT_SOURCE_DIR}/cmake/script.tmpl
${PROJECT_BINARY_DIR}/${script_source})
endforeach()
foreach (script_source ${WEBSITESCRIPTS})
configure_file(${PROJECT_SOURCE_DIR}/cmake/website.tmpl
${PROJECT_BINARY_DIR}/${script_source})
endforeach()
foreach (script_source ${WEBSITESCRIPTS})
configure_file(${PROJECT_SOURCE_DIR}/cmake/website.tmpl
${PROJECT_BINARY_DIR}/${script_source})
endforeach()
set(WEBPATHS css images js)
foreach (wp ${WEBPATHS})
execute_process(
COMMAND ln -sf ${PROJECT_SOURCE_DIR}/website/${wp} ${PROJECT_BINARY_DIR}/website/
)
endforeach()
endif()
#-----------------------------------------------------------------------------
# default settings
#-----------------------------------------------------------------------------
configure_file(${PROJECT_SOURCE_DIR}/settings/defaults.php
${PROJECT_BINARY_DIR}/settings/settings.php)
set(WEBPATHS css images js)
foreach (wp ${WEBPATHS})
execute_process(
COMMAND ln -sf ${PROJECT_SOURCE_DIR}/website/${wp} ${PROJECT_BINARY_DIR}/website/
)
endforeach()
#-----------------------------------------------------------------------------
#
# Tests
#
#-----------------------------------------------------------------------------
if (NOT ONLY_DOCS)
if (BUILD_TESTS)
include(CTest)
set(TEST_BDD db osm2pgsql api)
foreach (test ${TEST_BDD})
add_test(NAME bdd_${test}
COMMAND lettuce features/${test}
WORKING_DIRECTORY ${PROJECT_SOURCE_DIR}/tests)
COMMAND behave ${test}
WORKING_DIRECTORY ${PROJECT_SOURCE_DIR}/test/bdd)
set_tests_properties(bdd_${test}
PROPERTIES ENVIRONMENT "NOMINATIM_DIR=${PROJECT_BINARY_DIR}")
endforeach()
add_test(NAME php
COMMAND phpunit ./
WORKING_DIRECTORY ${PROJECT_SOURCE_DIR}/tests-php)
WORKING_DIRECTORY ${PROJECT_SOURCE_DIR}/test/php)
add_test(NAME phpcs
COMMAND phpcs --report-width=120 --colors lib website utils
WORKING_DIRECTORY ${PROJECT_SOURCE_DIR})
endif()
#-----------------------------------------------------------------------------
# Postgres module
#-----------------------------------------------------------------------------
if (NOT ONLY_DOCS)
if (BUILD_MODULE)
add_subdirectory(module)
add_subdirectory(nominatim)
endif()
add_subdirectory(docs)
#-----------------------------------------------------------------------------
# Documentation
#-----------------------------------------------------------------------------
if (BUILD_DOCS)
add_subdirectory(docs)
endif()

View File

@@ -15,9 +15,9 @@ Please make sure to add the following information:
* the result you are getting
* the expected result, preferably a link to the OSM object you want to find,
otherwise an address that is as precise as possible
To get the link to the OSM object, you can try the following:
To get the link to the OSM object, you can try the following:
* go to https://openstreetmap.org
* zoom to the area of the map where you expect the result and
zoom in as much as possible
@@ -26,7 +26,7 @@ Please make sure to add the following information:
* find the object of interest in the list that appears on the left side
* click on the object and report the URL back that the browser shows
### When Reporting Problems with your Installation...
### When Reporting Bugs...
Please add the following information to your issue:
@@ -38,6 +38,9 @@ Please add the following information to your issue:
if you run from the git repo, the output of `git rev-parse HEAD`)
* (if applicable) exact command line of the command that was causing the issue
Bug reports that do not include extensive information about your system,
about the problem and about what you have been trying to debug the problem
will be closed.
## Workflow for Pull Requests

View File

@@ -1,11 +1,61 @@
3.5.2
* ensure that wikipedia tags are imported for all styles
* reinstate verbosity for indexing during updates
* make house number reappear in display name on named POIs
* introduce batch processing in indexer to avoid transaction ID overrun
* increase splitting for large geometries to improve indexing speed
* remove deprecated get_magic_quotes_gpc() function
* make sure that all postcodes have an entry in word and are thus searchable
* remove use of ST_Covers in conjunction woth ST_Intersects,
causes bad query planning and slow updates in Postgis3
* update osm2pgsql
3.5.1
* disable jit and parallel processing in PostgreSQL for osm2pgsql
* update libosmium to 2.15.6 (fixes an issue with processing hanging
on large multipolygons)
3.5.0
* structured select on HTML search page
* new PHP Nominatim\Shell class to wrap shell escaping
* remove polygon parameter from all API calls
* improve handling of postcode areas
* reorganise place linking algorithm, now using wikidata tag as well
* remove linkees from search_name and larger_area tables
* introduce country-specific address ranks
* reorganise rank address computation
* cleanup of partition function
* improve parenting for large POIs
* add support for Postgresql 12 and Postgis 3
* add earlier cleanup when --drop is given, to reduce meory usage
* remove use of place_id in URLs
* replace C nominatim indexer with a simpler Python implementation
* split up the huge sql/functions.sql file
* move osm2pgsql tests to osm2pgsql
* add new extratags style which imports all tags from OSM
* add new script for checking the import after completion
* update osm2pgsql, reducing memory usage
* use new wikipedia importance and add processing of wikidata tags
* add search form for details page
* use ExtraDataPath for country_grid table
* remove short_name from list of names to be displayed
* split up CMakeFile, so that all parts can be built separately
* update installation instructions for CentOS and Ubuntu
* add script for importing/updating multiple country extracts
* various documentation improvements
3.4.2
* security fix: fix possible SQL injection via details API
* fix security bug in /details endpoint where user input was not
properly sanitized
3.4.1
* update osm2pgsql
* move deletion to copy thread (fixes deadlock in updates)
* fix filtering where valid address objects got dropped
* fix typo in import styles
* update osm2pgsql to fix hans during updates and lost address numbers
during updates
3.4.0

View File

@@ -1,4 +1,4 @@
[![Build Status](https://travis-ci.org/openstreetmap/Nominatim.svg?branch=master)](https://travis-ci.org/openstreetmap/Nominatim)
[![Build Status](https://travis-ci.org/osm-search/Nominatim.svg?branch=master)](https://travis-ci.org/osm-search/Nominatim)
Nominatim
=========
@@ -19,8 +19,16 @@ https://nominatim.org/release-docs/develop/ .
Installation
============
**Nominatim is a complex piece of software and runs in a complex environment.
Installing and running Nominatim is something for experienced system
administrators only who can do some trouble-shooting themselves. We are sorry,
but we can not provide installation support. We are all doing this in our free
time and there is just so much of that time to go around. Do not open issues in
our bug tracker if you need help. You can ask questions on the mailing list
(see below) or on [help.openstreetmap.org](https://help.openstreetmap.org/).**
The latest stable release can be downloaded from https://nominatim.org.
There you can also find [installation instructions for the release](https://nominatim.org/release-docs/latest/admin/Installation).
There you can also find [installation instructions for the release](https://nominatim.org/release-docs/latest/admin/Installation), as well as an extensive [Troubleshooting/FAQ section](https://nominatim.org/release-docs/latest/admin/Faq/).
Detailed installation instructions for the development version can be
found at [nominatim.org](https://nominatim.org/release-docs/develop/admin/Installation)

View File

@@ -141,7 +141,7 @@ No. Long running Nominatim installations will differ once new import features (o
bug fixes) get added since those usually only get applied to new/changed data.
Also this document skips the optional Wikipedia data import which affects ranking
of search results. See [Nominatim installation](http://nominatim.org/release-docs/latest/Installation) for details.
of search results. See [Nominatim installation](https://nominatim.org/release-docs/latest/admin/Installation) for details.
##### Why Ubuntu? Can I test CentOS/Fedora/CoreOS/FreeBSD?

21
Vagrantfile vendored
View File

@@ -15,6 +15,15 @@ Vagrant.configure("2") do |config|
end
config.vm.define "ubuntu", primary: true do |sub|
sub.vm.box = "bento/ubuntu-20.04"
sub.vm.provision :shell do |s|
s.path = "vagrant/Install-on-Ubuntu-20.sh"
s.privileged = false
s.args = [checkout]
end
end
config.vm.define "ubuntu18", primary: true do |sub|
sub.vm.box = "bento/ubuntu-18.04"
sub.vm.provision :shell do |s|
s.path = "vagrant/Install-on-Ubuntu-18.sh"
@@ -61,6 +70,18 @@ Vagrant.configure("2") do |config|
sub.vm.synced_folder ".", "/vagrant", disabled: true
end
config.vm.define "centos8" do |sub|
sub.vm.box = "generic/centos8"
sub.vm.provision :shell do |s|
s.path = "vagrant/Install-on-Centos-8.sh"
s.privileged = false
s.args = "yes"
end
sub.vm.synced_folder ".", "/home/vagrant/Nominatim", disabled: true
sub.vm.synced_folder ".", "/vagrant", disabled: true
end
config.vm.provider "virtualbox" do |vb|
vb.gui = false
vb.memory = 2048

View File

@@ -1,6 +1,6 @@
# US TIGER address data
Convert [TIGER](https://www.census.gov/geo/maps-data/data/tiger.html)/Line dataset of the US Census Bureau to SQL files which can be imported by Nominatim. The created tables in the Nominatim database are separate from OpenStreetMap tables and get queried at search time separately.
Convert [TIGER](https://www.census.gov/geographies/mapping-files/time-series/geo/tiger-line-file.html)/Line dataset of the US Census Bureau to SQL files which can be imported by Nominatim. The created tables in the Nominatim database are separate from OpenStreetMap tables and get queried at search time separately.
The dataset gets updated once per year. Downloading is prone to be slow (can take a full day) and converting them can take hours as well.

View File

@@ -26,7 +26,7 @@ This step downloads and converts [Wikipedia](https://dumps.wikimedia.org/) page
To download, convert, and import the data, then process summary statistics and compute importance scores, run:
```
./wikipedia_import.sh
./import_wikipedia.sh
```
---
Wikidata
@@ -54,5 +54,5 @@ By including Wikidata in the wikipedia_articles table, new connections can be ma
To download, convert, and import the data, then process required items, run:
```
./wikidata_import.sh
./import_wikidata.sh
```

View File

@@ -1,37 +1,66 @@
#!/bin/bash
psqlcmd() {
psql wikiprocessingdb
psql --quiet wikiprocessingdb
}
mysql2pgsqlcmd() {
./mysql2pgsql.perl /dev/stdin /dev/stdout
}
download() {
echo "Downloading $1"
wget --quiet --no-clobber --tries 3 "$1"
}
# list the languages to process (refer to List of Wikipedias here: https://en.wikipedia.org/wiki/List_of_Wikipedias)
language=( "ar" "bg" "ca" "cs" "da" "de" "en" "es" "eo" "eu" "fa" "fr" "ko" "hi" "hr" "id" "it" "he" "lt" "hu" "ms" "nl" "ja" "no" "pl" "pt" "kk" "ro" "ru" "sk" "sl" "sr" "fi" "sv" "tr" "uk" "vi" "vo" "war" "zh" )
# languages to process (refer to List of Wikipedias here: https://en.wikipedia.org/wiki/List_of_Wikipedias)
# requires Bash 4.0
readarray -t LANGUAGES < languages.txt
# get a few wikidata dump tables
wget https://dumps.wikimedia.org/wikidatawiki/latest/wikidatawiki-latest-geo_tags.sql.gz
wget https://dumps.wikimedia.org/wikidatawiki/latest/wikidatawiki-latest-page.sql.gz
wget https://dumps.wikimedia.org/wikidatawiki/latest/wikidatawiki-latest-wb_items_per_site.sql.gz
echo "====================================================================="
echo "Download wikidata dump tables"
echo "====================================================================="
# 114M wikidatawiki-latest-geo_tags.sql.gz
# 1.7G wikidatawiki-latest-page.sql.gz
# 1.2G wikidatawiki-latest-wb_items_per_site.sql.gz
download https://dumps.wikimedia.org/wikidatawiki/latest/wikidatawiki-latest-geo_tags.sql.gz
download https://dumps.wikimedia.org/wikidatawiki/latest/wikidatawiki-latest-page.sql.gz
download https://dumps.wikimedia.org/wikidatawiki/latest/wikidatawiki-latest-wb_items_per_site.sql.gz
# import wikidata tables
gzip -dc wikidatawiki-latest-geo_tags.sql.gz | mysql2pgsqlcmd | psqlcmd
gzip -dc wikidatawiki-latest-page.sql.gz | mysql2pgsqlcmd | psqlcmd
echo "====================================================================="
echo "Import wikidata dump tables"
echo "====================================================================="
echo "Importing wikidatawiki-latest-geo_tags"
gzip -dc wikidatawiki-latest-geo_tags.sql.gz | mysql2pgsqlcmd | psqlcmd
echo "Importing wikidatawiki-latest-page"
gzip -dc wikidatawiki-latest-page.sql.gz | mysql2pgsqlcmd | psqlcmd
echo "Importing wikidatawiki-latest-wb_items_per_site"
gzip -dc wikidatawiki-latest-wb_items_per_site.sql.gz | mysql2pgsqlcmd | psqlcmd
# get wikidata places from wikidata query API
echo "====================================================================="
echo "Get wikidata places from wikidata query API"
echo "====================================================================="
echo "Number of place types:"
wc -l wikidata_place_types.txt
while read F ; do
wget "https://query.wikidata.org/bigdata/namespace/wdq/sparql?format=json&query=SELECT ?item WHERE{?item wdt:P31*/wdt:P279*wd:$F;}" -O $F.json
echo "Querying for place type $F..."
wget --quiet "https://query.wikidata.org/bigdata/namespace/wdq/sparql?format=json&query=SELECT ?item WHERE{?item wdt:P31*/wdt:P279*wd:$F;}" -O $F.json
jq -r '.results | .[] | .[] | [.item.value] | @csv' $F.json >> $F.txt
awk -v qid=$F '{print $0 ","qid}' $F.txt | sed -e 's!"http://www.wikidata.org/entity/!!' | sed 's/"//g' >> $F.csv
cat $F.csv >> wikidata_place_dump.csv
@@ -39,57 +68,207 @@ while read F ; do
done < wikidata_place_types.txt
# import wikidata places
echo "CREATE TABLE wikidata_place_dump (item text, instance_of text);" | psqlcmd
echo "COPY wikidata_place_dump (item, instance_of) FROM '/srv/nominatim/Nominatim/data-sources/wikipedia-wikidata/wikidata_place_dump.csv' DELIMITER ',' CSV;" | psqlcmd
echo "CREATE TABLE wikidata_place_type_levels (place_type text, level integer);" | psqlcmd
echo "COPY wikidata_place_type_levels (place_type, level) FROM '/srv/nominatim/Nominatim/data-sources/wikipedia-wikidata/wikidata_place_type_levels.csv' DELIMITER ',' CSV HEADER;" | psqlcmd
# create derived tables
echo "====================================================================="
echo "Import wikidata places"
echo "====================================================================="
echo "CREATE TABLE geo_earth_primary AS SELECT gt_page_id, gt_lat, gt_lon FROM geo_tags WHERE gt_globe = 'earth' AND gt_primary = 1 AND NOT( gt_lat < -90 OR gt_lat > 90 OR gt_lon < -180 OR gt_lon > 180 OR gt_lat=0 OR gt_lon=0) ;" | psqlcmd
echo "CREATE TABLE geo_earth_wikidata AS SELECT DISTINCT geo_earth_primary.gt_page_id, geo_earth_primary.gt_lat, geo_earth_primary.gt_lon, page.page_title, page.page_namespace FROM geo_earth_primary LEFT OUTER JOIN page ON (geo_earth_primary.gt_page_id = page.page_id) ORDER BY geo_earth_primary.gt_page_id;" | psqlcmd
echo "CREATE TABLE wikidata_place_dump (
item text,
instance_of text
);" | psqlcmd
echo "ALTER TABLE wikidata_place_dump ADD COLUMN ont_level integer, ADD COLUMN lat numeric(11,8), ADD COLUMN lon numeric(11,8);" | psqlcmd
echo "UPDATE wikidata_place_dump SET ont_level = wikidata_place_type_levels.level FROM wikidata_place_type_levels WHERE wikidata_place_dump.instance_of = wikidata_place_type_levels.place_type;" | psqlcmd
echo "COPY wikidata_place_dump (item, instance_of)
FROM '/srv/nominatim/Nominatim/data-sources/wikipedia-wikidata/wikidata_place_dump.csv'
DELIMITER ','
CSV
;" | psqlcmd
echo "CREATE TABLE wikidata_places AS SELECT DISTINCT ON (item) item, instance_of, MAX(ont_level) AS ont_level, lat, lon FROM wikidata_place_dump GROUP BY item, instance_of, ont_level, lat, lon ORDER BY item;" | psqlcmd
echo "UPDATE wikidata_places SET lat = geo_earth_wikidata.gt_lat, lon = geo_earth_wikidata.gt_lon FROM geo_earth_wikidata WHERE wikidata_places.item = geo_earth_wikidata.page_title" | psqlcmd
echo "CREATE TABLE wikidata_place_type_levels (
place_type text,
level integer
);" | psqlcmd
echo "COPY wikidata_place_type_levels (place_type, level)
FROM '/srv/nominatim/Nominatim/data-sources/wikipedia-wikidata/wikidata_place_type_levels.csv'
DELIMITER ','
CSV
HEADER
;" | psqlcmd
# process language pages
echo "CREATE TABLE wikidata_pages (item text, instance_of text, lat numeric(11,8), lon numeric(11,8), ips_site_page text, language text );" | psqlcmd
for i in "${language[@]}"
echo "====================================================================="
echo "Create derived tables"
echo "====================================================================="
echo "CREATE TABLE geo_earth_primary AS
SELECT gt_page_id,
gt_lat,
gt_lon
FROM geo_tags
WHERE gt_globe = 'earth'
AND gt_primary = 1
AND NOT( gt_lat < -90
OR gt_lat > 90
OR gt_lon < -180
OR gt_lon > 180
OR gt_lat=0
OR gt_lon=0)
;" | psqlcmd
echo "CREATE TABLE geo_earth_wikidata AS
SELECT DISTINCT geo_earth_primary.gt_page_id,
geo_earth_primary.gt_lat,
geo_earth_primary.gt_lon,
page.page_title,
page.page_namespace
FROM geo_earth_primary
LEFT OUTER JOIN page
ON (geo_earth_primary.gt_page_id = page.page_id)
ORDER BY geo_earth_primary.gt_page_id
;" | psqlcmd
echo "ALTER TABLE wikidata_place_dump
ADD COLUMN ont_level integer,
ADD COLUMN lat numeric(11,8),
ADD COLUMN lon numeric(11,8)
;" | psqlcmd
echo "UPDATE wikidata_place_dump
SET ont_level = wikidata_place_type_levels.level
FROM wikidata_place_type_levels
WHERE wikidata_place_dump.instance_of = wikidata_place_type_levels.place_type
;" | psqlcmd
echo "CREATE TABLE wikidata_places
AS
SELECT DISTINCT ON (item) item,
instance_of,
MAX(ont_level) AS ont_level,
lat,
lon
FROM wikidata_place_dump
GROUP BY item,
instance_of,
ont_level,
lat,
lon
ORDER BY item
;" | psqlcmd
echo "UPDATE wikidata_places
SET lat = geo_earth_wikidata.gt_lat,
lon = geo_earth_wikidata.gt_lon
FROM geo_earth_wikidata
WHERE wikidata_places.item = geo_earth_wikidata.page_title
;" | psqlcmd
echo "====================================================================="
echo "Process language pages"
echo "====================================================================="
echo "CREATE TABLE wikidata_pages (
item text,
instance_of text,
lat numeric(11,8),
lon numeric(11,8),
ips_site_page text,
language text
);" | psqlcmd
for i in "${LANGUAGES[@]}"
do
echo "CREATE TABLE wikidata_${i}_pages as select wikidata_places.item, wikidata_places.instance_of, wikidata_places.lat, wikidata_places.lon, wb_items_per_site.ips_site_page FROM wikidata_places LEFT JOIN wb_items_per_site ON (CAST (( LTRIM(wikidata_places.item, 'Q')) AS INTEGER) = wb_items_per_site.ips_item_id) WHERE ips_site_id = '${i}wiki' AND LEFT(wikidata_places.item,1) = 'Q' order by wikidata_places.item;" | psqlcmd
echo "ALTER TABLE wikidata_${i}_pages ADD COLUMN language text;" | psqlcmd
echo "UPDATE wikidata_${i}_pages SET language = '${i}';" | psqlcmd
echo "INSERT INTO wikidata_pages SELECT item, instance_of, lat, lon, ips_site_page, language FROM wikidata_${i}_pages;" | psqlcmd
echo "CREATE TABLE wikidata_${i}_pages AS
SELECT wikidata_places.item,
wikidata_places.instance_of,
wikidata_places.lat,
wikidata_places.lon,
wb_items_per_site.ips_site_page
FROM wikidata_places
LEFT JOIN wb_items_per_site
ON (CAST (( LTRIM(wikidata_places.item, 'Q')) AS INTEGER) = wb_items_per_site.ips_item_id)
WHERE ips_site_id = '${i}wiki'
AND LEFT(wikidata_places.item,1) = 'Q'
ORDER BY wikidata_places.item
;" | psqlcmd
echo "ALTER TABLE wikidata_${i}_pages
ADD COLUMN language text
;" | psqlcmd
echo "UPDATE wikidata_${i}_pages
SET language = '${i}'
;" | psqlcmd
echo "INSERT INTO wikidata_pages
SELECT item,
instance_of,
lat,
lon,
ips_site_page,
language
FROM wikidata_${i}_pages
;" | psqlcmd
done
echo "ALTER TABLE wikidata_pages ADD COLUMN wp_page_title text;" | psqlcmd
echo "UPDATE wikidata_pages SET wp_page_title = REPLACE(ips_site_page, ' ', '_');" | psqlcmd
echo "ALTER TABLE wikidata_pages DROP COLUMN ips_site_page;" | psqlcmd
echo "ALTER TABLE wikidata_pages
ADD COLUMN wp_page_title text
;" | psqlcmd
echo "UPDATE wikidata_pages
SET wp_page_title = REPLACE(ips_site_page, ' ', '_')
;" | psqlcmd
echo "ALTER TABLE wikidata_pages
DROP COLUMN ips_site_page
;" | psqlcmd
# add wikidata to wikipedia_article table
echo "UPDATE wikipedia_article SET lat = wikidata_pages.lat, lon = wikidata_pages.lon, wd_page_title = wikidata_pages.item, instance_of = wikidata_pages.instance_of FROM wikidata_pages WHERE wikipedia_article.language = wikidata_pages.language AND wikipedia_article.title = wikidata_pages.wp_page_title;" | psqlcmd
echo "CREATE TABLE wikipedia_article_slim AS SELECT * FROM wikipedia_article WHERE wikidata_id IS NOT NULL;" | psqlcmd
echo "ALTER TABLE wikipedia_article RENAME TO wikipedia_article_full;" | psqlcmd
echo "ALTER TABLE wikipedia_article_slim RENAME TO wikipedia_article;" | psqlcmd
# clean up intermediate tables
echo "====================================================================="
echo "Add wikidata to wikipedia_article table"
echo "====================================================================="
echo "UPDATE wikipedia_article
SET lat = wikidata_pages.lat,
lon = wikidata_pages.lon,
wd_page_title = wikidata_pages.item,
instance_of = wikidata_pages.instance_of
FROM wikidata_pages
WHERE wikipedia_article.language = wikidata_pages.language
AND wikipedia_article.title = wikidata_pages.wp_page_title
;" | psqlcmd
echo "CREATE TABLE wikipedia_article_slim
AS
SELECT * FROM wikipedia_article
WHERE wikidata_id IS NOT NULL
;" | psqlcmd
echo "ALTER TABLE wikipedia_article
RENAME TO wikipedia_article_full
;" | psqlcmd
echo "ALTER TABLE wikipedia_article_slim
RENAME TO wikipedia_article
;" | psqlcmd
echo "====================================================================="
echo "Dropping intermediate tables"
echo "====================================================================="
echo "DROP TABLE wikidata_place_dump;" | psqlcmd
echo "DROP TABLE geo_earth_primary;" | psqlcmd
for i in "${language[@]}"
for i in "${LANGUAGES[@]}"
do
echo "DROP TABLE wikidata_${i}_pages;" | psqlcmd
done

View File

@@ -1,77 +1,297 @@
#!/bin/bash
psqlcmd() {
psql wikiprocessingdb
psql --quiet wikiprocessingdb |& \
grep -v 'does not exist, skipping' |& \
grep -v 'violates check constraint' |& \
grep -vi 'Failing row contains'
}
mysql2pgsqlcmd() {
./mysql2pgsql.perl /dev/stdin /dev/stdout
./mysql2pgsql.perl --nodrop /dev/stdin /dev/stdout
}
download() {
echo "Downloading $1"
wget --quiet --no-clobber --tries=3 "$1"
}
# list the languages to process (refer to List of Wikipedias here: https://en.wikipedia.org/wiki/List_of_Wikipedias)
language=( "ar" "bg" "ca" "cs" "da" "de" "en" "es" "eo" "eu" "fa" "fr" "ko" "hi" "hr" "id" "it" "he" "lt" "hu" "ms" "nl" "ja" "no" "pl" "pt" "kk" "ro" "ru" "sk" "sl" "sr" "fi" "sv" "tr" "uk" "vi" "vo" "war" "zh" )
# languages to process (refer to List of Wikipedias here: https://en.wikipedia.org/wiki/List_of_Wikipedias)
# requires Bash 4.0
readarray -t LANGUAGES < languages.txt
# create wikipedia calculation tables
echo "CREATE TABLE linkcounts (language text, title text, count integer, sumcount integer, lat double precision, lon double precision);" | psqlcmd
echo "CREATE TABLE wikipedia_article (language text NOT NULL, title text NOT NULL, langcount integer, othercount integer, totalcount integer, lat double precision, lon double precision, importance double precision, title_en text, osm_type character(1), osm_id bigint );" | psqlcmd
echo "CREATE TABLE wikipedia_redirect (language text, from_title text, to_title text );" | psqlcmd
echo "====================================================================="
echo "Create wikipedia calculation tables"
echo "====================================================================="
echo "CREATE TABLE linkcounts (
language text,
title text,
count integer,
sumcount integer,
lat double precision,
lon double precision
);" | psqlcmd
echo "CREATE TABLE wikipedia_article (
language text NOT NULL,
title text NOT NULL,
langcount integer,
othercount integer,
totalcount integer,
lat double precision,
lon double precision,
importance double precision,
title_en text,
osm_type character(1),
osm_id bigint
);" | psqlcmd
echo "CREATE TABLE wikipedia_redirect (
language text,
from_title text,
to_title text
);" | psqlcmd
# download individual wikipedia language tables
for i in "${language[@]}"
echo "====================================================================="
echo "Download individual wikipedia language tables"
echo "====================================================================="
for i in "${LANGUAGES[@]}"
do
wget https://dumps.wikimedia.org/${i}wiki/latest/${i}wiki-latest-page.sql.gz
wget https://dumps.wikimedia.org/${i}wiki/latest/${i}wiki-latest-pagelinks.sql.gz
wget https://dumps.wikimedia.org/${i}wiki/latest/${i}wiki-latest-langlinks.sql.gz
wget https://dumps.wikimedia.org/${i}wiki/latest/${i}wiki-latest-redirect.sql.gz
echo "Language: $i"
# english is the largest
# 1.7G enwiki-latest-page.sql.gz
# 6.2G enwiki-latest-pagelinks.sql.gz
# 355M enwiki-latest-langlinks.sql.gz
# 128M enwiki-latest-redirect.sql.gz
# example of smaller languge turkish
# 53M trwiki-latest-page.sql.gz
# 176M trwiki-latest-pagelinks.sql.gz
# 106M trwiki-latest-langlinks.sql.gz
# 3.2M trwiki-latest-redirect.sql.gz
download https://dumps.wikimedia.org/${i}wiki/latest/${i}wiki-latest-page.sql.gz
download https://dumps.wikimedia.org/${i}wiki/latest/${i}wiki-latest-pagelinks.sql.gz
download https://dumps.wikimedia.org/${i}wiki/latest/${i}wiki-latest-langlinks.sql.gz
download https://dumps.wikimedia.org/${i}wiki/latest/${i}wiki-latest-redirect.sql.gz
done
# import individual wikipedia language tables
for i in "${language[@]}"
echo "====================================================================="
echo "Import individual wikipedia language tables"
echo "====================================================================="
for i in "${LANGUAGES[@]}"
do
gzip -dc ${i}wiki-latest-pagelinks.sql.gz | sed "s/\`pagelinks\`/\`${i}pagelinks\`/g" | mysql2pgsqlcmd | psqlcmd
gzip -dc ${i}wiki-latest-page.sql.gz | sed "s/\`page\`/\`${i}page\`/g" | mysql2pgsqlcmd | psqlcmd
gzip -dc ${i}wiki-latest-langlinks.sql.gz | sed "s/\`langlinks\`/\`${i}langlinks\`/g" | mysql2pgsqlcmd | psqlcmd
gzip -dc ${i}wiki-latest-redirect.sql.gz | sed "s/\`redirect\`/\`${i}redirect\`/g" | mysql2pgsqlcmd | psqlcmd
echo "Language: $i"
# We pre-create the table schema. This allows us to
# 1. Skip index creation. Most queries we do are full table scans
# 2. Add constrain to only import namespace=0 (wikipedia articles)
# Both cuts down data size considerably (50%+)
echo "Importing ${i}wiki-latest-pagelinks"
echo "DROP TABLE IF EXISTS ${i}pagelinks;" | psqlcmd
echo "CREATE TABLE ${i}pagelinks (
pl_from int NOT NULL DEFAULT '0',
pl_namespace int NOT NULL DEFAULT '0',
pl_title text NOT NULL DEFAULT '',
pl_from_namespace int NOT NULL DEFAULT '0'
);" | psqlcmd
time \
gzip -dc ${i}wiki-latest-pagelinks.sql.gz | \
sed "s/\`pagelinks\`/\`${i}pagelinks\`/g" | \
mysql2pgsqlcmd | \
grep -v '^CREATE INDEX ' | \
psqlcmd
echo "Importing ${i}wiki-latest-page"
# autoincrement serial8 4byte
echo "DROP TABLE IF EXISTS ${i}page;" | psqlcmd
echo "CREATE TABLE ${i}page (
page_id int NOT NULL,
page_namespace int NOT NULL DEFAULT '0',
page_title text NOT NULL DEFAULT '',
page_restrictions text NOT NULL,
page_is_redirect smallint NOT NULL DEFAULT '0',
page_is_new smallint NOT NULL DEFAULT '0',
page_random double precision NOT NULL DEFAULT '0',
page_touched text NOT NULL DEFAULT '',
page_links_updated text DEFAULT NULL,
page_latest int NOT NULL DEFAULT '0',
page_len int NOT NULL DEFAULT '0',
page_content_model text DEFAULT NULL,
page_lang text DEFAULT NULL
);" | psqlcmd
time \
gzip -dc ${i}wiki-latest-page.sql.gz | \
sed "s/\`page\`/\`${i}page\`/g" | \
mysql2pgsqlcmd | \
grep -v '^CREATE INDEX ' | \
psqlcmd
echo "Importing ${i}wiki-latest-langlinks"
echo "DROP TABLE IF EXISTS ${i}langlinks;" | psqlcmd
echo "CREATE TABLE ${i}langlinks (
ll_from int NOT NULL DEFAULT '0',
ll_lang text NOT NULL DEFAULT '',
ll_title text NOT NULL DEFAULT ''
);" | psqlcmd
time \
gzip -dc ${i}wiki-latest-langlinks.sql.gz | \
sed "s/\`langlinks\`/\`${i}langlinks\`/g" | \
mysql2pgsqlcmd | \
grep -v '^CREATE INDEX ' | \
psqlcmd
echo "Importing ${i}wiki-latest-redirect"
echo "DROP TABLE IF EXISTS ${i}redirect;" | psqlcmd
echo "CREATE TABLE ${i}redirect (
rd_from int NOT NULL DEFAULT '0',
rd_namespace int NOT NULL DEFAULT '0',
rd_title text NOT NULL DEFAULT '',
rd_interwiki text DEFAULT NULL,
rd_fragment text DEFAULT NULL
);" | psqlcmd
time \
gzip -dc ${i}wiki-latest-redirect.sql.gz | \
sed "s/\`redirect\`/\`${i}redirect\`/g" | \
mysql2pgsqlcmd | \
grep -v '^CREATE INDEX ' | \
psqlcmd
done
# process language tables and associated pagelink counts
for i in "${language[@]}"
echo "====================================================================="
echo "Process language tables and associated pagelink counts"
echo "====================================================================="
for i in "${LANGUAGES[@]}"
do
echo "create table ${i}pagelinkcount as select pl_title as title,count(*) as count from ${i}pagelinks where pl_namespace = 0 group by pl_title;" | psqlcmd
echo "insert into linkcounts select '${i}',pl_title,count(*) from ${i}pagelinks where pl_namespace = 0 group by pl_title;" | psqlcmd
echo "insert into wikipedia_redirect select '${i}',page_title,rd_title from ${i}redirect join ${i}page on (rd_from = page_id) where page_namespace = 0 and rd_namespace = 0;" | psqlcmd
echo "alter table ${i}pagelinkcount add column othercount integer;" | psqlcmd
echo "update ${i}pagelinkcount set othercount = 0;" | psqlcmd
for j in "${language[@]}"
echo "Language: $i"
echo "CREATE TABLE ${i}pagelinkcount
AS
SELECT pl_title AS title,
COUNT(*) AS count,
0::bigint as othercount
FROM ${i}pagelinks
WHERE pl_namespace = 0
GROUP BY pl_title
;" | psqlcmd
echo "INSERT INTO linkcounts
SELECT '${i}',
pl_title,
COUNT(*)
FROM ${i}pagelinks
WHERE pl_namespace = 0
GROUP BY pl_title
;" | psqlcmd
echo "INSERT INTO wikipedia_redirect
SELECT '${i}',
page_title,
rd_title
FROM ${i}redirect
JOIN ${i}page ON (rd_from = page_id)
WHERE page_namespace = 0
AND rd_namespace = 0
;" | psqlcmd
done
for i in "${LANGUAGES[@]}"
do
for j in "${LANGUAGES[@]}"
do
echo "update ${i}pagelinkcount set othercount = ${i}pagelinkcount.othercount + x.count from (select page_title as title,count from ${i}langlinks join ${i}page on (ll_from = page_id) join ${j}pagelinkcount on (ll_lang = '${j}' and ll_title = title)) as x where x.title = ${i}pagelinkcount.title;" | psqlcmd
echo "UPDATE ${i}pagelinkcount
SET othercount = ${i}pagelinkcount.othercount + x.count
FROM (
SELECT page_title AS title,
count
FROM ${i}langlinks
JOIN ${i}page ON (ll_from = page_id)
JOIN ${j}pagelinkcount ON (ll_lang = '${j}' AND ll_title = title)
) AS x
WHERE x.title = ${i}pagelinkcount.title
;" | psqlcmd
done
echo "insert into wikipedia_article select '${i}', title, count, othercount, count+othercount from ${i}pagelinkcount;" | psqlcmd
echo "INSERT INTO wikipedia_article
SELECT '${i}',
title,
count,
othercount,
count + othercount
FROM ${i}pagelinkcount
;" | psqlcmd
done
# calculate importance score for each wikipedia page
echo "update wikipedia_article set importance = log(totalcount)/log((select max(totalcount) from wikipedia_article))" | psqlcmd
# clean up intermediate tables to conserve space
for i in "${language[@]}"
echo "====================================================================="
echo "Calculate importance score for each wikipedia page"
echo "====================================================================="
echo "UPDATE wikipedia_article
SET importance = LOG(totalcount)/LOG((SELECT MAX(totalcount) FROM wikipedia_article))
;" | psqlcmd
echo "====================================================================="
echo "Clean up intermediate tables to conserve space"
echo "====================================================================="
for i in "${LANGUAGES[@]}"
do
echo "DROP TABLE ${i}pagelinks;" | psqlcmd
echo "DROP TABLE ${i}page;" | psqlcmd
echo "DROP TABLE ${i}langlinks;" | psqlcmd
echo "DROP TABLE ${i}redirect;" | psqlcmd
echo "DROP TABLE ${i}pagelinks;" | psqlcmd
echo "DROP TABLE ${i}page;" | psqlcmd
echo "DROP TABLE ${i}langlinks;" | psqlcmd
echo "DROP TABLE ${i}redirect;" | psqlcmd
echo "DROP TABLE ${i}pagelinkcount;" | psqlcmd
done
echo "all done."

View File

@@ -0,0 +1,39 @@
ar
bg
ca
cs
da
de
en
es
eo
eu
fa
fr
ko
hi
hr
id
it
he
lt
hu
ms
nl
ja
no
pl
pt
kk
ro
ru
sk
sl
sr
fi
sv
tr
uk
vi
war
zh

View File

@@ -7,21 +7,43 @@ configure_file(mkdocs.yml ../mkdocs.yml)
file(MAKE_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}/appendix)
file(MAKE_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}/data-sources)
ADD_CUSTOM_TARGET(doc
COMMAND ${CMAKE_COMMAND} -E create_symlink ${CMAKE_CURRENT_SOURCE_DIR}/admin ${CMAKE_CURRENT_BINARY_DIR}/admin
COMMAND ${CMAKE_COMMAND} -E create_symlink ${CMAKE_CURRENT_SOURCE_DIR}/develop ${CMAKE_CURRENT_BINARY_DIR}/develop
COMMAND ${CMAKE_COMMAND} -E create_symlink ${CMAKE_CURRENT_SOURCE_DIR}/api ${CMAKE_CURRENT_BINARY_DIR}/api
COMMAND ${CMAKE_COMMAND} -E create_symlink ${CMAKE_CURRENT_SOURCE_DIR}/index.md ${CMAKE_CURRENT_BINARY_DIR}/index.md
COMMAND ${CMAKE_COMMAND} -E create_symlink ${CMAKE_CURRENT_SOURCE_DIR}/extra.css ${CMAKE_CURRENT_BINARY_DIR}/extra.css
COMMAND ${CMAKE_COMMAND} -E create_symlink ${CMAKE_CURRENT_SOURCE_DIR}/data-sources/overview.md ${CMAKE_CURRENT_BINARY_DIR}/data-sources/overview.md
set (DOC_SOURCES
admin
develop
api
index.md
extra.css
styles.css
data-sources/overview.md
)
foreach (src ${DOC_SOURCES})
execute_process(
COMMAND ${CMAKE_COMMAND} -E create_symlink ${CMAKE_CURRENT_SOURCE_DIR}/${src} ${CMAKE_CURRENT_BINARY_DIR}/${src}
)
endforeach()
execute_process(
COMMAND ${CMAKE_COMMAND} -E create_symlink ${PROJECT_SOURCE_DIR}/data-sources/us-tiger/README.md ${CMAKE_CURRENT_BINARY_DIR}/data-sources/US-Tiger.md
)
execute_process(
COMMAND ${CMAKE_COMMAND} -E create_symlink ${PROJECT_SOURCE_DIR}/data-sources/gb-postcodes/README.md ${CMAKE_CURRENT_BINARY_DIR}/data-sources/GB-Postcodes.md
)
execute_process(
COMMAND ${CMAKE_COMMAND} -E create_symlink ${PROJECT_SOURCE_DIR}/data-sources/country-grid/README.md ${CMAKE_CURRENT_BINARY_DIR}/data-sources/Country-Grid.md
)
execute_process(
COMMAND ${CMAKE_COMMAND} -E create_symlink ${PROJECT_SOURCE_DIR}/data-sources/country-grid/mexico.quad.png ${CMAKE_CURRENT_BINARY_DIR}/data-sources/mexico.quad.png
)
execute_process(
COMMAND ${CMAKE_COMMAND} -E create_symlink ${PROJECT_SOURCE_DIR}/data-sources/wikipedia-wikidata/README.md ${CMAKE_CURRENT_BINARY_DIR}/data-sources/Wikipedia-Wikidata.md
)
ADD_CUSTOM_TARGET(doc
COMMAND ${CMAKE_CURRENT_SOURCE_DIR}/bash2md.sh ${PROJECT_SOURCE_DIR}/vagrant/Install-on-Centos-7.sh ${CMAKE_CURRENT_BINARY_DIR}/appendix/Install-on-Centos-7.md
COMMAND ${CMAKE_CURRENT_SOURCE_DIR}/bash2md.sh ${PROJECT_SOURCE_DIR}/vagrant/Install-on-Ubuntu-16.sh ${CMAKE_CURRENT_BINARY_DIR}/appendix/Install-on-Ubuntu-16.md
COMMAND ${CMAKE_CURRENT_SOURCE_DIR}/bash2md.sh ${PROJECT_SOURCE_DIR}/vagrant/Install-on-Centos-8.sh ${CMAKE_CURRENT_BINARY_DIR}/appendix/Install-on-Centos-8.md
COMMAND ${CMAKE_CURRENT_SOURCE_DIR}/bash2md.sh ${PROJECT_SOURCE_DIR}/vagrant/Install-on-Ubuntu-18.sh ${CMAKE_CURRENT_BINARY_DIR}/appendix/Install-on-Ubuntu-18.md
COMMAND ${CMAKE_CURRENT_SOURCE_DIR}/bash2md.sh ${PROJECT_SOURCE_DIR}/vagrant/Install-on-Ubuntu-20.sh ${CMAKE_CURRENT_BINARY_DIR}/appendix/Install-on-Ubuntu-20.md
COMMAND mkdocs build -d ${CMAKE_CURRENT_BINARY_DIR}/../site-html -f ${CMAKE_CURRENT_BINARY_DIR}/../mkdocs.yml
)

View File

@@ -0,0 +1,109 @@
# Advanced installations
This page contains instructions for setting up multiple countries in
your Nominatim database. It is assumed that you have already successfully
installed the Nominatim software itself, if not return to the
[installation page](Installation.md).
## Importing multiple regions
To import multiple regions in your database, you need to configure and run `utils/import_multiple_regions.sh` file. This script will set up the update directory which has the following structure:
```bash
update
   ├── europe
   │   ├── andorra
   │   │   └── sequence.state
   │   └── monaco
   │   └── sequence.state
   └── tmp
├── combined.osm.pbf
└── europe
├── andorra-latest.osm.pbf
└── monaco-latest.osm.pbf
```
The `sequence.state` files will contain the sequence ID, which will be used by pyosmium to get updates. The tmp folder is used for import dump.
### Configuring multiple regions
The file `import_multiple_regions.sh` needs to be edited as per your requirement:
1. List of countries. eg:
COUNTRIES="europe/monaco europe/andorra"
2. Path to Build directory. eg:
NOMINATIMBUILD="/srv/nominatim/build"
3. Path to Update directory. eg:
UPDATEDIR="/srv/nominatim/update"
4. Replication URL. eg:
BASEURL="https://download.geofabrik.de"
DOWNCOUNTRYPOSTFIX="-latest.osm.pbf"
!!! tip
If your database already exists and you want to add more countries, replace the setting up part
`${SETUPFILE} --osm-file ${UPDATEDIR}/tmp/combined.osm.pbf --all 2>&1`
with `${UPDATEFILE} --import-file ${UPDATEDIR}/tmp/combined.osm.pbf 2>&1`.
### Setting up multiple regions
Run the following command from your Nominatim directory after configuring the file.
bash ./utils/import_multiple_regions.sh
!!! danger "Important"
This file uses osmium-tool. It must be installed before executing the import script.
Installation instructions can be found [here](https://osmcode.org/osmium-tool/manual.html#installation).
## Updating multiple regions
To import multiple regions in your database, you need to configure and run ```utils/update_database.sh```.
This uses the update directory set up while setting up the DB.
### Configuring multiple regions
The file `update_database.sh` needs to be edited as per your requirement:
1. List of countries. eg:
COUNTRIES="europe/monaco europe/andorra"
2. Path to Build directory. eg:
NOMINATIMBUILD="/srv/nominatim/build"
3. Path to Update directory. eg:
UPDATEDIR="/srv/nominatim/update"
4. Replication URL. eg:
BASEURL="https://download.geofabrik.de"
DOWNCOUNTRYPOSTFIX="-updates"
5. Followup can be set according to your installation. eg: For Photon,
FOLLOWUP="curl http://localhost:2322/nominatim-update"
will handle the indexing.
### Updating the database
Run the following command from your Nominatim directory after configuring the file.
bash ./utils/update_database.sh
This will get diffs from the replication server, import diffs and index the database. The default replication server in the script([Geofabrik](https://download.geofabrik.de)) provides daily updates.
## Verification and further setup
Instructions for import verification and other details like importing Wikidata can be found in [import and update page](Import-and-Update.md)

View File

@@ -26,14 +26,16 @@ If the reported rank is 26 or higher, you can also safely add `--index-noanalyse
PHP Warning: file_get_contents(): open_basedir restriction in effect.
You need to adjust the [open_basedir](https://www.php.net/manual/en/ini.core.php#ini.open-basedir) setting
in your PHP configuration (`php.ini file`). By default this setting may look like this:
You need to adjust the
[open_basedir](https://www.php.net/manual/en/ini.core.php#ini.open-basedir)
setting in your PHP configuration (`php.ini` file). By default this setting may
look like this:
open_basedir = /srv/http/:/home/:/tmp/:/usr/share/pear/
Either add reported directories to the list or disable this setting temporarily by
dding ";" at the beginning of the line. Don't forget to enable this setting again
once you are done with the PHP command line operations.
Either add reported directories to the list or disable this setting temporarily
by adding ";" at the beginning of the line. Don't forget to enable this setting
again once you are done with the PHP command line operations.
### PHP timzeone warnings
@@ -66,7 +68,7 @@ server development libraries (`postgresql-server-dev-9.5` on Ubuntu)
and recompile (`cmake .. && make`).
## I see the error "ERROR: permission denied for language c"
### I see the error "ERROR: permission denied for language c"
`nominatim.so`, written in C, is required to be installed on the database
server. Some managed database (cloud) services like Amazon RDS do not allow
@@ -89,6 +91,11 @@ If you are using a flatnode file, then it may also be that the underlying
filesystem does not fully support 'mmap'. A notable candidate is virtualbox's
vboxfs.
### I see the error: "clang: Command not found" on CentOS
On CentOS 7 users reported `/opt/rh/llvm-toolset-7/root/usr/bin/clang: Command not found`.
Double-check clang is installed. Instead of `make` try running `make CLANG=true`.
### nominatim UPDATE failed: ERROR: buffer 179261 is not owned by resource owner Portal
Several users [reported this](https://github.com/openstreetmap/Nominatim/issues/1168) during the initial import of the database. It's
@@ -107,10 +114,11 @@ to get the full error message.
`could not connect to server: No such file or directory`
On CentOS v7 the PostgreSQL server is started with `systemd`.
Check if `/usr/lib/systemd/system/httpd.service` contains a line `PrivateTmp=true`.
If so then Apache cannot see the `/tmp/.s.PGSQL.5432` file. It's a good security feature,
so use the [preferred solution](../appendix/Install-on-Centos-7/#adding-selinux-security-settings).
On CentOS v7 the PostgreSQL server is started with `systemd`. Check if
`/usr/lib/systemd/system/httpd.service` contains a line `PrivateTmp=true`. If
so then Apache cannot see the `/tmp/.s.PGSQL.5432` file. It's a good security
feature, so use the
[preferred solution](../appendix/Install-on-Centos-7/#adding-selinux-security-settings).
However, you can solve this the quick and dirty way by commenting out that line and then run
@@ -118,14 +126,12 @@ However, you can solve this the quick and dirty way by commenting out that line
sudo systemctl restart httpd
### "must be an array or an object that implements Countable" warning in /usr/share/pear/DB.php
The warning started with PHP 7.2. Make sure you have at least [version 1.9.3 of PEAR DB](https://github.com/pear/DB/releases)
installed.
### Website reports "DB Error: insufficient permissions"
The user the webserver, e.g. Apache, runs under needs to have access to the Nominatim database. You can find the user like [this](https://serverfault.com/questions/125865/finding-out-what-user-apache-is-running-as), for default Ubuntu operating system for example it's `www-data`.
The user the webserver, e.g. Apache, runs under needs to have access to the
Nominatim database. You can find the user like
[this](https://serverfault.com/questions/125865/finding-out-what-user-apache-is-running-as),
for default Ubuntu operating system for example it's `www-data`.
1. Repeat the `createuser` step of the installation instructions.
@@ -164,19 +170,9 @@ When running SELinux, make sure that the
### Setup.php fails with "DB Error: extension not found"
Make sure you have the PostgreSQL extensions "hstore" and "postgis" installed.
See the installation instruction for a full list of required packages.
See the installation instructions for a full list of required packages.
### Setup.php reports "Cannot redeclare getDB()"
`Cannot redeclare getDB() (previously declared in /your/path/Nominatim/lib/db.php:4)`
The message is a bit misleading as PHP needs to load the file `DB.php` and
instead re-loads Nominatim's `db.php`. To solve this make sure you
have the [Pear module 'DB'](https://pear.php.net/package/DB/) installed.
sudo pear install DB
### I forgot to delete the flatnodes file before starting an import.
That's fine. For each import the flatnodes file get overwritten.
@@ -186,22 +182,6 @@ for more information.
## Running your own instance
### Can I import multiple countries and keep them up to date?
You should use the extracts and updates from https://download.geofabrik.de.
For the initial import, download the countries you need and merge them.
See [OSM Help](https://help.openstreetmap.org/questions/48843/merging-two-or-more-geographical-areas-to-import-two-or-more-osm-files-in-nominatim)
for examples how to do that. Use the resulting single osm file when
running `setup.php`.
For updates you need to download the change files for each country
once per day and apply them **separately** using
./utils/update.php --import-diff <filename> --index
See [this issue](https://github.com/openstreetmap/Nominatim/issues/60#issuecomment-18679446)
for a script that runs the updates using osmosis.
### Can I import negative OSM ids into Nominatim?
See [this question of Stackoverflow](https://help.openstreetmap.org/questions/64662/nominatim-flatnode-with-negative-id).

View File

@@ -29,11 +29,11 @@ Add to your `settings/local.php`:
@define('CONST_Osm2pgsql_Flatnode_File', '/path/to/flatnode.file');
Replace the second part with a suitable path on your system and make sure
the directory exists. There should be at least 40GB of free space.
the directory exists. There should be at least 64GB of free space.
## Downloading additional data
### Wikipedia rankings
### Wikipedia/Wikidata rankings
Wikipedia can be used as an optional auxiliary data source to help indicate
the importance of OSM features. Nominatim will work without this information
@@ -41,15 +41,15 @@ but it will improve the quality of the results if this is installed.
This data is available as a binary download:
cd $NOMINATIM_SOURCE_DIR/data
wget https://www.nominatim.org/data/wikipedia_article.sql.bin
wget https://www.nominatim.org/data/wikipedia_redirect.sql.bin
wget https://www.nominatim.org/data/wikimedia-importance.sql.gz
Combined the 2 files are around 1.5GB and add around 30GB to the install
size of Nominatim. They also increase the install time by an hour or so.
The file is about 400MB and adds around 4GB to Nominatim database.
*NOTE:* you'll need to download the Wikipedia rankings before performing
the initial import of the data if you want the rankings applied to the
loaded data.
!!! tip
If you forgot to download the wikipedia rankings, you can also add
importances after the import. Download the files, then run
`./utils/setup.php --import-wikipedia-articles`
and `./utils/update.php --recompute-importance`.
### Great Britain, USA postcodes
@@ -64,7 +64,7 @@ involve a GB or US postcode. This data can be optionally downloaded:
In its default setup Nominatim is configured to import the full OSM data
set for the entire planet. Such a setup requires a powerful machine with
at least 32GB of RAM and around 800GB of SSD hard disks. Depending on your
at least 64GB of RAM and around 800GB of SSD hard disks. Depending on your
use case there are various ways to reduce the amount of data imported. This
section discusses these methods. They can also be combined.
@@ -120,13 +120,16 @@ import styles available which only read selected data:
Import all data necessary to compute addresses down to house number level.
* **settings/import-full.style**
Default style that also includes points of interest.
* **settings/import-extratags.style**
Like the full style but also adds most of the OSM tags into the extratags
column.
The style can be changed with the configuration `CONST_Import_Style`.
To give you an idea of the impact of using the different styles, the table
below gives rough estimates of the final database size after import of a
2018 planet and after using the `--drop` option. It also shows the time
needed for the import on a machine with 32GB RAM, 4 CPUS and SSDs. Note that
needed for the import on a machine with 64GB RAM, 4 CPUS and SSDs. Note that
the given sizes are just an estimate meant for comparison of style requirements.
Your planet import is likely to be larger as the OSM data grows with time.
@@ -136,31 +139,68 @@ admin | 5h | 190 GB | 20 GB
street | 42h | 400 GB | 180 GB
address | 59h | 500 GB | 260 GB
full | 80h | 575 GB | 300 GB
extratags | 80h | 585 GB | 310 GB
You can also customize the styles further. For an description of the
You can also customize the styles further. For a description of the
style format see [the development section](../develop/Import.md).
## Initial import of the data
**Important:** first try the import with a small extract, for example from
[Geofabrik](https://download.geofabrik.de).
!!! danger "Important"
First try the import with a small extract, for example from
[Geofabrik](https://download.geofabrik.de).
Download the data to import and load the data with the following command
from the build directory:
```sh
./utils/setup.php --osm-file <data file> --all [--osm2pgsql-cache 28000] 2>&1 | tee setup.log
./utils/setup.php --osm-file <data file> --all 2>&1 | tee setup.log
```
The `--osm2pgsql-cache` parameter is optional but strongly recommended for
planet imports. It sets the node cache size for the osm2pgsql import part
(see `-C` parameter in osm2pgsql help). As a rule of thumb, this should be
about the same size as the file you are importing but never more than
2/3 of RAM available. If your machine starts swapping reduce the size.
***Note for full planet imports:*** Even on a perfectly configured machine
the import of a full planet takes at least 2 days. Once you see messages
with `Rank .. ETA` appear, the indexing process has started. This part takes
the most time. There are 30 ranks to process. Rank 26 and 30 are the most complex.
They take each about a third of the total import time. If you have not reached
rank 26 after two days of import, it is worth revisiting your system
configuration as it may not be optimal for the import.
Computing word frequency for search terms can improve the performance of
forward geocoding in particular under high load as it helps PostgreSQL's query
planner to make the right decisions. To recompute word counts run:
### Notes on memory usage
In the first step of the import Nominatim uses osm2pgsql to load the OSM data
into the PostgreSQL database. This step is very demanding in terms of RAM usage.
osm2pgsql and PostgreSQL are running in parallel at this point. PostgreSQL
blocks at least the part of RAM that has been configured with the
`shared_buffers` parameter during [PostgreSQL tuning](Installation#postgresql-tuning)
and needs some memory on top of that. osm2pgsql needs at least 2GB of RAM for
its internal data structures, potentially more when it has to process very large
relations. In addition it needs to maintain a cache for node locations. The size
of this cache can be configured with the parameter `--osm2pgsql-cache`.
When importing with a flatnode file, it is best to disable the node cache
completely and leave the memory for the flatnode file. Nominatim will do this
by default, so you do not need to configure anything in this case.
For imports without a flatnode file, set `--osm2pgsql-cache` approximately to
the size of the OSM pbf file (in MB) you are importing. Make sure you leave
enough RAM for PostgreSQL and osm2pgsql as mentioned above. If the system starts
swapping or you are getting out-of-memory errors, reduce the cache size or
even consider using a flatnode file.
### Verify import finished
Run this script to verify all required tables and indices got created successfully.
```sh
./utils/check_import_finished.php
```
## Tuning the database
Accurate word frequency information for search terms helps PostgreSQL's query
planner to make the right decisions. Recomputing them can improve the performance
of forward geocoding in particular under high load. To recompute word counts run:
```sh
./utils/update.php --recompute-word-counts
@@ -178,12 +218,13 @@ you also need to enable these key phrases like this:
./utils/specialphrases.php --wiki-import > specialphrases.sql
psql -d nominatim -f specialphrases.sql
Note that this command downloads the phrases from the wiki link above.
Note that this command downloads the phrases from the wiki link above. You
need internet access for the step.
## Installing Tiger housenumber data for the US
Nominatim is able to use the official [TIGER](https://www.census.gov/geo/maps-data/data/tiger.html)
Nominatim is able to use the official [TIGER](https://www.census.gov/geographies/mapping-files/time-series/geo/tiger-line-file.html)
address set to complement the OSM house number data in the US. You can add
TIGER data to your own Nominatim instance by following these steps. The
entire US adds about 10GB to your database.
@@ -218,6 +259,10 @@ There are many different ways to update your Nominatim database.
The following section describes how to keep it up-to-date with Pyosmium.
For a list of other methods see the output of `./utils/update.php --help`.
!!! warning
If you have configured a flatnode file for the import, then you
need to keep this flatnode file around for updates as well.
#### Installing the newest version of Pyosmium
It is recommended to install Pyosmium via pip. Make sure to use python3.
@@ -268,9 +313,10 @@ The following command will keep your database constantly up to date:
./utils/update.php --import-osmosis-all
(Note that even though the old name "import-osmosis-all" has been kept for compatibility reasons, Osmosis is not required to run this - it uses pyosmium behind the scenes.)
(Note that even though the old name "import-osmosis-all" has been kept for
compatibility reasons, Osmosis is not required to run this - it uses pyosmium
behind the scenes.)
If you have imported multiple country extracts and want to keep them
up-to-date, have a look at the script in
[issue #60](https://github.com/openstreetmap/Nominatim/issues/60).
up-to-date, [Advanced installations section](Advanced-Installations.md) contains instructions
to set up and update multiple country extracts.

View File

@@ -4,8 +4,9 @@ This page contains generic installation instructions for Nominatim and its
prerequisites. There are also step-by-step instructions available for
the following operating systems:
* [Ubuntu 20.04](../appendix/Install-on-Ubuntu-20.md)
* [Ubuntu 18.04](../appendix/Install-on-Ubuntu-18.md)
* [Ubuntu 16.04](../appendix/Install-on-Ubuntu-16.md)
* [CentOS 8](../appendix/Install-on-Centos-8.md)
* [CentOS 7.2](../appendix/Install-on-Centos-7.md)
These OS-specific instructions can also be found in executable form
@@ -25,45 +26,47 @@ and can't offer support.
For compiling:
* [cmake](https://cmake.org/)
* [libxml2](http://xmlsoft.org/)
* a recent C++ compiler
Nominatim comes with its own version of osm2pgsql. See the
osm2pgsql README for additional dependencies required for compiling osm2pgsql.
For running tests:
* [behave](http://pythonhosted.org/behave/)
* [Psycopg2](https://initd.org/psycopg)
* [nose](https://nose.readthedocs.io)
* [phpunit](https://phpunit.de)
* [expat](https://libexpat.github.io/)
* [proj](https://proj.org/)
* [bzip2](http://www.bzip.org/)
* [zlib](https://www.zlib.net/)
* [Boost libraries](https://www.boost.org/), including system and filesystem
* PostgreSQL client libraries
* a recent C++ compiler (gcc 5+ or Clang 3.8+)
For running Nominatim:
* [PostgreSQL](https://www.postgresql.org) (9.3 or later)
* [PostGIS](https://postgis.org) (2.2 or later)
* [PostgreSQL](https://www.postgresql.org) (9.3+)
* [PostGIS](https://postgis.org) (2.2+)
* [Python 3](https://www.python.org/)
* [Psycopg2](https://initd.org/psycopg)
* [PHP](https://php.net) (7.0 or later)
* PHP-pgsql
* PHP-intl (bundled with PHP)
* [PEAR::DB](https://pear.php.net/package/DB)
* a webserver (apache or nginx are recommended)
For running continuous updates:
* [pyosmium](https://osmcode.org/pyosmium/) (with Python 3)
For running tests:
* [behave](http://pythonhosted.org/behave/)
* [nose](https://nose.readthedocs.io)
* [phpunit](https://phpunit.de) >= 7.3
### Hardware
A minimum of 2GB of RAM is required or installation will fail. For a full
planet import 32GB of RAM or more are strongly recommended.
planet import 64GB of RAM or more are strongly recommended. Do not report
out of memory problems if you have less than 64GB RAM.
For a full planet install you will need at least 700GB of hard disk space
For a full planet install you will need at least 800GB of hard disk space
(take into account that the OSM database is growing fast). SSD disks
will help considerably to speed up import and queries.
On a 6-core machine with 32GB RAM and SSDs the import of a full planet takes
a bit more than 2 days. Without SSDs 7-8 days are more realistic.
Even on a well configured machine the import of a full planet takes
at least 2 days. Without SSDs 7-8 days are more realistic.
## Setup of the server
@@ -73,17 +76,30 @@ You might want to tune your PostgreSQL installation so that the later steps
make best use of your hardware. You should tune the following parameters in
your `postgresql.conf` file.
shared_buffers (2GB)
maintenance_work_mem (10GB)
work_mem (50MB)
effective_cache_size (24GB)
shared_buffers = 2GB
maintenance_work_mem = (10GB)
autovacuum_work_mem = 2GB
work_mem = (50MB)
effective_cache_size = (24GB)
synchronous_commit = off
checkpoint_segments = 100 # only for postgresql <= 9.4
max_wal_size = 1GB # postgresql > 9.4
checkpoint_timeout = 10min
checkpoint_completion_target = 0.9
The numbers in brackets behind some parameters seem to work fine for
32GB RAM machine. Adjust to your setup.
64GB RAM machine. Adjust to your setup. A higher number for `max_wal_size`
means that PostgreSQL needs to run checkpoints less often but it does require
the additional space on your disk.
Autovacuum must not be switched off because it ensures that the
tables are frequently analysed. If your machine has very little memory,
you might consider setting:
autovacuum_max_workers = 1
and even reduce `autovacuum_work_mem` further. This will reduce the amount
of memory that autovacuum takes away from the import process.
For the initial import, you should also set:
@@ -91,8 +107,8 @@ For the initial import, you should also set:
full_page_writes = off
Don't forget to reenable them after the initial import or you risk database
corruption. Autovacuum must not be switched off because it ensures that the
tables are frequently analysed.
corruption.
### Webserver setup
@@ -105,13 +121,15 @@ from there.
Make sure your Apache configuration contains the required permissions for the
directory and create an alias:
<Directory "/srv/nominatim/build/website">
Options FollowSymLinks MultiViews
AddType text/html .php
DirectoryIndex search.php
Require all granted
</Directory>
Alias /nominatim /srv/nominatim/build/website
``` apache
<Directory "/srv/nominatim/build/website">
Options FollowSymLinks MultiViews
AddType text/html .php
DirectoryIndex search.php
Require all granted
</Directory>
Alias /nominatim /srv/nominatim/build/website
```
`/srv/nominatim/build` should be replaced with the location of your
build directory.
@@ -139,17 +157,32 @@ follows:
Tell nginx that php files are special and to fastcgi_pass to the php-fpm
unix socket by adding the location definition to the default configuration.
root /srv/nominatim/build/website;
index search.php index.html;
location ~ [^/]\.php(/|$) {
fastcgi_split_path_info ^(.+?\.php)(/.*)$;
if (!-f $document_root$fastcgi_script_name) {
return 404;
}
fastcgi_pass unix:/var/run/php5-fpm.sock;
fastcgi_index search.php;
include fastcgi.conf;
``` nginx
root /srv/nominatim/build/website;
index search.php;
location / {
try_files $uri $uri/ @php;
}
location @php {
fastcgi_param SCRIPT_FILENAME "$document_root$uri.php";
fastcgi_param PATH_TRANSLATED "$document_root$uri.php";
fastcgi_param QUERY_STRING $args;
fastcgi_pass unix:/var/run/php/php7.3-fpm.sock;
fastcgi_index index.php;
include fastcgi_params;
}
location ~ [^/]\.php(/|$) {
fastcgi_split_path_info ^(.+?\.php)(/.*)$;
if (!-f $document_root$fastcgi_script_name) {
return 404;
}
fastcgi_pass unix:/var/run/php7.3-fpm.sock;
fastcgi_index search.php;
include fastcgi.conf;
}
```
Restart the nginx and php5-fpm services and the website should now be available
at `http://localhost/`.

View File

@@ -6,6 +6,21 @@ to newer versions of Nominatim.
SQL statements should be executed from the PostgreSQL commandline. Execute
`psql nominatim` to enter command line mode.
## 3.4.0 -> 3.5.0
### New Wikipedia/Wikidata importance tables
The `wikipedia_*` tables have a new format that also includes references to
Wikidata. You need to update the computation functions and the tables as
follows:
* download the new Wikipedia tables as described in the import section
* reimport the tables: `./utils/setup.php --import-wikipedia-articles`
* update the functions: `./utils/setup.php --create-functions --enable-diff-updates`
* compute importance: `./utils/update.php --recompute-importance`
The last step takes about 10 hours on the full planet.
## 3.3.0 -> 3.4.0
### Reorganisation of location_area_country table

View File

@@ -46,15 +46,16 @@ a single place (for reverse) of the following format:
The possible fields are:
* `place_id` - reference to the Nominatim internal database ID (see notes below)
* `place_id` - reference to the Nominatim internal database ID ([see notes](#place_id-is-not-a-persistent-id))
* `osm_type`, `osm_id` - reference to the OSM object
* `boundingbox` - area of corner coordinates
* `boundingbox` - area of corner coordinates ([see notes](#boundingbox))
* `lat`, `lon` - latitude and longitude of the centroid of the object
* `display_name` - full comma-separated address
* `class`, `type` - key and value of the main OSM tag
* `importance` - computed importance rank
* `icon` - link to class icon (if available)
* `address` - dictionary of address details (only with `addressdetails=1`)
* `address` - dictionary of address details (only with `addressdetails=1`,
[see notes](#addressdetails))
* `extratags` - dictionary with additional useful tags like website or maxspeed
(only with `extratags=1`)
* `namedetails` - dictionary with full list of available names including ref etc.
@@ -75,14 +76,15 @@ a bounding box (`bbox`).
The feature list has the following fields:
* `place_id` - reference to the Nominatim internal database ID (see notes below)
* `place_id` - reference to the Nominatim internal database ID ([see notes](#place_id-is-not-a-persistent-id))
* `osm_type`, `osm_id` - reference to the OSM object
* `category`, `type` - key and value of the main OSM tag
* `display_name` - full comma-separated address
* `place_rank` - class search rank
* `importance` - computed importance rank
* `icon` - link to class icon (if available)
* `address` - dictionary of address details (only with `addressdetails=1`)
* `address` - dictionary of address details (only with `addressdetails=1`,
[see notes](#addressdetails))
* `extratags` - dictionary with additional useful tags like `website` or `maxspeed`
(only with `extratags=1`)
* `namedetails` - dictionary with full list of available names including ref etc.
@@ -100,12 +102,9 @@ The following feature attributes are implemented:
* `type` - value of the main tag of the object (e.g. residential, restaurant, ...)
* `label` - full comma-separated address
* `name` - localised name of the place
* `housenumber`, `street`, `locality`, `postcode`, `city`,
`district`, `county`, `state`, `country` -
* `housenumber`, `street`, `locality`, `district`, `postcode`, `city`,
`county`, `state`, `country` -
provided when it can be determined from the address
(see [this issue](https://github.com/openstreetmap/Nominatim/issues/1080) for
current limitations on the correctness of the address) and `addressdetails=1`
was given
* `admin` - list of localised names of administrative boundaries (only with `addressdetails=1`)
Use `polygon_geojson` to output the full geometry of the object instead
@@ -148,11 +147,11 @@ attribution to OSM and the original querystring.
The place information can be found in the `result` element. The attributes of that element contain:
* `place_id` - reference to the Nominatim internal database ID (see notes below)
* `place_id` - reference to the Nominatim internal database ID ([see notes](#place_id-is-not-a-persistent-id))
* `osm_type`, `osm_id` - reference to the OSM object
* `ref` - content of `ref` tag if it exists
* `lat`, `lon` - latitude and longitude of the centroid of the object
* `boundingbox` - comma-separated list of corner coordinates
* `boundingbox` - comma-separated list of corner coordinates ([see notes](#boundingbox))
The full address of the result can be found in the content of the
`result` element as a comma-separated list.
@@ -203,11 +202,11 @@ generic information about the query:
The place information can be found in the `place` elements, of which there may
be more than one. The attributes of that element contain:
* `place_id` - reference to the Nominatim internal database ID (see notes below)
* `place_id` - reference to the Nominatim internal database ID ([see notes](#place_id-is-not-a-persistent-id))
* `osm_type`, `osm_id` - reference to the OSM object
* `ref` - content of `ref` tag if it exists
* `lat`, `lon` - latitude and longitude of the centroid of the object
* `boundingbox` - comma-separated list of corner coordinates
* `boundingbox` - comma-separated list of corner coordinates ([see notes](#boundingbox))
* `place_rank` - class search rank
* `display_name` - full comma-separated address
* `class`, `type` - key and value of the main OSM tag
@@ -244,3 +243,32 @@ relation) so `osm_type`+`osm_id`+`class_name` would be more unique.
Comma separated list of min latitude, max latitude, min longitude, max longitude.
The whole planet would be `-90,90,-180,180`.
Can we used to pan and center the map on the result, for example with leafletjs
mapping library
`map.fitBounds([[bbox[0],bbox[2]],[bbox[1],bbox[3]]], {padding: [20, 20], maxzoom: 16});`
Bounds crossing the antimeridian have a min latitude -180 and max latitude 180,
essentially covering the planet (See [issue 184](https://github.com/openstreetmap/Nominatim/issues/184)).
### addressdetails
Address details in the xml and json formats return a list of names together
with a designation label. Per default the following labels may appear:
* continent
* country, country_code
* region, state, state_district, county
* municipality, city, town, village
* city_district, district, borough, suburb, subdivision
* hamlet, croft, isolated_dwelling
* neighbourhood, allotments, quarter
* city_block, residental, farm, farmyard, industrial, commercial, retail
* road
* house_number, house_name
* emergency, historic, military, natural, landuse, place, railway,
man_made, aerialway, boundary, amenity, aeroway, club, craft, leisure,
office, mountain_pass, shop, tourism, bridge, tunnel, waterway
They roughly correspond to the classification of the OpenStreetMap data
according to either the `place` tag or the main key of the object.

View File

@@ -7,7 +7,7 @@ Its API has the following endpoints for querying the data:
* __[/search](Search.md)__ - search OSM objects by name or type
* __[/reverse](Reverse.md)__ - search OSM object by their location
* __[/lookup](Lookup.md)__ - look up address details for OSM objects by their ID
* __/status__ - query the status of the server
* __[/status](Status.md)__ - query the status of the server
* __/deletable__ - list objects that have been deleted in OSM but are held
back in Nominatim in case the deletion was accidental
* __/polygons__ - list of broken polygons detected by Nominatim

View File

@@ -92,8 +92,12 @@ comma-separated list of language codes.
* `countrycodes=<countrycode>[,<countrycode>][,<countrycode>]...`
Limit search results to one or more countries. `<countrycode>` must be the
ISO 3166-1alpha2 code, e.g. `gb` for the United Kingdom, `de` for Germany.
[ISO 3166-1alpha2](https://en.wikipedia.org/wiki/ISO_3166-1_alpha-2) code,
e.g. `gb` for the United Kingdom, `de` for Germany.
Each place in Nominatim is assigned to one country code based
on `admin_level=2` tags, in rare cases to none (for example in
international waters outside any country).
* `exclude_place_ids=<place_id,[place_id],[place_id]`
@@ -168,21 +172,27 @@ This overrides the specified machine readable format. (Default: 0)
## Examples
##### XML with polygon points
##### XML with kml polygon
* [https://nominatim.openstreetmap.org/search?q=135+pilkington+avenue,+birmingham&format=xml&polygon=1&addressdetails=1](https://nominatim.openstreetmap.org/search?q=135+pilkington+avenue,+birmingham&format=xml&polygon=1&addressdetails=1)
* [https://nominatim.openstreetmap.org/search/gb/birmingham/pilkington%20avenue/135?format=xml&polygon=1&addressdetails=1](https://nominatim.openstreetmap.org/search/gb/birmingham/pilkington%20avenue/135?format=xml&polygon=1&addressdetails=1)
* [https://nominatim.openstreetmap.org/search/135%20pilkington%20avenue,%20birmingham?format=xml&polygon=1&addressdetails=1](https://nominatim.openstreetmap.org/search/135%20pilkington%20avenue,%20birmingham?format=xml&polygon=1&addressdetails=1)
* [https://nominatim.openstreetmap.org/search?q=135+pilkington+avenue,+birmingham&format=xml&polygon_geojson=1&addressdetails=1](https://nominatim.openstreetmap.org/search?q=135+pilkington+avenue,+birmingham&format=xml&polygon_geojson=1&addressdetails=1)
```xml
<searchresults timestamp="Sat, 07 Nov 09 14:42:10 +0000" querystring="135 pilkington, avenue birmingham" polygon="true">
<place
place_id="1620612" osm_type="node" osm_id="452010817"
boundingbox="52.548641204834,52.5488433837891,-1.81612110137939,-1.81592094898224"
polygonpoints="[['-1.81592098644987','52.5487429714954'],['-1.81592290792183','52.5487234624632'],...]"
lat="52.5487429714954" lon="-1.81602098644987"
display_name="135, Pilkington Avenue, Wylde Green, City of Birmingham, West Midlands (county), B72, United Kingdom"
class="place" type="house">
<geokml>
<Polygon>
<outerBoundaryIs>
<LinearRing>
<coordinates>-1.816513,52.548756599999997 -1.816434,52.548747300000002 -1.816429,52.5487629 -1.8163717,52.548756099999999 -1.8163464,52.548834599999999 -1.8164599,52.548848100000001 -1.8164685,52.5488213 -1.8164913,52.548824000000003 -1.816513,52.548756599999997</coordinates>
</LinearRing>
</outerBoundaryIs>
</Polygon>
</geokml>
<house_number>135</house_number>
<road>Pilkington Avenue</road>
<village>Wylde Green</village>

60
docs/api/Status.md Normal file
View File

@@ -0,0 +1,60 @@
# Status
Useful for checking if the service and database is running. The JSON output also shows
when the database was last updated.
## Parameters
* `format=[text|json]` (defaults to 'text')
## Output
#### Text format
```
https://nominatim.openstreetmap.org/status.php
```
will return HTTP status code 200 and print `OK`.
On error it will return HTTP status code 500 and print a message, e.g.
`ERROR: Database connection failed`.
#### JSON format
```
https://nominatim.openstreetmap.org/status.php?format=json
```
will return HTTP code 200 and a structure
```json
{
"status": 0,
"message": "OK",
"data_updated": "2020-05-04T14:47:00+00:00"
}
```
On error will also return HTTP status code 200 and a structure with error
code and message, e.g.
```json
{
"status": 700,
"message": "Database connection failed"
}
```
Possible status codes are
| | message | notes |
|-----|----------------------|---------------------------------------------------|
| 700 | "No database" | connection failed |
| 701 | "Module failed" | database could not load nominatim.so |
| 702 | "Module call failed" | nominatim.so loaded but calling a function failed |
| 703 | "Query failed" | test query against a database table failed |
| 704 | "No value" | test query worked but returned no results |

View File

@@ -1,36 +1,36 @@
# Documentation Pages
The [Nominatim documentation](https://nominatim.org/release-docs/develop/) is built using the [MkDocs](https://www.mkdocs.org/) static site generation framework. The master branch is automatically deployed every night on under [https://nominatim.org/release-docs/develop/]()
The [Nominatim documentation](https://nominatim.org/release-docs/develop/) is built using the [MkDocs](https://www.mkdocs.org/) static site generation framework. The master branch is automatically deployed every night on under [https://nominatim.org/release-docs/develop/](https://nominatim.org/release-docs/develop/)
To preview local changes:
To preview local changes, first install MkDocs
1. Install MkDocs
```
pip3 install --user mkdocs
```
```
pip3 install --user mkdocs
```
2. In build directory run
Then go to the build directory and run
```
make doc
INFO - Cleaning site directory
INFO - Building documentation to directory: /home/vagrant/build/site-html
```
```
make doc
INFO - Cleaning site directory
INFO - Building documentation to directory: /home/vagrant/build/site-html
```
This runs `mkdocs build` plus extra transformion of some files and adds symlinks (see `CMakeLists.txt` for the exact steps).
This runs `mkdocs build` plus extra transformation of some files and adds
symlinks (see `CMakeLists.txt` for the exact steps).
Now you can start webserver for local testing
3. Start webserver for local testing
```
build> mkdocs serve
[server:296] Serving on http://127.0.0.1:8000
[handlers:62] Start watching changes
```
```
mkdocs serve
[server:296] Serving on http://127.0.0.1:8000
[handlers:62] Start watching changes
```
If you develop inside a Vagrant virtual machine:
If you develop inside a Vagrant virtual machine:
* add port forwarding to your Vagrantfile, e.g. `config.vm.network "forwarded_port", guest: 8000, host: 8000`
* use `mkdocs serve --dev-addr 0.0.0.0:8000` because the default localhost
IP does not get forwarded.
* add port forwarding to your Vagrantfile,
e.g. `config.vm.network "forwarded_port", guest: 8000, host: 8000`
* use `mkdocs serve --dev-addr 0.0.0.0:8000` because the default localhost
IP does not get forwarded.

45
docs/develop/Postcodes.md Normal file
View File

@@ -0,0 +1,45 @@
# Postcodes in Nominatim
The blog post
[Nominatim and Postcodes](https://www.openstreetmap.org/user/lonvia/diary/43143)
describes the handling implemented since Nominatim 3.1.
Postcode centroids (aka 'calculated postcodes') are generated by looking at all
postcodes of a country, grouping them and calculating the geometric centroid.
There is currently no logic to deal with extreme outliers (typos or other
mistakes in OSM data). There is also no check if a postcodes adheres to a
country's format, e.g. if Swiss postcodes are 4 digits.
## Regular updating calculated postcodes
The script to rerun the calculation is
`build/utils/update.php --calculate-postcodes`
and runs once per night on nominatim.openstreetmap.org.
## Finding places that share a specific postcode
In the Nominatim database run
```sql
SELECT address->'postcode' as pc,
osm_type, osm_id, class, type,
st_x(centroid) as lon, st_y(centroid) as lat
FROM placex
WHERE country_code='fr'
AND upper(trim (both ' ' from address->'postcode')) = '33210';
```
Alternatively on [Overpass](https://overpass-turbo.eu/) run the following query
```
[out:json][timeout:250];
area["name"="France"]->.boundaryarea;
(
nwr(area.boundaryarea)["addr:postcode"="33210"];
);
out body;
>;
out skel qt;
```

View File

@@ -1,7 +1,7 @@
site_name: Nominatim Documentation
theme: readthedocs
docs_dir: ${CMAKE_CURRENT_BINARY_DIR}
site_url: http://nominatim.org
site_url: https://nominatim.org
repo_url: https://github.com/openstreetmap/Nominatim
pages:
- 'Introduction' : 'index.md'
@@ -11,17 +11,20 @@ pages:
- 'Reverse': 'api/Reverse.md'
- 'Address Lookup': 'api/Lookup.md'
- 'Details' : 'api/Details.md'
- 'Status' : 'api/Status.md'
- 'Place Output Formats': 'api/Output.md'
- 'FAQ': 'api/Faq.md'
- 'Administration Guide':
- 'Basic Installation': 'admin/Installation.md'
- 'Importing and Updating' : 'admin/Import-and-Update.md'
- 'Advanced Installations' : 'admin/Advanced-Installations.md'
- 'Migration from older Versions' : 'admin/Migration.md'
- 'Troubleshooting' : 'admin/Faq.md'
- 'Developers Guide':
- 'Overview' : 'develop/overview.md'
- 'OSM Data Import' : 'develop/Import.md'
- 'Place Ranking' : 'develop/Ranking.md'
- 'Postcodes' : 'develop/Postcodes.md'
- 'Documentation' : 'develop/Documentation.md'
- 'External Data Sources':
- 'Overview' : 'data-sources/overview.md'
@@ -31,11 +34,12 @@ pages:
- 'Wikipedia & Wikidata': 'data-sources/Wikipedia-Wikidata.md'
- 'Appendix':
- 'Installation on CentOS 7' : 'appendix/Install-on-Centos-7.md'
- 'Installation on Ubuntu 16' : 'appendix/Install-on-Ubuntu-16.md'
- 'Installation on CentOS 8' : 'appendix/Install-on-Centos-8.md'
- 'Installation on Ubuntu 18' : 'appendix/Install-on-Ubuntu-18.md'
- 'Installation on Ubuntu 20' : 'appendix/Install-on-Ubuntu-20.md'
markdown_extensions:
- codehilite:
use_pygments: False
- codehilite
- admonition
- toc:
permalink:
extra_css: [extra.css]
extra_css: [extra.css, styles.css]

69
docs/styles.css Normal file
View File

@@ -0,0 +1,69 @@
.codehilite .hll { background-color: #ffffcc }
.codehilite { background: #f0f0f0; }
.codehilite .c { color: #60a0b0; font-style: italic } /* Comment */
.codehilite .err { /* border: 1px solid #FF0000 */ } /* Error */
.codehilite .k { color: #007020; font-weight: bold } /* Keyword */
.codehilite .o { color: #666666 } /* Operator */
.codehilite .ch { color: #60a0b0; font-style: italic } /* Comment.Hashbang */
.codehilite .cm { color: #60a0b0; font-style: italic } /* Comment.Multiline */
.codehilite .cp { color: #007020 } /* Comment.Preproc */
.codehilite .cpf { color: #60a0b0; font-style: italic } /* Comment.PreprocFile */
.codehilite .c1 { color: #60a0b0; font-style: italic } /* Comment.Single */
.codehilite .cs { color: #60a0b0; background-color: #fff0f0 } /* Comment.Special */
.codehilite .gd { color: #A00000 } /* Generic.Deleted */
.codehilite .ge { font-style: italic } /* Generic.Emph */
.codehilite .gr { color: #FF0000 } /* Generic.Error */
.codehilite .gh { color: #000080; font-weight: bold } /* Generic.Heading */
.codehilite .gi { color: #00A000 } /* Generic.Inserted */
.codehilite .go { color: #888888 } /* Generic.Output */
.codehilite .gp { color: #c65d09; font-weight: bold } /* Generic.Prompt */
.codehilite .gs { font-weight: bold } /* Generic.Strong */
.codehilite .gu { color: #800080; font-weight: bold } /* Generic.Subheading */
.codehilite .gt { color: #0044DD } /* Generic.Traceback */
.codehilite .kc { color: #007020; font-weight: bold } /* Keyword.Constant */
.codehilite .kd { color: #007020; font-weight: bold } /* Keyword.Declaration */
.codehilite .kn { color: #007020; font-weight: bold } /* Keyword.Namespace */
.codehilite .kp { color: #007020 } /* Keyword.Pseudo */
.codehilite .kr { color: #007020; font-weight: bold } /* Keyword.Reserved */
.codehilite .kt { color: #902000 } /* Keyword.Type */
.codehilite .m { color: #40a070 } /* Literal.Number */
.codehilite .s { color: #4070a0 } /* Literal.String */
.codehilite .na { color: #4070a0 } /* Name.Attribute */
.codehilite .nb { color: #007020 } /* Name.Builtin */
.codehilite .nc { color: #0e84b5; font-weight: bold } /* Name.Class */
.codehilite .no { color: #60add5 } /* Name.Constant */
.codehilite .nd { color: #555555; font-weight: bold } /* Name.Decorator */
.codehilite .ni { color: #d55537; font-weight: bold } /* Name.Entity */
.codehilite .ne { color: #007020 } /* Name.Exception */
.codehilite .nf { color: #06287e } /* Name.Function */
.codehilite .nl { color: #002070; font-weight: bold } /* Name.Label */
.codehilite .nn { color: #0e84b5; font-weight: bold } /* Name.Namespace */
.codehilite .nt { color: #062873; font-weight: bold } /* Name.Tag */
.codehilite .nv { color: #bb60d5 } /* Name.Variable */
.codehilite .ow { color: #007020; font-weight: bold } /* Operator.Word */
.codehilite .w { color: #bbbbbb } /* Text.Whitespace */
.codehilite .mb { color: #40a070 } /* Literal.Number.Bin */
.codehilite .mf { color: #40a070 } /* Literal.Number.Float */
.codehilite .mh { color: #40a070 } /* Literal.Number.Hex */
.codehilite .mi { color: #40a070 } /* Literal.Number.Integer */
.codehilite .mo { color: #40a070 } /* Literal.Number.Oct */
.codehilite .sa { color: #4070a0 } /* Literal.String.Affix */
.codehilite .sb { color: #4070a0 } /* Literal.String.Backtick */
.codehilite .sc { color: #4070a0 } /* Literal.String.Char */
.codehilite .dl { color: #4070a0 } /* Literal.String.Delimiter */
.codehilite .sd { color: #4070a0; font-style: italic } /* Literal.String.Doc */
.codehilite .s2 { color: #4070a0 } /* Literal.String.Double */
.codehilite .se { color: #4070a0; font-weight: bold } /* Literal.String.Escape */
.codehilite .sh { color: #4070a0 } /* Literal.String.Heredoc */
.codehilite .si { color: #70a0d0; font-style: italic } /* Literal.String.Interpol */
.codehilite .sx { color: #c65d09 } /* Literal.String.Other */
.codehilite .sr { color: #235388 } /* Literal.String.Regex */
.codehilite .s1 { color: #4070a0 } /* Literal.String.Single */
.codehilite .ss { color: #517918 } /* Literal.String.Symbol */
.codehilite .bp { color: #007020 } /* Name.Builtin.Pseudo */
.codehilite .fm { color: #06287e } /* Name.Function.Magic */
.codehilite .vc { color: #bb60d5 } /* Name.Variable.Class */
.codehilite .vg { color: #bb60d5 } /* Name.Variable.Global */
.codehilite .vi { color: #bb60d5 } /* Name.Variable.Instance */
.codehilite .vm { color: #bb60d5 } /* Name.Variable.Magic */
.codehilite .il { color: #40a070 } /* Literal.Number.Integer.Long */

View File

@@ -9,10 +9,13 @@ require_once(CONST_BasePath.'/lib/ClassTypes.php');
*/
class AddressDetails
{
private $iPlaceID;
private $aAddressLines;
public function __construct(&$oDB, $iPlaceID, $sHousenumber, $mLangPref)
{
$this->iPlaceID = $iPlaceID;
if (is_array($mLangPref)) {
$mLangPref = $oDB->getArraySQL($oDB->getDBQuotedList($mLangPref));
}
@@ -58,48 +61,94 @@ class AddressDetails
return join(', ', $aParts);
}
public function getAddressNames()
public function getAddressNames($sCountry = null)
{
$aAddress = array();
$aFallback = array();
foreach ($this->aAddressLines as $aLine) {
if (!self::isAddress($aLine)) {
continue;
}
$bFallback = false;
$aTypeLabel = ClassTypes\getInfo($aLine);
$sTypeLabel = ClassTypes\getLabelTag($aLine);
if ($aTypeLabel === false) {
$aTypeLabel = ClassTypes\getFallbackInfo($aLine);
$bFallback = true;
}
$sName = false;
if (isset($aLine['localname']) && $aLine['localname']) {
$sName = null;
if (isset($aLine['localname']) && $aLine['localname']!=='') {
$sName = $aLine['localname'];
} elseif (isset($aLine['housenumber']) && $aLine['housenumber']) {
} elseif (isset($aLine['housenumber']) && $aLine['housenumber']!=='') {
$sName = $aLine['housenumber'];
}
if ($sName) {
$sTypeLabel = strtolower(isset($aTypeLabel['simplelabel']) ? $aTypeLabel['simplelabel'] : $aTypeLabel['label']);
$sTypeLabel = str_replace(' ', '_', $sTypeLabel);
if (isset($sName)) {
$sTypeLabel = strtolower(str_replace(' ', '_', $sTypeLabel));
if (!isset($aAddress[$sTypeLabel])
|| isset($aFallback[$sTypeLabel])
|| $aLine['class'] == 'place'
) {
$aAddress[$sTypeLabel] = $sName;
if ($bFallback) {
$aFallback[$sTypeLabel] = $bFallback;
}
}
}
}
return $aAddress;
}
/**
* Annotates the given json with geocodejson address information fields.
*
* @param array $aJson Json hash to add the fields to.
*
* Geocodejson has the following fields:
* street, locality, postcode, city, district,
* county, state, country
*
* Postcode and housenumber are added by type, district is not used.
* All other fields are set according to address rank.
*/
public function addGeocodeJsonAddressParts(&$aJson)
{
foreach (array_reverse($this->aAddressLines) as $aLine) {
if (!$aLine['isaddress']) {
continue;
}
if (!isset($aLine['localname']) || $aLine['localname'] == '') {
continue;
}
if ($aLine['type'] == 'postcode' || $aLine['type'] == 'postal_code') {
$aJson['postcode'] = $aLine['localname'];
continue;
}
if ($aLine['type'] == 'house_number') {
$aJson['housenumber'] = $aLine['localname'];
continue;
}
if ($this->iPlaceID == $aLine['place_id']) {
continue;
}
$iRank = (int)$aLine['rank_address'];
if ($iRank > 25 && $iRank < 28) {
$aJson['street'] = $aLine['localname'];
} elseif ($iRank >= 22 && $iRank <= 25) {
$aJson['locality'] = $aLine['localname'];
} elseif ($iRank >= 17 && $iRank <= 21) {
$aJson['district'] = $aLine['localname'];
} elseif ($iRank >= 13 && $iRank <= 16) {
$aJson['city'] = $aLine['localname'];
} elseif ($iRank >= 10 && $iRank <= 12) {
$aJson['county'] = $aLine['localname'];
} elseif ($iRank >= 5 && $iRank <= 9) {
$aJson['state'] = $aLine['localname'];
} elseif ($iRank == 4) {
$aJson['country'] = $aLine['localname'];
}
}
}
public function getAdminLevels()
{
$aAddress = array();

View File

@@ -2,373 +2,559 @@
namespace Nominatim\ClassTypes;
function getInfo($aPlace)
/**
* Create a label tag for the given place that can be used as an XML name.
*
* @param array[] $aPlace Information about the place to label.
*
* A label tag groups various object types together under a common
* label. The returned value is lower case and has no spaces
*/
function getLabelTag($aPlace, $sCountry = null)
{
$aClassType = getList();
if (isset($aPlace['admin_level'])) {
$sName = $aPlace['class'].':'.$aPlace['type'].':'.$aPlace['admin_level'];
if (isset($aClassType[$sName])) {
return $aClassType[$sName];
}
$iRank = (int) ($aPlace['rank_address'] ?? 30);
$sLabel;
if (isset($aPlace['place_type'])) {
$sLabel = $aPlace['place_type'];
} elseif ($aPlace['class'] == 'boundary' && $aPlace['type'] == 'administrative') {
$sLabel = getBoundaryLabel($iRank/2, $sCountry);
} elseif ($iRank < 26) {
$sLabel = $aPlace['type'];
} elseif ($iRank < 28) {
$sLabel = 'road';
} elseif ($aPlace['class'] == 'place'
&& ($aPlace['type'] == 'house_number' ||
$aPlace['type'] == 'house_name' ||
$aPlace['type'] == 'country_code')
) {
$sLabel = $aPlace['type'];
} else {
$sLabel = $aPlace['class'];
}
$sName = $aPlace['class'].':'.$aPlace['type'];
if (isset($aClassType[$sName])) {
return $aClassType[$sName];
}
return false;
return strtolower(str_replace(' ', '_', $sLabel));
}
function getFallbackInfo($aPlace)
/**
* Create a label for the given place.
*
* @param array[] $aPlace Information about the place to label.
*/
function getLabel($aPlace, $sCountry = null)
{
$aClassType = getList();
$sFallback = 'boundary:administrative:'.((int)($aPlace['rank_address']/2));
if (isset($aClassType[$sFallback])) {
return $aClassType[$sFallback];
if (isset($aPlace['place_type'])) {
return ucwords(str_replace('_', ' ', $aPlace['place_type']));
}
return array('simplelabel' => 'address'.$aPlace['rank_address']);
if ($aPlace['class'] == 'boundary' && $aPlace['type'] == 'administrative') {
return getBoundaryLabel(($aPlace['rank_address'] ?? 30)/2, $sCountry ?? null);
}
// Return a label only for 'important' class/type combinations
if (getImportance($aPlace) !== null) {
return ucwords(str_replace('_', ' ', $aPlace['type']));
}
return null;
}
function getProperty($aPlace, $sProp, $mDefault = false)
/**
* Return a simple label for an administrative boundary for the given country.
*
* @param int $iAdminLevel Content of admin_level tag.
* @param string $sCountry Country code of the country where the object is
* in. May be null, in which case a world-wide
* fallback is used.
* @param string $sFallback String to return if no explicit string is listed.
*
* @return string
*/
function getBoundaryLabel($iAdminLevel, $sCountry, $sFallback = 'Administrative')
{
$aClassType = getList();
static $aBoundaryList = array (
'default' => array (
1 => 'Continent',
2 => 'Country',
3 => 'Region',
4 => 'State',
5 => 'State District',
6 => 'County',
7 => 'Municipality',
8 => 'City',
9 => 'City District',
10 => 'Suburb',
11 => 'Neighbourhood'
),
'no' => array (
3 => 'State',
4 => 'County'
),
'se' => array (
3 => 'State',
4 => 'County'
)
);
if (isset($aPlace['admin_level'])) {
$sName = $aPlace['class'].':'.$aPlace['type'].':'.$aPlace['admin_level'];
if (isset($aClassType[$sName]) && isset($aClassType[$sName][$sProp])) {
return $aClassType[$sName][$sProp];
}
if (isset($aBoundaryList[$sCountry])
&& isset($aBoundaryList[$sCountry][$iAdminLevel])
) {
return $aBoundaryList[$sCountry][$iAdminLevel];
}
$sName = $aPlace['class'].':'.$aPlace['type'];
if (isset($aClassType[$sName]) && isset($aClassType[$sName][$sProp])) {
return $aClassType[$sName][$sProp];
}
return $mDefault;
return $aBoundaryList['default'][$iAdminLevel] ?? $sFallback;
}
function getListWithImportance()
/**
* Return an estimated radius of how far the object node extends.
*
* @param array[] $aPlace Information about the place. This must be a node
* feature.
*
* @return float The radius around the feature in degrees.
*/
function getDefRadius($aPlace)
{
static $aOrders = null;
if ($aOrders === null) {
$aOrders = getList();
$i = 1;
foreach ($aOrders as $sID => $a) {
$aOrders[$sID]['importance'] = $i++;
}
}
return $aOrders;
}
function getList()
{
return array(
'boundary:administrative:1' => array('label' => 'Continent', 'frequency' => 0, 'icon' => 'poi_boundary_administrative', 'defdiameter' => 0.32),
'boundary:administrative:2' => array('label' => 'Country', 'frequency' => 0, 'icon' => 'poi_boundary_administrative', 'defdiameter' => 0.32),
'place:country' => array('label' => 'Country', 'frequency' => 0, 'icon' => 'poi_boundary_administrative', 'defzoom' => 6, 'defdiameter' => 15),
'boundary:administrative:3' => array('label' => 'State', 'frequency' => 0, 'icon' => 'poi_boundary_administrative', 'defdiameter' => 0.32),
'boundary:administrative:4' => array('label' => 'State', 'frequency' => 0, 'icon' => 'poi_boundary_administrative', 'defdiameter' => 0.32),
'place:state' => array('label' => 'State', 'frequency' => 0, 'icon' => 'poi_boundary_administrative', 'defzoom' => 8, 'defdiameter' => 5.12),
'boundary:administrative:5' => array('label' => 'State District', 'frequency' => 0, 'icon' => 'poi_boundary_administrative', 'defdiameter' => 0.32),
'boundary:administrative:6' => array('label' => 'County', 'frequency' => 0, 'icon' => 'poi_boundary_administrative', 'defdiameter' => 0.32),
'boundary:administrative:7' => array('label' => 'County', 'frequency' => 0, 'icon' => 'poi_boundary_administrative', 'defdiameter' => 0.32),
'place:county' => array('label' => 'County', 'frequency' => 108, 'icon' => 'poi_boundary_administrative', 'defzoom' => 10, 'defdiameter' => 1.28),
'boundary:administrative:8' => array('label' => 'City', 'frequency' => 0, 'icon' => 'poi_boundary_administrative', 'defdiameter' => 0.32),
'place:city' => array('label' => 'City', 'frequency' => 66, 'icon' => 'poi_place_city', 'defzoom' => 12, 'defdiameter' => 0.32),
'boundary:administrative:9' => array('label' => 'City District', 'frequency' => 0, 'icon' => 'poi_boundary_administrative', 'defdiameter' => 0.32),
'boundary:administrative:10' => array('label' => 'Suburb', 'frequency' => 0, 'icon' => 'poi_boundary_administrative', 'defdiameter' => 0.32),
'boundary:administrative:11' => array('label' => 'Neighbourhood', 'frequency' => 0, 'icon' => 'poi_boundary_administrative', 'defdiameter' => 0.32),
'place:region' => array('label' => 'Region', 'frequency' => 0, 'icon' => 'poi_boundary_administrative', 'defzoom' => 8, 'defdiameter' => 0.04),
'place:island' => array('label' => 'Island', 'frequency' => 288, 'defzoom' => 11, 'defdiameter' => 0.64),
'boundary:administrative' => array('label' => 'Administrative', 'frequency' => 413, 'icon' => 'poi_boundary_administrative', 'defdiameter' => 0.32),
'boundary:postal_code' => array('label' => 'Postcode', 'frequency' => 413, 'icon' => 'poi_boundary_administrative', 'defdiameter' => 0.32),
'place:town' => array('label' => 'Town', 'frequency' => 1497, 'icon' => 'poi_place_town', 'defzoom' => 14, 'defdiameter' => 0.08),
'place:village' => array('label' => 'Village', 'frequency' => 11230, 'icon' => 'poi_place_village', 'defzoom' => 15, 'defdiameter' => 0.04),
'place:hamlet' => array('label' => 'Hamlet', 'frequency' => 7075, 'icon' => 'poi_place_village', 'defzoom' => 15, 'defdiameter' => 0.04),
'place:suburb' => array('label' => 'Suburb', 'frequency' => 2528, 'icon' => 'poi_place_village', 'defdiameter' => 0.04),
'place:locality' => array('label' => 'Locality', 'frequency' => 4113, 'icon' => 'poi_place_village', 'defdiameter' => 0.02),
'landuse:farm' => array('label' => 'Farm', 'frequency' => 1201, 'defdiameter' => 0.02),
'place:farm' => array('label' => 'Farm', 'frequency' => 1162, 'defdiameter' => 0.02),
'highway:motorway_junction' => array('label' => 'Motorway Junction', 'frequency' => 1126, 'simplelabel' => 'Junction'),
'highway:motorway' => array('label' => 'Motorway', 'frequency' => 4627, 'simplelabel' => 'Road'),
'highway:trunk' => array('label' => 'Trunk', 'frequency' => 23084, 'simplelabel' => 'Road'),
'highway:primary' => array('label' => 'Primary', 'frequency' => 32138, 'simplelabel' => 'Road'),
'highway:secondary' => array('label' => 'Secondary', 'frequency' => 25807, 'simplelabel' => 'Road'),
'highway:tertiary' => array('label' => 'Tertiary', 'frequency' => 29829, 'simplelabel' => 'Road'),
'highway:residential' => array('label' => 'Residential', 'frequency' => 361498, 'simplelabel' => 'Road'),
'highway:unclassified' => array('label' => 'Unclassified', 'frequency' => 66441, 'simplelabel' => 'Road'),
'highway:living_street' => array('label' => 'Living Street', 'frequency' => 710, 'simplelabel' => 'Road'),
'highway:service' => array('label' => 'Service', 'frequency' => 9963, 'simplelabel' => 'Road'),
'highway:track' => array('label' => 'Track', 'frequency' => 2565, 'simplelabel' => 'Road'),
'highway:road' => array('label' => 'Road', 'frequency' => 591, 'simplelabel' => 'Road'),
'highway:byway' => array('label' => 'Byway', 'frequency' => 346, 'simplelabel' => 'Road'),
'highway:bridleway' => array('label' => 'Bridleway', 'frequency' => 1556),
'highway:cycleway' => array('label' => 'Cycleway', 'frequency' => 2419),
'highway:pedestrian' => array('label' => 'Pedestrian', 'frequency' => 2757),
'highway:footway' => array('label' => 'Footway', 'frequency' => 15008),
'highway:steps' => array('label' => 'Steps', 'frequency' => 444, 'simplelabel' => 'Footway'),
'highway:motorway_link' => array('label' => 'Motorway Link', 'frequency' => 795, 'simplelabel' => 'Road'),
'highway:trunk_link' => array('label' => 'Trunk Link', 'frequency' => 1258, 'simplelabel' => 'Road'),
'highway:primary_link' => array('label' => 'Primary Link', 'frequency' => 313, 'simplelabel' => 'Road'),
'landuse:industrial' => array('label' => 'Industrial', 'frequency' => 1062),
'landuse:residential' => array('label' => 'Residential', 'frequency' => 886),
'landuse:retail' => array('label' => 'Retail', 'frequency' => 754),
'landuse:commercial' => array('label' => 'Commercial', 'frequency' => 657),
'place:airport' => array('label' => 'Airport', 'frequency' => 36, 'icon' => 'transport_airport2', 'defdiameter' => 0.03),
'aeroway:aerodrome' => array('label' => 'Aerodrome', 'frequency' => 36, 'icon' => 'transport_airport2', 'defdiameter' => 0.03),
'aeroway' => array('label' => 'Aeroway', 'frequency' => 36, 'icon' => 'transport_airport2', 'defdiameter' => 0.03),
'railway:station' => array('label' => 'Station', 'frequency' => 3431, 'icon' => 'transport_train_station2', 'defdiameter' => 0.01),
'amenity:place_of_worship' => array('label' => 'Place Of Worship', 'frequency' => 9049, 'icon' => 'place_of_worship_unknown3'),
'amenity:pub' => array('label' => 'Pub', 'frequency' => 18969, 'icon' => 'food_pub'),
'amenity:bar' => array('label' => 'Bar', 'frequency' => 164, 'icon' => 'food_bar'),
'amenity:university' => array('label' => 'University', 'frequency' => 607, 'icon' => 'education_university'),
'tourism:museum' => array('label' => 'Museum', 'frequency' => 543, 'icon' => 'tourist_museum'),
'amenity:arts_centre' => array('label' => 'Arts Centre', 'frequency' => 136, 'icon' => 'tourist_art_gallery2'),
'tourism:zoo' => array('label' => 'Zoo', 'frequency' => 47, 'icon' => 'tourist_zoo'),
'tourism:theme_park' => array('label' => 'Theme Park', 'frequency' => 24, 'icon' => 'poi_point_of_interest'),
'tourism:attraction' => array('label' => 'Attraction', 'frequency' => 1463, 'icon' => 'poi_point_of_interest'),
'leisure:golf_course' => array('label' => 'Golf Course', 'frequency' => 712, 'icon' => 'sport_golf'),
'historic:castle' => array('label' => 'Castle', 'frequency' => 316, 'icon' => 'tourist_castle'),
'amenity:hospital' => array('label' => 'Hospital', 'frequency' => 879, 'icon' => 'health_hospital'),
'amenity:school' => array('label' => 'School', 'frequency' => 8192, 'icon' => 'education_school'),
'amenity:theatre' => array('label' => 'Theatre', 'frequency' => 371, 'icon' => 'tourist_theatre'),
'amenity:public_building' => array('label' => 'Public Building', 'frequency' => 985),
'amenity:library' => array('label' => 'Library', 'frequency' => 794, 'icon' => 'amenity_library'),
'amenity:townhall' => array('label' => 'Townhall', 'frequency' => 242),
'amenity:community_centre' => array('label' => 'Community Centre', 'frequency' => 157),
'amenity:fire_station' => array('label' => 'Fire Station', 'frequency' => 221, 'icon' => 'amenity_firestation3'),
'amenity:police' => array('label' => 'Police', 'frequency' => 334, 'icon' => 'amenity_police2'),
'amenity:bank' => array('label' => 'Bank', 'frequency' => 1248, 'icon' => 'money_bank2'),
'amenity:post_office' => array('label' => 'Post Office', 'frequency' => 859, 'icon' => 'amenity_post_office'),
'leisure:park' => array('label' => 'Park', 'frequency' => 2378),
'amenity:park' => array('label' => 'Park', 'frequency' => 53),
'landuse:park' => array('label' => 'Park', 'frequency' => 50),
'landuse:recreation_ground' => array('label' => 'Recreation Ground', 'frequency' => 517),
'tourism:hotel' => array('label' => 'Hotel', 'frequency' => 2150, 'icon' => 'accommodation_hotel2'),
'tourism:motel' => array('label' => 'Motel', 'frequency' => 43),
'amenity:cinema' => array('label' => 'Cinema', 'frequency' => 277, 'icon' => 'tourist_cinema'),
'tourism:artwork' => array('label' => 'Artwork', 'frequency' => 171, 'icon' => 'tourist_art_gallery2'),
'historic:archaeological_site' => array('label' => 'Archaeological Site', 'frequency' => 407, 'icon' => 'tourist_archaeological2'),
'amenity:doctors' => array('label' => 'Doctors', 'frequency' => 581, 'icon' => 'health_doctors'),
'leisure:sports_centre' => array('label' => 'Sports Centre', 'frequency' => 767, 'icon' => 'sport_leisure_centre'),
'leisure:swimming_pool' => array('label' => 'Swimming Pool', 'frequency' => 24, 'icon' => 'sport_swimming_outdoor'),
'shop:supermarket' => array('label' => 'Supermarket', 'frequency' => 2673, 'icon' => 'shopping_supermarket'),
'shop:convenience' => array('label' => 'Convenience', 'frequency' => 1469, 'icon' => 'shopping_convenience'),
'amenity:restaurant' => array('label' => 'Restaurant', 'frequency' => 3179, 'icon' => 'food_restaurant'),
'amenity:fast_food' => array('label' => 'Fast Food', 'frequency' => 2289, 'icon' => 'food_fastfood'),
'amenity:cafe' => array('label' => 'Cafe', 'frequency' => 1780, 'icon' => 'food_cafe'),
'tourism:guest_house' => array('label' => 'Guest House', 'frequency' => 223, 'icon' => 'accommodation_bed_and_breakfast'),
'amenity:pharmacy' => array('label' => 'Pharmacy', 'frequency' => 733, 'icon' => 'health_pharmacy_dispensing'),
'amenity:fuel' => array('label' => 'Fuel', 'frequency' => 1308, 'icon' => 'transport_fuel'),
'natural:peak' => array('label' => 'Peak', 'frequency' => 3212, 'icon' => 'poi_peak'),
'waterway:waterfall' => array('label' => 'Waterfall', 'frequency' => 24),
'natural:wood' => array('label' => 'Wood', 'frequency' => 1845, 'icon' => 'landuse_coniferous_and_deciduous'),
'natural:water' => array('label' => 'Water', 'frequency' => 1790),
'landuse:forest' => array('label' => 'Forest', 'frequency' => 467),
'landuse:cemetery' => array('label' => 'Cemetery', 'frequency' => 463),
'landuse:allotments' => array('label' => 'Allotments', 'frequency' => 408),
'landuse:farmyard' => array('label' => 'Farmyard', 'frequency' => 397),
'railway:rail' => array('label' => 'Rail', 'frequency' => 4894),
'waterway:canal' => array('label' => 'Canal', 'frequency' => 1723),
'waterway:river' => array('label' => 'River', 'frequency' => 4089),
'waterway:stream' => array('label' => 'Stream', 'frequency' => 2684),
'shop:bicycle' => array('label' => 'Bicycle', 'frequency' => 349, 'icon' => 'shopping_bicycle'),
'shop:clothes' => array('label' => 'Clothes', 'frequency' => 315, 'icon' => 'shopping_clothes'),
'shop:hairdresser' => array('label' => 'Hairdresser', 'frequency' => 312, 'icon' => 'shopping_hairdresser'),
'shop:doityourself' => array('label' => 'Doityourself', 'frequency' => 247, 'icon' => 'shopping_diy'),
'shop:estate_agent' => array('label' => 'Estate Agent', 'frequency' => 162, 'icon' => 'shopping_estateagent2'),
'shop:car' => array('label' => 'Car', 'frequency' => 159, 'icon' => 'shopping_car'),
'shop:garden_centre' => array('label' => 'Garden Centre', 'frequency' => 143, 'icon' => 'shopping_garden_centre'),
'shop:car_repair' => array('label' => 'Car Repair', 'frequency' => 141, 'icon' => 'shopping_car_repair'),
'shop:newsagent' => array('label' => 'Newsagent', 'frequency' => 132),
'shop:bakery' => array('label' => 'Bakery', 'frequency' => 129, 'icon' => 'shopping_bakery'),
'shop:furniture' => array('label' => 'Furniture', 'frequency' => 124),
'shop:butcher' => array('label' => 'Butcher', 'frequency' => 105, 'icon' => 'shopping_butcher'),
'shop:apparel' => array('label' => 'Apparel', 'frequency' => 98, 'icon' => 'shopping_clothes'),
'shop:electronics' => array('label' => 'Electronics', 'frequency' => 96),
'shop:department_store' => array('label' => 'Department Store', 'frequency' => 86),
'shop:books' => array('label' => 'Books', 'frequency' => 85),
'shop:yes' => array('label' => 'Shop', 'frequency' => 68),
'shop:outdoor' => array('label' => 'Outdoor', 'frequency' => 67),
'shop:mall' => array('label' => 'Mall', 'frequency' => 63),
'shop:florist' => array('label' => 'Florist', 'frequency' => 61),
'shop:charity' => array('label' => 'Charity', 'frequency' => 60),
'shop:hardware' => array('label' => 'Hardware', 'frequency' => 59),
'shop:laundry' => array('label' => 'Laundry', 'frequency' => 51, 'icon' => 'shopping_laundrette'),
'shop:shoes' => array('label' => 'Shoes', 'frequency' => 49),
'shop:beverages' => array('label' => 'Beverages', 'frequency' => 48, 'icon' => 'shopping_alcohol'),
'shop:dry_cleaning' => array('label' => 'Dry Cleaning', 'frequency' => 46),
'shop:carpet' => array('label' => 'Carpet', 'frequency' => 45),
'shop:computer' => array('label' => 'Computer', 'frequency' => 44),
'shop:alcohol' => array('label' => 'Alcohol', 'frequency' => 44, 'icon' => 'shopping_alcohol'),
'shop:optician' => array('label' => 'Optician', 'frequency' => 55, 'icon' => 'health_opticians'),
'shop:chemist' => array('label' => 'Chemist', 'frequency' => 42, 'icon' => 'health_pharmacy'),
'shop:gallery' => array('label' => 'Gallery', 'frequency' => 38, 'icon' => 'tourist_art_gallery2'),
'shop:mobile_phone' => array('label' => 'Mobile Phone', 'frequency' => 37),
'shop:sports' => array('label' => 'Sports', 'frequency' => 37),
'shop:jewelry' => array('label' => 'Jewelry', 'frequency' => 32, 'icon' => 'shopping_jewelry'),
'shop:pet' => array('label' => 'Pet', 'frequency' => 29),
'shop:beauty' => array('label' => 'Beauty', 'frequency' => 28),
'shop:stationery' => array('label' => 'Stationery', 'frequency' => 25),
'shop:shopping_centre' => array('label' => 'Shopping Centre', 'frequency' => 25),
'shop:general' => array('label' => 'General', 'frequency' => 25),
'shop:electrical' => array('label' => 'Electrical', 'frequency' => 25),
'shop:toys' => array('label' => 'Toys', 'frequency' => 23),
'shop:jeweller' => array('label' => 'Jeweller', 'frequency' => 23),
'shop:betting' => array('label' => 'Betting', 'frequency' => 23),
'shop:household' => array('label' => 'Household', 'frequency' => 21),
'shop:travel_agency' => array('label' => 'Travel Agency', 'frequency' => 21),
'shop:hifi' => array('label' => 'Hifi', 'frequency' => 21),
'amenity:shop' => array('label' => 'Shop', 'frequency' => 61),
'tourism:information' => array('label' => 'Information', 'frequency' => 224, 'icon' => 'amenity_information'),
'place:house' => array('label' => 'House', 'frequency' => 2086, 'defzoom' => 18),
'place:house_name' => array('label' => 'House', 'frequency' => 2086, 'defzoom' => 18),
'place:house_number' => array('label' => 'House Number', 'frequency' => 2086, 'defzoom' => 18),
'place:country_code' => array('label' => 'Country Code', 'frequency' => 2086, 'defzoom' => 18),
//
'leisure:pitch' => array('label' => 'Pitch', 'frequency' => 762),
'highway:unsurfaced' => array('label' => 'Unsurfaced', 'frequency' => 492),
'historic:ruins' => array('label' => 'Ruins', 'frequency' => 483, 'icon' => 'tourist_ruin'),
'amenity:college' => array('label' => 'College', 'frequency' => 473, 'icon' => 'education_school'),
'historic:monument' => array('label' => 'Monument', 'frequency' => 470, 'icon' => 'tourist_monument'),
'railway:subway' => array('label' => 'Subway', 'frequency' => 385),
'historic:memorial' => array('label' => 'Memorial', 'frequency' => 382, 'icon' => 'tourist_monument'),
'leisure:nature_reserve' => array('label' => 'Nature Reserve', 'frequency' => 342),
'leisure:common' => array('label' => 'Common', 'frequency' => 322),
'waterway:lock_gate' => array('label' => 'Lock Gate', 'frequency' => 321),
'natural:fell' => array('label' => 'Fell', 'frequency' => 308),
'amenity:nightclub' => array('label' => 'Nightclub', 'frequency' => 292),
'highway:path' => array('label' => 'Path', 'frequency' => 287),
'leisure:garden' => array('label' => 'Garden', 'frequency' => 285),
'landuse:reservoir' => array('label' => 'Reservoir', 'frequency' => 276),
'leisure:playground' => array('label' => 'Playground', 'frequency' => 264),
'leisure:stadium' => array('label' => 'Stadium', 'frequency' => 212),
'historic:mine' => array('label' => 'Mine', 'frequency' => 193, 'icon' => 'poi_mine'),
'natural:cliff' => array('label' => 'Cliff', 'frequency' => 193),
'tourism:caravan_site' => array('label' => 'Caravan Site', 'frequency' => 183, 'icon' => 'accommodation_caravan_park'),
'amenity:bus_station' => array('label' => 'Bus Station', 'frequency' => 181, 'icon' => 'transport_bus_station'),
'amenity:kindergarten' => array('label' => 'Kindergarten', 'frequency' => 179),
'highway:construction' => array('label' => 'Construction', 'frequency' => 176),
'amenity:atm' => array('label' => 'Atm', 'frequency' => 172, 'icon' => 'money_atm2'),
'amenity:emergency_phone' => array('label' => 'Emergency Phone', 'frequency' => 164),
'waterway:lock' => array('label' => 'Lock', 'frequency' => 146),
'waterway:riverbank' => array('label' => 'Riverbank', 'frequency' => 143),
'natural:coastline' => array('label' => 'Coastline', 'frequency' => 142),
'tourism:viewpoint' => array('label' => 'Viewpoint', 'frequency' => 140, 'icon' => 'tourist_view_point'),
'tourism:hostel' => array('label' => 'Hostel', 'frequency' => 140),
'tourism:bed_and_breakfast' => array('label' => 'Bed And Breakfast', 'frequency' => 140, 'icon' => 'accommodation_bed_and_breakfast'),
'railway:halt' => array('label' => 'Halt', 'frequency' => 135),
'railway:platform' => array('label' => 'Platform', 'frequency' => 134),
'railway:tram' => array('label' => 'Tram', 'frequency' => 130, 'icon' => 'transport_tram_stop'),
'amenity:courthouse' => array('label' => 'Courthouse', 'frequency' => 129, 'icon' => 'amenity_court'),
'amenity:recycling' => array('label' => 'Recycling', 'frequency' => 126, 'icon' => 'amenity_recycling'),
'amenity:dentist' => array('label' => 'Dentist', 'frequency' => 124, 'icon' => 'health_dentist'),
'natural:beach' => array('label' => 'Beach', 'frequency' => 121, 'icon' => 'tourist_beach'),
'place:moor' => array('label' => 'Moor', 'frequency' => 118),
'amenity:grave_yard' => array('label' => 'Grave Yard', 'frequency' => 110),
'waterway:drain' => array('label' => 'Drain', 'frequency' => 108),
'landuse:grass' => array('label' => 'Grass', 'frequency' => 106),
'landuse:village_green' => array('label' => 'Village Green', 'frequency' => 106),
'natural:bay' => array('label' => 'Bay', 'frequency' => 102),
'railway:tram_stop' => array('label' => 'Tram Stop', 'frequency' => 101, 'icon' => 'transport_tram_stop'),
'leisure:marina' => array('label' => 'Marina', 'frequency' => 98),
'highway:stile' => array('label' => 'Stile', 'frequency' => 97),
'natural:moor' => array('label' => 'Moor', 'frequency' => 95),
'railway:light_rail' => array('label' => 'Light Rail', 'frequency' => 91),
'railway:narrow_gauge' => array('label' => 'Narrow Gauge', 'frequency' => 90),
'natural:land' => array('label' => 'Land', 'frequency' => 86),
'amenity:village_hall' => array('label' => 'Village Hall', 'frequency' => 82),
'waterway:dock' => array('label' => 'Dock', 'frequency' => 80),
'amenity:veterinary' => array('label' => 'Veterinary', 'frequency' => 79),
'landuse:brownfield' => array('label' => 'Brownfield', 'frequency' => 77),
'leisure:track' => array('label' => 'Track', 'frequency' => 76),
'railway:historic_station' => array('label' => 'Historic Station', 'frequency' => 74),
'landuse:construction' => array('label' => 'Construction', 'frequency' => 72),
'amenity:prison' => array('label' => 'Prison', 'frequency' => 71, 'icon' => 'amenity_prison'),
'landuse:quarry' => array('label' => 'Quarry', 'frequency' => 71),
'amenity:telephone' => array('label' => 'Telephone', 'frequency' => 70),
'highway:traffic_signals' => array('label' => 'Traffic Signals', 'frequency' => 66),
'natural:heath' => array('label' => 'Heath', 'frequency' => 62),
'historic:house' => array('label' => 'House', 'frequency' => 61),
'amenity:social_club' => array('label' => 'Social Club', 'frequency' => 61),
'landuse:military' => array('label' => 'Military', 'frequency' => 61),
'amenity:health_centre' => array('label' => 'Health Centre', 'frequency' => 59),
'historic:building' => array('label' => 'Building', 'frequency' => 58),
'amenity:clinic' => array('label' => 'Clinic', 'frequency' => 57),
'highway:services' => array('label' => 'Services', 'frequency' => 56),
'amenity:ferry_terminal' => array('label' => 'Ferry Terminal', 'frequency' => 55),
'natural:marsh' => array('label' => 'Marsh', 'frequency' => 55),
'natural:hill' => array('label' => 'Hill', 'frequency' => 54),
'highway:raceway' => array('label' => 'Raceway', 'frequency' => 53),
'amenity:taxi' => array('label' => 'Taxi', 'frequency' => 47),
'amenity:take_away' => array('label' => 'Take Away', 'frequency' => 45),
'amenity:car_rental' => array('label' => 'Car Rental', 'frequency' => 44),
'place:islet' => array('label' => 'Islet', 'frequency' => 44),
'amenity:nursery' => array('label' => 'Nursery', 'frequency' => 44),
'amenity:nursing_home' => array('label' => 'Nursing Home', 'frequency' => 43),
'amenity:toilets' => array('label' => 'Toilets', 'frequency' => 38),
'amenity:hall' => array('label' => 'Hall', 'frequency' => 38),
'waterway:boatyard' => array('label' => 'Boatyard', 'frequency' => 36),
'highway:mini_roundabout' => array('label' => 'Mini Roundabout', 'frequency' => 35),
'historic:manor' => array('label' => 'Manor', 'frequency' => 35),
'tourism:chalet' => array('label' => 'Chalet', 'frequency' => 34),
'amenity:bicycle_parking' => array('label' => 'Bicycle Parking', 'frequency' => 34),
'amenity:hotel' => array('label' => 'Hotel', 'frequency' => 34),
'waterway:weir' => array('label' => 'Weir', 'frequency' => 33),
'natural:wetland' => array('label' => 'Wetland', 'frequency' => 33),
'natural:cave_entrance' => array('label' => 'Cave Entrance', 'frequency' => 32),
'amenity:crematorium' => array('label' => 'Crematorium', 'frequency' => 31),
'tourism:picnic_site' => array('label' => 'Picnic Site', 'frequency' => 31),
'landuse:wood' => array('label' => 'Wood', 'frequency' => 30),
'landuse:basin' => array('label' => 'Basin', 'frequency' => 30),
'natural:tree' => array('label' => 'Tree', 'frequency' => 30),
'leisure:slipway' => array('label' => 'Slipway', 'frequency' => 29),
'landuse:meadow' => array('label' => 'Meadow', 'frequency' => 29),
'landuse:piste' => array('label' => 'Piste', 'frequency' => 28),
'amenity:care_home' => array('label' => 'Care Home', 'frequency' => 28),
'amenity:club' => array('label' => 'Club', 'frequency' => 28),
'amenity:medical_centre' => array('label' => 'Medical Centre', 'frequency' => 27),
'historic:roman_road' => array('label' => 'Roman Road', 'frequency' => 27),
'historic:fort' => array('label' => 'Fort', 'frequency' => 26),
'railway:subway_entrance' => array('label' => 'Subway Entrance', 'frequency' => 26),
'historic:yes' => array('label' => 'Historic', 'frequency' => 25),
'highway:gate' => array('label' => 'Gate', 'frequency' => 25),
'leisure:fishing' => array('label' => 'Fishing', 'frequency' => 24),
'historic:museum' => array('label' => 'Museum', 'frequency' => 24),
'amenity:car_wash' => array('label' => 'Car Wash', 'frequency' => 24),
'railway:level_crossing' => array('label' => 'Level Crossing', 'frequency' => 23),
'leisure:bird_hide' => array('label' => 'Bird Hide', 'frequency' => 23),
'natural:headland' => array('label' => 'Headland', 'frequency' => 21),
'tourism:apartments' => array('label' => 'Apartments', 'frequency' => 21),
'amenity:shopping' => array('label' => 'Shopping', 'frequency' => 21),
'natural:scrub' => array('label' => 'Scrub', 'frequency' => 20),
'natural:fen' => array('label' => 'Fen', 'frequency' => 20),
'building:yes' => array('label' => 'Building', 'frequency' => 200),
'mountain_pass:yes' => array('label' => 'Mountain Pass', 'frequency' => 200),
'amenity:parking' => array('label' => 'Parking', 'frequency' => 3157),
'highway:bus_stop' => array('label' => 'Bus Stop', 'frequency' => 35777, 'icon' => 'transport_bus_stop2'),
'place:postcode' => array('label' => 'Postcode', 'frequency' => 27267),
'amenity:post_box' => array('label' => 'Post Box', 'frequency' => 9613),
'place:houses' => array('label' => 'Houses', 'frequency' => 85),
'railway:preserved' => array('label' => 'Preserved', 'frequency' => 227),
'waterway:derelict_canal' => array('label' => 'Derelict Canal', 'frequency' => 21),
'amenity:dead_pub' => array('label' => 'Dead Pub', 'frequency' => 20),
'railway:disused_station' => array('label' => 'Disused Station', 'frequency' => 114),
'railway:abandoned' => array('label' => 'Abandoned', 'frequency' => 641),
'railway:disused' => array('label' => 'Disused', 'frequency' => 72),
$aSpecialRadius = array(
'place:continent' => 25,
'place:country' => 7,
'place:state' => 2.6,
'place:province' => 2.6,
'place:region' => 1.0,
'place:county' => 0.7,
'place:city' => 0.16,
'place:municipality' => 0.16,
'place:island' => 0.32,
'place:postcode' => 0.16,
'place:town' => 0.04,
'place:village' => 0.02,
'place:hamlet' => 0.02,
'place:district' => 0.02,
'place:borough' => 0.02,
'place:suburb' => 0.02,
'place:locality' => 0.01,
'place:neighbourhood'=> 0.01,
'place:quarter' => 0.01,
'place:city_block' => 0.01,
'landuse:farm' => 0.01,
'place:farm' => 0.01,
'place:airport' => 0.015,
'aeroway:aerodrome' => 0.015,
'railway:station' => 0.005
);
$sClassPlace = $aPlace['class'].':'.$aPlace['type'];
return $aSpecialRadius[$sClassPlace] ?? 0.00005;
}
/**
* Get the icon to use with the given object.
*/
function getIcon($aPlace)
{
$aIcons = array(
'boundary:administrative' => 'poi_boundary_administrative',
'place:city' => 'poi_place_city',
'place:town' => 'poi_place_town',
'place:village' => 'poi_place_village',
'place:hamlet' => 'poi_place_village',
'place:suburb' => 'poi_place_village',
'place:locality' => 'poi_place_village',
'place:airport' => 'transport_airport2',
'aeroway:aerodrome' => 'transport_airport2',
'railway:station' => 'transport_train_station2',
'amenity:place_of_worship' => 'place_of_worship_unknown3',
'amenity:pub' => 'food_pub',
'amenity:bar' => 'food_bar',
'amenity:university' => 'education_university',
'tourism:museum' => 'tourist_museum',
'amenity:arts_centre' => 'tourist_art_gallery2',
'tourism:zoo' => 'tourist_zoo',
'tourism:theme_park' => 'poi_point_of_interest',
'tourism:attraction' => 'poi_point_of_interest',
'leisure:golf_course' => 'sport_golf',
'historic:castle' => 'tourist_castle',
'amenity:hospital' => 'health_hospital',
'amenity:school' => 'education_school',
'amenity:theatre' => 'tourist_theatre',
'amenity:library' => 'amenity_library',
'amenity:fire_station' => 'amenity_firestation3',
'amenity:police' => 'amenity_police2',
'amenity:bank' => 'money_bank2',
'amenity:post_office' => 'amenity_post_office',
'tourism:hotel' => 'accommodation_hotel2',
'amenity:cinema' => 'tourist_cinema',
'tourism:artwork' => 'tourist_art_gallery2',
'historic:archaeological_site' => 'tourist_archaeological2',
'amenity:doctors' => 'health_doctors',
'leisure:sports_centre' => 'sport_leisure_centre',
'leisure:swimming_pool' => 'sport_swimming_outdoor',
'shop:supermarket' => 'shopping_supermarket',
'shop:convenience' => 'shopping_convenience',
'amenity:restaurant' => 'food_restaurant',
'amenity:fast_food' => 'food_fastfood',
'amenity:cafe' => 'food_cafe',
'tourism:guest_house' => 'accommodation_bed_and_breakfast',
'amenity:pharmacy' => 'health_pharmacy_dispensing',
'amenity:fuel' => 'transport_fuel',
'natural:peak' => 'poi_peak',
'natural:wood' => 'landuse_coniferous_and_deciduous',
'shop:bicycle' => 'shopping_bicycle',
'shop:clothes' => 'shopping_clothes',
'shop:hairdresser' => 'shopping_hairdresser',
'shop:doityourself' => 'shopping_diy',
'shop:estate_agent' => 'shopping_estateagent2',
'shop:car' => 'shopping_car',
'shop:garden_centre' => 'shopping_garden_centre',
'shop:car_repair' => 'shopping_car_repair',
'shop:bakery' => 'shopping_bakery',
'shop:butcher' => 'shopping_butcher',
'shop:apparel' => 'shopping_clothes',
'shop:laundry' => 'shopping_laundrette',
'shop:beverages' => 'shopping_alcohol',
'shop:alcohol' => 'shopping_alcohol',
'shop:optician' => 'health_opticians',
'shop:chemist' => 'health_pharmacy',
'shop:gallery' => 'tourist_art_gallery2',
'shop:jewelry' => 'shopping_jewelry',
'tourism:information' => 'amenity_information',
'historic:ruins' => 'tourist_ruin',
'amenity:college' => 'education_school',
'historic:monument' => 'tourist_monument',
'historic:memorial' => 'tourist_monument',
'historic:mine' => 'poi_mine',
'tourism:caravan_site' => 'accommodation_caravan_park',
'amenity:bus_station' => 'transport_bus_station',
'amenity:atm' => 'money_atm2',
'tourism:viewpoint' => 'tourist_view_point',
'tourism:guesthouse' => 'accommodation_bed_and_breakfast',
'railway:tram' => 'transport_tram_stop',
'amenity:courthouse' => 'amenity_court',
'amenity:recycling' => 'amenity_recycling',
'amenity:dentist' => 'health_dentist',
'natural:beach' => 'tourist_beach',
'railway:tram_stop' => 'transport_tram_stop',
'amenity:prison' => 'amenity_prison',
'highway:bus_stop' => 'transport_bus_stop2'
);
$sClassPlace = $aPlace['class'].':'.$aPlace['type'];
return $aIcons[$sClassPlace] ?? null;
}
/**
* Get an icon for the given object with its full URL.
*/
function getIconFile($aPlace)
{
$sIcon = getIcon($aPlace);
if (!isset($sIcon)) {
return null;
}
return CONST_Website_BaseURL.'images/mapicons/'.$sIcon.'.p.20.png';
}
/**
* Return a class importance value for the given place.
*
* @param array[] $aPlace Information about the place.
*
* @return int An importance value. The lower the value, the more
* important the class.
*/
function getImportance($aPlace)
{
static $aWithImportance = null;
if ($aWithImportance === null) {
$aWithImportance = array_flip(array(
'place:country',
'place:state',
'place:province',
'place:county',
'place:city',
'place:region',
'place:island',
'place:town',
'place:village',
'place:hamlet',
'place:suburb',
'place:locality',
'landuse:farm',
'place:farm',
'highway:motorway_junction',
'highway:motorway',
'highway:trunk',
'highway:primary',
'highway:secondary',
'highway:tertiary',
'highway:residential',
'highway:unclassified',
'highway:living_street',
'highway:service',
'highway:track',
'highway:road',
'highway:byway',
'highway:bridleway',
'highway:cycleway',
'highway:pedestrian',
'highway:footway',
'highway:steps',
'highway:motorway_link',
'highway:trunk_link',
'highway:primary_link',
'landuse:industrial',
'landuse:residential',
'landuse:retail',
'landuse:commercial',
'place:airport',
'aeroway:aerodrome',
'railway:station',
'amenity:place_of_worship',
'amenity:pub',
'amenity:bar',
'amenity:university',
'tourism:museum',
'amenity:arts_centre',
'tourism:zoo',
'tourism:theme_park',
'tourism:attraction',
'leisure:golf_course',
'historic:castle',
'amenity:hospital',
'amenity:school',
'amenity:theatre',
'amenity:public_building',
'amenity:library',
'amenity:townhall',
'amenity:community_centre',
'amenity:fire_station',
'amenity:police',
'amenity:bank',
'amenity:post_office',
'leisure:park',
'amenity:park',
'landuse:park',
'landuse:recreation_ground',
'tourism:hotel',
'tourism:motel',
'amenity:cinema',
'tourism:artwork',
'historic:archaeological_site',
'amenity:doctors',
'leisure:sports_centre',
'leisure:swimming_pool',
'shop:supermarket',
'shop:convenience',
'amenity:restaurant',
'amenity:fast_food',
'amenity:cafe',
'tourism:guest_house',
'amenity:pharmacy',
'amenity:fuel',
'natural:peak',
'waterway:waterfall',
'natural:wood',
'natural:water',
'landuse:forest',
'landuse:cemetery',
'landuse:allotments',
'landuse:farmyard',
'railway:rail',
'waterway:canal',
'waterway:river',
'waterway:stream',
'shop:bicycle',
'shop:clothes',
'shop:hairdresser',
'shop:doityourself',
'shop:estate_agent',
'shop:car',
'shop:garden_centre',
'shop:car_repair',
'shop:newsagent',
'shop:bakery',
'shop:furniture',
'shop:butcher',
'shop:apparel',
'shop:electronics',
'shop:department_store',
'shop:books',
'shop:yes',
'shop:outdoor',
'shop:mall',
'shop:florist',
'shop:charity',
'shop:hardware',
'shop:laundry',
'shop:shoes',
'shop:beverages',
'shop:dry_cleaning',
'shop:carpet',
'shop:computer',
'shop:alcohol',
'shop:optician',
'shop:chemist',
'shop:gallery',
'shop:mobile_phone',
'shop:sports',
'shop:jewelry',
'shop:pet',
'shop:beauty',
'shop:stationery',
'shop:shopping_centre',
'shop:general',
'shop:electrical',
'shop:toys',
'shop:jeweller',
'shop:betting',
'shop:household',
'shop:travel_agency',
'shop:hifi',
'amenity:shop',
'tourism:information',
'place:house',
'place:house_name',
'place:house_number',
'place:country_code',
'leisure:pitch',
'highway:unsurfaced',
'historic:ruins',
'amenity:college',
'historic:monument',
'railway:subway',
'historic:memorial',
'leisure:nature_reserve',
'leisure:common',
'waterway:lock_gate',
'natural:fell',
'amenity:nightclub',
'highway:path',
'leisure:garden',
'landuse:reservoir',
'leisure:playground',
'leisure:stadium',
'historic:mine',
'natural:cliff',
'tourism:caravan_site',
'amenity:bus_station',
'amenity:kindergarten',
'highway:construction',
'amenity:atm',
'amenity:emergency_phone',
'waterway:lock',
'waterway:riverbank',
'natural:coastline',
'tourism:viewpoint',
'tourism:hostel',
'tourism:bed_and_breakfast',
'railway:halt',
'railway:platform',
'railway:tram',
'amenity:courthouse',
'amenity:recycling',
'amenity:dentist',
'natural:beach',
'place:moor',
'amenity:grave_yard',
'waterway:drain',
'landuse:grass',
'landuse:village_green',
'natural:bay',
'railway:tram_stop',
'leisure:marina',
'highway:stile',
'natural:moor',
'railway:light_rail',
'railway:narrow_gauge',
'natural:land',
'amenity:village_hall',
'waterway:dock',
'amenity:veterinary',
'landuse:brownfield',
'leisure:track',
'railway:historic_station',
'landuse:construction',
'amenity:prison',
'landuse:quarry',
'amenity:telephone',
'highway:traffic_signals',
'natural:heath',
'historic:house',
'amenity:social_club',
'landuse:military',
'amenity:health_centre',
'historic:building',
'amenity:clinic',
'highway:services',
'amenity:ferry_terminal',
'natural:marsh',
'natural:hill',
'highway:raceway',
'amenity:taxi',
'amenity:take_away',
'amenity:car_rental',
'place:islet',
'amenity:nursery',
'amenity:nursing_home',
'amenity:toilets',
'amenity:hall',
'waterway:boatyard',
'highway:mini_roundabout',
'historic:manor',
'tourism:chalet',
'amenity:bicycle_parking',
'amenity:hotel',
'waterway:weir',
'natural:wetland',
'natural:cave_entrance',
'amenity:crematorium',
'tourism:picnic_site',
'landuse:wood',
'landuse:basin',
'natural:tree',
'leisure:slipway',
'landuse:meadow',
'landuse:piste',
'amenity:care_home',
'amenity:club',
'amenity:medical_centre',
'historic:roman_road',
'historic:fort',
'railway:subway_entrance',
'historic:yes',
'highway:gate',
'leisure:fishing',
'historic:museum',
'amenity:car_wash',
'railway:level_crossing',
'leisure:bird_hide',
'natural:headland',
'tourism:apartments',
'amenity:shopping',
'natural:scrub',
'natural:fen',
'building:yes',
'mountain_pass:yes',
'amenity:parking',
'highway:bus_stop',
'place:postcode',
'amenity:post_box',
'place:houses',
'railway:preserved',
'waterway:derelict_canal',
'amenity:dead_pub',
'railway:disused_station',
'railway:abandoned',
'railway:disused'
));
}
$sClassPlace = $aPlace['class'].':'.$aPlace['type'];
return $aWithImportance[$sClassPlace] ?? null;
}

View File

@@ -135,7 +135,7 @@ class DB
try {
$stmt = $this->getQueryStatement($sSQL, $aInputVars, $sErrMessage);
while ($val = $stmt->fetchColumn(0)) { // returns first column or false
while (($val = $stmt->fetchColumn(0)) !== false) { // returns first column or false
$aVals[] = $val;
}
} catch (\PDOException $e) {
@@ -241,11 +241,103 @@ class DB
}
/**
* Since the DSN includes the database name, checks if the connection works.
* Returns a list of table names in the database
*
* @return array[]
*/
public function getListOfTables()
{
return $this->getCol("SELECT tablename FROM pg_tables WHERE schemaname='public'");
}
/**
* Deletes a table. Returns true if deleted or didn't exist.
*
* @param string $sTableName
*
* @return boolean
*/
public function databaseExists()
public function deleteTable($sTableName)
{
return $this->exec('DROP TABLE IF EXISTS '.$sTableName.' CASCADE') == 0;
}
/**
* Check if an index exists in the database. Optional filtered by tablename
*
* @param string $sTableName
*
* @return boolean
*/
public function indexExists($sIndexName, $sTableName = null)
{
return in_array($sIndexName, $this->getListOfIndices($sTableName));
}
/**
* Returns a list of index names in the database, optional filtered by tablename
*
* @param string $sTableName
*
* @return array
*/
public function getListOfIndices($sTableName = null)
{
// table_name | index_name | column_name
// -----------------------+---------------------------------+--------------
// country_name | idx_country_name_country_code | country_code
// country_osm_grid | idx_country_osm_grid_geometry | geometry
// import_polygon_delete | idx_import_polygon_delete_osmid | osm_id
// import_polygon_delete | idx_import_polygon_delete_osmid | osm_type
// import_polygon_error | idx_import_polygon_error_osmid | osm_id
// import_polygon_error | idx_import_polygon_error_osmid | osm_type
$sSql = <<< END
SELECT
t.relname as table_name,
i.relname as index_name,
a.attname as column_name
FROM
pg_class t,
pg_class i,
pg_index ix,
pg_attribute a
WHERE
t.oid = ix.indrelid
and i.oid = ix.indexrelid
and a.attrelid = t.oid
and a.attnum = ANY(ix.indkey)
and t.relkind = 'r'
and i.relname NOT LIKE 'pg_%'
FILTERS
ORDER BY
t.relname,
i.relname,
a.attname
END;
$aRows = null;
if ($sTableName) {
$sSql = str_replace('FILTERS', 'and t.relname = :tablename', $sSql);
$aRows = $this->getAll($sSql, array(':tablename' => $sTableName));
} else {
$sSql = str_replace('FILTERS', '', $sSql);
$aRows = $this->getAll($sSql);
}
$aIndexNames = array_unique(array_map(function ($aRow) {
return $aRow['index_name'];
}, $aRows));
sort($aIndexNames);
return $aIndexNames;
}
/**
* Tries to connect to the database but on failure doesn't throw an exception.
*
* @return boolean
*/
public function checkConnection()
{
$bExists = true;
try {
@@ -280,11 +372,18 @@ class DB
return (float) ($aMatches[1].'.'.$aMatches[2]);
}
/**
* Returns an associate array of postgresql database connection settings. Keys can
* be 'database', 'hostspec', 'port', 'username', 'password'.
* Returns empty array on failure, thus check if at least 'database' is set.
*
* @return array[]
*/
public static function parseDSN($sDSN)
{
// https://secure.php.net/manual/en/ref.pdo-pgsql.connection.php
$aInfo = array();
if (preg_match('/^pgsql:(.+)/', $sDSN, $aMatches)) {
if (preg_match('/^pgsql:(.+)$/', $sDSN, $aMatches)) {
foreach (explode(';', $aMatches[1]) as $sKeyVal) {
list($sKey, $sVal) = explode('=', $sKeyVal, 2);
if ($sKey == 'host') $sKey = 'hostspec';
@@ -295,4 +394,28 @@ class DB
}
return $aInfo;
}
/**
* Takes an array of settings and return the DNS string. Key names can be
* 'database', 'hostspec', 'port', 'username', 'password' but aliases
* 'dbname', 'host' and 'user' are also supported.
*
* @return string
*
*/
public static function generateDSN($aInfo)
{
$sDSN = sprintf(
'pgsql:host=%s;port=%s;dbname=%s;user=%s;password=%s;',
$aInfo['host'] ?? $aInfo['hostspec'] ?? '',
$aInfo['port'] ?? '',
$aInfo['dbname'] ?? $aInfo['database'] ?? '',
$aInfo['user'] ?? '',
$aInfo['password'] ?? ''
);
$sDSN = preg_replace('/\b\w+=;/', '', $sDSN);
$sDSN = preg_replace('/;\Z/', '', $sDSN);
return $sDSN;
}
}

View File

@@ -245,7 +245,6 @@ class Geocode
}
$this->oPlaceLookup->loadParamArray($oParams, $sForceGeometryType);
$this->oPlaceLookup->setIncludePolygonAsPoints($oParams->getBool('polygon'));
$this->oPlaceLookup->setIncludeAddressDetails($oParams->getBool('addressdetails', false));
}
@@ -808,9 +807,7 @@ class Geocode
$sSQL .= 'WHERE place_id in ('.$sPlaceIds.') ';
$sSQL .= ' AND (';
$sSQL .= " placex.rank_address between $this->iMinAddressRank and $this->iMaxAddressRank ";
if (14 >= $this->iMinAddressRank && 14 <= $this->iMaxAddressRank) {
$sSQL .= " OR (extratags->'place') = 'city'";
}
$sSQL .= " OR placex.rank_search between $this->iMinAddressRank and $this->iMaxAddressRank ";
if ($this->aAddressRankList) {
$sSQL .= ' OR placex.rank_address in ('.join(',', $this->aAddressRankList).')';
}
@@ -890,7 +887,6 @@ class Geocode
$aSearchResults = $this->oPlaceLookup->lookup($aResults);
$aClassType = ClassTypes\getListWithImportance();
$aRecheckWords = preg_split('/\b[\s,\\-]*/u', $sQuery);
foreach ($aRecheckWords as $i => $sWord) {
if (!preg_match('/[\pL\pN]/', $sWord)) unset($aRecheckWords[$i]);
@@ -899,33 +895,23 @@ class Geocode
Debug::printVar('Recheck words', $aRecheckWords);
foreach ($aSearchResults as $iIdx => $aResult) {
// Default
$fDiameter = ClassTypes\getProperty($aResult, 'defdiameter', 0.0001);
$fRadius = ClassTypes\getDefRadius($aResult);
$aOutlineResult = $this->oPlaceLookup->getOutlines($aResult['place_id'], $aResult['lon'], $aResult['lat'], $fDiameter/2);
$aOutlineResult = $this->oPlaceLookup->getOutlines($aResult['place_id'], $aResult['lon'], $aResult['lat'], $fRadius);
if ($aOutlineResult) {
$aResult = array_merge($aResult, $aOutlineResult);
}
if ($aResult['extra_place'] == 'city') {
$aResult['class'] = 'place';
$aResult['type'] = 'city';
$aResult['rank_search'] = 16;
}
// Is there an icon set for this type of result?
$aClassInfo = ClassTypes\getInfo($aResult);
if ($aClassInfo) {
if (isset($aClassInfo['icon'])) {
$aResult['icon'] = CONST_Website_BaseURL.'images/mapicons/'.$aClassInfo['icon'].'.p.20.png';
}
if (isset($aClassInfo['label'])) {
$aResult['label'] = $aClassInfo['label'];
}
$sIcon = ClassTypes\getIconFile($aResult);
if (isset($sIcon)) {
$aResult['icon'] = $sIcon;
}
$sLabel = ClassTypes\getLabel($aResult);
if (isset($sLabel)) {
$aResult['label'] = $sLabel;
}
$aResult['name'] = $aResult['langaddress'];
if ($oCtx->hasNearPoint()) {
@@ -955,10 +941,9 @@ class Geocode
// - number of exact matches from the query
$aResult['foundorder'] -= $aResults[$aResult['place_id']]->iExactMatches;
// - importance of the class/type
if (isset($aClassType[$aResult['class'].':'.$aResult['type']]['importance'])
&& $aClassType[$aResult['class'].':'.$aResult['type']]['importance']
) {
$aResult['foundorder'] += 0.0001 * $aClassType[$aResult['class'].':'.$aResult['type']]['importance'];
$iClassImportance = ClassTypes\getImportance($aResult);
if (isset($iClassImportance)) {
$aResult['foundorder'] += 0.0001 * $iClassImportance;
} else {
$aResult['foundorder'] += 0.01;
}

View File

@@ -104,18 +104,29 @@ class ParameterParser
}
foreach ($aLanguages as $sLanguage => $fLanguagePref) {
$aLangPrefOrder['short_name:'.$sLanguage] = 'short_name:'.$sLanguage;
$aLangPrefOrder['name:'.$sLanguage] = 'name:'.$sLanguage;
}
$aLangPrefOrder['short_name'] = 'short_name';
$aLangPrefOrder['name'] = 'name';
$aLangPrefOrder['brand'] = 'brand';
foreach ($aLanguages as $sLanguage => $fLanguagePref) {
$aLangPrefOrder['official_name:'.$sLanguage] = 'official_name:'.$sLanguage;
$aLangPrefOrder['short_name:'.$sLanguage] = 'short_name:'.$sLanguage;
}
$aLangPrefOrder['official_name'] = 'official_name';
$aLangPrefOrder['short_name'] = 'short_name';
$aLangPrefOrder['ref'] = 'ref';
$aLangPrefOrder['type'] = 'type';
return $aLangPrefOrder;
}
public function hasSetAny($aParamNames)
{
foreach ($aParamNames as $sName) {
if ($this->getBool($sName)) {
return true;
}
}
return false;
}
}

View File

@@ -15,7 +15,6 @@ class PlaceLookup
protected $bExtraTags = false;
protected $bNameDetails = false;
protected $bIncludePolygonAsPoints = false;
protected $bIncludePolygonAsText = false;
protected $bIncludePolygonAsGeoJSON = false;
protected $bIncludePolygonAsKML = false;
@@ -38,11 +37,6 @@ class PlaceLookup
return $this->bDeDupe;
}
public function setIncludePolygonAsPoints($b = true)
{
$this->bIncludePolygonAsPoints = $b;
}
public function setIncludeAddressDetails($b)
{
$this->bAddressDetails = $b;
@@ -61,7 +55,6 @@ class PlaceLookup
if ($sGeomType === null || $sGeomType == 'geojson') {
$this->bIncludePolygonAsGeoJSON = $oParams->getBool('polygon_geojson');
$this->bIncludePolygonAsPoints = false;
}
if ($oParams->getString('format', '') !== 'geojson') {
@@ -100,7 +93,6 @@ class PlaceLookup
if ($this->bExtraTags) $aParams['extratags'] = '1';
if ($this->bNameDetails) $aParams['namedetails'] = '1';
if ($this->bIncludePolygonAsPoints) $aParams['polygon'] = '1';
if ($this->bIncludePolygonAsText) $aParams['polygon_text'] = '1';
if ($this->bIncludePolygonAsGeoJSON) $aParams['polygon_geojson'] = '1';
if ($this->bIncludePolygonAsKML) $aParams['polygon_kml'] = '1';
@@ -215,7 +207,7 @@ class PlaceLookup
'ST_Collect(centroid)',
'min(CASE WHEN placex.rank_search < 28 THEN placex.place_id ELSE placex.parent_place_id END)'
);
$sSQL .= " (extratags->'place') AS extra_place ";
$sSQL .= " COALESCE(extratags->'place', extratags->'linked_place') AS extra_place ";
$sSQL .= ' FROM placex';
$sSQL .= " WHERE place_id in ($sPlaceIDs) ";
$sSQL .= ' AND (';
@@ -248,7 +240,7 @@ class PlaceLookup
$sSQL .= ' ref, ';
if ($this->bExtraTags) $sSQL .= 'extratags, ';
if ($this->bNameDetails) $sSQL .= 'name, ';
$sSQL .= " extratags->'place' ";
$sSQL .= ' extra_place ';
$aSubSelects[] = $sSQL;
}
@@ -456,10 +448,9 @@ class PlaceLookup
}
}
$aPlace['addresstype'] = ClassTypes\getProperty(
$aPlace['addresstype'] = ClassTypes\getLabelTag(
$aPlace,
'simplelabel',
$aPlace['class']
$aPlace['country_code']
);
}
@@ -500,7 +491,7 @@ class PlaceLookup
if ($this->bIncludePolygonAsGeoJSON) $sSQL .= ',ST_AsGeoJSON(geometry) as asgeojson';
if ($this->bIncludePolygonAsKML) $sSQL .= ',ST_AsKML(geometry) as askml';
if ($this->bIncludePolygonAsSVG) $sSQL .= ',ST_AsSVG(geometry) as assvg';
if ($this->bIncludePolygonAsText || $this->bIncludePolygonAsPoints) $sSQL .= ',ST_AsText(geometry) as astext';
if ($this->bIncludePolygonAsText) $sSQL .= ',ST_AsText(geometry) as astext';
if ($fLonReverse != null && $fLatReverse != null) {
$sFrom = ' from (SELECT * , CASE WHEN (class = \'highway\') AND (ST_GeometryType(geometry) = \'ST_LineString\') THEN ';
$sFrom .=' ST_ClosestPoint(geometry, ST_SetSRID(ST_Point('.$fLatReverse.','.$fLonReverse.'),4326))';
@@ -527,8 +518,6 @@ class PlaceLookup
if ($this->bIncludePolygonAsKML) $aOutlineResult['askml'] = $aPointPolygon['askml'];
if ($this->bIncludePolygonAsSVG) $aOutlineResult['assvg'] = $aPointPolygon['assvg'];
if ($this->bIncludePolygonAsText) $aOutlineResult['astext'] = $aPointPolygon['astext'];
if ($this->bIncludePolygonAsPoints) $aOutlineResult['aPolyPoints'] = geometryText2Points($aPointPolygon['astext'], $fRadius);
if (abs($aPointPolygon['minlat'] - $aPointPolygon['maxlat']) < 0.0000001) {
$aPointPolygon['minlat'] = $aPointPolygon['minlat'] - $fRadius;
@@ -551,17 +540,12 @@ class PlaceLookup
// as a fallback we generate a bounding box without knowing the size of the geometry
if ((!isset($aOutlineResult['aBoundingBox'])) && isset($fLon)) {
//
if ($this->bIncludePolygonAsPoints) {
$sGeometryText = 'POINT('.$fLon.','.$fLat.')';
$aOutlineResult['aPolyPoints'] = geometryText2Points($sGeometryText, $fRadius);
}
$aBounds = array();
$aBounds['minlat'] = $fLat - $fRadius;
$aBounds['maxlat'] = $fLat + $fRadius;
$aBounds['minlon'] = $fLon - $fRadius;
$aBounds['maxlon'] = $fLon + $fRadius;
$aBounds = array(
'minlat' => $fLat - $fRadius,
'maxlat' => $fLat + $fRadius,
'minlon' => $fLon - $fRadius,
'maxlon' => $fLon + $fRadius
);
$aOutlineResult['aBoundingBox'] = array(
(string)$aBounds['minlat'],

View File

@@ -203,7 +203,7 @@ class SearchContext
}
/**
* Get an SQL snipped for computing the distance from the reference point.
* Get an SQL snippet for computing the distance from the reference point.
*
* @param string $sObj SQL variable name to compute the distance from.
*
@@ -215,7 +215,7 @@ class SearchContext
}
/**
* Get an SQL snipped for checking if something is within range of the
* Get an SQL snippet for checking if something is within range of the
* reference point.
*
* @param string $sObj SQL variable name to compute if it is within range.
@@ -228,14 +228,14 @@ class SearchContext
}
/**
* Get an SQL snipped of the importance factor of the viewbox.
* Get an SQL snippet of the importance factor of the viewbox.
*
* The importance factor is computed by checking if an object is within
* the viewbox and/or the extended version of the viewbox.
*
* @param string $sObj SQL variable name of object to weight the importance
*
* @return string SQL snipped of the factor with a leading multiply sign.
* @return string SQL snippet of the factor with a leading multiply sign.
*/
public function viewboxImportanceSQL($sObj)
{
@@ -252,7 +252,7 @@ class SearchContext
}
/**
* SQL snipped checking if a place ID should be excluded.
* SQL snippet checking if a place ID should be excluded.
*
* @param string $sVariable SQL variable name of place ID to check,
* potentially prefixed with more SQL.

View File

@@ -447,8 +447,8 @@ class SearchDescription
$iLimit
);
//now search for housenumber, if housenumber provided
if ($this->sHouseNumber && !empty($aResults)) {
// Now search for housenumber, if housenumber provided. Can be zero.
if (($this->sHouseNumber || $this->sHouseNumber === '0') && !empty($aResults)) {
// Downgrade the rank of the street results, they are missing
// the housenumber.
foreach ($aResults as $oRes) {
@@ -660,10 +660,7 @@ class SearchDescription
$aTerms[] = 'address_rank between 16 and 27';
} elseif (!$this->sClass || $this->iOperator == Operator::NAME) {
if ($iMinAddressRank > 0) {
$aTerms[] = 'address_rank >= '.$iMinAddressRank;
}
if ($iMaxAddressRank < 30) {
$aTerms[] = 'address_rank <= '.$iMaxAddressRank;
$aTerms[] = "((address_rank between $iMinAddressRank and $iMaxAddressRank) or (search_rank between $iMinAddressRank and $iMaxAddressRank))";
}
}

80
lib/Shell.php Normal file
View File

@@ -0,0 +1,80 @@
<?php
namespace Nominatim;
class Shell
{
public function __construct($sBaseCmd, ...$aParams)
{
if (!$sBaseCmd) {
throw new Exception('Command missing in new() call');
}
$this->baseCmd = $sBaseCmd;
$this->aParams = array();
$this->aEnv = null; // null = use the same environment as the current PHP process
$this->stdoutString = null;
foreach ($aParams as $sParam) {
$this->addParams($sParam);
}
}
public function addParams(...$aParams)
{
foreach ($aParams as $sParam) {
if (isset($sParam) && $sParam !== null && $sParam !== '') {
array_push($this->aParams, $sParam);
}
}
return $this;
}
public function addEnvPair($sKey, $sVal)
{
if (isset($sKey) && $sKey && isset($sVal)) {
if (!isset($this->aEnv)) $this->aEnv = $_ENV;
$this->aEnv = array_merge($this->aEnv, array($sKey => $sVal), $_ENV);
}
return $this;
}
public function escapedCmd()
{
$aEscaped = array_map(function ($sParam) {
return $this->escapeParam($sParam);
}, array_merge(array($this->baseCmd), $this->aParams));
return join(' ', $aEscaped);
}
public function run()
{
$sCmd = $this->escapedCmd();
// $aEnv does not need escaping, proc_open seems to handle it fine
$aFDs = array(
0 => array('pipe', 'r'),
1 => STDOUT,
2 => STDERR
);
$aPipes = null;
$hProc = @proc_open($sCmd, $aFDs, $aPipes, null, $this->aEnv);
if (!is_resource($hProc)) {
throw new \Exception('Unable to run command: ' . $sCmd);
}
fclose($aPipes[0]); // no stdin
$iStat = proc_close($hProc);
return $iStat;
}
private function escapeParam($sParam)
{
if (preg_match('/^-*\w+$/', $sParam)) return $sParam;
return escapeshellarg($sParam);
}
}

View File

@@ -1,5 +1,6 @@
<?php
require_once(CONST_BasePath.'/lib/Shell.php');
function getCmdOpt($aArg, $aSpec, &$aResult, $bExitOnError = false, $bExitOnUnknown = false)
{
@@ -148,30 +149,33 @@ function runSQLScript($sScript, $bfatal = true, $bVerbose = false, $bIgnoreError
// Convert database DSN to psql parameters
$aDSNInfo = \Nominatim\DB::parseDSN(CONST_Database_DSN);
if (!isset($aDSNInfo['port']) || !$aDSNInfo['port']) $aDSNInfo['port'] = 5432;
$sCMD = 'psql -p '.$aDSNInfo['port'].' -d '.$aDSNInfo['database'];
$oCmd = new \Nominatim\Shell('psql');
$oCmd->addParams('--port', $aDSNInfo['port']);
$oCmd->addParams('--dbname', $aDSNInfo['database']);
if (isset($aDSNInfo['hostspec']) && $aDSNInfo['hostspec']) {
$sCMD .= ' -h ' . $aDSNInfo['hostspec'];
$oCmd->addParams('--host', $aDSNInfo['hostspec']);
}
if (isset($aDSNInfo['username']) && $aDSNInfo['username']) {
$sCMD .= ' -U ' . $aDSNInfo['username'];
$oCmd->addParams('--username', $aDSNInfo['username']);
}
$aProcEnv = null;
if (isset($aDSNInfo['password']) && $aDSNInfo['password']) {
$aProcEnv = array_merge(array('PGPASSWORD' => $aDSNInfo['password']), $_ENV);
if (isset($aDSNInfo['password'])) {
$oCmd->addEnvPair('PGPASSWORD', $aDSNInfo['password']);
}
if (!$bVerbose) {
$sCMD .= ' -q';
$oCmd->addParams('--quiet');
}
if ($bfatal && !$bIgnoreErrors) {
$sCMD .= ' -v ON_ERROR_STOP=1';
$oCmd->addParams('-v', 'ON_ERROR_STOP=1');
}
$aDescriptors = array(
0 => array('pipe', 'r'),
1 => STDOUT,
2 => STDERR
);
$ahPipes = null;
$hProcess = @proc_open($sCMD, $aDescriptors, $ahPipes, null, $aProcEnv);
$hProcess = @proc_open($oCmd->escapedCmd(), $aDescriptors, $ahPipes, null, $oCmd->aEnv);
if (!is_resource($hProcess)) {
fail('unable to start pgsql');
}
@@ -191,23 +195,3 @@ function runSQLScript($sScript, $bfatal = true, $bVerbose = false, $bIgnoreError
fail("pgsql returned with error code ($iReturn)");
}
}
function runWithEnv($sCmd, $aEnv)
{
$aFDs = array(
0 => array('pipe', 'r'),
1 => STDOUT,
2 => STDERR
);
$aPipes = null;
$hProc = @proc_open($sCmd, $aFDs, $aPipes, null, $aEnv);
if (!is_resource($hProc)) {
fail('unable to run command:' . $sCmd);
}
fclose($aPipes[0]); // no stdin
$iStat = proc_close($hProc);
return $iStat;
}

View File

@@ -2,8 +2,3 @@
require_once(CONST_BasePath.'/lib/lib.php');
require_once(CONST_BasePath.'/lib/DB.php');
if (get_magic_quotes_gpc()) {
echo "Please disable magic quotes in your php.ini configuration\n";
exit;
}

View File

@@ -95,8 +95,8 @@ function parseLatLon($sQuery)
$fQueryLat = null;
$fQueryLon = null;
if (preg_match('/\\s*([NS])[ ]+([0-9]+[0-9.]*)[° ]+([0-9.]+)?[\']*[, ]+([EW])[ ]+([0-9]+)[° ]+([0-9]+[0-9.]*)[\']*\\s*/', $sQuery, $aData)) {
/* 1 2 3 4 5 6
if (preg_match('/\\s*([NS])[\s]+([0-9]+[0-9.]*)[°\s]+([0-9.]+)?[\']*[,\s]+([EW])[\s]+([0-9]+)[°\s]+([0-9]+[0-9.]*)[\']*\\s*/', $sQuery, $aData)) {
/* 1 2 3 4 5 6
* degrees decimal minutes
* N 40 26.767, W 79 58.933
* N 40°26.767, W 79°58.933
@@ -104,8 +104,8 @@ function parseLatLon($sQuery)
$sFound = $aData[0];
$fQueryLat = ($aData[1]=='N'?1:-1) * ($aData[2] + $aData[3]/60);
$fQueryLon = ($aData[4]=='E'?1:-1) * ($aData[5] + $aData[6]/60);
} elseif (preg_match('/\\s*([0-9]+)[° ]+([0-9]+[0-9.]*)?[\']*[ ]+([NS])[, ]+([0-9]+)[° ]+([0-9]+[0-9.]*)?[\' ]+([EW])\\s*/', $sQuery, $aData)) {
/* 1 2 3 4 5 6
} elseif (preg_match('/\\s*([0-9]+)[°\s]+([0-9]+[0-9.]*)?[\']*[\s]+([NS])[,\s]+([0-9]+)[°\s]+([0-9]+[0-9.]*)?[\'\s]+([EW])\\s*/', $sQuery, $aData)) {
/* 1 2 3 4 5 6
* degrees decimal minutes
* 40 26.767 N, 79 58.933 W
* 40° 26.767 N 79° 58.933 W
@@ -113,8 +113,8 @@ function parseLatLon($sQuery)
$sFound = $aData[0];
$fQueryLat = ($aData[3]=='N'?1:-1) * ($aData[1] + $aData[2]/60);
$fQueryLon = ($aData[6]=='E'?1:-1) * ($aData[4] + $aData[5]/60);
} elseif (preg_match('/\\s*([NS])[ ]([0-9]+)[° ]+([0-9]+)[\' ]+([0-9]+)[″"]*[, ]+([EW])[ ]([0-9]+)[° ]+([0-9]+)[\' ]+([0-9]+)[″"]*\\s*/', $sQuery, $aData)) {
/* 1 2 3 4 5 6 7 8
} elseif (preg_match('/\\s*([NS])[\s]+([0-9]+)[°\s]+([0-9]+)[\'\s]+([0-9]+)[″"]*[,\s]+([EW])[\s]+([0-9]+)[°\s]+([0-9]+)[\'\s]+([0-9]+)[″"]*\\s*/', $sQuery, $aData)) {
/* 1 2 3 4 5 6 7 8
* degrees decimal seconds
* N 40 26 46 W 79 58 56
* N 40° 26 46″, W 79° 58 56″
@@ -122,8 +122,8 @@ function parseLatLon($sQuery)
$sFound = $aData[0];
$fQueryLat = ($aData[1]=='N'?1:-1) * ($aData[2] + $aData[3]/60 + $aData[4]/3600);
$fQueryLon = ($aData[5]=='E'?1:-1) * ($aData[6] + $aData[7]/60 + $aData[8]/3600);
} elseif (preg_match('/\\s*([0-9]+)[° ]+([0-9]+)[\' ]+([0-9]+[0-9.]*)[″" ]+([NS])[, ]+([0-9]+)[° ]+([0-9]+)[\' ]+([0-9]+[0-9.]*)[″" ]+([EW])\\s*/', $sQuery, $aData)) {
/* 1 2 3 4 5 6 7 8
} elseif (preg_match('/\\s*([0-9]+)[°\s]+([0-9]+)[\'\s]+([0-9]+[0-9.]*)[″"\s]+([NS])[,\s]+([0-9]+)[°\s]+([0-9]+)[\'\s]+([0-9]+[0-9.]*)[″"\s]+([EW])\\s*/', $sQuery, $aData)) {
/* 1 2 3 4 5 6 7 8
* degrees decimal seconds
* 40 26 46 N 79 58 56 W
* 40° 26 46″ N, 79° 58 56″ W
@@ -132,24 +132,24 @@ function parseLatLon($sQuery)
$sFound = $aData[0];
$fQueryLat = ($aData[4]=='N'?1:-1) * ($aData[1] + $aData[2]/60 + $aData[3]/3600);
$fQueryLon = ($aData[8]=='E'?1:-1) * ($aData[5] + $aData[6]/60 + $aData[7]/3600);
} elseif (preg_match('/\\s*([NS])[ ]([0-9]+[0-9]*\\.[0-9]+)[°]*[, ]+([EW])[ ]([0-9]+[0-9]*\\.[0-9]+)[°]*\\s*/', $sQuery, $aData)) {
/* 1 2 3 4
} elseif (preg_match('/\\s*([NS])[\s]+([0-9]+[0-9]*\\.[0-9]+)[°]*[,\s]+([EW])[\s]+([0-9]+[0-9]*\\.[0-9]+)[°]*\\s*/', $sQuery, $aData)) {
/* 1 2 3 4
* degrees decimal
* N 40.446° W 79.982°
*/
$sFound = $aData[0];
$fQueryLat = ($aData[1]=='N'?1:-1) * ($aData[2]);
$fQueryLon = ($aData[3]=='E'?1:-1) * ($aData[4]);
} elseif (preg_match('/\\s*([0-9]+[0-9]*\\.[0-9]+)[° ]+([NS])[, ]+([0-9]+[0-9]*\\.[0-9]+)[° ]+([EW])\\s*/', $sQuery, $aData)) {
/* 1 2 3 4
} elseif (preg_match('/\\s*([0-9]+[0-9]*\\.[0-9]+)[°\s]+([NS])[,\s]+([0-9]+[0-9]*\\.[0-9]+)[°\s]+([EW])\\s*/', $sQuery, $aData)) {
/* 1 2 3 4
* degrees decimal
* 40.446° N 79.982° W
*/
$sFound = $aData[0];
$fQueryLat = ($aData[2]=='N'?1:-1) * ($aData[1]);
$fQueryLon = ($aData[4]=='E'?1:-1) * ($aData[3]);
} elseif (preg_match('/(\\s*\\[|^\\s*|\\s*)(-?[0-9]+[0-9]*\\.[0-9]+)[, ]+(-?[0-9]+[0-9]*\\.[0-9]+)(\\]\\s*|\\s*$|\\s*)/', $sQuery, $aData)) {
/* 1 2 3 4
} elseif (preg_match('/(\\s*\\[|^\\s*|\\s*)(-?[0-9]+[0-9]*\\.[0-9]+)[,\s]+(-?[0-9]+[0-9]*\\.[0-9]+)(\\]\\s*|\\s*$|\\s*)/', $sQuery, $aData)) {
/* 1 2 3 4
* degrees decimal
* 12.34, 56.78
* 12.34 56.78
@@ -165,39 +165,6 @@ function parseLatLon($sQuery)
return array($sFound, $fQueryLat, $fQueryLon);
}
function geometryText2Points($geometry_as_text, $fRadius)
{
$aPolyPoints = null;
if (preg_match('#POLYGON\\(\\(([- 0-9.,]+)#', $geometry_as_text, $aMatch)) {
//
preg_match_all('/(-?[0-9.]+) (-?[0-9.]+)/', $aMatch[1], $aPolyPoints, PREG_SET_ORDER);
//
} elseif (preg_match('#LINESTRING\\(([- 0-9.,]+)#', $geometry_as_text, $aMatch)) {
//
preg_match_all('/(-?[0-9.]+) (-?[0-9.]+)/', $aMatch[1], $aPolyPoints, PREG_SET_ORDER);
//
} elseif (preg_match('#MULTIPOLYGON\\(\\(\\(([- 0-9.,]+)#', $geometry_as_text, $aMatch)) {
//
preg_match_all('/(-?[0-9.]+) (-?[0-9.]+)/', $aMatch[1], $aPolyPoints, PREG_SET_ORDER);
//
} elseif (preg_match('#POINT\\((-?[0-9.]+) (-?[0-9.]+)\\)#', $geometry_as_text, $aMatch)) {
//
$aPolyPoints = createPointsAroundCenter($aMatch[1], $aMatch[2], $fRadius);
//
}
if (isset($aPolyPoints)) {
$aResultPoints = array();
foreach ($aPolyPoints as $aPoint) {
$aResultPoints[] = array($aPoint[1], $aPoint[2]);
}
return $aResultPoints;
}
return;
}
function createPointsAroundCenter($fLon, $fLat, $fRadius)
{
$iSteps = max(8, min(100, ($fRadius * 40000)^2));

View File

@@ -12,6 +12,8 @@ function formatOSMType($sType, $bIncludeExternal = true)
if ($sType == 'T') return 'way';
if ($sType == 'I') return 'way';
// not handled: P, L
return '';
}
@@ -33,20 +35,39 @@ function wikipediaLink($aFeature)
return '';
}
function detailsLink($aFeature, $sTitle = false)
function detailsLink($aFeature, $sTitle = false, $sExtraProperties = false)
{
if (!$aFeature['place_id']) return '';
return '<a href="details.php?place_id='.$aFeature['place_id'].'">'.($sTitle?$sTitle:$aFeature['place_id']).'</a>';
$sHtml = '<a ';
if ($sExtraProperties) {
$sHtml .= $sExtraProperties.' ';
}
$sHtml .= 'href="details.php?place_id='.$aFeature['place_id'].'">'.($sTitle?$sTitle:$aFeature['place_id']).'</a>';
return $sHtml;
}
function detailsPermaLink($aFeature, $sRefText = false)
function detailsPermaLink($aFeature, $sRefText = false, $sExtraProperties = false)
{
$sOSMType = formatOSMType($aFeature['osm_type'], false);
if ($sOSMType) {
$sLabel = $sRefText ? $sRefText : $sOSMType.' '.$aFeature['osm_id'];
return '<a href="details.php?osmtype='.$aFeature['osm_type'].'&osmid='.$aFeature['osm_id'].'&class='.$aFeature['class'].'">'.$sLabel.'</a>';
$sHtml = '<a ';
if ($sExtraProperties) {
$sHtml .= $sExtraProperties.' ';
}
$sHtml .= 'href="details.php?osmtype='.$aFeature['osm_type']
.'&osmid='.$aFeature['osm_id'].'&class='.$aFeature['class'].'">';
if ($sRefText) {
$sHtml .= $sRefText.'</a>';
} else {
$sHtml .= $sOSMType.' '.$aFeature['osm_id'].'</a>';
}
return $sHtml;
}
return '';
return detailsLink($aFeature, $sRefText, $sExtraProperties);
}

View File

@@ -3,6 +3,7 @@
namespace Nominatim\Setup;
require_once(CONST_BasePath.'/lib/setup/AddressLevelParser.php');
require_once(CONST_BasePath.'/lib/Shell.php');
class SetupFunctions
{
@@ -10,11 +11,13 @@ class SetupFunctions
protected $iInstances;
protected $sModulePath;
protected $aDSNInfo;
protected $bQuiet;
protected $bVerbose;
protected $sIgnoreErrors;
protected $bEnableDiffUpdates;
protected $bEnableDebugStatements;
protected $bNoPartitions;
protected $bDrop;
protected $oDB = null;
public function __construct(array $aCMDResult)
@@ -29,10 +32,13 @@ class SetupFunctions
warn('resetting threads to '.$this->iInstances);
}
// Assume we can steal all the cache memory in the box (unless told otherwise)
if (isset($aCMDResult['osm2pgsql-cache'])) {
$this->iCacheMemory = $aCMDResult['osm2pgsql-cache'];
} elseif (!is_null(CONST_Osm2pgsql_Flatnode_File)) {
// When flatnode files are enabled then disable cache per default.
$this->iCacheMemory = 0;
} else {
// Otherwise: Assume we can steal all the cache memory in the box.
$this->iCacheMemory = getCacheMemoryMB();
}
@@ -46,6 +52,7 @@ class SetupFunctions
}
// setting member variables based on command line options stored in $aCMDResult
$this->bQuiet = isset($aCMDResult['quiet']) && $aCMDResult['quiet'];
$this->bVerbose = $aCMDResult['verbose'];
//setting default values which are not set by the update.php array
@@ -69,6 +76,8 @@ class SetupFunctions
} else {
$this->bEnableDiffUpdates = false;
}
$this->bDrop = isset($aCMDResult['drop']) && $aCMDResult['drop'];
}
public function createDB()
@@ -76,21 +85,27 @@ class SetupFunctions
info('Create DB');
$oDB = new \Nominatim\DB;
if ($oDB->databaseExists()) {
if ($oDB->checkConnection()) {
fail('database already exists ('.CONST_Database_DSN.')');
}
$sCreateDBCmd = 'createdb -E UTF-8 -p '.$this->aDSNInfo['port'].' '.$this->aDSNInfo['database'];
$oCmd = (new \Nominatim\Shell('createdb'))
->addParams('-E', 'UTF-8')
->addParams('-p', $this->aDSNInfo['port']);
if (isset($this->aDSNInfo['username'])) {
$sCreateDBCmd .= ' -U '.$this->aDSNInfo['username'];
$oCmd->addParams('-U', $this->aDSNInfo['username']);
}
if (isset($this->aDSNInfo['password'])) {
$oCmd->addEnvPair('PGPASSWORD', $this->aDSNInfo['password']);
}
if (isset($this->aDSNInfo['hostspec'])) {
$sCreateDBCmd .= ' -h '.$this->aDSNInfo['hostspec'];
$oCmd->addParams('-h', $this->aDSNInfo['hostspec']);
}
$oCmd->addParams($this->aDSNInfo['database']);
$result = $this->runWithPgEnv($sCreateDBCmd);
if ($result != 0) fail('Error executing external command: '.$sCreateDBCmd);
$result = $oCmd->run();
if ($result != 0) fail('Error executing external command: '.$oCmd->escapedCmd());
}
public function connect()
@@ -137,7 +152,7 @@ class SetupFunctions
exit(1);
}
$this->pgsqlRunScriptFile(CONST_BasePath.'/data/country_name.sql');
$this->pgsqlRunScriptFile(CONST_BasePath.'/data/country_osm_grid.sql.gz');
$this->pgsqlRunScriptFile(CONST_ExtraDataPath.'/country_osm_grid.sql.gz');
$this->pgsqlRunScriptFile(CONST_BasePath.'/data/gb_postcode_table.sql');
$this->pgsqlRunScriptFile(CONST_BasePath.'/data/us_postcode_table.sql');
@@ -158,56 +173,64 @@ class SetupFunctions
if ($this->bNoPartitions) {
$this->pgsqlRunScript('update country_name set partition = 0');
}
// the following will be needed by createFunctions later but
// is only defined in the subsequently called createTables
// Create dummies here that will be overwritten by the proper
// versions in create-tables.
$this->pgsqlRunScript('CREATE TABLE IF NOT EXISTS place_boundingbox ()');
$this->pgsqlRunScript('CREATE TYPE wikipedia_article_match AS ()', false);
}
public function importData($sOSMFile)
{
info('Import data');
$osm2pgsql = CONST_Osm2pgsql_Binary;
if (!file_exists($osm2pgsql)) {
if (!file_exists(CONST_Osm2pgsql_Binary)) {
echo "Check CONST_Osm2pgsql_Binary in your local settings file.\n";
echo "Normally you should not need to set this manually.\n";
fail("osm2pgsql not found in '$osm2pgsql'");
fail("osm2pgsql not found in '".CONST_Osm2pgsql_Binary."'");
}
$osm2pgsql .= ' -S '.CONST_Import_Style;
$oCmd = new \Nominatim\Shell(CONST_Osm2pgsql_Binary);
$oCmd->addParams('--style', CONST_Import_Style);
if (!is_null(CONST_Osm2pgsql_Flatnode_File) && CONST_Osm2pgsql_Flatnode_File) {
$osm2pgsql .= ' --flat-nodes '.CONST_Osm2pgsql_Flatnode_File;
$oCmd->addParams('--flat-nodes', CONST_Osm2pgsql_Flatnode_File);
}
if (CONST_Tablespace_Osm2pgsql_Data) {
$oCmd->addParams('--tablespace-slim-data', CONST_Tablespace_Osm2pgsql_Data);
}
if (CONST_Tablespace_Osm2pgsql_Index) {
$oCmd->addParams('--tablespace-slim-index', CONST_Tablespace_Osm2pgsql_Index);
}
if (CONST_Tablespace_Place_Data) {
$oCmd->addParams('--tablespace-main-data', CONST_Tablespace_Place_Data);
}
if (CONST_Tablespace_Place_Index) {
$oCmd->addParams('--tablespace-main-index', CONST_Tablespace_Place_Index);
}
$oCmd->addParams('--latlong', '--slim', '--create');
$oCmd->addParams('--output', 'gazetteer');
$oCmd->addParams('--hstore');
$oCmd->addParams('--number-processes', 1);
$oCmd->addParams('--cache', $this->iCacheMemory);
$oCmd->addParams('--port', $this->aDSNInfo['port']);
if (CONST_Tablespace_Osm2pgsql_Data)
$osm2pgsql .= ' --tablespace-slim-data '.CONST_Tablespace_Osm2pgsql_Data;
if (CONST_Tablespace_Osm2pgsql_Index)
$osm2pgsql .= ' --tablespace-slim-index '.CONST_Tablespace_Osm2pgsql_Index;
if (CONST_Tablespace_Place_Data)
$osm2pgsql .= ' --tablespace-main-data '.CONST_Tablespace_Place_Data;
if (CONST_Tablespace_Place_Index)
$osm2pgsql .= ' --tablespace-main-index '.CONST_Tablespace_Place_Index;
$osm2pgsql .= ' -lsc -O gazetteer --hstore --number-processes 1';
$osm2pgsql .= ' -C '.$this->iCacheMemory;
$osm2pgsql .= ' -P '.$this->aDSNInfo['port'];
if (isset($this->aDSNInfo['username'])) {
$osm2pgsql .= ' -U '.$this->aDSNInfo['username'];
$oCmd->addParams('--username', $this->aDSNInfo['username']);
}
if (isset($this->aDSNInfo['password'])) {
$oCmd->addEnvPair('PGPASSWORD', $this->aDSNInfo['password']);
}
if (isset($this->aDSNInfo['hostspec'])) {
$osm2pgsql .= ' -H '.$this->aDSNInfo['hostspec'];
$oCmd->addParams('--host', $this->aDSNInfo['hostspec']);
}
$osm2pgsql .= ' -d '.$this->aDSNInfo['database'].' '.$sOSMFile;
$this->runWithPgEnv($osm2pgsql);
$oCmd->addParams('--database', $this->aDSNInfo['database']);
$oCmd->addParams($sOSMFile);
$oCmd->run();
if (!$this->sIgnoreErrors && !$this->oDB->getRow('select * from place limit 1')) {
fail('No Data');
}
if ($this->bDrop) {
$this->dropTable('planet_osm_nodes');
$this->removeFlatnodeFile();
}
}
public function createFunctions()
@@ -225,88 +248,34 @@ class SetupFunctions
info('Create Tables');
$sTemplate = file_get_contents(CONST_BasePath.'/sql/tables.sql');
$sTemplate = str_replace('{www-user}', CONST_Database_Web_User, $sTemplate);
$sTemplate = $this->replaceTablespace(
'{ts:address-data}',
CONST_Tablespace_Address_Data,
$sTemplate
);
$sTemplate = $this->replaceTablespace(
'{ts:address-index}',
CONST_Tablespace_Address_Index,
$sTemplate
);
$sTemplate = $this->replaceTablespace(
'{ts:search-data}',
CONST_Tablespace_Search_Data,
$sTemplate
);
$sTemplate = $this->replaceTablespace(
'{ts:search-index}',
CONST_Tablespace_Search_Index,
$sTemplate
);
$sTemplate = $this->replaceTablespace(
'{ts:aux-data}',
CONST_Tablespace_Aux_Data,
$sTemplate
);
$sTemplate = $this->replaceTablespace(
'{ts:aux-index}',
CONST_Tablespace_Aux_Index,
$sTemplate
);
$sTemplate = $this->replaceSqlPatterns($sTemplate);
$this->pgsqlRunScript($sTemplate, false);
if ($bReverseOnly) {
$this->pgExec('DROP TABLE search_name');
$this->dropTable('search_name');
}
$oAlParser = new AddressLevelParser(CONST_Address_Level_Config);
$oAlParser->createTable($this->oDB, 'address_levels');
}
public function createTableTriggers()
{
info('Create Tables');
$sTemplate = file_get_contents(CONST_BasePath.'/sql/table-triggers.sql');
$sTemplate = $this->replaceSqlPatterns($sTemplate);
$this->pgsqlRunScript($sTemplate, false);
}
public function createPartitionTables()
{
info('Create Partition Tables');
$sTemplate = file_get_contents(CONST_BasePath.'/sql/partition-tables.src.sql');
$sTemplate = $this->replaceTablespace(
'{ts:address-data}',
CONST_Tablespace_Address_Data,
$sTemplate
);
$sTemplate = $this->replaceTablespace(
'{ts:address-index}',
CONST_Tablespace_Address_Index,
$sTemplate
);
$sTemplate = $this->replaceTablespace(
'{ts:search-data}',
CONST_Tablespace_Search_Data,
$sTemplate
);
$sTemplate = $this->replaceTablespace(
'{ts:search-index}',
CONST_Tablespace_Search_Index,
$sTemplate
);
$sTemplate = $this->replaceTablespace(
'{ts:aux-data}',
CONST_Tablespace_Aux_Data,
$sTemplate
);
$sTemplate = $this->replaceTablespace(
'{ts:aux-index}',
CONST_Tablespace_Aux_Index,
$sTemplate
);
$sTemplate = $this->replaceSqlPatterns($sTemplate);
$this->pgsqlRunPartitionScript($sTemplate);
}
@@ -321,19 +290,14 @@ class SetupFunctions
public function importWikipediaArticles()
{
$sWikiArticlesFile = CONST_Wikipedia_Data_Path.'/wikipedia_article.sql.bin';
$sWikiRedirectsFile = CONST_Wikipedia_Data_Path.'/wikipedia_redirect.sql.bin';
$sWikiArticlesFile = CONST_Wikipedia_Data_Path.'/wikimedia-importance.sql.gz';
if (file_exists($sWikiArticlesFile)) {
info('Importing wikipedia articles');
$this->pgsqlRunDropAndRestore($sWikiArticlesFile);
info('Importing wikipedia articles and redirects');
$this->dropTable('wikipedia_article');
$this->dropTable('wikipedia_redirect');
$this->pgsqlRunScriptFile($sWikiArticlesFile);
} else {
warn('wikipedia article dump file not found - places will have default importance');
}
if (file_exists($sWikiRedirectsFile)) {
info('Importing wikipedia redirects');
$this->pgsqlRunDropAndRestore($sWikiRedirectsFile);
} else {
warn('wikipedia redirect dump file not found - some place importance values may be missing');
warn('wikipedia importance dump file not found - places will have default importance');
}
}
@@ -341,27 +305,25 @@ class SetupFunctions
{
info('Drop old Data');
$this->pgExec('TRUNCATE word');
$this->oDB->exec('TRUNCATE word');
echo '.';
$this->pgExec('TRUNCATE placex');
$this->oDB->exec('TRUNCATE placex');
echo '.';
$this->pgExec('TRUNCATE location_property_osmline');
$this->oDB->exec('TRUNCATE location_property_osmline');
echo '.';
$this->pgExec('TRUNCATE place_addressline');
$this->oDB->exec('TRUNCATE place_addressline');
echo '.';
$this->pgExec('TRUNCATE place_boundingbox');
echo '.';
$this->pgExec('TRUNCATE location_area');
$this->oDB->exec('TRUNCATE location_area');
echo '.';
if (!$this->dbReverseOnly()) {
$this->pgExec('TRUNCATE search_name');
$this->oDB->exec('TRUNCATE search_name');
echo '.';
}
$this->pgExec('TRUNCATE search_name_blank');
$this->oDB->exec('TRUNCATE search_name_blank');
echo '.';
$this->pgExec('DROP SEQUENCE seq_place');
$this->oDB->exec('DROP SEQUENCE seq_place');
echo '.';
$this->pgExec('CREATE SEQUENCE seq_place start 100000');
$this->oDB->exec('CREATE SEQUENCE seq_place start 100000');
echo '.';
$sSQL = 'select distinct partition from country_name';
@@ -369,14 +331,14 @@ class SetupFunctions
if (!$this->bNoPartitions) $aPartitions[] = 0;
foreach ($aPartitions as $sPartition) {
$this->pgExec('TRUNCATE location_road_'.$sPartition);
$this->oDB->exec('TRUNCATE location_road_'.$sPartition);
echo '.';
}
// used by getorcreate_word_id to ignore frequent partial words
$sSQL = 'CREATE OR REPLACE FUNCTION get_maxwordfreq() RETURNS integer AS ';
$sSQL .= '$$ SELECT '.CONST_Max_Word_Frequency.' as maxwordfreq; $$ LANGUAGE SQL IMMUTABLE';
$this->pgExec($sSQL);
$this->oDB->exec($sSQL);
echo ".\n";
// pre-create the word list
@@ -468,18 +430,15 @@ class SetupFunctions
{
info('Import Tiger data');
$aFilenames = glob(CONST_Tiger_Data_Path.'/*.sql');
info('Found '.count($aFilenames).' SQL files in path '.CONST_Tiger_Data_Path);
if (empty($aFilenames)) {
warn('Tiger data import selected but no files found in path '.CONST_Tiger_Data_Path);
return;
}
$sTemplate = file_get_contents(CONST_BasePath.'/sql/tiger_import_start.sql');
$sTemplate = str_replace('{www-user}', CONST_Database_Web_User, $sTemplate);
$sTemplate = $this->replaceTablespace(
'{ts:aux-data}',
CONST_Tablespace_Aux_Data,
$sTemplate
);
$sTemplate = $this->replaceTablespace(
'{ts:aux-index}',
CONST_Tablespace_Aux_Index,
$sTemplate
);
$sTemplate = $this->replaceSqlPatterns($sTemplate);
$this->pgsqlRunScript($sTemplate, false);
$aDBInstances = array();
@@ -492,7 +451,7 @@ class SetupFunctions
pg_ping($aDBInstances[$i]);
}
foreach (glob(CONST_Tiger_Data_Path.'/*.sql') as $sFile) {
foreach ($aFilenames as $sFile) {
echo $sFile.': ';
$hFile = fopen($sFile, 'r');
$sSQL = fgets($hFile, 100000);
@@ -532,24 +491,15 @@ class SetupFunctions
info('Creating indexes on Tiger data');
$sTemplate = file_get_contents(CONST_BasePath.'/sql/tiger_import_finish.sql');
$sTemplate = str_replace('{www-user}', CONST_Database_Web_User, $sTemplate);
$sTemplate = $this->replaceTablespace(
'{ts:aux-data}',
CONST_Tablespace_Aux_Data,
$sTemplate
);
$sTemplate = $this->replaceTablespace(
'{ts:aux-index}',
CONST_Tablespace_Aux_Index,
$sTemplate
);
$sTemplate = $this->replaceSqlPatterns($sTemplate);
$this->pgsqlRunScript($sTemplate, false);
}
public function calculatePostcodes($bCMDResultAll)
{
info('Calculate Postcodes');
$this->pgExec('TRUNCATE location_postcode');
$this->oDB->exec('TRUNCATE location_postcode');
$sSQL = 'INSERT INTO location_postcode';
$sSQL .= ' (place_id, indexed_status, country_code, postcode, geometry) ';
@@ -560,7 +510,7 @@ class SetupFunctions
$sSQL .= " WHERE address ? 'postcode' AND address->'postcode' NOT SIMILAR TO '%(,|;)%'";
$sSQL .= ' AND geometry IS NOT null';
$sSQL .= ' GROUP BY country_code, pc';
$this->pgExec($sSQL);
$this->oDB->exec($sSQL);
// only add postcodes that are not yet available in OSM
$sSQL = 'INSERT INTO location_postcode';
@@ -570,7 +520,7 @@ class SetupFunctions
$sSQL .= ' FROM us_postcode WHERE postcode NOT IN';
$sSQL .= ' (SELECT postcode FROM location_postcode';
$sSQL .= " WHERE country_code = 'us')";
$this->pgExec($sSQL);
$this->oDB->exec($sSQL);
// add missing postcodes for GB (if available)
$sSQL = 'INSERT INTO location_postcode';
@@ -579,80 +529,94 @@ class SetupFunctions
$sSQL .= ' FROM gb_postcode WHERE postcode NOT IN';
$sSQL .= ' (SELECT postcode FROM location_postcode';
$sSQL .= " WHERE country_code = 'gb')";
$this->pgExec($sSQL);
$this->oDB->exec($sSQL);
if (!$bCMDResultAll) {
$sSQL = "DELETE FROM word WHERE class='place' and type='postcode'";
$sSQL .= 'and word NOT IN (SELECT postcode FROM location_postcode)';
$this->pgExec($sSQL);
$this->oDB->exec($sSQL);
}
$sSQL = 'SELECT count(getorcreate_postcode_id(v)) FROM ';
$sSQL .= '(SELECT distinct(postcode) as v FROM location_postcode) p';
$this->pgExec($sSQL);
$this->oDB->exec($sSQL);
}
public function index($bIndexNoanalyse)
{
$sOutputFile = '';
$sBaseCmd = CONST_InstallPath.'/nominatim/nominatim -i -d '.$this->aDSNInfo['database'].' -P '
.$this->aDSNInfo['port'].' -t '.$this->iInstances.$sOutputFile;
$oBaseCmd = (new \Nominatim\Shell(CONST_BasePath.'/nominatim/nominatim.py'))
->addParams('--database', $this->aDSNInfo['database'])
->addParams('--port', $this->aDSNInfo['port'])
->addParams('--threads', $this->iInstances);
if (!$this->bQuiet) {
$oBaseCmd->addParams('-v');
}
if ($this->bVerbose) {
$oBaseCmd->addParams('-v');
}
if (isset($this->aDSNInfo['hostspec'])) {
$sBaseCmd .= ' -H '.$this->aDSNInfo['hostspec'];
$oBaseCmd->addParams('--host', $this->aDSNInfo['hostspec']);
}
if (isset($this->aDSNInfo['username'])) {
$sBaseCmd .= ' -U '.$this->aDSNInfo['username'];
$oBaseCmd->addParams('--user', $this->aDSNInfo['username']);
}
if (isset($this->aDSNInfo['password'])) {
$oBaseCmd->addEnvPair('PGPASSWORD', $this->aDSNInfo['password']);
}
info('Index ranks 0 - 4');
$iStatus = $this->runWithPgEnv($sBaseCmd.' -R 4');
$oCmd = (clone $oBaseCmd)->addParams('--maxrank', 4);
echo $oCmd->escapedCmd();
$iStatus = $oCmd->run();
if ($iStatus != 0) {
fail('error status ' . $iStatus . ' running nominatim!');
}
if (!$bIndexNoanalyse) $this->pgsqlRunScript('ANALYSE');
info('Index ranks 5 - 25');
$iStatus = $this->runWithPgEnv($sBaseCmd.' -r 5 -R 25');
$oCmd = (clone $oBaseCmd)->addParams('--minrank', 5, '--maxrank', 25);
$iStatus = $oCmd->run();
if ($iStatus != 0) {
fail('error status ' . $iStatus . ' running nominatim!');
}
if (!$bIndexNoanalyse) $this->pgsqlRunScript('ANALYSE');
info('Index ranks 26 - 30');
$iStatus = $this->runWithPgEnv($sBaseCmd.' -r 26');
$oCmd = (clone $oBaseCmd)->addParams('--minrank', 26);
$iStatus = $oCmd->run();
if ($iStatus != 0) {
fail('error status ' . $iStatus . ' running nominatim!');
}
info('Index postcodes');
$sSQL = 'UPDATE location_postcode SET indexed_status = 0';
$this->pgExec($sSQL);
$this->oDB->exec($sSQL);
}
public function createSearchIndices()
{
info('Create Search indices');
$sSQL = 'SELECT relname FROM pg_class, pg_index ';
$sSQL .= 'WHERE pg_index.indisvalid = false AND pg_index.indexrelid = pg_class.oid';
$aInvalidIndices = $this->oDB->getCol($sSQL);
foreach ($aInvalidIndices as $sIndexName) {
info("Cleaning up invalid index $sIndexName");
$this->oDB->exec("DROP INDEX $sIndexName;");
}
$sTemplate = file_get_contents(CONST_BasePath.'/sql/indices.src.sql');
if (!$this->bDrop) {
$sTemplate .= file_get_contents(CONST_BasePath.'/sql/indices_updates.src.sql');
}
if (!$this->dbReverseOnly()) {
$sTemplate .= file_get_contents(CONST_BasePath.'/sql/indices_search.src.sql');
}
$sTemplate = str_replace('{www-user}', CONST_Database_Web_User, $sTemplate);
$sTemplate = $this->replaceTablespace(
'{ts:address-index}',
CONST_Tablespace_Address_Index,
$sTemplate
);
$sTemplate = $this->replaceTablespace(
'{ts:search-index}',
CONST_Tablespace_Search_Index,
$sTemplate
);
$sTemplate = $this->replaceTablespace(
'{ts:aux-index}',
CONST_Tablespace_Aux_Index,
$sTemplate
);
$sTemplate = $this->replaceSqlPatterns($sTemplate);
$this->pgsqlRunScript($sTemplate);
}
@@ -711,7 +675,7 @@ class SetupFunctions
);
$aDropTables = array();
$aHaveTables = $this->oDB->getCol("SELECT tablename FROM pg_tables WHERE schemaname='public'");
$aHaveTables = $this->oDB->getListOfTables();
foreach ($aHaveTables as $sTable) {
$bFound = false;
@@ -724,10 +688,14 @@ class SetupFunctions
if (!$bFound) array_push($aDropTables, $sTable);
}
foreach ($aDropTables as $sDrop) {
if ($this->bVerbose) echo "Dropping table $sDrop\n";
$this->oDB->exec("DROP TABLE IF EXISTS $sDrop CASCADE");
$this->dropTable($sDrop);
}
$this->removeFlatnodeFile();
}
private function removeFlatnodeFile()
{
if (!is_null(CONST_Osm2pgsql_Flatnode_File) && CONST_Osm2pgsql_Flatnode_File) {
if (file_exists(CONST_Osm2pgsql_Flatnode_File)) {
if ($this->bVerbose) echo 'Deleting '.CONST_Osm2pgsql_Flatnode_File."\n";
@@ -736,22 +704,6 @@ class SetupFunctions
}
}
private function pgsqlRunDropAndRestore($sDumpFile)
{
$sCMD = 'pg_restore -p '.$this->aDSNInfo['port'].' -d '.$this->aDSNInfo['database'].' --no-owner -Fc --clean '.$sDumpFile;
if ($this->oDB->getPostgresVersion() >= 9.04) {
$sCMD .= ' --if-exists';
}
if (isset($this->aDSNInfo['hostspec'])) {
$sCMD .= ' -h '.$this->aDSNInfo['hostspec'];
}
if (isset($this->aDSNInfo['username'])) {
$sCMD .= ' -U '.$this->aDSNInfo['username'];
}
$this->runWithPgEnv($sCMD);
}
private function pgsqlRunScript($sScript, $bfatal = true)
{
runSQLScript(
@@ -764,7 +716,22 @@ class SetupFunctions
private function createSqlFunctions()
{
$sTemplate = file_get_contents(CONST_BasePath.'/sql/functions.sql');
$sBasePath = CONST_BasePath.'/sql/functions/';
$sTemplate = file_get_contents($sBasePath.'utils.sql');
$sTemplate .= file_get_contents($sBasePath.'normalization.sql');
$sTemplate .= file_get_contents($sBasePath.'ranking.sql');
$sTemplate .= file_get_contents($sBasePath.'importance.sql');
$sTemplate .= file_get_contents($sBasePath.'address_lookup.sql');
$sTemplate .= file_get_contents($sBasePath.'interpolation.sql');
if ($this->oDB->tableExists('place')) {
$sTemplate .= file_get_contents($sBasePath.'place_triggers.sql');
}
if ($this->oDB->tableExists('placex')) {
$sTemplate .= file_get_contents($sBasePath.'placex_triggers.sql');
}
if ($this->oDB->tableExists('location_postcode')) {
$sTemplate .= file_get_contents($sBasePath.'postcode_triggers.sql');
}
$sTemplate = str_replace('{modulepath}', $this->sModulePath, $sTemplate);
if ($this->bEnableDiffUpdates) {
$sTemplate = str_replace('RETURN NEW; -- %DIFFUPDATES%', '--', $sTemplate);
@@ -810,19 +777,21 @@ class SetupFunctions
{
if (!file_exists($sFilename)) fail('unable to find '.$sFilename);
$sCMD = 'psql -p '.$this->aDSNInfo['port'].' -d '.$this->aDSNInfo['database'];
$oCmd = (new \Nominatim\Shell('psql'))
->addParams('--port', $this->aDSNInfo['port'])
->addParams('--dbname', $this->aDSNInfo['database']);
if (!$this->bVerbose) {
$sCMD .= ' -q';
$oCmd->addParams('--quiet');
}
if (isset($this->aDSNInfo['hostspec'])) {
$sCMD .= ' -h '.$this->aDSNInfo['hostspec'];
$oCmd->addParams('--host', $this->aDSNInfo['hostspec']);
}
if (isset($this->aDSNInfo['username'])) {
$sCMD .= ' -U '.$this->aDSNInfo['username'];
$oCmd->addParams('--username', $this->aDSNInfo['username']);
}
$aProcEnv = null;
if (isset($this->aDSNInfo['password'])) {
$aProcEnv = array_merge(array('PGPASSWORD' => $this->aDSNInfo['password']), $_ENV);
$oCmd->addEnvPair('PGPASSWORD', $this->aDSNInfo['password']);
}
$ahGzipPipes = null;
if (preg_match('/\\.gz$/', $sFilename)) {
@@ -831,12 +800,14 @@ class SetupFunctions
1 => array('pipe', 'w'),
2 => array('file', '/dev/null', 'a')
);
$hGzipProcess = proc_open('zcat '.$sFilename, $aDescriptors, $ahGzipPipes);
$oZcatCmd = new \Nominatim\Shell('zcat', $sFilename);
$hGzipProcess = proc_open($oZcatCmd->escapedCmd(), $aDescriptors, $ahGzipPipes);
if (!is_resource($hGzipProcess)) fail('unable to start zcat');
$aReadPipe = $ahGzipPipes[1];
fclose($ahGzipPipes[0]);
} else {
$sCMD .= ' -f '.$sFilename;
$oCmd->addParams('--file', $sFilename);
$aReadPipe = array('pipe', 'r');
}
$aDescriptors = array(
@@ -845,7 +816,8 @@ class SetupFunctions
2 => array('file', '/dev/null', 'a')
);
$ahPipes = null;
$hProcess = proc_open($sCMD, $aDescriptors, $ahPipes, null, $aProcEnv);
$hProcess = proc_open($oCmd->escapedCmd(), $aDescriptors, $ahPipes, null, $oCmd->aEnv);
if (!is_resource($hProcess)) fail('unable to start pgsql');
// TODO: error checking
while (!feof($ahPipes[1])) {
@@ -862,43 +834,43 @@ class SetupFunctions
}
}
private function replaceTablespace($sTemplate, $sTablespace, $sSql)
private function replaceSqlPatterns($sSql)
{
if ($sTablespace) {
$sSql = str_replace($sTemplate, 'TABLESPACE "'.$sTablespace.'"', $sSql);
} else {
$sSql = str_replace($sTemplate, '', $sSql);
$sSql = str_replace('{www-user}', CONST_Database_Web_User, $sSql);
$aPatterns = array(
'{ts:address-data}' => CONST_Tablespace_Address_Data,
'{ts:address-index}' => CONST_Tablespace_Address_Index,
'{ts:search-data}' => CONST_Tablespace_Search_Data,
'{ts:search-index}' => CONST_Tablespace_Search_Index,
'{ts:aux-data}' => CONST_Tablespace_Aux_Data,
'{ts:aux-index}' => CONST_Tablespace_Aux_Index,
);
foreach ($aPatterns as $sPattern => $sTablespace) {
if ($sTablespace) {
$sSql = str_replace($sPattern, 'TABLESPACE "'.$sTablespace.'"', $sSql);
} else {
$sSql = str_replace($sPattern, '', $sSql);
}
}
return $sSql;
}
private function runWithPgEnv($sCmd)
{
if ($this->bVerbose) {
echo "Execute: $sCmd\n";
}
$aProcEnv = null;
if (isset($this->aDSNInfo['password'])) {
$aProcEnv = array_merge(array('PGPASSWORD' => $this->aDSNInfo['password']), $_ENV);
}
return runWithEnv($sCmd, $aProcEnv);
}
/**
* Execute the SQL command on the open database.
* Drop table with the given name if it exists.
*
* @param string $sSQL SQL command to execute.
* @param string $sName Name of table to remove.
*
* @return null
*
* @pre connect() must have been called.
*/
private function pgExec($sSQL)
private function dropTable($sName)
{
$this->oDB->exec($sSQL);
if ($this->bVerbose) echo "Dropping table $sName\n";
$this->oDB->deleteTable($sName);
}
/**

View File

@@ -30,27 +30,14 @@ if (empty($aPlace)) {
$aFilteredPlaces['properties']['geocoding']['label'] = $aPlace['langaddress'];
$aFilteredPlaces['properties']['geocoding']['name'] = $aPlace['placename'];
if ($aPlace['placename'] !== null) {
$aFilteredPlaces['properties']['geocoding']['name'] = $aPlace['placename'];
}
if (isset($aPlace['address'])) {
$aFieldMappings = array(
'house_number' => 'housenumber',
'road' => 'street',
'locality' => 'locality',
'postcode' => 'postcode',
'city' => 'city',
'district' => 'district',
'county' => 'county',
'state' => 'state',
'country' => 'country'
);
$aAddressNames = $aPlace['address']->getAddressNames();
foreach ($aFieldMappings as $sFrom => $sTo) {
if (isset($aAddressNames[$sFrom])) {
$aFilteredPlaces['properties']['geocoding'][$sTo] = $aAddressNames[$sFrom];
}
}
$aPlace['address']->addGeocodeJsonAddressParts(
$aFilteredPlaces['properties']['geocoding']
);
$aFilteredPlaces['properties']['geocoding']['admin']
= $aPlace['address']->getAdminLevels();

View File

@@ -9,7 +9,7 @@
<body id="reverse-page">
<?php include(CONST_BasePath.'/lib/template/includes/html-top-navigation.php'); ?>
<div class="top-bar">
<form class="form-inline" role="search" accept-charset="UTF-8" action="<?php echo CONST_Website_BaseURL; ?>reverse.php">
<div class="form-group">
<input name="format" type="hidden" value="html">
@@ -64,7 +64,7 @@
<a href="<?php echo CONST_Website_BaseURL; ?>search.php">forward search</a>
</div>
</form>
</div>
<div id="content">
@@ -85,7 +85,7 @@
else
echo ' <span class="type">('.ucwords(str_replace('_',' ',$aResult['type'])).')</span>';
echo '<p>'.$aResult['lat'].','.$aResult['lon'].'</p>';
echo ' <a class="btn btn-default btn-xs details" href="details.php?place_id='.$aResult['place_id'].'">details</a>';
echo detailsPermaLink($aResult, 'details', 'class="btn btn-default btn-xs details"');
echo '</div>';
?>
</div>

View File

@@ -0,0 +1,48 @@
<?php
header("content-type: text/html; charset=UTF-8");
include(CONST_BasePath.'/lib/template/includes/html-header.php');
?>
<title>Nominatim Deleted Data</title>
<meta name="description" content="List of OSM data that has been deleted" lang="en-US" />
</head>
<body>
<div class="container">
<h1>Deletable</h1>
<p>
<?php echo sizeof($aPolygons) ?> objects have been deleted in OSM but are still in the Nominatim database.
Also available in <a href="<?php echo CONST_Website_BaseURL; ?>deletable.php?format=json">JSON format</a>.
</p>
<table class="table table-striped table-hover">
<?php
if (!empty($aPolygons)) {
echo '<tr>';
foreach (array_keys($aPolygons[0]) as $sCol) {
echo '<th>'.$sCol.'</th>';
}
echo '</tr>';
foreach ($aPolygons as $aRow) {
echo '<tr>';
foreach ($aRow as $sCol => $sVal) {
switch ($sCol) {
case 'osm_id':
echo '<td>'.osmLink($aRow).'</td>';
break;
case 'place_id':
echo '<td>'.detailsLink($aRow).'</td>';
break;
default:
echo '<td>'.($sVal?$sVal:'&nbsp;').'</td>';
break;
}
}
echo '</tr>';
}
}
?>
</table>
</div>
</body>
</html>

View File

@@ -20,20 +20,25 @@
}
function format_distance($fDistance)
function format_distance($fDistance, $bInMeters = false)
{
// $fDistance is in meters
if ($fDistance < 1)
{
return '0';
}
elseif ($fDistance < 1000)
{
return'<abbr class="distance" title="'.$fDistance.'">~'.(round($fDistance,0)).' m</abbr>';
}
else
{
return'<abbr class="distance" title="'.$fDistance.'">~'.(round($fDistance/1000,1)).' km</abbr>';
if ($bInMeters) {
// $fDistance is in meters
if ($fDistance < 1) {
return '0';
}
elseif ($fDistance < 1000) {
return '<abbr class="distance" title="'.$fDistance.' meters">~'.(round($fDistance,0)).' m</abbr>';
}
else {
return '<abbr class="distance" title="'.$fDistance.' meters">~'.(round($fDistance/1000,1)).' km</abbr>';
}
} else {
if ($fDistance == 0) {
return '0';
} else {
return '<abbr class="distance" title="spheric distance '.$fDistance.'">'.(round($fDistance,4)).'</abbr>';
}
}
}
@@ -52,25 +57,33 @@
return $sHTML;
}
function map_icon($sIcon)
function map_icon($aPlace)
{
if ($sIcon){
echo '<img id="mapicon" src="'.CONST_Website_BaseURL.'images/mapicons/'.$sIcon.'.n.32.png'.'" alt="'.$sIcon.'" />';
$sIcon = Nominatim\ClassTypes\getIconFile($aPlace);
if (isset($sIcon)) {
$sLabel = Nominatim\ClassTypes\getIcon($aPlace);
echo '<img id="mapicon" src="'.$sIcon.'" alt="'.$sLabel.'" />';
}
}
function _one_row($aAddressLine){
function _one_row($aAddressLine, $bDistanceInMeters = false){
$bNotUsed = isset($aAddressLine['isaddress']) && !$aAddressLine['isaddress'];
echo '<tr class="' . ($bNotUsed?'notused':'') . '">'."\n";
echo ' <td class="name">'.(trim($aAddressLine['localname'])?$aAddressLine['localname']:'<span class="noname">No Name</span>')."</td>\n";
echo ' <td>' . $aAddressLine['class'].':'.$aAddressLine['type'] . "</td>\n";
echo ' <td class="name">'.(trim($aAddressLine['localname'])!==null?$aAddressLine['localname']:'<span class="noname">No Name</span>')."</td>\n";
echo ' <td>' . $aAddressLine['class'].':'.$aAddressLine['type'];
if ($aAddressLine['type'] == 'administrative'
&& isset($aAddressLine['place_type']))
{
echo '('.$aAddressLine['place_type'].')';
}
echo "</td>\n";
echo ' <td>' . osmLink($aAddressLine) . "</td>\n";
echo ' <td>' . (isset($aAddressLine['rank_address']) ? $aAddressLine['rank_address'] : '') . "</td>\n";
echo ' <td>' . ($aAddressLine['admin_level'] < 15 ? $aAddressLine['admin_level'] : '') . "</td>\n";
echo ' <td>' . format_distance($aAddressLine['distance'])."</td>\n";
echo ' <td>' . detailsLink($aAddressLine,'details &gt;') . "</td>\n";
echo ' <td>' . format_distance($aAddressLine['distance'], $bDistanceInMeters)."</td>\n";
echo ' <td>' . detailsPermaLink($aAddressLine,'details &gt;') . "</td>\n";
echo "</tr>\n";
}
@@ -98,11 +111,10 @@
<div class="col-sm-10">
<h1>
<?php echo $aPointDetails['localname'] ?>
<small><?php echo detailsPermaLink($aPointDetails, 'link to this page') ?></small>
</h1>
</div>
<div class="col-sm-2 text-right">
<?php map_icon($aPointDetails['icon']) ?>
<?php map_icon($aPointDetails) ?>
</div>
</div>
<div class="row">
@@ -122,6 +134,8 @@
kv('Coverage' , ($aPointDetails['isarea']?'Polygon':'Point') );
kv('Centre Point' , $aPointDetails['lat'].','.$aPointDetails['lon'] );
kv('OSM' , osmLink($aPointDetails) );
kv('Place Id (<a href="https://nominatim.org/release-docs/develop/api/Output/#place_id-is-not-a-persistent-id">on this server</a>)'
, $aPointDetails['place_id'] );
if ($aPointDetails['wikipedia'])
{
kv('Wikipedia Calculated' , wikipediaLink($aPointDetails) );
@@ -173,7 +187,7 @@
{
headline('Linked Places');
foreach ($aLinkedLines as $aAddressLine) {
_one_row($aAddressLine);
_one_row($aAddressLine, true);
}
}
@@ -212,7 +226,7 @@
headline3($sGroupHeading);
foreach ($aHierarchyLines as $aAddressLine) {
_one_row($aAddressLine);
_one_row($aAddressLine, true);
}
}
if (count($aHierarchyLines) >= 500) {

View File

@@ -0,0 +1,55 @@
<?php
header("content-type: text/html; charset=UTF-8");
?>
<?php include(CONST_BasePath.'/lib/template/includes/html-header.php'); ?>
<link href="css/common.css" rel="stylesheet" type="text/css" />
<link href="css/details.css" rel="stylesheet" type="text/css" />
</head>
<body id="details-index-page">
<div class="container">
<div class="row">
<div class="col-md-12">
<h1>Show details for place</h1>
<div class="search-form">
<h4>Search by place id</h4>
<form class="form-inline" action="details.php">
<input type="edit" class="form-control input-sm" pattern="^[0-9]+$" name="place_id" placeholder="12345" />
<input type="submit" class="btn btn-primary btn-sm" value="Show" />
</form>
</div>
<div class="search-form">
<h4>Search by OSM type and OSM id</h4>
<form id="form-by-type-and-id" class="form-inline" action="details.php">
<input type="edit" class="form-control input-sm" pattern="^[NWR][0-9]+$" placeholder="N123 or W123 or R123" />
<input type="hidden" name="osmtype" />
<input type="hidden" name="osmid" />
<input type="submit" class="btn btn-primary btn-sm" value="Show" />
</form>
</div>
<div class="search-form">
<h4>Search by openstreetmap.org URL</h4>
<form id="form-by-osm-url" class="form-inline" action="details.php">
<input type="edit" class="form-control input-sm" pattern=".*openstreetmap.*" placeholder="https://www.openstreetmap.org/relation/123" />
<input type="hidden" name="osmtype" />
<input type="hidden" name="osmid" />
<input type="submit" class="btn btn-primary btn-sm" value="Show" />
</form>
</div>
</div>
</div>
</div>
<?php include(CONST_BasePath.'/lib/template/includes/html-footer.php'); ?>
</body>
</html>

View File

@@ -26,8 +26,9 @@ $aPlaceDetails['calculated_importance'] = (float) $aPointDetails['calculated_imp
$aPlaceDetails['extratags'] = $aPointDetails['aExtraTags'];
$aPlaceDetails['calculated_wikipedia'] = $aPointDetails['wikipedia'];
if ($aPointDetails['icon']) {
$aPlaceDetails['icon'] = CONST_Website_BaseURL.'images/mapicons/'.$aPointDetails['icon'].'.n.32.png';
$sIcon = Nominatim\ClassTypes\getIconFile($aPointDetails);
if (isset($sIcon)) {
$aPlaceDetails['icon'] = $sIcon;
}
$aPlaceDetails['rank_address'] = (int) $aPointDetails['rank_address'];
@@ -47,11 +48,13 @@ $funcMapAddressLine = function ($aFull) {
'place_id' => isset($aFull['place_id']) ? (int) $aFull['place_id'] : null,
'osm_id' => isset($aFull['osm_id']) ? (int) $aFull['osm_id'] : null,
'osm_type' => isset($aFull['osm_type']) ? $aFull['osm_type'] : null,
'place_type' => isset($aFull['place_type']) ? $aFull['place_type'] : null,
'class' => $aFull['class'],
'type' => $aFull['type'],
'admin_level' => isset($aFull['admin_level']) ? (int) $aFull['admin_level'] : null,
'rank_address' => $aFull['rank_address'] ? (int) $aFull['rank_address'] : null,
'distance' => (float) $aFull['distance']
'distance' => (float) $aFull['distance'],
'isaddress' => isset($aFull['isaddress']) ? (bool) $aFull['isaddress'] : null
);
return $aMapped;

View File

@@ -6,7 +6,6 @@
<meta name="viewport" content="width=device-width, initial-scale=1">
<base href="<?php echo CONST_Website_BaseURL;?>" />
<link href="nominatim.xml" rel="search" title="Nominatim Search" type="application/opensearchdescription+xml" />
<link href="css/leaflet.css" rel="stylesheet" />
<link href="css/Control.Minimap.min.css" rel="stylesheet" />
<link href="css/bootstrap-theme.min.css" rel="stylesheet" />

View File

@@ -21,8 +21,10 @@
About &amp; Help <span class="caret"></span>
</button>
<ul class="dropdown-menu dropdown-menu-right">
<li><a href="https://wiki.openstreetmap.org/wiki/Nominatim" target="_blank">Documentation</a></li>
<li><a href="https://wiki.openstreetmap.org/wiki/Nominatim/FAQ" target="_blank">FAQ</a></li>
<li><a href="https://nominatim.org/release-docs/develop/api/Overview/" target="_blank">API Reference</a></li>
<li><a href="https://nominatim.org/release-docs/develop/api/Faq/" target="_blank">FAQ</a></li>
<li><a href="https://help.openstreetmap.org/tags/nominatim/">OpenStreetMap Help</a></li>
<li><a href="https://github.com/openstreetmap/Nominatim">Nominatim on Github</a></li>
<li role="separator" class="divider"></li>
<li><a href="#" class="" data-toggle="modal" data-target="#report-modal">Report problem with results</a></li>
</ul>

View File

@@ -6,4 +6,4 @@ look up data by its geographic coordinate (reverse search). Each result comes wi
link to a details page where you can inspect what data about the object is saved in
the database and investigate how the address of the object has been computed.</p>
For more information visit the <a href="https://wiki.openstreetmap.org/wiki/Nominatim">Nominatim wiki page</a>.
For more information visit the <a href="https://nominatim.org">Nominatim home page</a>.

View File

@@ -1,24 +1,42 @@
<p>
Before reporting problems please read the <a target="_blank" href="https://wiki.openstreetmap.org/wiki/Nominatim">user documentation</a>
and
<a target="_blank" href="https://wiki.openstreetmap.org/wiki/Nominatim/FAQ">FAQ</a>.
Before reporting problems please read the <a target="_blank" href="https://nominatim.org/release-docs/develop/api/Overview">user documentation</a>.
<h4>Finding the expected result</h4>
First of all, please make sure that the result that you expect is
available in the OpenStreetMap data.
To find the OpenStreetMap data, do the following:
<ul>
<li>Go to <a href="https://openstreetmap.org">https://openstreetmap.org</a>.</li>
<li>Go to the area of the map where you expect the result
and zoom in until you see the object you are looking for.</li>
<li>Click on the question mark on the right side of the map,
then with the question cursor on the map where your object is located.</li>
<li>Find the object of interest in the list that appears on the left side.</li>
<li>Click on the object and note down the URL that the browser shows.</li>
</ul>
If you cannot find the data you are looking for, there is a good chance
that it has not been entered yet. You should <a href="https://www.openstreetmap.org/fixthemap">report or fix the problem in OpenStreetMap</a> directly.
<h4>Reporting bad searches</h4>
Problems may be reported at the <a target="_blank" href="https://github.com/openstreetmap/nominatim/issues">issue tracker on github</a>. Please read through
the open tickets first and check if your problem has not already been
reported.
When reporting a problem, include the following:
<ul>
<li>A full description of the problem, including the exact term you
were searching for.</li>
<li>The result you get.</li>
<li>The OpenStreetMap object you expect to find (see above).</li>
</ul>
For general questions about installing and searching in Nominatim, please
use <a href="https://help.openstreetmap.org/tags/nominatim/">Help OpenStreetMap</a>.
If your problem relates to the address of a particular search result please use the 'details' link
to check how the address was generated before reporting a problem.
</p>
<p>
Use <a target="_blank" href="https://github.com/openstreetmap/nominatim/issues">Nominatim issues on github</a>
to report problems.
<!-- You can search for existing bug reports
<a href="https://trac.openstreetmap.org/query?status=new&amp;status=assigned&amp;status=reopened&amp;component=nominatim&amp;order=priority">here</a>.</p>
-->
</p>
<p>
Please ensure that you include a full description of the problem, including the search
query that you used, the problem with the result and, if the problem relates to missing data,
the osm type (node, way, relation) and id of the item that is missing.
</p>
<p>
Problems that contain enough detail are likely to get looked at before ones that require
significant research.
</p>

View File

@@ -0,0 +1,71 @@
<?php
header("content-type: text/html; charset=UTF-8");
include(CONST_BasePath.'/lib/template/includes/html-header.php');
?>
<title>Nominatim Broken Polygon Data</title>
<meta name="description" content="List of broken OSM polygon data by date" lang="en-US" />
</head>
<body>
<div class="container">
<h1>Broken polygons</h1>
<p>
Total number of broken polygons: <?php echo $iTotalBroken ?>.
Also available in <a href="<?php echo CONST_Website_BaseURL; ?>polygons.php?format=json">JSON format</a>.
</p>
<table class="table table-striped table-hover">
<?php
if (!empty($aPolygons)) {
echo '<tr>';
//var_dump($aPolygons[0]);
foreach (array_keys($aPolygons[0]) as $sCol) {
echo '<th>'.$sCol.'</th>';
}
echo '<th>&nbsp;</th>';
echo '</tr>';
$aSeen = array();
foreach ($aPolygons as $aRow) {
if (isset($aSeen[$aRow['osm_type'].$aRow['osm_id']])) continue;
$aSeen[$aRow['osm_type'].$aRow['osm_id']] = 1;
echo '<tr>';
$sOSMType = formatOSMType($aRow['osm_type']);
foreach ($aRow as $sCol => $sVal) {
switch ($sCol) {
case 'errormessage':
if (preg_match('/Self-intersection\\[([0-9.\\-]+) ([0-9.\\-]+)\\]/', $sVal, $aMatch)) {
$aRow['lat'] = $aMatch[2];
$aRow['lon'] = $aMatch[1];
$sUrl = sprintf('https://www.openstreetmap.org/?lat=%f&lon=%f&zoom=18&layers=M&%s=%d',
$aRow['lat'],
$aRow['lon'],
$sOSMType,
$aRow['osm_id']);
echo '<td><a href="'.$sUrl.'">'.($sVal?$sVal:'&nbsp;').'</a></td>';
} else {
echo '<td>'.($sVal?$sVal:'&nbsp;').'</td>';
}
break;
case 'osm_id':
echo '<td>'.osmLink(array('osm_type' => $aRow['osm_type'], 'osm_id' => $aRow['osm_id'])).'</td>';
break;
default:
echo '<td>'.($sVal?$sVal:'&nbsp;').'</td>';
break;
}
}
$sJosmUrl = 'http://localhost:8111/import?url=https://www.openstreetmap.org/api/0.6/'.$sOSMType.'/'.$aRow['osm_id'].'/full';
echo '<td><a href="'.$sJosmUrl.'" target="josm">josm</a></td>';
echo '</tr>';
}
echo '</table>';
}
?>
</div>
</body>
</html>

View File

@@ -25,10 +25,6 @@ foreach ($aBatchResults as $aSearchResults) {
$aPointDetails['aBoundingBox'][2],
$aPointDetails['aBoundingBox'][3]
);
if (isset($aPointDetails['aPolyPoints']) && $bShowPolygons) {
$aPlace['polygonpoints'] = $aPointDetails['aPolyPoints'];
}
}
if (isset($aPointDetails['zoom'])) {

View File

@@ -20,27 +20,14 @@ foreach ($aSearchResults as $iResNum => $aPointDetails) {
$aPlace['properties']['geocoding']['label'] = $aPointDetails['langaddress'];
$aPlace['properties']['geocoding']['name'] = $aPointDetails['placename'];
if ($aPointDetails['placename'] !== null) {
$aPlace['properties']['geocoding']['name'] = $aPointDetails['placename'];
}
if (isset($aPointDetails['address'])) {
$aFieldMappings = array(
'house_number' => 'housenumber',
'road' => 'street',
'locality' => 'locality',
'postcode' => 'postcode',
'city' => 'city',
'district' => 'district',
'county' => 'county',
'state' => 'state',
'country' => 'country'
);
$aAddrNames = $aPointDetails['address']->getAddressNames();
foreach ($aFieldMappings as $sFrom => $sTo) {
if (isset($aAddrNames[$sFrom])) {
$aPlace['properties']['geocoding'][$sTo] = $aAddrNames[$sFrom];
}
}
$aPointDetails['address']->addGeocodeJsonAddressParts(
$aPlace['properties']['geocoding']
);
$aPlace['properties']['geocoding']['admin']
= $aPointDetails['address']->getAdminLevels();

View File

@@ -10,26 +10,46 @@
<?php include(CONST_BasePath.'/lib/template/includes/html-top-navigation.php'); ?>
<form class="form-inline" role="search" accept-charset="UTF-8" action="<?php echo CONST_Website_BaseURL; ?>search.php">
<div class="form-group">
<input id="q" name="q" type="text" class="form-control input-sm" placeholder="Search" value="<?php echo htmlspecialchars($sQuery); ?>" >
<div class="top-bar" id="structured-query-selector">
<div class="search-type-link">
<a id="switch-to-reverse" href="<?php echo CONST_Website_BaseURL; ?>reverse.php?format=html">reverse search</a>
</div>
<div class="radio-inline">
<input type="radio" name="query-selector" id="simple" value="simple">
<label for="simple">simple</label>
</div>
<div class="radio-inline">
<input type="radio" name="query-selector" id="structured" value="structured">
<label for="structured">structured</label>
</div>
<form role="search" accept-charset="UTF-8" action="<?php echo CONST_Website_BaseURL; ?>search.php">
<div class="form-group-simple">
<input id="q" name="q" type="text" class="form-control input-sm" placeholder="Search" value="<?php echo htmlspecialchars($aMoreParams['q'] ?? ''); ?>" >
</div>
<div class="form-group-structured">
<div class="form-inline">
<input id="street" name="street" type="text" class="form-control input-sm" placeholder="House number/Street" value="<?php echo htmlspecialchars($aMoreParams['street'] ?? ''); ?>" >
<input id="city" name="city" type="text" class="form-control input-sm" placeholder="City" value="<?php echo htmlspecialchars($aMoreParams['city'] ?? ''); ?>" >
<input id="county" name="county" type="text" class="form-control input-sm" placeholder="County" value="<?php echo htmlspecialchars($aMoreParams['county'] ?? ''); ?>" >
<input id="state" name="state" type="text" class="form-control input-sm" placeholder="State" value="<?php echo htmlspecialchars($aMoreParams['state'] ?? ''); ?>" >
<input id="country" name="country" type="text" class="form-control input-sm" placeholder="Country" value="<?php echo htmlspecialchars($aMoreParams['country'] ?? ''); ?>" >
<input id="postalcode" name="postalcode" type="text" class="form-control input-sm" placeholder="Postal Code" value="<?php echo htmlspecialchars($aMoreParams['postalcode'] ?? ''); ?>" >
</div></div>
<div class="form-group search-button-group">
<button type="submit" class="btn btn-primary btn-sm">Search</button>
<?php if (CONST_Search_AreaPolygons) { ?>
<input type="hidden" value="1" name="polygon_geojson" />
<?php } ?>
<input type="hidden" name="viewbox" value="<?php if (isset($aMoreParams['viewbox'])) echo ($aMoreParams['viewbox']); ?>" />
<input type="hidden" name="viewbox" value="<?php echo htmlspecialchars($aMoreParams['viewbox'] ?? ''); ?>" />
<div class="checkbox-inline">
<input type="checkbox" id="use_viewbox" <?php if (isset($aMoreParams['viewbox'])) echo "checked='checked'"; ?>>
<input type="checkbox" id="use_viewbox" <?php if (!empty($aMoreParams['viewbox'])) echo "checked='checked'"; ?>>
<label for="use_viewbox">apply viewbox</label>
</div>
</div>
<div class="search-type-link">
<a id="switch-to-reverse" href="<?php echo CONST_Website_BaseURL; ?>reverse.php?format=html">reverse search</a>
</div>
</form>
</div>
<div id="content">
@@ -53,7 +73,7 @@
echo ' <span class="type">('.ucwords(str_replace('_',' ',$aResult['class'])).')</span>';
else
echo ' <span class="type">('.ucwords(str_replace('_',' ',$aResult['type'])).')</span>';
echo ' <a class="btn btn-default btn-xs details" href="details.php?place_id='.$aResult['place_id'].'">details</a>';
echo detailsPermaLink($aResult, 'details', 'class="btn btn-default btn-xs details"');
echo '</div>';
$i = $i+1;
}
@@ -89,10 +109,6 @@
<script type="text/javascript">
<?php
@@ -105,7 +121,16 @@
);
echo 'var nominatim_map_init = ' . json_encode($aNominatimMapInit, JSON_PRETTY_PRINT) . ';';
echo 'var nominatim_results = ' . json_encode($aSearchResults, JSON_PRETTY_PRINT) . ';';
echo 'var nominatim_results = ' . json_encode($aSearchResults, JSON_PRETTY_PRINT) . ';';
$sStructuredQuery = (empty($aMoreParams['q'])
&& !(empty($aMoreParams['street'])
&& empty($aMoreParams['city'])
&& empty($aMoreParams['county'])
&& empty($aMoreParams['state'])
&& empty($aMoreParams['country'])
&& empty($aMoreParams['postalcode'])))
? 'true' : 'false';
echo 'var nominatim_structured_query = '.$sStructuredQuery.';';
?>
</script>
<?php include(CONST_BasePath.'/lib/template/includes/html-footer.php'); ?>

View File

@@ -15,10 +15,6 @@ foreach ($aSearchResults as $iResNum => $aPointDetails) {
if (isset($aPointDetails['aBoundingBox'])) {
$aPlace['boundingbox'] = $aPointDetails['aBoundingBox'];
if (isset($aPointDetails['aPolyPoints'])) {
$aPlace['polygonpoints'] = $aPointDetails['aPolyPoints'];
}
}
if (isset($aPointDetails['zoom'])) {

View File

@@ -11,7 +11,6 @@ echo " timestamp='".date(DATE_RFC822)."'";
echo " attribution='Data © OpenStreetMap contributors, ODbL 1.0. http://www.openstreetmap.org/copyright'";
echo " querystring='".htmlspecialchars($sQuery, ENT_QUOTES)."'";
if (isset($aMoreParams['viewbox'])) echo " viewbox='".htmlspecialchars($aMoreParams['viewbox'], ENT_QUOTES)."'";
echo " polygon='".(isset($aMoreParams['polygon'])?'true':'false')."'";
if (isset($aMoreParams['exclude_place_ids'])) {
echo " exclude_place_ids='".htmlspecialchars($aMoreParams['exclude_place_ids'])."'";
}
@@ -31,12 +30,6 @@ foreach ($aSearchResults as $iResNum => $aResult) {
echo ' boundingbox="';
echo join(',', $aResult['aBoundingBox']);
echo '"';
if (isset($aResult['aPolyPoints'])) {
echo ' polygonpoints=\'';
echo json_encode($aResult['aPolyPoints']);
echo '\'';
}
}
if (isset($aResult['asgeojson'])) {

View File

@@ -1,4 +1,13 @@
# just use the pgxs makefile
find_program(PG_CONFIG pg_config)
execute_process(COMMAND ${PG_CONFIG} --pgxs
OUTPUT_VARIABLE PGXS
OUTPUT_STRIP_TRAILING_WHITESPACE)
if (NOT EXISTS "${PGXS}")
message(FATAL_ERROR "Postgresql server package not found.")
endif()
ADD_CUSTOM_COMMAND( OUTPUT ${CMAKE_CURRENT_SOURCE_DIR}/dummy
COMMAND PGXS=${PGXS} PG_CONFIG=${PG_CONFIG} MODSRCDIR=${CMAKE_CURRENT_SOURCE_DIR} $(MAKE) -f ${CMAKE_CURRENT_SOURCE_DIR}/Makefile
COMMENT "Running external makefile ${PGXS}"

File diff suppressed because one or more lines are too long

View File

@@ -1,12 +0,0 @@
add_executable(nominatim export.c geometry.cpp import.c index.c input.c nominatim.c postgresql.c sprompt.c)
CHECK_SYMBOL_EXISTS(bswap_32 "byteswap.h" HAVE_BYTESWAP)
CHECK_SYMBOL_EXISTS(bswap32 "sys/endian.h" HAVE_SYS_ENDIAN)
target_compile_definitions(nominatim
PRIVATE HAVE_BYTESWAP=$<BOOL:${HAVE_BYTESWAP}>
PRIVATE HAVE_SYS_ENDIAN=$<BOOL:${HAVE_SYS_ENDIAN}>
)
target_link_libraries(nominatim ${LIBXML2_LIBRARIES} ${ZLIB_LIBRARIES} ${BZIP2_LIBRARIES} ${PostgreSQL_LIBRARIES} ${CMAKE_THREAD_LIBS_INIT})

View File

@@ -1,558 +0,0 @@
/*
*/
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <string.h>
#include <assert.h>
#include <time.h>
#include <stdint.h>
#include <pthread.h>
#include <libpq-fe.h>
#include "nominatim.h"
#include "export.h"
#include "postgresql.h"
extern int verbose;
int mode = 0;
void nominatim_export(int rank_min, int rank_max, const char *conninfo, const char *structuredoutputfile)
{
xmlTextWriterPtr writer;
int rankTotalDone;
PGconn *conn;
PGresult * res;
PGresult * resSectors;
PGresult * resPlaces;
int rank;
int i;
int iSector;
int tuples;
const char *paramValues[2];
int paramLengths[2];
int paramFormats[2];
uint32_t paramRank;
uint32_t paramSector;
uint32_t sector;
Oid pg_prepare_params[2];
conn = PQconnectdb(conninfo);
if (PQstatus(conn) != CONNECTION_OK)
{
fprintf(stderr, "Connection to database failed: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
pg_prepare_params[0] = PG_OID_INT4;
res = PQprepare(conn, "index_sectors",
"select geometry_sector,count(*) from placex where rank_search = $1 and indexed_status = 0 group by geometry_sector order by geometry_sector",
1, pg_prepare_params);
if (PQresultStatus(res) != PGRES_COMMAND_OK) exit(EXIT_FAILURE);
PQclear(res);
pg_prepare_params[0] = PG_OID_INT4;
pg_prepare_params[1] = PG_OID_INT4;
res = PQprepare(conn, "index_sector_places",
"select place_id from placex where rank_search = $1 and geometry_sector = $2",
2, pg_prepare_params);
if (PQresultStatus(res) != PGRES_COMMAND_OK) exit(EXIT_FAILURE);
PQclear(res);
nominatim_exportCreatePreparedQueries(conn);
// Create the output file
writer = nominatim_exportXMLStart(structuredoutputfile);
for (rank = rank_min; rank <= rank_max; rank++)
{
printf("Starting rank %d\n", rank);
paramRank = PGint32(rank);
paramValues[0] = (char *)&paramRank;
paramLengths[0] = sizeof(paramRank);
paramFormats[0] = 1;
resSectors = PQexecPrepared(conn, "index_sectors", 1, paramValues, paramLengths, paramFormats, 1);
if (PQresultStatus(resSectors) != PGRES_TUPLES_OK)
{
fprintf(stderr, "index_sectors: SELECT failed: %s", PQerrorMessage(conn));
PQclear(resSectors);
exit(EXIT_FAILURE);
}
if (PQftype(resSectors, 0) != PG_OID_INT4)
{
fprintf(stderr, "Sector value has unexpected type\n");
PQclear(resSectors);
exit(EXIT_FAILURE);
}
if (PQftype(resSectors, 1) != PG_OID_INT8)
{
fprintf(stderr, "Sector value has unexpected type\n");
PQclear(resSectors);
exit(EXIT_FAILURE);
}
rankTotalDone = 0;
for (iSector = 0; iSector < PQntuples(resSectors); iSector++)
{
sector = PGint32(*((uint32_t *)PQgetvalue(resSectors, iSector, 0)));
// Get all the place_id's for this sector
paramRank = PGint32(rank);
paramValues[0] = (char *)&paramRank;
paramLengths[0] = sizeof(paramRank);
paramFormats[0] = 1;
paramSector = PGint32(sector);
paramValues[1] = (char *)&paramSector;
paramLengths[1] = sizeof(paramSector);
paramFormats[1] = 1;
resPlaces = PQexecPrepared(conn, "index_sector_places", 2, paramValues, paramLengths, paramFormats, 1);
if (PQresultStatus(resPlaces) != PGRES_TUPLES_OK)
{
fprintf(stderr, "index_sector_places: SELECT failed: %s", PQerrorMessage(conn));
PQclear(resPlaces);
exit(EXIT_FAILURE);
}
if (PQftype(resPlaces, 0) != PG_OID_INT8)
{
fprintf(stderr, "Place_id value has unexpected type\n");
PQclear(resPlaces);
exit(EXIT_FAILURE);
}
tuples = PQntuples(resPlaces);
for (i = 0; i < tuples; i++)
{
nominatim_exportPlace(PGint64(*((uint64_t *)PQgetvalue(resPlaces, i, 0))), conn, writer, NULL, NULL);
rankTotalDone++;
if (rankTotalDone%1000 == 0) printf("Done %i (k)\n", rankTotalDone/1000);
}
PQclear(resPlaces);
}
PQclear(resSectors);
}
nominatim_exportXMLEnd(writer);
PQfinish(conn);
}
void nominatim_exportCreatePreparedQueries(PGconn * conn)
{
Oid pg_prepare_params[2];
PGresult * res;
pg_prepare_params[0] = PG_OID_INT8;
res = PQprepare(conn, "placex_details",
"select placex.osm_type, placex.osm_id, placex.class, placex.type, placex.name, placex.housenumber, placex.country_code, ST_AsText(placex.geometry), placex.admin_level, placex.rank_address, placex.rank_search, placex.parent_place_id, parent.osm_type, parent.osm_id, placex.indexed_status, placex.linked_place_id from placex left outer join placex as parent on (placex.parent_place_id = parent.place_id) where placex.place_id = $1",
1, pg_prepare_params);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Error preparing placex_details: %s", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
PQclear(res);
pg_prepare_params[0] = PG_OID_INT8;
res = PQprepare(conn, "placex_address",
"select osm_type,osm_id,class,type,distance,cached_rank_address,isaddress from place_addressline join placex on (address_place_id = placex.place_id) where place_addressline.place_id = $1 and address_place_id != place_addressline.place_id order by cached_rank_address asc,osm_type,osm_id",
1, pg_prepare_params);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Error preparing placex_address: %s", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
PQclear(res);
pg_prepare_params[0] = PG_OID_INT8;
res = PQprepare(conn, "placex_names",
"select (each(name)).key,(each(name)).value from (select name from placex where place_id = $1) as x order by (each(name)).key",
1, pg_prepare_params);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Error preparing placex_names: %s", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
PQclear(res);
pg_prepare_params[0] = PG_OID_INT8;
res = PQprepare(conn, "placex_extratags",
"select (each(extratags)).key,(each(extratags)).value from (select extratags from placex where place_id = $1) as x order by (each(extratags)).key",
1, pg_prepare_params);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Error preparing placex_extratags: %s", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
PQclear(res);
}
xmlTextWriterPtr nominatim_exportXMLStart(const char *structuredoutputfile)
{
xmlTextWriterPtr writer;
writer = xmlNewTextWriterFilename(structuredoutputfile, 0);
if (writer==NULL)
{
fprintf(stderr, "Unable to open %s\n", structuredoutputfile);
exit(EXIT_FAILURE);
}
xmlTextWriterSetIndent(writer, 1);
if (xmlTextWriterStartDocument(writer, NULL, "UTF8", NULL) < 0)
{
fprintf(stderr, "xmlTextWriterStartDocument failed\n");
exit(EXIT_FAILURE);
}
if (xmlTextWriterStartElement(writer, BAD_CAST "osmStructured") < 0)
{
fprintf(stderr, "xmlTextWriterStartElement failed\n");
exit(EXIT_FAILURE);
}
if (xmlTextWriterWriteAttribute(writer, BAD_CAST "version", BAD_CAST "0.1") < 0)
{
fprintf(stderr, "xmlTextWriterWriteAttribute failed\n");
exit(EXIT_FAILURE);
}
if (xmlTextWriterWriteAttribute(writer, BAD_CAST "generator", BAD_CAST "Nominatim") < 0)
{
fprintf(stderr, "xmlTextWriterWriteAttribute failed\n");
exit(EXIT_FAILURE);
}
mode = 0;
return writer;
}
void nominatim_exportXMLEnd(xmlTextWriterPtr writer)
{
nominatim_exportEndMode(writer);
// End <osmStructured>
if (xmlTextWriterEndElement(writer) < 0)
{
fprintf(stderr, "xmlTextWriterEndElement failed\n");
exit(EXIT_FAILURE);
}
if (xmlTextWriterEndDocument(writer) < 0)
{
fprintf(stderr, "xmlTextWriterEndDocument failed\n");
exit(EXIT_FAILURE);
}
xmlFreeTextWriter(writer);
}
void nominatim_exportStartMode(xmlTextWriterPtr writer, int newMode)
{
if (mode == newMode) return;
nominatim_exportEndMode(writer);
switch(newMode)
{
case 0:
break;
case 1:
if (xmlTextWriterStartElement(writer, BAD_CAST "add") < 0)
{
fprintf(stderr, "xmlTextWriterStartElement failed\n");
exit(EXIT_FAILURE);
}
break;
case 2:
if (xmlTextWriterStartElement(writer, BAD_CAST "update") < 0)
{
fprintf(stderr, "xmlTextWriterStartElement failed\n");
exit(EXIT_FAILURE);
}
break;
case 3:
if (xmlTextWriterStartElement(writer, BAD_CAST "delete") < 0)
{
fprintf(stderr, "xmlTextWriterStartElement failed\n");
exit(EXIT_FAILURE);
}
break;
}
mode = newMode;
}
void nominatim_exportEndMode(xmlTextWriterPtr writer)
{
if (!mode) return;
if (xmlTextWriterEndElement(writer) < 0)
{
fprintf(stderr, "xmlTextWriterEndElement failed\n");
exit(EXIT_FAILURE);
}
}
void nominatim_exportPlaceQueries(uint64_t place_id, PGconn * conn, struct export_data * querySet)
{
const char * paramValues[1];
int paramLengths[1];
int paramFormats[1];
uint64_t paramPlaceID;
paramPlaceID = PGint64(place_id);
paramValues[0] = (char *)&paramPlaceID;
paramLengths[0] = sizeof(paramPlaceID);
paramFormats[0] = 1;
querySet->res = PQexecPrepared(conn, "placex_details", 1, paramValues, paramLengths, paramFormats, 0);
if (PQresultStatus(querySet->res) != PGRES_TUPLES_OK)
{
fprintf(stderr, "placex_details: SELECT failed: %s", PQerrorMessage(conn));
PQclear(querySet->res);
exit(EXIT_FAILURE);
}
querySet->resNames = PQexecPrepared(conn, "placex_names", 1, paramValues, paramLengths, paramFormats, 0);
if (PQresultStatus(querySet->resNames) != PGRES_TUPLES_OK)
{
fprintf(stderr, "placex_names: SELECT failed: %s", PQerrorMessage(conn));
PQclear(querySet->resNames);
exit(EXIT_FAILURE);
}
querySet->resAddress = PQexecPrepared(conn, "placex_address", 1, paramValues, paramLengths, paramFormats, 0);
if (PQresultStatus(querySet->resAddress) != PGRES_TUPLES_OK)
{
fprintf(stderr, "placex_address: SELECT failed: %s", PQerrorMessage(conn));
PQclear(querySet->resAddress);
exit(EXIT_FAILURE);
}
querySet->resExtraTags = PQexecPrepared(conn, "placex_extratags", 1, paramValues, paramLengths, paramFormats, 0);
if (PQresultStatus(querySet->resExtraTags) != PGRES_TUPLES_OK)
{
fprintf(stderr, "placex_extratags: SELECT failed: %s", PQerrorMessage(conn));
PQclear(querySet->resExtraTags);
exit(EXIT_FAILURE);
}
}
void nominatim_exportFreeQueries(struct export_data * querySet)
{
PQclear(querySet->res);
PQclear(querySet->resNames);
PQclear(querySet->resAddress);
PQclear(querySet->resExtraTags);
}
/*
* Requirements: the prepared queries must exist
*/
void nominatim_exportPlace(uint64_t place_id, PGconn * conn,
xmlTextWriterPtr writer, pthread_mutex_t * writer_mutex, struct export_data * prevQuerySet)
{
struct export_data querySet;
int i;
nominatim_exportPlaceQueries(place_id, conn, &querySet);
// Add, modify or delete?
if (prevQuerySet)
{
if ((PQgetvalue(prevQuerySet->res, 0, 14) && strcmp(PQgetvalue(prevQuerySet->res, 0, 14), "100") == 0) || PQntuples(querySet.res) == 0)
{
// Delete
if (writer_mutex) pthread_mutex_lock( writer_mutex );
nominatim_exportStartMode(writer, 3);
xmlTextWriterStartElement(writer, BAD_CAST "feature");
xmlTextWriterWriteFormatAttribute(writer, BAD_CAST "place_id", "%li", place_id);
xmlTextWriterWriteAttribute(writer, BAD_CAST "type", BAD_CAST PQgetvalue(prevQuerySet->res, 0, 0));
xmlTextWriterWriteAttribute(writer, BAD_CAST "id", BAD_CAST PQgetvalue(prevQuerySet->res, 0, 1));
xmlTextWriterWriteAttribute(writer, BAD_CAST "key", BAD_CAST PQgetvalue(prevQuerySet->res, 0, 2));
xmlTextWriterWriteAttribute(writer, BAD_CAST "value", BAD_CAST PQgetvalue(prevQuerySet->res, 0, 3));
xmlTextWriterEndElement(writer);
if (writer_mutex) pthread_mutex_unlock( writer_mutex );
nominatim_exportFreeQueries(&querySet);
return;
}
if (PQgetvalue(prevQuerySet->res, 0, 14) && strcmp(PQgetvalue(prevQuerySet->res, 0, 14), "1") == 0)
{
// Add
if (writer_mutex) pthread_mutex_lock( writer_mutex );
nominatim_exportStartMode(writer, 1);
}
else
{
// Update, but only if something has changed
// TODO: detect changes
if (writer_mutex) pthread_mutex_lock( writer_mutex );
nominatim_exportStartMode(writer, 2);
}
}
else
{
// Add
if (writer_mutex) pthread_mutex_lock( writer_mutex );
nominatim_exportStartMode(writer, 1);
}
xmlTextWriterStartElement(writer, BAD_CAST "feature");
xmlTextWriterWriteFormatAttribute(writer, BAD_CAST "place_id", "%li", place_id);
xmlTextWriterWriteAttribute(writer, BAD_CAST "type", BAD_CAST PQgetvalue(querySet.res, 0, 0));
xmlTextWriterWriteAttribute(writer, BAD_CAST "id", BAD_CAST PQgetvalue(querySet.res, 0, 1));
xmlTextWriterWriteAttribute(writer, BAD_CAST "key", BAD_CAST PQgetvalue(querySet.res, 0, 2));
xmlTextWriterWriteAttribute(writer, BAD_CAST "value", BAD_CAST PQgetvalue(querySet.res, 0, 3));
xmlTextWriterWriteAttribute(writer, BAD_CAST "rank", BAD_CAST PQgetvalue(querySet.res, 0, 9));
xmlTextWriterWriteAttribute(writer, BAD_CAST "importance", BAD_CAST PQgetvalue(querySet.res, 0, 10));
xmlTextWriterWriteAttribute(writer, BAD_CAST "parent_place_id", BAD_CAST PQgetvalue(querySet.res, 0, 11));
xmlTextWriterWriteAttribute(writer, BAD_CAST "parent_type", BAD_CAST PQgetvalue(querySet.res, 0, 12));
xmlTextWriterWriteAttribute(writer, BAD_CAST "parent_id", BAD_CAST PQgetvalue(querySet.res, 0, 13));
xmlTextWriterWriteAttribute(writer, BAD_CAST "linked_place_id", BAD_CAST PQgetvalue(querySet.res, 0, 15));
if (PQntuples(querySet.resNames))
{
xmlTextWriterStartElement(writer, BAD_CAST "names");
for (i = 0; i < PQntuples(querySet.resNames); i++)
{
xmlTextWriterStartElement(writer, BAD_CAST "name");
xmlTextWriterWriteAttribute(writer, BAD_CAST "type", BAD_CAST PQgetvalue(querySet.resNames, i, 0));
xmlTextWriterWriteString(writer, BAD_CAST PQgetvalue(querySet.resNames, i, 1));
xmlTextWriterEndElement(writer);
}
xmlTextWriterEndElement(writer);
}
if (PQgetvalue(querySet.res, 0, 5) && strlen(PQgetvalue(querySet.res, 0, 5)))
{
xmlTextWriterStartElement(writer, BAD_CAST "houseNumber");
xmlTextWriterWriteString(writer, BAD_CAST PQgetvalue(querySet.res, 0, 5));
xmlTextWriterEndElement(writer);
}
if (PQgetvalue(querySet.res, 0, 8) && strlen(PQgetvalue(querySet.res, 0, 8)))
{
xmlTextWriterStartElement(writer, BAD_CAST "adminLevel");
xmlTextWriterWriteString(writer, BAD_CAST PQgetvalue(querySet.res, 0, 8));
xmlTextWriterEndElement(writer);
}
if (PQgetvalue(querySet.res, 0, 6) && strlen(PQgetvalue(querySet.res, 0, 6)))
{
xmlTextWriterStartElement(writer, BAD_CAST "countryCode");
xmlTextWriterWriteString(writer, BAD_CAST PQgetvalue(querySet.res, 0, 6));
xmlTextWriterEndElement(writer);
}
if (PQntuples(querySet.resAddress) > 0)
{
xmlTextWriterStartElement(writer, BAD_CAST "address");
for (i = 0; i < PQntuples(querySet.resAddress); i++)
{
xmlTextWriterStartElement(writer, BAD_CAST getRankLabel(atoi(PQgetvalue(querySet.resAddress, i, 5))));
xmlTextWriterWriteAttribute(writer, BAD_CAST "rank", BAD_CAST PQgetvalue(querySet.resAddress, i, 5));
xmlTextWriterWriteAttribute(writer, BAD_CAST "type", BAD_CAST PQgetvalue(querySet.resAddress, i, 0));
xmlTextWriterWriteAttribute(writer, BAD_CAST "id", BAD_CAST PQgetvalue(querySet.resAddress, i, 1));
xmlTextWriterWriteAttribute(writer, BAD_CAST "key", BAD_CAST PQgetvalue(querySet.resAddress, i, 2));
xmlTextWriterWriteAttribute(writer, BAD_CAST "value", BAD_CAST PQgetvalue(querySet.resAddress, i, 3));
xmlTextWriterWriteAttribute(writer, BAD_CAST "distance", BAD_CAST PQgetvalue(querySet.resAddress, i, 4));
xmlTextWriterWriteAttribute(writer, BAD_CAST "isaddress", BAD_CAST PQgetvalue(querySet.resAddress, i, 6));
xmlTextWriterEndElement(writer);
}
xmlTextWriterEndElement(writer);
}
if (PQntuples(querySet.resExtraTags))
{
xmlTextWriterStartElement(writer, BAD_CAST "tags");
for (i = 0; i < PQntuples(querySet.resExtraTags); i++)
{
xmlTextWriterStartElement(writer, BAD_CAST "tag");
xmlTextWriterWriteAttribute(writer, BAD_CAST "type", BAD_CAST PQgetvalue(querySet.resExtraTags, i, 0));
xmlTextWriterWriteString(writer, BAD_CAST PQgetvalue(querySet.resExtraTags, i, 1));
xmlTextWriterEndElement(writer);
}
xmlTextWriterEndElement(writer);
}
xmlTextWriterStartElement(writer, BAD_CAST "osmGeometry");
xmlTextWriterWriteString(writer, BAD_CAST PQgetvalue(querySet.res, 0, 7));
xmlTextWriterEndElement(writer);
xmlTextWriterEndElement(writer); // </feature>
if (writer_mutex) pthread_mutex_unlock( writer_mutex );
nominatim_exportFreeQueries(&querySet);
}
const char * getRankLabel(int rank)
{
switch (rank)
{
case 0:
case 1:
return "continent";
case 2:
case 3:
return "sea";
case 4:
case 5:
case 6:
case 7:
return "country";
case 8:
case 9:
case 10:
case 11:
return "state";
case 12:
case 13:
case 14:
case 15:
return "county";
case 16:
return "city";
case 17:
return "town";
case 18:
return "village";
case 19:
return "unknown";
case 20:
return "suburb";
case 21:
return "postcode";
case 22:
return "neighborhood";
case 23:
return "postcode";
case 24:
return "unknown";
case 25:
return "postcode";
case 26:
return "street";
case 27:
return "access";
case 28:
return "building";
case 29:
default:
return "other";
}
}

View File

@@ -1,31 +0,0 @@
#ifndef EXPORT_H
#define EXPORT_H
#include <libxml/encoding.h>
#include <libxml/xmlwriter.h>
#include <stdint.h>
struct export_data
{
PGresult * res;
PGresult * resNames;
PGresult * resAddress;
PGresult * resExtraTags;
};
void nominatim_export(int rank_min, int rank_max, const char *conninfo, const char *structuredoutputfile);
void nominatim_exportCreatePreparedQueries(PGconn * conn);
xmlTextWriterPtr nominatim_exportXMLStart(const char *structuredoutputfile);
void nominatim_exportXMLEnd(xmlTextWriterPtr writer);
void nominatim_exportEndMode(xmlTextWriterPtr writer);
void nominatim_exportPlaceQueries(uint64_t place_id, PGconn * conn, struct export_data * querySet);
void nominatim_exportFreeQueries(struct export_data * querySet);
void nominatim_exportPlace(uint64_t place_id, PGconn * conn,
xmlTextWriterPtr writer, pthread_mutex_t * writer_mutex, struct export_data * prevQuerySet);
const char * getRankLabel(int rank);
#endif

View File

View File

@@ -1,856 +0,0 @@
/*
*/
#include <stdlib.h>
#include <string.h>
#include <libpq-fe.h>
#include <libxml/xmlstring.h>
#include <libxml/xmlreader.h>
#include <libxml/hash.h>
#include "nominatim.h"
#include "import.h"
#include "input.h"
typedef enum { FILETYPE_NONE, FILETYPE_STRUCTUREDV0P1 } filetypes_t;
typedef enum { FILEMODE_NONE, FILEMODE_ADD, FILEMODE_UPDATE, FILEMODE_DELETE } filemodes_t;
#define MAX_FEATUREADDRESS 5000
#define MAX_FEATURENAMES 10000
#define MAX_FEATUREEXTRATAGS 10000
#define MAX_FEATURENAMESTRING 1000000
#define MAX_FEATUREEXTRATAGSTRING 500000
struct feature_address
{
int place_id;
int rankAddress;
char isAddress[2];
xmlChar * type;
xmlChar * id;
xmlChar * key;
xmlChar * value;
xmlChar * distance;
};
struct feature_tag
{
xmlChar * type;
xmlChar * value;
};
struct feature
{
xmlChar * placeID;
xmlChar * type;
xmlChar * id;
xmlChar * key;
xmlChar * value;
xmlChar * rankAddress;
xmlChar * rankSearch;
xmlChar * countryCode;
xmlChar * parentPlaceID;
xmlChar * parentType;
xmlChar * parentID;
xmlChar * adminLevel;
xmlChar * houseNumber;
xmlChar * geometry;
} feature;
int fileType = FILETYPE_NONE;
int fileMode = FILEMODE_ADD;
PGconn * conn;
struct feature_address featureAddress[MAX_FEATUREADDRESS];
struct feature_tag featureName[MAX_FEATURENAMES];
struct feature_tag featureExtraTag[MAX_FEATUREEXTRATAGS];
struct feature feature;
int featureAddressLines = 0;
int featureNameLines = 0;
int featureExtraTagLines = 0;
int featureCount = 0;
xmlHashTablePtr partionTableTagsHash;
xmlHashTablePtr partionTableTagsHashDelete;
char featureNameString[MAX_FEATURENAMESTRING];
char featureExtraTagString[MAX_FEATUREEXTRATAGSTRING];
extern int verbose;
void StartElement(xmlTextReaderPtr reader, const xmlChar *name)
{
char * value;
float version;
int isAddressLine;
if (fileType == FILETYPE_NONE)
{
// Potential to handle other file types in the future / versions
if (xmlStrEqual(name, BAD_CAST "osmStructured"))
{
value = (char*)xmlTextReaderGetAttribute(reader, BAD_CAST "version");
version = strtof(value, NULL);
xmlFree(value);
if (version == (float)0.1)
{
fileType = FILETYPE_STRUCTUREDV0P1;
fileMode = FILEMODE_ADD;
}
else
{
fprintf( stderr, "Unknown osmStructured version %f (%s)\n", version, value );
exit_nicely();
}
}
else
{
fprintf( stderr, "Unknown XML document type: %s\n", name );
exit_nicely();
}
return;
}
if (xmlStrEqual(name, BAD_CAST "add"))
{
fileMode = FILEMODE_ADD;
return;
}
if (xmlStrEqual(name, BAD_CAST "update"))
{
fileMode = FILEMODE_UPDATE;
return;
}
if (xmlStrEqual(name, BAD_CAST "delete"))
{
fileMode = FILEMODE_DELETE;
return;
}
if (fileMode == FILEMODE_NONE)
{
fprintf( stderr, "Unknown import mode in: %s\n", name );
exit_nicely();
}
if (xmlStrEqual(name, BAD_CAST "feature"))
{
feature.placeID = xmlTextReaderGetAttribute(reader, BAD_CAST "place_id");
feature.type = xmlTextReaderGetAttribute(reader, BAD_CAST "type");
feature.id = xmlTextReaderGetAttribute(reader, BAD_CAST "id");
feature.key = xmlTextReaderGetAttribute(reader, BAD_CAST "key");
feature.value = xmlTextReaderGetAttribute(reader, BAD_CAST "value");
feature.rankAddress = xmlTextReaderGetAttribute(reader, BAD_CAST "rank");
feature.rankSearch = xmlTextReaderGetAttribute(reader, BAD_CAST "importance");
feature.parentPlaceID = xmlTextReaderGetAttribute(reader, BAD_CAST "parent_place_id");
/*
if (strlen(feature.parentPlaceID) == 0)
{
xmlFree(feature.parentPlaceID);
feature.parentPlaceID = NULL;
}
*/
feature.parentType = xmlTextReaderGetAttribute(reader, BAD_CAST "parent_type");
feature.parentID = xmlTextReaderGetAttribute(reader, BAD_CAST "parent_id");
feature.countryCode = NULL;
feature.adminLevel = NULL;
feature.houseNumber = NULL;
feature.geometry = NULL;
featureAddressLines = 0;
featureNameLines = 0;
featureExtraTagLines = 0;
return;
}
if (xmlStrEqual(name, BAD_CAST "names")) return;
if (xmlStrEqual(name, BAD_CAST "name"))
{
if (featureNameLines < MAX_FEATURENAMES)
{
featureName[featureNameLines].type = xmlTextReaderGetAttribute(reader, BAD_CAST "type");
featureName[featureNameLines].value = xmlTextReaderReadString(reader);
featureNameLines++;
}
else
{
fprintf( stderr, "Too many name elements (%s%s)\n", feature.type, feature.id);
// exit_nicely();
}
return;
}
if (xmlStrEqual(name, BAD_CAST "tags")) return;
if (xmlStrEqual(name, BAD_CAST "tag"))
{
if (featureExtraTagLines < MAX_FEATUREEXTRATAGS)
{
featureExtraTag[featureExtraTagLines].type = xmlTextReaderGetAttribute(reader, BAD_CAST "type");
featureExtraTag[featureExtraTagLines].value = xmlTextReaderReadString(reader);
featureExtraTagLines++;
}
else
{
fprintf( stderr, "Too many extra tag elements (%s%s)\n", feature.type, feature.id);
// exit_nicely();
}
return;
}
if (xmlStrEqual(name, BAD_CAST "osmGeometry"))
{
feature.geometry = xmlTextReaderReadString(reader);
return;
}
if (xmlStrEqual(name, BAD_CAST "adminLevel"))
{
feature.adminLevel = xmlTextReaderReadString(reader);
return;
}
if (xmlStrEqual(name, BAD_CAST "countryCode"))
{
feature.countryCode = xmlTextReaderReadString(reader);
return;
}
if (xmlStrEqual(name, BAD_CAST "houseNumber"))
{
feature.houseNumber = xmlTextReaderReadString(reader);
return;
}
if (xmlStrEqual(name, BAD_CAST "address"))
{
featureAddressLines = 0;
return;
}
isAddressLine = 0;
if (xmlStrEqual(name, BAD_CAST "continent"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "sea"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "country"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "state"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "county"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "city"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "town"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "village"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "unknown"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "suburb"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "postcode"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "neighborhood"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "street"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "access"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "building"))
{
isAddressLine = 1;
}
else if (xmlStrEqual(name, BAD_CAST "other"))
{
isAddressLine = 1;
}
if (isAddressLine)
{
if (featureAddressLines < MAX_FEATUREADDRESS)
{
value = (char*)xmlTextReaderGetAttribute(reader, BAD_CAST "rank");
if (!value)
{
fprintf( stderr, "Address element missing rank\n");
exit_nicely();
}
featureAddress[featureAddressLines].rankAddress = atoi(value);
xmlFree(value);
value = (char*)xmlTextReaderGetAttribute(reader, BAD_CAST "isaddress");
if (!value)
{
fprintf( stderr, "Address element missing rank\n");
exit_nicely();
}
if (*value == 't') strcpy(featureAddress[featureAddressLines].isAddress, "t");
else strcpy(featureAddress[featureAddressLines].isAddress, "f");
xmlFree(value);
featureAddress[featureAddressLines].type = xmlTextReaderGetAttribute(reader, BAD_CAST "type");
featureAddress[featureAddressLines].id = xmlTextReaderGetAttribute(reader, BAD_CAST "id");
featureAddress[featureAddressLines].key = xmlTextReaderGetAttribute(reader, BAD_CAST "key");
featureAddress[featureAddressLines].value = xmlTextReaderGetAttribute(reader, BAD_CAST "value");
featureAddress[featureAddressLines].distance = xmlTextReaderGetAttribute(reader, BAD_CAST "distance");
featureAddressLines++;
}
else
{
fprintf( stderr, "Too many address elements (%s%s)\n", feature.type, feature.id);
// exit_nicely();
}
return;
}
fprintf(stderr, "%s: Unknown element name: %s\n", __FUNCTION__, name);
}
void EndElement(xmlTextReaderPtr reader, const xmlChar *name)
{
PGresult * res;
const char * paramValues[14];
char * place_id;
char * partionQueryName;
int i, namePos, lineTypeLen, lineValueLen;
if (xmlStrEqual(name, BAD_CAST "feature"))
{
featureCount++;
if (featureCount % 1000 == 0) printf("feature %i(k)\n", featureCount/1000);
/*
if (fileMode == FILEMODE_ADD)
{
resPlaceID = PQexecPrepared(conn, "get_new_place_id", 0, NULL, NULL, NULL, 0);
if (PQresultStatus(resPlaceID) != PGRES_TUPLES_OK)
{
fprintf(stderr, "get_place_id: INSERT failed: %s", PQerrorMessage(conn));
PQclear(resPlaceID);
exit(EXIT_FAILURE);
}
}
else
{
paramValues[0] = (const char *)feature.type;
paramValues[1] = (const char *)feature.id;
paramValues[2] = (const char *)feature.key;
paramValues[3] = (const char *)feature.value;
resPlaceID = PQexecPrepared(conn, "get_new_place_id", 4, paramValues, NULL, NULL, 0);
if (PQresultStatus(resPlaceID) != PGRES_TUPLES_OK)
{
fprintf(stderr, "index_placex: INSERT failed: %s", PQerrorMessage(conn));
PQclear(resPlaceID);
exit(EXIT_FAILURE);
}
}
*/
place_id = (char *)feature.placeID;
if (fileMode == FILEMODE_UPDATE || fileMode == FILEMODE_DELETE || fileMode == FILEMODE_ADD)
{
paramValues[0] = (const char *)place_id;
if (verbose) fprintf(stderr, "placex_delete: %s\n", paramValues[0]);
res = PQexecPrepared(conn, "placex_delete", 1, paramValues, NULL, NULL, 0);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "placex_delete: DELETE failed: %s", PQerrorMessage(conn));
PQclear(res);
exit(EXIT_FAILURE);
}
PQclear(res);
if (verbose) fprintf(stderr, "search_name_delete: %s\n", paramValues[0]);
res = PQexecPrepared(conn, "search_name_delete", 1, paramValues, NULL, NULL, 0);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "search_name_delete: DELETE failed: %s", PQerrorMessage(conn));
PQclear(res);
exit(EXIT_FAILURE);
}
PQclear(res);
if (verbose) fprintf(stderr, "place_addressline_delete: %s\n", paramValues[0]);
res = PQexecPrepared(conn, "place_addressline_delete", 1, paramValues, NULL, NULL, 0);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "place_addressline_delete: DELETE failed: %s", PQerrorMessage(conn));
PQclear(res);
exit(EXIT_FAILURE);
}
PQclear(res);
partionQueryName = xmlHashLookup2(partionTableTagsHashDelete, feature.key, feature.value);
if (partionQueryName)
{
res = PQexecPrepared(conn, partionQueryName, 1, paramValues, NULL, NULL, 0);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "%s: DELETE failed: %s", partionQueryName, PQerrorMessage(conn));
PQclear(res);
exit(EXIT_FAILURE);
}
PQclear(res);
}
}
if (fileMode == FILEMODE_UPDATE || fileMode == FILEMODE_ADD)
{
// Insert into placex
paramValues[0] = (const char *)place_id;
paramValues[1] = (const char *)feature.type;
paramValues[2] = (const char *)feature.id;
paramValues[3] = (const char *)feature.key;
paramValues[4] = (const char *)feature.value;
featureNameString[0] = 0;
if (featureNameLines)
{
namePos = 0;
lineTypeLen = 0;
lineValueLen = 0;
for (i = 0; i < featureNameLines; i++)
{
lineTypeLen = (int)strlen((char *) featureName[i].type);
lineValueLen = (int)strlen((char *) featureName[i].value);
if (namePos+lineTypeLen+lineValueLen+7 > MAX_FEATURENAMESTRING)
{
fprintf(stderr, "feature name too long: %s", (const char *)featureName[i].value);
break;
}
if (namePos) strcpy(featureNameString+(namePos++), ",");
strcpy(featureNameString+(namePos++), "\"");
strcpy(featureNameString+namePos, (char*) featureName[i].type);
namePos += lineTypeLen;
strcpy(featureNameString+namePos, "\"=>\"");
namePos += 4;
strcpy(featureNameString+namePos, (char *) featureName[i].value);
namePos += lineValueLen;
strcpy(featureNameString+(namePos++), "\"");
xmlFree(featureName[i].type);
xmlFree(featureName[i].value);
}
}
paramValues[5] = (const char *)featureNameString;
paramValues[6] = (const char *)feature.countryCode;
featureExtraTagString[0] = 0;
if (featureExtraTagLines)
{
namePos = 0;
lineTypeLen = 0;
lineValueLen = 0;
for (i = 0; i < featureExtraTagLines; i++)
{
lineTypeLen = strlen((char *) featureExtraTag[i].type);
lineValueLen = strlen((char *) featureExtraTag[i].value);
if (namePos+lineTypeLen+lineValueLen+7 > MAX_FEATUREEXTRATAGSTRING)
{
fprintf(stderr, "feature extra tag too long: %s", (const char *)featureExtraTag[i].value);
break;
}
if (namePos) strcpy(featureExtraTagString+(namePos++),",");
strcpy(featureExtraTagString+(namePos++), "\"");
strcpy(featureExtraTagString+namePos, (char *) featureExtraTag[i].type);
namePos += lineTypeLen;
strcpy(featureExtraTagString+namePos, "\"=>\"");
namePos += 4;
strcpy(featureExtraTagString+namePos, (char *) featureExtraTag[i].value);
namePos += lineValueLen;
strcpy(featureExtraTagString+(namePos++), "\"");
xmlFree(featureExtraTag[i].type);
xmlFree(featureExtraTag[i].value);
}
}
paramValues[7] = (const char *)featureExtraTagString;
if (xmlStrlen(feature.parentPlaceID) == 0)
paramValues[8] = "0";
else
paramValues[8] = (const char *)feature.parentPlaceID;
paramValues[9] = (const char *)feature.adminLevel;
paramValues[10] = (const char *)feature.houseNumber;
paramValues[11] = (const char *)feature.rankAddress;
paramValues[12] = (const char *)feature.rankSearch;
paramValues[13] = (const char *)feature.geometry;
if (strlen(paramValues[3]) && strlen(paramValues[13]))
{
if (verbose) fprintf(stderr, "placex_insert: %s\n", paramValues[0]);
res = PQexecPrepared(conn, "placex_insert", 14, paramValues, NULL, NULL, 0);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "index_placex: INSERT failed: %s", PQerrorMessage(conn));
fprintf(stderr, "index_placex: INSERT failed: %s %s %s", paramValues[0], paramValues[1], paramValues[2]);
PQclear(res);
exit(EXIT_FAILURE);
}
PQclear(res);
}
for (i = 0; i < featureAddressLines; i++)
{
// insert into place_address
paramValues[0] = (const char *)place_id;
paramValues[1] = (const char *)featureAddress[i].distance;
if (paramValues[1] == NULL || strlen(paramValues[1]) == 0) paramValues[1] = "0";
paramValues[2] = (const char *)featureAddress[i].type;
paramValues[3] = (const char *)featureAddress[i].id;
paramValues[4] = (const char *)featureAddress[i].key;
paramValues[5] = (const char *)featureAddress[i].value;
paramValues[6] = (const char *)featureAddress[i].isAddress;
if (verbose) fprintf(stderr, "placex_insert: %s %s\n", paramValues[2], paramValues[3]);
res = PQexecPrepared(conn, "place_addressline_insert", 7, paramValues, NULL, NULL, 0);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "place_addressline_insert: INSERT failed: %s", PQerrorMessage(conn));
fprintf(stderr, "(%s,%s,%s,%s,%s,%s,%s)",paramValues[0],paramValues[1],paramValues[2],paramValues[3],paramValues[4],paramValues[5],paramValues[6]);
PQclear(res);
exit(EXIT_FAILURE);
}
PQclear(res);
xmlFree(featureAddress[i].type);
xmlFree(featureAddress[i].id);
xmlFree(featureAddress[i].key);
xmlFree(featureAddress[i].value);
xmlFree(featureAddress[i].distance);
}
if (featureNameLines)
{
if (xmlStrlen(feature.parentPlaceID) > 0 && featureAddressLines == 0)
{
paramValues[0] = (const char *)place_id;
paramValues[1] = (const char *)feature.parentPlaceID;
if (verbose) fprintf(stderr, "search_name_from_parent_insert: INSERT %s %s\n", paramValues[0], paramValues[1]);
res = PQexecPrepared(conn, "search_name_from_parent_insert", 2, paramValues, NULL, NULL, 0);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "search_name_from_parent_insert: INSERT failed: %s", PQerrorMessage(conn));
PQclear(res);
exit(EXIT_FAILURE);
}
PQclear(res);
}
else
{
paramValues[0] = (const char *)place_id;
if (verbose) fprintf(stderr, "search_name_insert: INSERT %s\n", paramValues[0]);
res = PQexecPrepared(conn, "search_name_insert", 1, paramValues, NULL, NULL, 0);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "search_name_insert: INSERT failed: %s", PQerrorMessage(conn));
PQclear(res);
exit(EXIT_FAILURE);
}
PQclear(res);
}
}
partionQueryName = xmlHashLookup2(partionTableTagsHash, feature.key, feature.value);
if (partionQueryName)
{
// insert into partition table
paramValues[0] = (const char *)place_id;
paramValues[1] = (const char *)feature.geometry;
res = PQexecPrepared(conn, partionQueryName, 2, paramValues, NULL, NULL, 0);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "%s: INSERT failed: %s", partionQueryName, PQerrorMessage(conn));
PQclear(res);
exit(EXIT_FAILURE);
}
PQclear(res);
}
}
else
{
for (i = 0; i < featureAddressLines; i++)
{
xmlFree(featureAddress[i].type);
xmlFree(featureAddress[i].id);
xmlFree(featureAddress[i].key);
xmlFree(featureAddress[i].value);
xmlFree(featureAddress[i].distance);
}
}
xmlFree(feature.placeID);
xmlFree(feature.type);
xmlFree(feature.id);
xmlFree(feature.key);
xmlFree(feature.value);
xmlFree(feature.rankAddress);
xmlFree(feature.rankSearch);
if (feature.countryCode) xmlFree(feature.countryCode);
if (feature.parentPlaceID) xmlFree(feature.parentPlaceID);
if (feature.parentType) xmlFree(feature.parentType);
if (feature.parentID) xmlFree(feature.parentID);
// if (feature.name) xmlFree(feature.name);
if (feature.adminLevel) xmlFree(feature.adminLevel);
if (feature.houseNumber) xmlFree(feature.houseNumber);
if (feature.geometry) xmlFree(feature.geometry);
// PQclear(resPlaceID);
}
}
static void processNode(xmlTextReaderPtr reader)
{
xmlChar *name;
name = xmlTextReaderName(reader);
if (name == NULL)
{
name = xmlStrdup(BAD_CAST "--");
}
switch (xmlTextReaderNodeType(reader))
{
case XML_READER_TYPE_ELEMENT:
StartElement(reader, name);
if (xmlTextReaderIsEmptyElement(reader))
EndElement(reader, name); /* No end_element for self closing tags! */
break;
case XML_READER_TYPE_END_ELEMENT:
EndElement(reader, name);
break;
case XML_READER_TYPE_TEXT:
case XML_READER_TYPE_CDATA:
case XML_READER_TYPE_SIGNIFICANT_WHITESPACE:
/* Ignore */
break;
default:
fprintf(stderr, "Unknown node type %d\n", xmlTextReaderNodeType(reader));
break;
}
xmlFree(name);
}
int nominatim_import(const char *conninfo, const char *partionTagsFilename, const char *filename)
{
xmlTextReaderPtr reader;
int ret = 0;
PGresult * res;
FILE * partionTagsFile;
char * partionQueryName;
char partionQuerySQL[1024];
conn = PQconnectdb(conninfo);
if (PQstatus(conn) != CONNECTION_OK)
{
fprintf(stderr, "Connection to database failed: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
partionTableTagsHash = xmlHashCreate(200);
partionTableTagsHashDelete = xmlHashCreate(200);
partionTagsFile = fopen(partionTagsFilename, "rt");
if (!partionTagsFile)
{
fprintf(stderr, "Unable to read partition tags file: %s\n", partionTagsFilename);
exit(EXIT_FAILURE);
}
char buffer[1024], osmkey[256], osmvalue[256];
int fields;
while (fgets(buffer, sizeof(buffer), partionTagsFile) != NULL)
{
fields = sscanf( buffer, "%23s %63s", osmkey, osmvalue );
if ( fields <= 0 ) continue;
if ( fields != 2 )
{
fprintf( stderr, "Error partition file\n");
exit_nicely();
}
partionQueryName = malloc(strlen("partition_insert_")+strlen(osmkey)+strlen(osmvalue)+2);
strcpy(partionQueryName, "partition_insert_");
strcat(partionQueryName, osmkey);
strcat(partionQueryName, "_");
strcat(partionQueryName, osmvalue);
strcpy(partionQuerySQL, "insert into place_classtype_");
strcat(partionQuerySQL, osmkey);
strcat(partionQuerySQL, "_");
strcat(partionQuerySQL, osmvalue);
strcat(partionQuerySQL, " (place_id, centroid) values ($1, ST_Centroid(st_setsrid($2, 4326)))");
res = PQprepare(conn, partionQueryName, partionQuerySQL, 2, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed to prepare %s: %s\n", partionQueryName, PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
xmlHashAddEntry2(partionTableTagsHash, BAD_CAST osmkey, BAD_CAST osmvalue, BAD_CAST partionQueryName);
partionQueryName = malloc(strlen("partition_delete_")+strlen(osmkey)+strlen(osmvalue)+2);
strcpy(partionQueryName, "partition_delete_");
strcat(partionQueryName, osmkey);
strcat(partionQueryName, "_");
strcat(partionQueryName, osmvalue);
strcpy(partionQuerySQL, "delete from place_classtype_");
strcat(partionQuerySQL, osmkey);
strcat(partionQuerySQL, "_");
strcat(partionQuerySQL, osmvalue);
strcat(partionQuerySQL, " where place_id = $1::integer");
res = PQprepare(conn, partionQueryName, partionQuerySQL, 1, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed to prepare %s: %s\n", partionQueryName, PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
xmlHashAddEntry2(partionTableTagsHashDelete, BAD_CAST osmkey, BAD_CAST osmvalue, BAD_CAST partionQueryName);
}
res = PQprepare(conn, "get_new_place_id",
"select nextval('seq_place')",
0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed to prepare get_new_place_id: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
res = PQprepare(conn, "get_place_id",
"select place_id from placex where osm_type = $1 and osm_id = $2 and class = $3 and type = $4",
4, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed to prepare get_place_id: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
res = PQprepare(conn, "placex_insert",
"insert into placex (place_id,osm_type,osm_id,class,type,name,country_code,extratags,parent_place_id,admin_level,housenumber,rank_address,rank_search,geometry) "
"values ($1, $2, $3, $4, $5, $6, $7, $8, $9, $10, $11, $12, $13, st_setsrid($14, 4326))",
12, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed to prepare placex_insert: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
res = PQprepare(conn, "search_name_insert",
"insert into search_name (place_id, search_rank, address_rank, country_code, name_vector, nameaddress_vector, centroid) "
"select place_id, rank_search, rank_address, country_code, make_keywords(name), "
"(select uniq(sort(array_agg(parent_search_name.name_vector))) from search_name as parent_search_name where place_id in "
"(select distinct address_place_id from place_addressline where place_addressline.place_id = $1 limit 1000)"
"), st_centroid(geometry) from placex "
"where place_id = $1",
1, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed to prepare search_name_insert: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
res = PQprepare(conn, "search_name_from_parent_insert",
"insert into search_name (place_id, search_rank, address_rank, country_code, name_vector, nameaddress_vector, centroid) "
"select place_id, rank_search, rank_address, country_code, make_keywords(name), "
"(select uniq(sort(name_vector+nameaddress_vector)) from search_name as parent_search_name "
"where parent_search_name.place_id = $2 ), st_centroid(geometry) from placex "
"where place_id = $1",
2, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed to prepare search_name_insert: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
res = PQprepare(conn, "place_addressline_insert",
"insert into place_addressline (place_id, address_place_id, fromarea, isaddress, distance, cached_rank_address) "
"select $1, place_id, false, $7, $2, rank_address from placex where osm_type = $3 and osm_id = $4 and class = $5 and type = $6",
7, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed to prepare place_addressline_insert: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
res = PQprepare(conn, "placex_delete",
"delete from placex where place_id = $1",
1, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed to prepare placex_delete: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
res = PQprepare(conn, "search_name_delete",
"delete from search_name where place_id = $1",
1, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed to prepare search_name_delete: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
res = PQprepare(conn, "place_addressline_delete",
"delete from place_addressline where place_id = $1",
1, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed to prepare place_addressline_delete: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
featureCount = 0;
reader = inputUTF8(filename);
if (reader == NULL)
{
fprintf(stderr, "Unable to open %s\n", filename);
return 1;
}
ret = xmlTextReaderRead(reader);
while (ret == 1)
{
processNode(reader);
ret = xmlTextReaderRead(reader);
}
if (ret != 0)
{
fprintf(stderr, "%s : failed to parse\n", filename);
return ret;
}
xmlFreeTextReader(reader);
xmlHashFree(partionTableTagsHash, NULL);
xmlHashFree(partionTableTagsHashDelete, NULL);
return 0;
}

View File

@@ -1,6 +0,0 @@
#ifndef IMPORT_H
#define IMPORT_H
int nominatim_import(const char *conninfo, const char *partionTagsFilename, const char *filename);
#endif

View File

@@ -1,547 +0,0 @@
/*
* triggers indexing (reparenting etc.) through setting resetting indexed_status: update placex/osmline set indexed_status = 0 where indexed_status > 0
* triggers placex_update and osmline_update
*/
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <string.h>
#include <assert.h>
#include <pthread.h>
#include <time.h>
#include <stdint.h>
#include <libpq-fe.h>
#include "nominatim.h"
#include "index.h"
#include "export.h"
#include "postgresql.h"
extern int verbose;
void run_indexing(int rank, int interpolation, PGconn *conn, int num_threads,
struct index_thread_data * thread_data, const char *structuredoutputfile)
{
int tuples, count, sleepcount;
pthread_mutex_t count_mutex = PTHREAD_MUTEX_INITIALIZER;
time_t rankStartTime;
int rankTotalTuples;
int rankCountTuples;
float rankPerSecond;
PGresult * resSectors;
PGresult * resPlaces;
PGresult * resNULL;
int i;
int iSector;
int iResult;
const char *paramValues[2];
int paramLengths[2];
int paramFormats[2];
uint32_t paramRank;
uint32_t paramSector;
uint32_t sector;
xmlTextWriterPtr writer;
pthread_mutex_t writer_mutex = PTHREAD_MUTEX_INITIALIZER;
// Create the output file
writer = NULL;
if (structuredoutputfile)
{
writer = nominatim_exportXMLStart(structuredoutputfile);
}
if (interpolation)
{
fprintf(stderr, "Starting interpolation lines (location_property_osmline)\n");
}
else
{
fprintf(stderr, "Starting rank %d\n", rank);
}
rankCountTuples = 0;
rankPerSecond = 0;
paramRank = PGint32(rank);
paramValues[0] = (char *)&paramRank;
paramLengths[0] = sizeof(paramRank);
paramFormats[0] = 1;
if (interpolation)
{
resSectors = PQexecPrepared(conn, "index_sectors_osmline", 0, NULL, 0, NULL, 1);
}
else
{
resSectors = PQexecPrepared(conn, "index_sectors", 1, paramValues, paramLengths, paramFormats, 1);
}
if (PQresultStatus(resSectors) != PGRES_TUPLES_OK)
{
fprintf(stderr, "index_sectors: SELECT failed: %s", PQerrorMessage(conn));
PQclear(resSectors);
exit(EXIT_FAILURE);
}
if (PQftype(resSectors, 0) != PG_OID_INT4)
{
fprintf(stderr, "Sector value has unexpected type\n");
PQclear(resSectors);
exit(EXIT_FAILURE);
}
if (PQftype(resSectors, 1) != PG_OID_INT8)
{
fprintf(stderr, "Sector value has unexpected type\n");
PQclear(resSectors);
exit(EXIT_FAILURE);
}
rankTotalTuples = 0;
for (iSector = 0; iSector < PQntuples(resSectors); iSector++)
{
rankTotalTuples += PGint64(*((uint64_t *)PQgetvalue(resSectors, iSector, 1)));
}
rankStartTime = time(0);
for (iSector = 0; iSector <= PQntuples(resSectors); iSector++)
{
if (iSector > 0)
{
resPlaces = PQgetResult(conn);
if (PQresultStatus(resPlaces) != PGRES_TUPLES_OK)
{
fprintf(stderr, "index_sector_places: SELECT failed: %s", PQerrorMessage(conn));
PQclear(resPlaces);
exit(EXIT_FAILURE);
}
if (PQftype(resPlaces, 0) != PG_OID_INT8)
{
fprintf(stderr, "Place_id value has unexpected type\n");
PQclear(resPlaces);
exit(EXIT_FAILURE);
}
resNULL = PQgetResult(conn);
if (resNULL != NULL)
{
fprintf(stderr, "Unexpected non-null response\n");
exit(EXIT_FAILURE);
}
}
if (iSector < PQntuples(resSectors))
{
sector = PGint32(*((uint32_t *)PQgetvalue(resSectors, iSector, 0)));
// fprintf(stderr, "\n Starting sector %d size %ld\n", sector, PGint64(*((uint64_t *)PQgetvalue(resSectors, iSector, 1))));
// Get all the place_id's for this sector
paramRank = PGint32(rank);
paramSector = PGint32(sector);
if (rankTotalTuples-rankCountTuples < num_threads*1000)
{
// no sectors
if (interpolation)
{
iResult = PQsendQueryPrepared(conn, "index_nosector_places_osmline", 0, NULL, 0, NULL, 1);
}
else
{
paramValues[0] = (char *)&paramRank;
paramLengths[0] = sizeof(paramRank);
paramFormats[0] = 1;
iResult = PQsendQueryPrepared(conn, "index_nosector_places", 1, paramValues, paramLengths, paramFormats, 1);
}
}
else
{
if (interpolation)
{
iResult = PQsendQueryPrepared(conn, "index_sector_places_osmline", 1, paramValues, paramLengths, paramFormats, 1);
paramValues[0] = (char *)&paramSector;
paramLengths[0] = sizeof(paramSector);
paramFormats[0] = 1;
}
else
{
paramValues[0] = (char *)&paramRank;
paramLengths[0] = sizeof(paramRank);
paramFormats[0] = 1;
paramValues[1] = (char *)&paramSector;
paramLengths[1] = sizeof(paramSector);
paramFormats[1] = 1;
iResult = PQsendQueryPrepared(conn, "index_sector_places", 2, paramValues, paramLengths, paramFormats, 1);
}
}
if (!iResult)
{
fprintf(stderr, "index_sector_places: SELECT failed: %s", PQerrorMessage(conn));
PQclear(resPlaces);
exit(EXIT_FAILURE);
}
}
if (iSector > 0)
{
count = 0;
rankPerSecond = 0;
tuples = PQntuples(resPlaces);
if (tuples > 0)
{
// Spawn threads
for (i = 0; i < num_threads; i++)
{
thread_data[i].res = resPlaces;
thread_data[i].tuples = tuples;
thread_data[i].count = &count;
thread_data[i].count_mutex = &count_mutex;
thread_data[i].writer = writer;
thread_data[i].writer_mutex = &writer_mutex;
if (interpolation)
{
thread_data[i].table = 0; // use interpolations table
}
else
{
thread_data[i].table = 1; // use placex table
}
pthread_create(&thread_data[i].thread, NULL, &nominatim_indexThread, (void *)&thread_data[i]);
}
// Monitor threads to give user feedback
sleepcount = 0;
while (count < tuples)
{
usleep(1000);
// Aim for one update per second
if (sleepcount++ > 1000)
{
rankPerSecond = ((float)rankCountTuples + (float)count) / MAX(difftime(time(0), rankStartTime),1);
if(interpolation)
{
fprintf(stderr, " Done %i in %i @ %f per second - Interpolation lines ETA (seconds): %f\n", (rankCountTuples + count), (int)(difftime(time(0), rankStartTime)), rankPerSecond, ((float)(rankTotalTuples - (rankCountTuples + count)))/rankPerSecond);
}
else
{
fprintf(stderr, " Done %i in %i @ %f per second - Rank %i ETA (seconds): %f\n", (rankCountTuples + count), (int)(difftime(time(0), rankStartTime)), rankPerSecond, rank, ((float)(rankTotalTuples - (rankCountTuples + count)))/rankPerSecond);
}
sleepcount = 0;
}
}
// Wait for everything to finish
for (i = 0; i < num_threads; i++)
{
pthread_join(thread_data[i].thread, NULL);
}
rankCountTuples += tuples;
}
// Finished sector
rankPerSecond = (float)rankCountTuples / MAX(difftime(time(0), rankStartTime),1);
fprintf(stderr, " Done %i in %i @ %f per second - ETA (seconds): %f\n", rankCountTuples, (int)(difftime(time(0), rankStartTime)), rankPerSecond, ((float)(rankTotalTuples - rankCountTuples))/rankPerSecond);
PQclear(resPlaces);
}
if (rankTotalTuples-rankCountTuples < num_threads*20 && iSector < PQntuples(resSectors))
{
iSector = PQntuples(resSectors) - 1;
}
}
// Finished rank
fprintf(stderr, "\r Done %i in %i @ %f per second - FINISHED\n\n", rankCountTuples, (int)(difftime(time(0), rankStartTime)), rankPerSecond);
PQclear(resSectors);
}
void nominatim_index(int rank_min, int rank_max, int num_threads, const char *conninfo, const char *structuredoutputfile)
{
struct index_thread_data *thread_data;
PGconn *conn;
PGresult *res;
int num_rows = 0, status_code = 0;
int db_has_locale = 0;
char *result_string = NULL;
int rank;
int i;
xmlTextWriterPtr writer;
pthread_mutex_t writer_mutex = PTHREAD_MUTEX_INITIALIZER;
Oid pg_prepare_params[2];
conn = PQconnectdb(conninfo);
if (PQstatus(conn) != CONNECTION_OK)
{
fprintf(stderr, "Connection to database failed: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
res = PQexec(conn, "SHOW lc_messages");
status_code = PQresultStatus(res);
if (status_code != PGRES_TUPLES_OK && status_code != PGRES_SINGLE_TUPLE) {
fprintf(stderr, "Failed determining database locale: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
num_rows = PQntuples(res);
if (num_rows > 0)
{
result_string = PQgetvalue(res, 0, 0);
if (result_string && (strlen(result_string) > 0) && (strcasecmp(result_string, "C") != 0))
{
// non-default locale if the result exists, is non-empty, and is not "C"
db_has_locale = 1;
}
}
pg_prepare_params[0] = PG_OID_INT4;
res = PQprepare(conn, "index_sectors",
"select geometry_sector,count(*) from placex where rank_search = $1 and indexed_status > 0 group by geometry_sector order by geometry_sector",
1, pg_prepare_params);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed preparing index_sectors: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
PQclear(res);
res = PQprepare(conn, "index_sectors_osmline",
"select geometry_sector,count(*) from location_property_osmline where indexed_status > 0 group by geometry_sector order by geometry_sector",
0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed preparing index_sectors: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
PQclear(res);
pg_prepare_params[0] = PG_OID_INT4;
res = PQprepare(conn, "index_nosectors",
"select 0::integer,count(*) from placex where rank_search = $1 and indexed_status > 0",
1, pg_prepare_params);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed preparing index_sectors: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
PQclear(res);
pg_prepare_params[0] = PG_OID_INT4;
pg_prepare_params[1] = PG_OID_INT4;
res = PQprepare(conn, "index_sector_places",
"select place_id from placex where rank_search = $1 and geometry_sector = $2 and indexed_status > 0",
2, pg_prepare_params);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed preparing index_sector_places: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
PQclear(res);
pg_prepare_params[0] = PG_OID_INT4;
res = PQprepare(conn, "index_nosector_places",
"select place_id from placex where rank_search = $1 and indexed_status > 0 order by geometry_sector",
1, pg_prepare_params);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed preparing index_nosector_places: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
PQclear(res);
pg_prepare_params[0] = PG_OID_INT4;
res = PQprepare(conn, "index_sector_places_osmline",
"select place_id from location_property_osmline where geometry_sector = $1 and indexed_status > 0",
1, pg_prepare_params);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed preparing index_sector_places: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
PQclear(res);
res = PQprepare(conn, "index_nosector_places_osmline",
"select place_id from location_property_osmline where indexed_status > 0 order by geometry_sector",
0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed preparing index_nosector_places: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
PQclear(res);
// Build the data for each thread
thread_data = (struct index_thread_data *)malloc(sizeof(struct index_thread_data)*num_threads);
for (i = 0; i < num_threads; i++)
{
thread_data[i].conn = PQconnectdb(conninfo);
if (PQstatus(thread_data[i].conn) != CONNECTION_OK)
{
fprintf(stderr, "Connection to database failed: %s\n", PQerrorMessage(thread_data[i].conn));
exit(EXIT_FAILURE);
}
pg_prepare_params[0] = PG_OID_INT8;
res = PQprepare(thread_data[i].conn, "index_placex",
"update placex set indexed_status = 0 where place_id = $1",
1, pg_prepare_params);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed preparing index_placex: %s\n", PQerrorMessage(thread_data[i].conn));
exit(EXIT_FAILURE);
}
PQclear(res);
pg_prepare_params[0] = PG_OID_INT8;
res = PQprepare(thread_data[i].conn, "index_osmline",
"update location_property_osmline set indexed_status = 0 where place_id = $1",
1, pg_prepare_params);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed preparing index_osmline: %s\n", PQerrorMessage(thread_data[i].conn));
exit(EXIT_FAILURE);
}
PQclear(res);
if (db_has_locale)
{
// Make sure the error message is not localized as we parse it later.
res = PQexec(thread_data[i].conn, "SET lc_messages TO 'C'");
if (PQresultStatus(res) != PGRES_COMMAND_OK)
{
fprintf(stderr, "Failed to set langauge: %s\n", PQerrorMessage(thread_data[i].conn));
exit(EXIT_FAILURE);
}
PQclear(res);
}
nominatim_exportCreatePreparedQueries(thread_data[i].conn);
}
fprintf(stderr, "Starting indexing rank (%i to %i) using %i threads\n", rank_min, rank_max, num_threads);
for (rank = rank_min; rank <= rank_max; rank++)
{
// OSMLINE: do reindexing (=> reparenting) for interpolation lines at rank 30, but before all other objects of rank 30
// reason: houses (rank 30) depend on the updated interpolation line, when reparenting (see placex_update in functions.sql)
if (rank == 30)
{
run_indexing(rank, 1, conn, num_threads, thread_data, structuredoutputfile);
}
run_indexing(rank, 0, conn, num_threads, thread_data, structuredoutputfile);
}
// Close all connections
for (i = 0; i < num_threads; i++)
{
PQfinish(thread_data[i].conn);
}
PQfinish(conn);
}
void *nominatim_indexThread(void * thread_data_in)
{
struct index_thread_data * thread_data = (struct index_thread_data * )thread_data_in;
struct export_data querySet;
PGresult *res;
const char *paramValues[1];
int paramLengths[1];
int paramFormats[1];
uint64_t paramPlaceID;
uint64_t place_id;
time_t updateStartTime;
unsigned table;
table = thread_data->table;
while (1)
{
pthread_mutex_lock( thread_data->count_mutex );
if (*(thread_data->count) >= thread_data->tuples)
{
pthread_mutex_unlock( thread_data->count_mutex );
break;
}
place_id = PGint64(*((uint64_t *)PQgetvalue(thread_data->res, *thread_data->count, 0)));
(*thread_data->count)++;
pthread_mutex_unlock( thread_data->count_mutex );
if (verbose) fprintf(stderr, " Processing place_id %ld\n", place_id);
updateStartTime = time(0);
int done = 0;
if (thread_data->writer)
{
nominatim_exportPlaceQueries(place_id, thread_data->conn, &querySet);
}
while(!done)
{
paramPlaceID = PGint64(place_id);
paramValues[0] = (char *)&paramPlaceID;
paramLengths[0] = sizeof(paramPlaceID);
paramFormats[0] = 1;
if (table == 1) // table=1 for placex
{
res = PQexecPrepared(thread_data->conn, "index_placex", 1, paramValues, paramLengths, paramFormats, 1);
}
else // table=0 for osmline
{
res = PQexecPrepared(thread_data->conn, "index_osmline", 1, paramValues, paramLengths, paramFormats, 1);
}
if (PQresultStatus(res) == PGRES_COMMAND_OK)
done = 1;
else
{
if (!strncmp(PQerrorMessage(thread_data->conn), "ERROR: deadlock detected", 25))
{
if (table == 1)
{
fprintf(stderr, "index_placex: UPDATE failed - deadlock, retrying (%ld)\n", place_id);
}
else
{
fprintf(stderr, "index_osmline: UPDATE failed - deadlock, retrying (%ld)\n", place_id);
}
PQclear(res);
sleep(rand() % 10);
}
else
{
if (table == 1)
{
fprintf(stderr, "index_placex: UPDATE failed: %s", PQerrorMessage(thread_data->conn));
}
else
{
fprintf(stderr, "index_osmline: UPDATE failed: %s", PQerrorMessage(thread_data->conn));
}
PQclear(res);
exit(EXIT_FAILURE);
}
}
}
PQclear(res);
if (difftime(time(0), updateStartTime) > 1) fprintf(stderr, " Slow place_id %ld\n", place_id);
if (thread_data->writer)
{
nominatim_exportPlace(place_id, thread_data->conn, thread_data->writer, thread_data->writer_mutex, &querySet);
nominatim_exportFreeQueries(&querySet);
}
}
return NULL;
}

View File

@@ -1,22 +0,0 @@
#ifndef INDEX_H
#define INDEX_H
#include <libxml/encoding.h>
#include <libxml/xmlwriter.h>
struct index_thread_data
{
pthread_t thread;
PGconn * conn;
PGresult * res;
int tuples;
int * count;
pthread_mutex_t * count_mutex;
xmlTextWriterPtr writer;
pthread_mutex_t * writer_mutex;
unsigned table;
};
void nominatim_index(int rank_min, int rank_max, int num_threads, const char *conninfo, const char *structuredoutputfile);
void *nominatim_indexThread(void * thread_data_in);
#endif

View File

@@ -0,0 +1,52 @@
# SPDX-License-Identifier: GPL-2.0-only
#
# This file is part of Nominatim.
# Copyright (C) 2020 Sarah Hoffmann
import logging
from datetime import datetime
log = logging.getLogger()
class ProgressLogger(object):
""" Tracks and prints progress for the indexing process.
`name` is the name of the indexing step being tracked.
`total` sets up the total number of items that need processing.
`log_interval` denotes the interval in seconds at which progres
should be reported.
"""
def __init__(self, name, total, log_interval=1):
self.name = name
self.total_places = total
self.done_places = 0
self.rank_start_time = datetime.now()
self.next_info = 100 if log.isEnabledFor(logging.INFO) else total + 1
def add(self, num=1):
""" Mark `num` places as processed. Print a log message if the
logging is at least info and the log interval has past.
"""
self.done_places += num
if self.done_places >= self.next_info:
now = datetime.now()
done_time = (now - self.rank_start_time).total_seconds()
places_per_sec = self.done_places / done_time
eta = (self.total_places - self.done_places)/places_per_sec
log.info("Done {} in {} @ {:.3f} per second - {} ETA (seconds): {:.2f}"
.format(self.done_places, int(done_time),
places_per_sec, self.name, eta))
self.next_info += int(places_per_sec)
def done(self):
""" Print final staticstics about the progress.
"""
rank_end_time = datetime.now()
diff_seconds = (rank_end_time-self.rank_start_time).total_seconds()
log.warning("Done {}/{} in {} @ {:.3f} per second - FINISHED {}\n".format(
self.done_places, self.total_places, int(diff_seconds),
self.done_places/diff_seconds, self.name))

View File

@@ -1,242 +0,0 @@
#define _FILE_OFFSET_BITS 64
#define _LARGEFILE64_SOURCE
#ifdef __MINGW_H
# include <windows.h>
#else
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <string.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <zlib.h>
#endif
#include <libxml/xmlreader.h>
#include <bzlib.h>
#include "input.h"
struct Input
{
char *name;
enum { plainFile, gzipFile, bzip2File } type;
void *fileHandle;
// needed by bzip2 when decompressing from multiple streams. other
// decompressors must ignore it.
FILE *systemHandle;
int eof;
char buf[4096];
int buf_ptr, buf_fill;
};
// tries to re-open the bz stream at the next stream start.
// returns 0 on success, -1 on failure.
int bzReOpen(struct Input *ctx, int *error)
{
// for copying out the last unused part of the block which
// has an EOS token in it. needed for re-initialising the
// next stream.
unsigned char unused[BZ_MAX_UNUSED];
void *unused_tmp_ptr = NULL;
int nUnused, i;
BZ2_bzReadGetUnused(error, (BZFILE *)(ctx->fileHandle), &unused_tmp_ptr, &nUnused);
if (*error != BZ_OK) return -1;
// when bzReadClose is called the unused buffer is deallocated,
// so it needs to be copied somewhere safe first.
for (i = 0; i < nUnused; ++i)
unused[i] = ((unsigned char *)unused_tmp_ptr)[i];
BZ2_bzReadClose(error, (BZFILE *)(ctx->fileHandle));
if (*error != BZ_OK) return -1;
// reassign the file handle
ctx->fileHandle = BZ2_bzReadOpen(error, ctx->systemHandle, 0, 0, unused, nUnused);
if (ctx->fileHandle == NULL || *error != BZ_OK) return -1;
return 0;
}
int readFile(void *context, char * buffer, int len)
{
struct Input *ctx = context;
void *f = ctx->fileHandle;
int l = 0, error = 0;
if (ctx->eof || (len == 0))
return 0;
switch (ctx->type)
{
case plainFile:
l = read(*(int *)f, buffer, len);
if (l <= 0) ctx->eof = 1;
break;
case gzipFile:
l = gzread((gzFile)f, buffer, len);
if (l <= 0) ctx->eof = 1;
break;
case bzip2File:
l = BZ2_bzRead(&error, (BZFILE *)f, buffer, len);
// error codes BZ_OK and BZ_STREAM_END are both "OK", but the stream
// end means the reader needs to be reset from the original handle.
if (error != BZ_OK)
{
// for stream errors, try re-opening the stream before admitting defeat.
if (error != BZ_STREAM_END || bzReOpen(ctx, &error) != 0)
{
l = 0;
ctx->eof = 1;
}
}
break;
default:
fprintf(stderr, "Bad file type\n");
break;
}
if (l < 0)
{
fprintf(stderr, "File reader received error %d (%d)\n", l, error);
l = 0;
}
return l;
}
char inputGetChar(void *context)
{
struct Input *ctx = context;
if (ctx->buf_ptr == ctx->buf_fill)
{
ctx->buf_fill = readFile(context, &ctx->buf[0], sizeof(ctx->buf));
ctx->buf_ptr = 0;
if (ctx->buf_fill == 0)
return 0;
if (ctx->buf_fill < 0)
{
perror("Error while reading file");
exit(1);
}
}
//readFile(context, &c, 1);
return ctx->buf[ctx->buf_ptr++];
}
int inputEof(void *context)
{
return ((struct Input *)context)->eof;
}
void *inputOpen(const char *name)
{
const char *ext = strrchr(name, '.');
struct Input *ctx = malloc (sizeof(*ctx));
if (!ctx)
return NULL;
memset(ctx, 0, sizeof(*ctx));
ctx->name = strdup(name);
if (ext && !strcmp(ext, ".gz"))
{
ctx->fileHandle = (void *)gzopen(name, "rb");
ctx->type = gzipFile;
}
else if (ext && !strcmp(ext, ".bz2"))
{
int error = 0;
ctx->systemHandle = fopen(name, "rb");
if (!ctx->systemHandle)
{
fprintf(stderr, "error while opening file %s\n", name);
exit(10);
}
ctx->fileHandle = (void *)BZ2_bzReadOpen(&error, ctx->systemHandle, 0, 0, NULL, 0);
ctx->type = bzip2File;
}
else
{
int *pfd = malloc(sizeof(pfd));
if (pfd)
{
if (!strcmp(name, "-"))
{
*pfd = STDIN_FILENO;
}
else
{
int flags = O_RDONLY;
#ifdef O_LARGEFILE
flags |= O_LARGEFILE;
#endif
*pfd = open(name, flags);
if (*pfd < 0)
{
free(pfd);
pfd = NULL;
}
}
}
ctx->fileHandle = (void *)pfd;
ctx->type = plainFile;
}
if (!ctx->fileHandle)
{
fprintf(stderr, "error while opening file %s\n", name);
exit(10);
}
ctx->buf_ptr = 0;
ctx->buf_fill = 0;
return (void *)ctx;
}
int inputClose(void *context)
{
struct Input *ctx = context;
void *f = ctx->fileHandle;
switch (ctx->type)
{
case plainFile:
close(*(int *)f);
free(f);
break;
case gzipFile:
gzclose((gzFile)f);
break;
case bzip2File:
BZ2_bzclose((BZFILE *)f);
break;
default:
fprintf(stderr, "Bad file type\n");
break;
}
free(ctx->name);
free(ctx);
return 0;
}
xmlTextReaderPtr inputUTF8(const char *name)
{
void *ctx = inputOpen(name);
if (!ctx)
{
fprintf(stderr, "Input reader create failed for: %s\n", name);
return NULL;
}
return xmlReaderForIO(readFile, inputClose, (void *)ctx, NULL, NULL, 0);
}

View File

@@ -1,11 +0,0 @@
#ifndef INPUT_H
#define INPUT_H
int readFile(void *context, char * buffer, int len);
int inputClose(void *context);
void *inputOpen(const char *name);
char inputGetChar(void *context);
int inputEof(void *context);
xmlTextReaderPtr inputUTF8(const char *name);
#endif

View File

@@ -1,255 +0,0 @@
/*
#-----------------------------------------------------------------------------
# nominatim - [description]
#-----------------------------------------------------------------------------
# Copyright 2010, Brian Quinion
# Based on osm2pgsql
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#-----------------------------------------------------------------------------
*/
#define _GNU_SOURCE
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <string.h>
#include <assert.h>
#include <getopt.h>
#include <libgen.h>
#include <pthread.h>
#include <time.h>
#include <libpq-fe.h>
#include "nominatim.h"
#include "postgresql.h"
#include "sprompt.h"
#include "index.h"
#include "export.h"
#include "import.h"
int verbose;
void exit_nicely(void)
{
fprintf(stderr, "Error occurred, cleaning up\n");
exit(1);
}
void short_usage(char *arg0)
{
const char *name = basename(arg0);
fprintf(stderr, "Usage error. For further information see:\n");
fprintf(stderr, "\t%s -h|--help\n", name);
}
static void long_usage(char *arg0)
{
const char *name = basename(arg0);
fprintf(stderr, "Usage:\n");
fprintf(stderr, "\t%s [options] planet.osms\n", name);
fprintf(stderr, "\nThis will import the structured osm data into a PostgreSQL database\n");
fprintf(stderr, "suitable for nominatim search engine\n");
fprintf(stderr, "\nOptions:\n");
fprintf(stderr, " -d|--database\tThe name of the PostgreSQL database to connect\n");
fprintf(stderr, " \tto (default: nominatim).\n");
fprintf(stderr, " -U|--username\tPostgresql user name.\n");
fprintf(stderr, " -W|--password\tForce password prompt.\n");
fprintf(stderr, " -H|--host\t\tDatabase server hostname or socket location.\n");
fprintf(stderr, " -P|--port\t\tDatabase server port.\n");
fprintf(stderr, " -i|--index\t\tIndex the database.\n");
fprintf(stderr, " -e|--export\t\tGenerate a structured file.\n");
fprintf(stderr, " -I|--import\t\tImport a structured file.\n");
fprintf(stderr, " -r|--minrank\t\tMinimum / starting rank. (default: 0))\n");
fprintf(stderr, " -R|--maxrank\t\tMaximum / finishing rank. (default: 30)\n");
fprintf(stderr, " -t|--threads\t\tNumber of threads to create for indexing.\n");
fprintf(stderr, " -F|--file\t\tfile to use (either to import or export).\n");
fprintf(stderr, " -T|--tagfile\t\tfile containing 'special' tag pairs\n");
fprintf(stderr, " \t(default: partitionedtags.def).\n");
fprintf(stderr, " -h|--help\t\tHelp information.\n");
fprintf(stderr, " -v|--verbose\t\tVerbose output.\n");
fprintf(stderr, "\n");
if (sizeof(int*) == 4)
{
fprintf(stderr, "\n\nYou are running this on 32bit system - this will not work\n");
}
}
int main(int argc, char *argv[])
{
int long_usage_bool=0;
int pass_prompt=0;
const char *db = "nominatim";
const char *username=NULL;
const char *host=NULL;
const char *password=NULL;
const char *port = "5432";
const char *conninfo = NULL;
int index = 0;
int export = 0;
int import = 0;
int minrank = 0;
int maxrank = 30;
int threads = 1;
const char *file = NULL;
const char *tagsfile = "partitionedtags.def";
//import = 1;
//structuredinputfile = "out.osms";
PGconn *conn;
fprintf(stderr, "nominatim version %s\n\n", NOMINATIM_VERSION);
while (1)
{
int c, option_index = 0;
static struct option long_options[] =
{
{"help", 0, 0, 'h'},
{"verbose", 0, 0, 'v'},
{"database", 1, 0, 'd'},
{"username", 1, 0, 'U'},
{"password", 0, 0, 'W'},
{"host", 1, 0, 'H'},
{"port", 1, 0, 'P'},
{"index", 0, 0, 'i'},
{"export", 0, 0, 'e'},
{"import", 1, 0, 'I'},
{"threads", 1, 0, 't'},
{"file", 1, 0, 'F'},
{"tagsfile", 1, 0, 'T'},
{"minrank", 1, 0, 'r'},
{"maxrank", 1, 0, 'R'},
{0, 0, 0, 0}
};
c = getopt_long(argc, argv, "vhd:U:WH:P:ieIt:F:T:r:R:", long_options, &option_index);
if (c == -1)
break;
switch (c)
{
case 'v':
verbose=1;
break;
case 'd':
db=optarg;
break;
case 'U':
username=optarg;
break;
case 'W':
pass_prompt=1;
break;
case 'H':
host=optarg;
break;
case 'P':
port=optarg;
break;
case 'h':
long_usage_bool=1;
break;
case 'i':
index=1;
break;
case 'e':
export=1;
break;
case 'I':
import=1;
break;
case 't':
threads=atoi(optarg);
break;
case 'r':
minrank=atoi(optarg);
break;
case 'R':
maxrank=atoi(optarg);
break;
case 'F':
file=optarg;
break;
case 'T':
tagsfile=optarg;
break;
case '?':
default:
short_usage(argv[0]);
exit(EXIT_FAILURE);
}
}
if (long_usage_bool)
{
long_usage(argv[0]);
exit(EXIT_FAILURE);
}
if (threads < 1) threads = 1;
/*
if (argc == optind) { // No non-switch arguments
short_usage(argv[0]);
exit(EXIT_FAILURE);
}
*/
if (index && import)
{
fprintf(stderr, "Error: --index and --import options can not be used on the same database!\n");
exit(EXIT_FAILURE);
}
if (pass_prompt)
password = simple_prompt("Password:", 100, 0);
else
{
password = getenv("PGPASS");
}
// Test the database connection
conninfo = build_conninfo(db, username, password, host, port);
conn = PQconnectdb(conninfo);
if (PQstatus(conn) != CONNECTION_OK)
{
fprintf(stderr, "Connection to database failed: %s\n", PQerrorMessage(conn));
exit(EXIT_FAILURE);
}
PQfinish(conn);
if (!index && !export && !import)
{
fprintf(stderr, "Please select index, export or import.\n");
exit(EXIT_FAILURE);
}
if (index) nominatim_index(minrank, maxrank, threads, conninfo, file);
if (export) nominatim_export(minrank, maxrank, conninfo, file);
if (import) nominatim_import(conninfo, tagsfile, file);
return 0;
}

View File

@@ -1,31 +0,0 @@
#ifndef NOMINATIM_H
#define NOMINATIM_H
#define MAX(x,y) (x > y?x:y)
#define MIN(x,y) (x < y?x:y)
struct output_options
{
const char *conninfo; /* Connection info string */
const char *prefix; /* prefix for table names */
int scale; /* scale for converting coordinates to fixed point */
int projection; /* SRS of projection */
int append; /* Append to existing data */
int slim; /* In slim mode */
int cache; /* Memory usable for cache in MB */
struct middle_t *mid; /* Mid storage to use */
const char *tblsindex; /* Pg Tablespace to store indexes */
const char *style; /* style file to use */
int expire_tiles_zoom; /* Zoom level for tile expiry list */
int expire_tiles_zoom_min; /* Minimum zoom level for tile expiry list */
const char *expire_tiles_filename; /* File name to output expired tiles list to */
int enable_hstore; /* add an additional hstore column with objects key/value pairs */
int enable_multi; /* Output multi-geometries instead of several simple geometries */
char** hstore_columns; /* list of columns that should be written into their own hstore column */
int n_hstore_columns; /* number of hstore columns */
};
void exit_nicely(void);
void short_usage(char *arg0);
#endif

326
nominatim/nominatim.py Executable file
View File

@@ -0,0 +1,326 @@
#! /usr/bin/env python3
#-----------------------------------------------------------------------------
# nominatim - [description]
#-----------------------------------------------------------------------------
#
# Indexing tool for the Nominatim database.
#
# Based on C version by Brian Quinion
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
#-----------------------------------------------------------------------------
from argparse import ArgumentParser, RawDescriptionHelpFormatter, ArgumentTypeError
import logging
import sys
import re
import getpass
from datetime import datetime
import psycopg2
from psycopg2.extras import wait_select
import select
from indexer.progress import ProgressLogger
log = logging.getLogger()
def make_connection(options, asynchronous=False):
params = {'dbname' : options.dbname,
'user' : options.user,
'password' : options.password,
'host' : options.host,
'port' : options.port,
'async' : asynchronous}
return psycopg2.connect(**params)
class RankRunner(object):
""" Returns SQL commands for indexing one rank within the placex table.
"""
def __init__(self, rank):
self.rank = rank
def name(self):
return "rank {}".format(self.rank)
def sql_count_objects(self):
return """SELECT count(*) FROM placex
WHERE rank_search = {} and indexed_status > 0
""".format(self.rank)
def sql_get_objects(self):
return """SELECT place_id FROM placex
WHERE indexed_status > 0 and rank_search = {}
ORDER BY geometry_sector""".format(self.rank)
def sql_index_place(self, ids):
return "UPDATE placex SET indexed_status = 0 WHERE place_id IN ({})"\
.format(','.join((str(i) for i in ids)))
class InterpolationRunner(object):
""" Returns SQL commands for indexing the address interpolation table
location_property_osmline.
"""
def name(self):
return "interpolation lines (location_property_osmline)"
def sql_count_objects(self):
return """SELECT count(*) FROM location_property_osmline
WHERE indexed_status > 0"""
def sql_get_objects(self):
return """SELECT place_id FROM location_property_osmline
WHERE indexed_status > 0
ORDER BY geometry_sector"""
def sql_index_place(self, ids):
return """UPDATE location_property_osmline
SET indexed_status = 0 WHERE place_id IN ({})"""\
.format(','.join((str(i) for i in ids)))
class DBConnection(object):
""" A single non-blocking database connection.
"""
def __init__(self, options):
self.current_query = None
self.current_params = None
self.conn = None
self.connect()
def connect(self):
if self.conn is not None:
self.cursor.close()
self.conn.close()
self.conn = make_connection(options, asynchronous=True)
self.wait()
self.cursor = self.conn.cursor()
# Disable JIT and parallel workers as they are known to cause problems.
# Update pg_settings instead of using SET because it does not yield
# errors on older versions of Postgres where the settings are not
# implemented.
self.perform(
""" UPDATE pg_settings SET setting = -1 WHERE name = 'jit_above_cost';
UPDATE pg_settings SET setting = 0
WHERE name = 'max_parallel_workers_per_gather';""")
self.wait()
def wait(self):
""" Block until any pending operation is done.
"""
while True:
try:
wait_select(self.conn)
self.current_query = None
return
except psycopg2.extensions.TransactionRollbackError as e:
if e.pgcode == '40P01':
log.info("Deadlock detected (params = {}), retry."
.format(self.current_params))
self.cursor.execute(self.current_query, self.current_params)
else:
raise
except psycopg2.errors.DeadlockDetected:
self.cursor.execute(self.current_query, self.current_params)
def perform(self, sql, args=None):
""" Send SQL query to the server. Returns immediately without
blocking.
"""
self.current_query = sql
self.current_params = args
self.cursor.execute(sql, args)
def fileno(self):
""" File descriptor to wait for. (Makes this class select()able.)
"""
return self.conn.fileno()
def is_done(self):
""" Check if the connection is available for a new query.
Also checks if the previous query has run into a deadlock.
If so, then the previous query is repeated.
"""
if self.current_query is None:
return True
try:
if self.conn.poll() == psycopg2.extensions.POLL_OK:
self.current_query = None
return True
except psycopg2.extensions.TransactionRollbackError as e:
if e.pgcode == '40P01':
log.info("Deadlock detected (params = {}), retry.".format(self.current_params))
self.cursor.execute(self.current_query, self.current_params)
else:
raise
except psycopg2.errors.DeadlockDetected:
self.cursor.execute(self.current_query, self.current_params)
return False
class Indexer(object):
""" Main indexing routine.
"""
def __init__(self, options):
self.minrank = max(0, options.minrank)
self.maxrank = min(30, options.maxrank)
self.conn = make_connection(options)
self.threads = [DBConnection(options) for i in range(options.threads)]
def run(self):
""" Run indexing over the entire database.
"""
log.warning("Starting indexing rank ({} to {}) using {} threads".format(
self.minrank, self.maxrank, len(self.threads)))
for rank in range(self.minrank, self.maxrank):
self.index(RankRunner(rank))
if self.maxrank == 30:
self.index(InterpolationRunner(), 20)
self.index(RankRunner(self.maxrank), 20)
def index(self, obj, batch=1):
""" Index a single rank or table. `obj` describes the SQL to use
for indexing. `batch` describes the number of objects that
should be processed with a single SQL statement
"""
log.warning("Starting {}".format(obj.name()))
cur = self.conn.cursor()
cur.execute(obj.sql_count_objects())
total_tuples = cur.fetchone()[0]
log.debug("Total number of rows: {}".format(total_tuples))
cur.close()
next_thread = self.find_free_thread()
progress = ProgressLogger(obj.name(), total_tuples)
cur = self.conn.cursor(name='places')
cur.execute(obj.sql_get_objects())
while True:
places = [p[0] for p in cur.fetchmany(batch)]
if len(places) == 0:
break
log.debug("Processing places: {}".format(places))
thread = next(next_thread)
thread.perform(obj.sql_index_place(places))
progress.add(len(places))
cur.close()
for t in self.threads:
t.wait()
progress.done()
def find_free_thread(self):
""" Generator that returns the next connection that is free for
sending a query.
"""
ready = self.threads
command_stat = 0
while True:
for thread in ready:
if thread.is_done():
command_stat += 1
yield thread
# refresh the connections occasionaly to avoid potential
# memory leaks in Postgresql.
if command_stat > 100000:
for t in self.threads:
while not t.is_done():
t.wait()
t.connect()
command_stat = 0
ready = self.threads
else:
ready, _, _ = select.select(self.threads, [], [])
assert False, "Unreachable code"
def nominatim_arg_parser():
""" Setup the command-line parser for the tool.
"""
def h(s):
return re.sub("\s\s+" , " ", s)
p = ArgumentParser(description="Indexing tool for Nominatim.",
formatter_class=RawDescriptionHelpFormatter)
p.add_argument('-d', '--database',
dest='dbname', action='store', default='nominatim',
help='Name of the PostgreSQL database to connect to.')
p.add_argument('-U', '--username',
dest='user', action='store',
help='PostgreSQL user name.')
p.add_argument('-W', '--password',
dest='password_prompt', action='store_true',
help='Force password prompt.')
p.add_argument('-H', '--host',
dest='host', action='store',
help='PostgreSQL server hostname or socket location.')
p.add_argument('-P', '--port',
dest='port', action='store',
help='PostgreSQL server port')
p.add_argument('-r', '--minrank',
dest='minrank', type=int, metavar='RANK', default=0,
help='Minimum/starting rank.')
p.add_argument('-R', '--maxrank',
dest='maxrank', type=int, metavar='RANK', default=30,
help='Maximum/finishing rank.')
p.add_argument('-t', '--threads',
dest='threads', type=int, metavar='NUM', default=1,
help='Number of threads to create for indexing.')
p.add_argument('-v', '--verbose',
dest='loglevel', action='count', default=0,
help='Increase verbosity')
return p
if __name__ == '__main__':
logging.basicConfig(stream=sys.stderr, format='%(levelname)s: %(message)s')
options = nominatim_arg_parser().parse_args(sys.argv[1:])
log.setLevel(max(3 - options.loglevel, 0) * 10)
options.password = None
if options.password_prompt:
password = getpass.getpass("Database password: ")
options.password = password
Indexer(options).run()

View File

@@ -1,55 +0,0 @@
%define svn @SVN@
Summary: Nominatim OpenStreetMap geocoding database
Name: @PACKAGE@
Group: Applications/Text
Version: @VERSION@
Release: 1.%{svn}%{?dist}
License: GPL
URL: http://svn.openstreetmap.org/applications/utils/nominatim
Source0: %{name}-%{version}-%{svn}.tar.bz2
Source1: nominatim-svn.sh
BuildRoot: %{_tmppath}/%{name}-%{version}-%{release}-root-%(%{__id_u} -n)
BuildRequires: geos-devel
BuildRequires: libxml2-devel
BuildRequires: postgresql-devel
BuildRequires: bzip2-devel
BuildRequires: proj-devel
%description
Processes data imported using osm2pgsql from the communtiy mapping project
at http://www.openstreetmap.org.
%prep
%setup -q -n %{name}
%build
export CFLAGS="$RPM_OPT_FLAGS"
export CXXFLAGS="$RPM_OPT_FLAGS"
make all
%install
rm -rf $RPM_BUILD_ROOT
install -D -p nominatim $RPM_BUILD_ROOT/usr/bin/nominatim
%clean
rm -rf $RPM_BUILD_ROOT
%files
%defattr(-,root,root)
%doc README.txt
%{_bindir}/nominatim
%changelog
* Fri Sep 09 2010 Brian Quinion <nominatim@brian.quinion.co.uk> 0.1-1.20070316svn
- Initial build

View File

@@ -1,41 +0,0 @@
/*
*/
#include <string.h>
#include "postgresql.h"
const char *build_conninfo(const char *db, const char *username, const char *password, const char *host, const char *port)
{
static char conninfo[1024];
conninfo[0]='\0';
strcat(conninfo, "dbname='");
strcat(conninfo, db);
strcat(conninfo, "'");
if (username)
{
strcat(conninfo, " user='");
strcat(conninfo, username);
strcat(conninfo, "'");
}
if (password)
{
strcat(conninfo, " password='");
strcat(conninfo, password);
strcat(conninfo, "'");
}
if (host)
{
strcat(conninfo, " host='");
strcat(conninfo, host);
strcat(conninfo, "'");
}
if (port)
{
strcat(conninfo, " port='");
strcat(conninfo, port);
strcat(conninfo, "'");
}
return conninfo;
}

View File

@@ -1,40 +0,0 @@
/*
*/
#ifndef POSTGRESQL_H
#define POSTGRESQL_H
#define PG_OID_INT8 20
#define PG_OID_INT4 23
#if HAVE_BYTESWAP
#include <byteswap.h>
#define PG_BSWAP32(x) bswap_32(x)
#define PG_BSWAP64(x) bswap_64(x)
#elif HAVE_SYS_ENDIAN
#include <sys/endian.h>
#define PG_BSWAP32(x) bswap32(x)
#define PG_BSWAP64(x) bswap64(x)
#else
#error "No appropriate byteswap found for your system."
#endif
#if defined(__BYTE_ORDER__) && (__BYTE_ORDER__ == __ORDER_BIG_ENDIAN__)
#define PGint32(x) (x)
#define PGint64(x) (x)
#elif defined(__BYTE_ORDER__) && (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
#define PGint32(x) PG_BSWAP32(x)
#define PGint64(x) PG_BSWAP64(x)
#elif defined(_BYTE_ORDER) && (_BYTE_ORDER == _BIG_ENDIAN)
#define PGint32(x) (x)
#define PGint64(x) (x)
#elif defined(_BYTE_ORDER) && (_BYTE_ORDER == _LITTLE_ENDIAN)
#define PGint32(x) PG_BSWAP32(x)
#define PGint64(x) PG_BSWAP64(x)
#else
#error "Cannot determine byte order."
#endif
const char *build_conninfo(const char *db, const char *username, const char *password, const char *host, const char *port);
#endif

View File

@@ -1,200 +0,0 @@
/*-------------------------------------------------------------------------
*
* sprompt.c
* simple_prompt() routine
*
* Portions Copyright (c) 1996-2006, PostgreSQL Global Development Group
* Portions Copyright (c) 1994, Regents of the University of California
*
*
* IDENTIFICATION
* $PostgreSQL: pgsql/src/port/sprompt.c,v 1.18 2006/10/04 00:30:14 momjian Exp $
*
*-------------------------------------------------------------------------
*
* PostgreSQL Database Management System
* (formerly known as Postgres, then as Postgres95)
*
* Portions Copyright (c) 1996-2006, PostgreSQL Global Development Group
*
* Portions Copyright (c) 1994, The Regents of the University of California
*
* Permission to use, copy, modify, and distribute this software and its
* documentation for any purpose, without fee, and without a written agreement
* is hereby granted, provided that the above copyright notice and this
* paragraph and the following two paragraphs appear in all copies.
*
* IN NO EVENT SHALL THE UNIVERSITY OF CALIFORNIA BE LIABLE TO ANY PARTY FOR
* DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES, INCLUDING
* LOST PROFITS, ARISING OUT OF THE USE OF THIS SOFTWARE AND ITS
* DOCUMENTATION, EVEN IF THE UNIVERSITY OF CALIFORNIA HAS BEEN ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*
* THE UNIVERSITY OF CALIFORNIA SPECIFICALLY DISCLAIMS ANY WARRANTIES,
* INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY
* AND FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE PROVIDED HEREUNDER IS
* ON AN "AS IS" BASIS, AND THE UNIVERSITY OF CALIFORNIA HAS NO OBLIGATIONS TO
* PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS.
*
*/
/*
* simple_prompt
*
* Generalized function especially intended for reading in usernames and
* password interactively. Reads from /dev/tty or stdin/stderr.
*
* prompt: The prompt to print
* maxlen: How many characters to accept
* echo: Set to false if you want to hide what is entered (for passwords)
*
* Returns a malloc()'ed string with the input (w/o trailing newline).
*/
#define DEVTTY "/dev/tty"
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <string.h>
#include <assert.h>
#include <libpq-fe.h>
#ifdef __MINGW_H
# include <windows.h>
#else
# define HAVE_TERMIOS_H
# include <termios.h>
#endif
/*
extern char *simple_prompt(const char *prompt, int maxlen, int echo);
*/
char *
simple_prompt(const char *prompt, int maxlen, int echo)
{
int length;
char *destination;
FILE *termin,
*termout;
#ifdef HAVE_TERMIOS_H
struct termios t_orig,
t;
#else
#ifdef WIN32
HANDLE t = NULL;
LPDWORD t_orig = NULL;
#endif
#endif
destination = (char *) malloc(maxlen + 1);
if (!destination)
return NULL;
/*
* Do not try to collapse these into one "w+" mode file. Doesn't work on
* some platforms (eg, HPUX 10.20).
*/
termin = fopen(DEVTTY, "r");
termout = fopen(DEVTTY, "w");
if (!termin || !termout
#ifdef WIN32
/* See DEVTTY comment for msys */
|| (getenv("OSTYPE") && strcmp(getenv("OSTYPE"), "msys") == 0)
#endif
)
{
if (termin)
fclose(termin);
if (termout)
fclose(termout);
termin = stdin;
termout = stderr;
}
#ifdef HAVE_TERMIOS_H
if (!echo)
{
tcgetattr(fileno(termin), &t);
t_orig = t;
t.c_lflag &= ~ECHO;
tcsetattr(fileno(termin), TCSAFLUSH, &t);
}
#else
#ifdef WIN32
if (!echo)
{
/* get a new handle to turn echo off */
t_orig = (LPDWORD) malloc(sizeof(DWORD));
t = GetStdHandle(STD_INPUT_HANDLE);
/* save the old configuration first */
GetConsoleMode(t, t_orig);
/* set to the new mode */
SetConsoleMode(t, ENABLE_LINE_INPUT | ENABLE_PROCESSED_INPUT);
}
#endif
#endif
if (prompt)
{
fputs(prompt, termout);
fflush(termout);
}
if (fgets(destination, maxlen + 1, termin) == NULL)
destination[0] = '\0';
length = strlen(destination);
if (length > 0 && destination[length - 1] != '\n')
{
/* eat rest of the line */
char buf[128];
int buflen;
do
{
if (fgets(buf, sizeof(buf), termin) == NULL)
break;
buflen = strlen(buf);
}
while (buflen > 0 && buf[buflen - 1] != '\n');
}
if (length > 0 && destination[length - 1] == '\n')
/* remove trailing newline */
destination[length - 1] = '\0';
#ifdef HAVE_TERMIOS_H
if (!echo)
{
tcsetattr(fileno(termin), TCSAFLUSH, &t_orig);
fputs("\n", termout);
fflush(termout);
}
#else
#ifdef WIN32
if (!echo)
{
/* reset to the original console mode */
SetConsoleMode(t, *t_orig);
fputs("\n", termout);
fflush(termout);
free(t_orig);
}
#endif
#endif
if (termin != stdin)
{
fclose(termin);
fclose(termout);
}
return destination;
}

View File

@@ -1,4 +0,0 @@
#ifndef SPROMPT_H
#define SPROMPT_H
char *simple_prompt(const char *prompt, int maxlen, int echo);
#endif

File diff suppressed because it is too large Load Diff

View File

@@ -1,281 +0,0 @@
/*
* The author of this software is Steven Fortune. Copyright (c) 1994 by AT&T
* Bell Laboratories.
* Permission to use, copy, modify, and distribute this software for any
* purpose without fee is hereby granted, provided that this entire notice
* is included in all copies of any software which is or includes a copy
* or modification of this software and in all copies of the supporting
* documentation for such software.
* THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
* WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR AT&T MAKE ANY
* REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
* OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
*/
/*
* This code was originally written by Stephan Fortune in C code. I, Shane O'Sullivan,
* have since modified it, encapsulating it in a C++ class and, fixing memory leaks and
* adding accessors to the Voronoi Edges.
* Permission to use, copy, modify, and distribute this software for any
* purpose without fee is hereby granted, provided that this entire notice
* is included in all copies of any software which is or includes a copy
* or modification of this software and in all copies of the supporting
* documentation for such software.
* THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
* WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR AT&T MAKE ANY
* REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
* OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
*/
#ifndef VORONOI_DIAGRAM_GENERATOR
#define VORONOI_DIAGRAM_GENERATOR
#include <math.h>
#include <stdlib.h>
#include <string.h>
#ifndef NULL
#define NULL 0
#endif
#define DELETED -2
#define le 0
#define re 1
struct SourcePoint
{
int id;
double weight;
double x;
double y;
};
struct Freenode
{
struct Freenode *nextfree;
};
struct FreeNodeArrayList
{
struct Freenode* memory;
struct FreeNodeArrayList* next;
};
struct Freelist
{
struct Freenode *head;
int nodesize;
};
struct Point
{
float x,y;
};
struct PolygonPoint
{
struct Point coord;
double angle;
int boundary;
};
struct Polygon
{
int sitenbr;
struct Point coord;
int numpoints;
struct PolygonPoint * pointlist;
int boundary;
};
// structure used both for sites and for vertices
struct Site
{
struct Point coord;
struct Point coordout;
double weight;
int sitenbr;
int refcnt;
};
struct Edge
{
float a,b,c;
struct Site *ep[2];
struct Site *reg[2];
int edgenbr;
};
struct GraphEdge
{
float x1,y1,x2,y2;
struct GraphEdge* next;
};
struct Halfedge
{
struct Halfedge *ELleft, *ELright;
struct Edge *ELedge;
int ELrefcnt;
char ELpm;
struct Site *vertex;
float ystar;
struct Halfedge *PQnext;
};
class VoronoiDiagramGenerator
{
public:
VoronoiDiagramGenerator();
~VoronoiDiagramGenerator();
bool generateVoronoi(struct SourcePoint* srcPoints, int numPoints, float minX, float maxX, float minY, float maxY, float minDist=0);
void getSitePoints(int sitenbr, int* numpoints, PolygonPoint** pS);
void resetIterator()
{
iteratorEdges = allEdges;
}
bool getNext(float& x1, float& y1, float& x2, float& y2)
{
if(iteratorEdges == 0)
return false;
x1 = iteratorEdges->x1;
x2 = iteratorEdges->x2;
y1 = iteratorEdges->y1;
y2 = iteratorEdges->y2;
iteratorEdges = iteratorEdges->next;
return true;
}
private:
void cleanup();
void cleanupEdges();
char *getfree(struct Freelist *fl);
struct Halfedge *PQfind();
int PQempty();
struct Halfedge **ELhash;
struct Halfedge *HEcreate(), *ELleft(), *ELright(), *ELleftbnd();
struct Halfedge *HEcreate(struct Edge *e,int pm);
struct Point PQ_min();
struct Halfedge *PQextractmin();
void freeinit(struct Freelist *fl,int size);
void makefree(struct Freenode *curr,struct Freelist *fl);
void geominit();
void plotinit();
bool voronoi(int triangulate);
void ref(struct Site *v);
void deref(struct Site *v);
void endpoint(struct Edge *e,int lr,struct Site * s);
void endpoint(struct Edge *e1,int lr,struct Site * s, struct Edge *e2, struct Edge *e3);
void ELdelete(struct Halfedge *he);
struct Halfedge *ELleftbnd(struct Point *p);
struct Halfedge *ELright(struct Halfedge *he);
void makevertex(struct Site *v);
void out_triple(struct Site *s1, struct Site *s2,struct Site * s3);
void PQinsert(struct Halfedge *he,struct Site * v, float offset);
void PQdelete(struct Halfedge *he);
bool ELinitialize();
void ELinsert(struct Halfedge *lb, struct Halfedge *newHe);
struct Halfedge * ELgethash(int b);
struct Halfedge *ELleft(struct Halfedge *he);
struct Site *leftreg(struct Halfedge *he);
void out_site(struct Site *s);
bool PQinitialize();
int PQbucket(struct Halfedge *he);
void pushpoint(int sitenbr, double x, double y, int boundary);
int ccw( Point p0, Point p1, Point p2 );
void clip_line(struct Edge *e);
char *myalloc(unsigned n);
int right_of(struct Halfedge *el,struct Point *p);
struct Site *rightreg(struct Halfedge *he);
struct Edge *bisect(struct Site *s1,struct Site *s2);
float dist(struct Site *s,struct Site *t);
struct Site *intersect(struct Halfedge *el1, struct Halfedge *el2, struct Point *p=0);
void out_bisector(struct Edge *e);
void out_ep(struct Edge *e);
void out_vertex(struct Site *v);
struct Site *nextone();
void pushGraphEdge(float x1, float y1, float x2, float y2);
void openpl();
void line(float x1, float y1, float x2, float y2);
void circle(float x, float y, float radius);
void range(float minX, float minY, float maxX, float maxY);
struct Freelist hfl;
struct Halfedge *ELleftend, *ELrightend;
int ELhashsize;
int triangulate, sorted, plot, debug;
float xmin, xmax, ymin, ymax, deltax, deltay;
struct Site *sites;
struct Polygon *polygons;
struct Point corners[4];
int nsites;
int siteidx;
int sqrt_nsites;
int nvertices;
struct Freelist sfl;
struct Site *bottomsite;
int nedges;
struct Freelist efl;
int PQhashsize;
struct Halfedge *PQhash;
int PQcount;
int PQmin;
int ntry, totalsearch;
float pxmin, pxmax, pymin, pymax, cradius;
int total_alloc;
float borderMinX, borderMaxX, borderMinY, borderMaxY;
FreeNodeArrayList* allMemoryList;
FreeNodeArrayList* currentMemoryBlock;
GraphEdge* allEdges;
GraphEdge* iteratorEdges;
float minDistanceBetweenSites;
};
int scomp(const void *p1,const void *p2);
int spcomp(const void *p1,const void *p2);
int anglecomp(const void * p1, const void * p2);
#endif

View File

@@ -1,86 +0,0 @@
/*
* The author of this software is Shane O'Sullivan.
* Permission to use, copy, modify, and distribute this software for any
* purpose without fee is hereby granted, provided that this entire notice
* is included in all copies of any software which is or includes a copy
* or modification of this software and in all copies of the supporting
* documentation for such software.
* THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
* WARRANTY. IN PARTICULAR, NEITHER THE AUTHORS NOR AT&T MAKE ANY
* REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
* OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
*/
#
#include <stdio.h>
#include <search.h>
#include <malloc.h>
#include "VoronoiDiagramGenerator.h"
int main(int argc, char **argv)
{
double xmin, xmax, ymin, ymax;
scanf("%lf %lf %lf %lf", &xmin, &xmax, &ymin, &ymax) ;
SourcePoint * sites;
long nsites;
nsites = 0;
sites = (SourcePoint *) malloc(4000 * sizeof(SourcePoint));
while (scanf("%d %lf %lf %lf", &sites[nsites].id, &sites[nsites].weight, &sites[nsites].x, &sites[nsites].y) != EOF)
{
nsites++;
if (nsites % 4000 == 0) {
sites = (SourcePoint *)realloc(sites,(nsites+4000)*sizeof(SourcePoint));
}
}
VoronoiDiagramGenerator * pvdg;
pvdg = new VoronoiDiagramGenerator();
pvdg->generateVoronoi(sites, nsites, xmin, xmax, ymin, ymax, 0);
// printf("sites %ld\n-------------------------------\n", nsites);
PolygonPoint* pSitePoints;
int numpoints, i, j;
for(i = 0; i < nsites; i++)
{
pvdg->getSitePoints(i, &numpoints, &pSitePoints);
if (numpoints == 0)
{
printf("-- no points for %d\n", i);
}
else
{
printf("update temp_child_4076440_0 set resultgeom = st_setsrid('POLYGON((");
for(j = 0; j < numpoints; j++)
{
printf("%.15lf %.15lf,", pSitePoints[j].coord.x, pSitePoints[j].coord.y, (pSitePoints[j].angle/M_PI)*180);
}
printf("%.15lf %.15lf", pSitePoints[0].coord.x, pSitePoints[0].coord.y, (pSitePoints[j].angle/M_PI)*180);
printf("))'::geometry,4326) where id = %d;\n", sites[i].id);
}
}
float x1,y1,x2,y2;
// printf("sites %ld\n-------------------------------\n", nsites);
pvdg->resetIterator();
while(pvdg->getNext(x1,y1,x2,y2))
{
printf("(%f %f,%f %f)\n",x1,y1,x2, y2);
}
delete pvdg;
free(sites);
return 0;
}

View File

@@ -5,21 +5,22 @@
"continent" : [2, 0],
"country" : [4, 0],
"state" : [8, 0],
"province" : [8, 0],
"region" : [18, 0],
"county" : 12,
"municipality" : [17, 14],
"city" : 16,
"island" : [17, 0],
"town" : [18, 16],
"village" : [19, 16],
"hamlet" : [19, 16],
"municipality" : [19, 16],
"district" : [19, 16],
"unincorporated_area" : [19, 16],
"borough" : [19, 16],
"borough" : [19, 18],
"hamlet" : 20,
"suburb" : 20,
"croft" : 20,
"subdivision" : 20,
"isolated_dwelling" : 20,
"allotments" : 20,
"farm" : [20, 0],
"locality" : [20, 0],
"islet" : [20, 0],
@@ -62,7 +63,11 @@
"sea" : [4, 0]
},
"waterway" : {
"" : [17, 0]
"river" : [19, 0],
"stream" : [22, 0],
"ditch" : [22, 0],
"drain" : [22, 0],
"" : [20, 0]
},
"highway" : {
"" : 26,
@@ -89,6 +94,7 @@
{ "countries" : [ "de" ],
"tags" : {
"place" : {
"region" : [10, 0],
"county" : [12, 0]
},
"boundary" : {
@@ -102,6 +108,16 @@
"administrative7" : [14, 0]
}
}
},
{ "countries" : ["se", "no"],
"tags" : {
"place" : {
},
"boundary" : {
"administrative3" : 8,
"administrative4" : 12
}
}
}
]

View File

@@ -5,6 +5,11 @@
"no" : "skip"
}
},
{ "keys" : ["wikipedia", "wikipedia:*", "wikidata"],
"values" : {
"" : "extra"
}
},
{
"keys" : ["name:prefix", "name:suffix", "name:botanical", "*wikidata"],
"values" : {
@@ -74,7 +79,7 @@
},
{
"keys" : ["country_code", "ISO3166-1", "is_in:country_code", "is_in:country",
"addr:country", "addr:country", "addr:country_code"],
"addr:country", "addr:country_code"],
"values" : {
"" : "country"
}

View File

@@ -1,4 +1,9 @@
[
{ "keys" : ["wikipedia", "wikipedia:*", "wikidata"],
"values" : {
"" : "extra"
}
},
{
"keys" : ["name:prefix", "name:suffix", "name:botanical", "*wikidata"],
"values" : {
@@ -43,7 +48,7 @@
},
{
"keys" : ["country_code", "ISO3166-1", "is_in:country_code", "is_in:country",
"addr:country", "addr:country", "addr:country_code"],
"addr:country", "addr:country_code"],
"values" : {
"" : "country"
}

View File

@@ -0,0 +1,238 @@
[
{
"keys" : ["*source"],
"values" : {
"" : "skip"
}
},
{
"keys" : ["name:prefix", "name:suffix", "name:botanical", "wikidata",
"*:wikidata"],
"values" : {
"" : "extra"
}
},
{
"keys" : ["ref", "int_ref", "nat_ref", "reg_ref", "loc_ref", "old_ref",
"iata", "icao", "pcode", "pcode:*"],
"values" : {
"" : "ref"
}
},
{
"keys" : ["name", "name:*", "int_name", "int_name:*", "nat_name", "nat_name:*",
"reg_name", "reg_name:*", "loc_name", "loc_name:*",
"old_name", "old_name:*", "alt_name", "alt_name:*", "alt_name_*",
"official_name", "official_name:*", "place_name", "place_name:*",
"short_name", "short_name:*", "brand"],
"values" : {
"" : "name"
}
},
{
"keys" : ["addr:housename"],
"values" : {
"" : "name,house"
}
},
{
"keys" : ["emergency"],
"values" : {
"fire_hydrant" : "skip",
"yes" : "skip",
"no" : "skip",
"" : "main"
}
},
{
"keys" : ["historic", "military"],
"values" : {
"no" : "skip",
"yes" : "skip",
"" : "main"
}
},
{
"keys" : ["natural"],
"values" : {
"yes" : "skip",
"no" : "skip",
"coastline" : "skip",
"" : "main,with_name"
}
},
{
"keys" : ["landuse"],
"values" : {
"cemetry" : "main,with_name",
"" : "main,fallback,with_name"
}
},
{
"keys" : ["highway"],
"values" : {
"no" : "skip",
"turning_circle" : "skip",
"mini_roundabout" : "skip",
"noexit" : "skip",
"crossing" : "skip",
"traffic_signals" : "main,with_name",
"service" : "main,with_name",
"cycleway" : "main,with_name",
"path" : "main,with_name",
"footway" : "main,with_name",
"steps" : "main,with_name",
"bridleway" : "main,with_name",
"track" : "main,with_name",
"byway": "main,with_name",
"motorway_link" : "main,with_name",
"trunk_link" : "main,with_name",
"primary_link" : "main,with_name",
"secondary_link" : "main,with_name",
"tertiary_link" : "main,with_name",
"" : "main"
}
},
{
"keys" : ["railway"],
"values" : {
"level_crossing" : "skip",
"no" : "skip",
"rail" : "extra",
"" : "main,with_name"
}
},
{
"keys" : ["man_made"],
"values" : {
"survey_point" : "skip",
"cutline" : "skip",
"" : "main"
}
},
{
"keys" : ["aerialway"],
"values" : {
"pylon" : "skip",
"no" : "skip",
"" : "main"
}
},
{
"keys" : ["boundary"],
"values" : {
"" : "main,with_name"
}
},
{
"keys" : ["amenity"],
"values" : {
"restaurant" : "main,operator",
"fuel" : "main,operator"
}
},
{
"keys" : ["aeroway", "amenity", "club", "craft", "leisure",
"office", "mountain_pass"],
"values" : {
"no" : "skip",
"" : "main"
}
},
{
"keys" : ["shop"],
"values" : {
"no" : "skip",
"" : "main,operator"
}
},
{
"keys" : ["tourism"],
"values" : {
"yes" : "skip",
"no" : "skip",
"" : "main,operator"
}
},
{
"keys" : ["bridge", "tunnel"],
"values" : {
"" : "main,with_name_key"
}
},
{
"keys" : ["waterway"],
"values" : {
"riverbank" : "skip",
"" : "main,with_name"
}
},
{
"keys" : ["place"],
"values" : {
"" : "main"
}
},
{
"keys" : ["junction"],
"values" : {
"" : "main,fallback,with_name"
}
},
{
"keys" : ["postal_code", "postcode", "addr:postcode",
"tiger:zip_left", "tiger:zip_right"],
"values" : {
"" : "postcode,fallback"
}
},
{
"keys" : ["country_code", "ISO3166-1", "is_in:country_code", "is_in:country",
"addr:country", "addr:country_code"],
"values" : {
"" : "country"
}
},
{
"keys" : ["addr:housenumber", "addr:conscriptionnumber", "addr:streetnumber"],
"values" : {
"" : "address,house"
}
},
{
"keys" : ["addr:interpolation"],
"values" : {
"" : "interpolation,address"
}
},
{
"keys" : ["addr:*", "is_in:*", "tiger:county", "is_in"],
"values" : {
"" : "address"
}
},
{
"keys" : ["building"],
"values" : {
"no" : "skip",
"" : "main,fallback,with_name"
}
},
{
"keys" : ["note", "note:*", "source", "source*", "attribution",
"comment", "fixme", "FIXME", "created_by", "tiger:*", "NHD:*",
"nhd:*", "gnis:*", "geobase:*", "KSJ2:*", "yh:*",
"osak:*", "naptan:*", "CLC:*", "import", "it:fvg:*",
"type", "lacounty:*", "ref:ruian:*", "building:ruian:type",
"ref:linz:*"],
"values" : {
"" : "skip"
}
},
{
"keys" : [""],
"values" : {
"" : "extra"
}
}
]

View File

@@ -98,6 +98,7 @@
"values" : {
"level_crossing" : "skip",
"no" : "skip",
"rail" : "skip",
"" : "main,with_name"
}
},
@@ -187,7 +188,7 @@
},
{
"keys" : ["country_code", "ISO3166-1", "is_in:country_code", "is_in:country",
"addr:country", "addr:country", "addr:country_code"],
"addr:country", "addr:country_code"],
"values" : {
"" : "country"
}

View File

@@ -1,4 +1,9 @@
[
{ "keys" : ["wikipedia", "wikipedia:*", "wikidata"],
"values" : {
"" : "extra"
}
},
{
"keys" : ["name:prefix", "name:suffix", "name:botanical", "*wikidata"],
"values" : {
@@ -43,7 +48,7 @@
},
{
"keys" : ["country_code", "ISO3166-1", "is_in:country_code", "is_in:country",
"addr:country", "addr:country", "addr:country_code"],
"addr:country", "addr:country_code"],
"values" : {
"" : "country"
}

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,289 @@
-- Functions for returning address information for a place.
DROP TYPE IF EXISTS addressline CASCADE;
CREATE TYPE addressline as (
place_id BIGINT,
osm_type CHAR(1),
osm_id BIGINT,
name HSTORE,
class TEXT,
type TEXT,
place_type TEXT,
admin_level INTEGER,
fromarea BOOLEAN,
isaddress BOOLEAN,
rank_address INTEGER,
distance FLOAT
);
CREATE OR REPLACE FUNCTION get_name_by_language(name hstore, languagepref TEXT[])
RETURNS TEXT
AS $$
DECLARE
result TEXT;
BEGIN
IF name is null THEN
RETURN null;
END IF;
FOR j IN 1..array_upper(languagepref,1) LOOP
IF name ? languagepref[j] THEN
result := trim(name->languagepref[j]);
IF result != '' THEN
return result;
END IF;
END IF;
END LOOP;
-- anything will do as a fallback - just take the first name type thing there is
RETURN trim((avals(name))[1]);
END;
$$
LANGUAGE plpgsql IMMUTABLE;
--housenumber only needed for tiger data
CREATE OR REPLACE FUNCTION get_address_by_language(for_place_id BIGINT,
housenumber INTEGER,
languagepref TEXT[])
RETURNS TEXT
AS $$
DECLARE
result TEXT[];
currresult TEXT;
prevresult TEXT;
location RECORD;
BEGIN
result := '{}';
prevresult := '';
FOR location IN
SELECT name,
CASE WHEN place_id = for_place_id THEN 99 ELSE rank_address END as rank_address
FROM get_addressdata(for_place_id, housenumber)
WHERE isaddress order by rank_address desc
LOOP
currresult := trim(get_name_by_language(location.name, languagepref));
IF currresult != prevresult AND currresult IS NOT NULL
AND result[(100 - location.rank_address)] IS NULL
THEN
result[(100 - location.rank_address)] := currresult;
prevresult := currresult;
END IF;
END LOOP;
RETURN array_to_string(result,', ');
END;
$$
LANGUAGE plpgsql STABLE;
-- Compute the list of address parts for the given place.
--
-- If in_housenumber is greator or equal 0, look for an interpolation.
CREATE OR REPLACE FUNCTION get_addressdata(in_place_id BIGINT, in_housenumber INTEGER)
RETURNS setof addressline
AS $$
DECLARE
for_place_id BIGINT;
result TEXT[];
search TEXT[];
found INTEGER;
location RECORD;
countrylocation RECORD;
searchcountrycode varchar(2);
searchhousenumber TEXT;
searchhousename HSTORE;
searchrankaddress INTEGER;
searchpostcode TEXT;
postcode_isexact BOOL;
searchclass TEXT;
searchtype TEXT;
countryname HSTORE;
BEGIN
-- The place ein question might not have a direct entry in place_addressline.
-- Look for the parent of such places then and save if in for_place_id.
postcode_isexact := false;
-- first query osmline (interpolation lines)
IF in_housenumber >= 0 THEN
SELECT parent_place_id, country_code, in_housenumber::text, 30, postcode,
null, 'place', 'house'
FROM location_property_osmline
WHERE place_id = in_place_id AND in_housenumber>=startnumber
AND in_housenumber <= endnumber
INTO for_place_id, searchcountrycode, searchhousenumber, searchrankaddress,
searchpostcode, searchhousename, searchclass, searchtype;
END IF;
--then query tiger data
-- %NOTIGERDATA% IF 0 THEN
IF for_place_id IS NULL AND in_housenumber >= 0 THEN
SELECT parent_place_id, 'us', in_housenumber::text, 30, postcode, null,
'place', 'house'
FROM location_property_tiger
WHERE place_id = in_place_id AND in_housenumber >= startnumber
AND in_housenumber <= endnumber
INTO for_place_id, searchcountrycode, searchhousenumber, searchrankaddress,
searchpostcode, searchhousename, searchclass, searchtype;
END IF;
-- %NOTIGERDATA% END IF;
-- %NOAUXDATA% IF 0 THEN
IF for_place_id IS NULL THEN
SELECT parent_place_id, 'us', housenumber, 30, postcode, null, 'place', 'house'
FROM location_property_aux
WHERE place_id = in_place_id
INTO for_place_id,searchcountrycode, searchhousenumber, searchrankaddress,
searchpostcode, searchhousename, searchclass, searchtype;
END IF;
-- %NOAUXDATA% END IF;
-- postcode table
IF for_place_id IS NULL THEN
SELECT parent_place_id, country_code, rank_search, postcode, 'place', 'postcode'
FROM location_postcode
WHERE place_id = in_place_id
INTO for_place_id, searchcountrycode, searchrankaddress, searchpostcode,
searchclass, searchtype;
END IF;
-- POI objects in the placex table
IF for_place_id IS NULL THEN
SELECT parent_place_id, country_code, housenumber, rank_search,
postcode, address is not null and address ? 'postcode',
name, class, type
FROM placex
WHERE place_id = in_place_id and rank_search > 27
INTO for_place_id, searchcountrycode, searchhousenumber, searchrankaddress,
searchpostcode, postcode_isexact, searchhousename, searchclass, searchtype;
END IF;
-- If for_place_id is still NULL at this point then the object has its own
-- entry in place_address line. However, still check if there is not linked
-- place we should be using instead.
IF for_place_id IS NULL THEN
select coalesce(linked_place_id, place_id), country_code,
housenumber, rank_search, postcode,
address is not null and address ? 'postcode', null
from placex where place_id = in_place_id
INTO for_place_id, searchcountrycode, searchhousenumber, searchrankaddress, searchpostcode, postcode_isexact, searchhousename;
END IF;
--RAISE WARNING '% % % %',searchcountrycode, searchhousenumber, searchrankaddress, searchpostcode;
found := 1000; -- the lowest rank_address included
-- Return the record for the base entry.
FOR location IN
SELECT placex.place_id, osm_type, osm_id, name,
class, type, admin_level,
type not in ('postcode', 'postal_code') as isaddress,
CASE WHEN rank_address = 0 THEN 100
WHEN rank_address = 11 THEN 5
ELSE rank_address END as rank_address,
0 as distance, country_code, postcode
FROM placex
WHERE place_id = for_place_id
LOOP
--RAISE WARNING '%',location;
IF searchcountrycode IS NULL AND location.country_code IS NOT NULL THEN
searchcountrycode := location.country_code;
END IF;
IF location.rank_address < 4 THEN
-- no country locations for ranks higher than country
searchcountrycode := NULL;
END IF;
countrylocation := ROW(location.place_id, location.osm_type, location.osm_id,
location.name, location.class, location.type, NULL,
location.admin_level, true, location.isaddress,
location.rank_address, location.distance)::addressline;
RETURN NEXT countrylocation;
found := location.rank_address;
END LOOP;
FOR location IN
SELECT placex.place_id, osm_type, osm_id, name, class, type,
coalesce(extratags->'linked_place', extratags->'place') as place_type,
admin_level, fromarea, isaddress,
CASE WHEN rank_address = 11 THEN 5 ELSE rank_address END as rank_address,
distance, country_code, postcode
FROM place_addressline join placex on (address_place_id = placex.place_id)
WHERE place_addressline.place_id = for_place_id
AND (cached_rank_address >= 4 AND cached_rank_address < searchrankaddress)
AND linked_place_id is null
AND (placex.country_code IS NULL OR searchcountrycode IS NULL
OR placex.country_code = searchcountrycode)
ORDER BY rank_address desc, isaddress desc, fromarea desc,
distance asc, rank_search desc
LOOP
--RAISE WARNING '%',location;
IF searchcountrycode IS NULL AND location.country_code IS NOT NULL THEN
searchcountrycode := location.country_code;
END IF;
IF location.type in ('postcode', 'postal_code')
AND searchpostcode is not null
THEN
-- If the place had a postcode assigned, take this one only
-- into consideration when it is an area and the place does not have
-- a postcode itself.
IF location.fromarea AND not postcode_isexact AND location.isaddress THEN
searchpostcode := null; -- remove the less exact postcode
ELSE
location.isaddress := false;
END IF;
END IF;
countrylocation := ROW(location.place_id, location.osm_type, location.osm_id,
location.name, location.class, location.type,
location.place_type,
location.admin_level, location.fromarea,
location.isaddress, location.rank_address,
location.distance)::addressline;
RETURN NEXT countrylocation;
found := location.rank_address;
END LOOP;
-- If no country was included yet, add the name information from country_name.
IF found > 4 THEN
SELECT name FROM country_name
WHERE country_code = searchcountrycode LIMIT 1 INTO countryname;
--RAISE WARNING '% % %',found,searchcountrycode,countryname;
IF countryname IS NOT NULL THEN
location := ROW(null, null, null, countryname, 'place', 'country', NULL,
null, true, true, 4, 0)::addressline;
RETURN NEXT location;
END IF;
END IF;
-- Finally add some artificial rows.
IF searchcountrycode IS NOT NULL THEN
location := ROW(null, null, null, hstore('ref', searchcountrycode),
'place', 'country_code', null, null, true, false, 4, 0)::addressline;
RETURN NEXT location;
END IF;
IF searchhousename IS NOT NULL THEN
location := ROW(in_place_id, null, null, searchhousename, searchclass,
searchtype, null, null, true, true, 29, 0)::addressline;
RETURN NEXT location;
END IF;
IF searchhousenumber IS NOT NULL THEN
location := ROW(null, null, null, hstore('ref', searchhousenumber),
'place', 'house_number', null, null, true, true, 28, 0)::addressline;
RETURN NEXT location;
END IF;
IF searchpostcode IS NOT NULL THEN
location := ROW(null, null, null, hstore('ref', searchpostcode), 'place',
'postcode', null, null, false, true, 5, 0)::addressline;
RETURN NEXT location;
END IF;
RETURN;
END;
$$
LANGUAGE plpgsql STABLE;

View File

@@ -0,0 +1,53 @@
-- Functions for adding external data (currently unused).
CREATE OR REPLACE FUNCTION aux_create_property(pointgeo GEOMETRY, in_housenumber TEXT,
in_street TEXT, in_isin TEXT,
in_postcode TEXT, in_countrycode char(2))
RETURNS INTEGER
AS $$
DECLARE
newpoints INTEGER;
place_centroid GEOMETRY;
out_partition INTEGER;
out_parent_place_id BIGINT;
location RECORD;
address_street_word_ids INTEGER[];
out_postcode TEXT;
BEGIN
place_centroid := ST_Centroid(pointgeo);
out_partition := get_partition(in_countrycode);
out_parent_place_id := null;
address_street_word_ids := word_ids_from_name(in_street);
IF address_street_word_ids IS NOT NULL THEN
out_parent_place_id := getNearestNamedRoadPlaceId(out_partition, place_centroid,
address_street_word_ids);
END IF;
IF out_parent_place_id IS NULL THEN
SELECT getNearestRoadPlaceId(out_partition, place_centroid)
INTO out_parent_place_id;
END LOOP;
END IF;
out_postcode := in_postcode;
IF out_postcode IS NULL THEN
SELECT postcode from placex where place_id = out_parent_place_id INTO out_postcode;
END IF;
-- XXX look into postcode table
newpoints := 0;
insert into location_property_aux (place_id, partition, parent_place_id,
housenumber, postcode, centroid)
values (nextval('seq_place'), out_partition, out_parent_place_id,
in_housenumber, out_postcode, place_centroid);
newpoints := newpoints + 1;
RETURN newpoints;
END;
$$
LANGUAGE plpgsql;

View File

@@ -0,0 +1,125 @@
-- Functions for interpreting wkipedia/wikidata tags and computing importance.
DROP TYPE IF EXISTS wikipedia_article_match CASCADE;
CREATE TYPE wikipedia_article_match as (
language TEXT,
title TEXT,
importance FLOAT
);
DROP TYPE IF EXISTS place_importance CASCADE;
CREATE TYPE place_importance as (
importance FLOAT,
wikipedia TEXT
);
-- See: http://stackoverflow.com/questions/6410088/how-can-i-mimic-the-php-urldecode-function-in-postgresql
CREATE OR REPLACE FUNCTION decode_url_part(p varchar)
RETURNS varchar
AS $$
SELECT convert_from(CAST(E'\\x' || array_to_string(ARRAY(
SELECT CASE WHEN length(r.m[1]) = 1 THEN encode(convert_to(r.m[1], 'SQL_ASCII'), 'hex') ELSE substring(r.m[1] from 2 for 2) END
FROM regexp_matches($1, '%[0-9a-f][0-9a-f]|.', 'gi') AS r(m)
), '') AS bytea), 'UTF8');
$$
LANGUAGE SQL IMMUTABLE STRICT;
CREATE OR REPLACE FUNCTION catch_decode_url_part(p varchar)
RETURNS varchar
AS $$
DECLARE
BEGIN
RETURN decode_url_part(p);
EXCEPTION
WHEN others THEN return null;
END;
$$
LANGUAGE plpgsql IMMUTABLE STRICT;
CREATE OR REPLACE FUNCTION get_wikipedia_match(extratags HSTORE, country_code varchar(2))
RETURNS wikipedia_article_match
AS $$
DECLARE
langs TEXT[];
i INT;
wiki_article TEXT;
wiki_article_title TEXT;
wiki_article_language TEXT;
result wikipedia_article_match;
BEGIN
langs := ARRAY['english','country','ar','bg','ca','cs','da','de','en','es','eo','eu','fa','fr','ko','hi','hr','id','it','he','lt','hu','ms','nl','ja','no','pl','pt','kk','ro','ru','sk','sl','sr','fi','sv','tr','uk','vi','vo','war','zh'];
i := 1;
WHILE langs[i] IS NOT NULL LOOP
wiki_article := extratags->(case when langs[i] in ('english','country') THEN 'wikipedia' ELSE 'wikipedia:'||langs[i] END);
IF wiki_article is not null THEN
wiki_article := regexp_replace(wiki_article,E'^(.*?)([a-z]{2,3}).wikipedia.org/wiki/',E'\\2:');
wiki_article := regexp_replace(wiki_article,E'^(.*?)([a-z]{2,3}).wikipedia.org/w/index.php\\?title=',E'\\2:');
wiki_article := regexp_replace(wiki_article,E'^(.*?)/([a-z]{2,3})/wiki/',E'\\2:');
--wiki_article := regexp_replace(wiki_article,E'^(.*?)([a-z]{2,3})[=:]',E'\\2:');
wiki_article := replace(wiki_article,' ','_');
IF strpos(wiki_article, ':') IN (3,4) THEN
wiki_article_language := lower(trim(split_part(wiki_article, ':', 1)));
wiki_article_title := trim(substr(wiki_article, strpos(wiki_article, ':')+1));
ELSE
wiki_article_title := trim(wiki_article);
wiki_article_language := CASE WHEN langs[i] = 'english' THEN 'en' WHEN langs[i] = 'country' THEN get_country_language_code(country_code) ELSE langs[i] END;
END IF;
select wikipedia_article.language,wikipedia_article.title,wikipedia_article.importance
from wikipedia_article
where language = wiki_article_language and
(title = wiki_article_title OR title = catch_decode_url_part(wiki_article_title) OR title = replace(catch_decode_url_part(wiki_article_title),E'\\',''))
UNION ALL
select wikipedia_article.language,wikipedia_article.title,wikipedia_article.importance
from wikipedia_redirect join wikipedia_article on (wikipedia_redirect.language = wikipedia_article.language and wikipedia_redirect.to_title = wikipedia_article.title)
where wikipedia_redirect.language = wiki_article_language and
(from_title = wiki_article_title OR from_title = catch_decode_url_part(wiki_article_title) OR from_title = replace(catch_decode_url_part(wiki_article_title),E'\\',''))
order by importance desc limit 1 INTO result;
IF result.language is not null THEN
return result;
END IF;
END IF;
i := i + 1;
END LOOP;
RETURN NULL;
END;
$$
LANGUAGE plpgsql STABLE;
CREATE OR REPLACE FUNCTION compute_importance(extratags HSTORE,
country_code varchar(2),
osm_type varchar(1), osm_id BIGINT)
RETURNS place_importance
AS $$
DECLARE
match RECORD;
result place_importance;
BEGIN
FOR match IN SELECT * FROM get_wikipedia_match(extratags, country_code)
WHERE language is not NULL
LOOP
result.importance := match.importance;
result.wikipedia := match.language || ':' || match.title;
RETURN result;
END LOOP;
IF extratags ? 'wikidata' THEN
FOR match IN SELECT * FROM wikipedia_article
WHERE wd_page_title = extratags->'wikidata'
ORDER BY language = 'en' DESC, langcount DESC LIMIT 1 LOOP
result.importance := match.importance;
result.wikipedia := match.language || ':' || match.title;
RETURN result;
END LOOP;
END IF;
RETURN null;
END;
$$
LANGUAGE plpgsql;

View File

@@ -0,0 +1,251 @@
-- Functions for address interpolation objects in location_property_osmline.
-- Splits the line at the given point and returns the two parts
-- in a multilinestring.
CREATE OR REPLACE FUNCTION split_line_on_node(line GEOMETRY, point GEOMETRY)
RETURNS GEOMETRY
AS $$
BEGIN
RETURN ST_Split(ST_Snap(line, point, 0.0005), point);
END;
$$
LANGUAGE plpgsql IMMUTABLE;
-- find the parent road of the cut road parts
CREATE OR REPLACE FUNCTION get_interpolation_parent(wayid BIGINT, street TEXT,
place TEXT, partition SMALLINT,
centroid GEOMETRY, geom GEOMETRY)
RETURNS BIGINT
AS $$
DECLARE
addr_street TEXT;
addr_place TEXT;
parent_place_id BIGINT;
waynodes BIGINT[];
location RECORD;
BEGIN
addr_street = street;
addr_place = place;
IF addr_street is null and addr_place is null THEN
select nodes from planet_osm_ways where id = wayid INTO waynodes;
FOR location IN SELECT placex.address from placex
where osm_type = 'N' and osm_id = ANY(waynodes)
and placex.address is not null
and (placex.address ? 'street' or placex.address ? 'place')
and indexed_status < 100
limit 1 LOOP
addr_street = location.address->'street';
addr_place = location.address->'place';
END LOOP;
END IF;
parent_place_id := find_parent_for_address(addr_street, addr_place,
partition, centroid);
IF parent_place_id is null THEN
FOR location IN SELECT place_id FROM placex
WHERE ST_DWithin(geom, placex.geometry, 0.001) and placex.rank_search = 26
ORDER BY (ST_distance(placex.geometry, ST_LineInterpolatePoint(geom,0))+
ST_distance(placex.geometry, ST_LineInterpolatePoint(geom,0.5))+
ST_distance(placex.geometry, ST_LineInterpolatePoint(geom,1))) ASC limit 1
LOOP
parent_place_id := location.place_id;
END LOOP;
END IF;
IF parent_place_id is null THEN
RETURN 0;
END IF;
RETURN parent_place_id;
END;
$$
LANGUAGE plpgsql STABLE;
CREATE OR REPLACE FUNCTION osmline_reinsert(node_id BIGINT, geom GEOMETRY)
RETURNS BOOLEAN
AS $$
DECLARE
existingline RECORD;
BEGIN
SELECT w.id FROM planet_osm_ways w, location_property_osmline p
WHERE p.linegeo && geom and p.osm_id = w.id and p.indexed_status = 0
and node_id = any(w.nodes) INTO existingline;
IF existingline.id is not NULL THEN
DELETE FROM location_property_osmline WHERE osm_id = existingline.id;
INSERT INTO location_property_osmline (osm_id, address, linegeo)
SELECT osm_id, address, geometry FROM place
WHERE osm_type = 'W' and osm_id = existingline.id;
END IF;
RETURN true;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION osmline_insert()
RETURNS TRIGGER
AS $$
BEGIN
NEW.place_id := nextval('seq_place');
NEW.indexed_date := now();
IF NEW.indexed_status IS NULL THEN
IF NEW.address is NULL OR NOT NEW.address ? 'interpolation'
OR NEW.address->'interpolation' NOT IN ('odd', 'even', 'all') THEN
-- other interpolation types than odd/even/all (e.g. numeric ones) are not supported
RETURN NULL;
END IF;
NEW.indexed_status := 1; --STATUS_NEW
NEW.country_code := lower(get_country_code(NEW.linegeo));
NEW.partition := get_partition(NEW.country_code);
NEW.geometry_sector := geometry_sector(NEW.partition, NEW.linegeo);
END IF;
RETURN NEW;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION osmline_update()
RETURNS TRIGGER
AS $$
DECLARE
place_centroid GEOMETRY;
waynodes BIGINT[];
prevnode RECORD;
nextnode RECORD;
startnumber INTEGER;
endnumber INTEGER;
housenum INTEGER;
linegeo GEOMETRY;
splitline GEOMETRY;
sectiongeo GEOMETRY;
interpol_postcode TEXT;
postcode TEXT;
BEGIN
-- deferred delete
IF OLD.indexed_status = 100 THEN
delete from location_property_osmline where place_id = OLD.place_id;
RETURN NULL;
END IF;
IF NEW.indexed_status != 0 OR OLD.indexed_status = 0 THEN
RETURN NEW;
END IF;
NEW.interpolationtype = NEW.address->'interpolation';
place_centroid := ST_PointOnSurface(NEW.linegeo);
NEW.parent_place_id = get_interpolation_parent(NEW.osm_id, NEW.address->'street',
NEW.address->'place',
NEW.partition, place_centroid, NEW.linegeo);
IF NEW.address is not NULL AND NEW.address ? 'postcode' AND NEW.address->'postcode' not similar to '%(,|;)%' THEN
interpol_postcode := NEW.address->'postcode';
housenum := getorcreate_postcode_id(NEW.address->'postcode');
ELSE
interpol_postcode := NULL;
END IF;
-- if the line was newly inserted, split the line as necessary
IF OLD.indexed_status = 1 THEN
select nodes from planet_osm_ways where id = NEW.osm_id INTO waynodes;
IF array_upper(waynodes, 1) IS NULL THEN
RETURN NEW;
END IF;
linegeo := NEW.linegeo;
startnumber := NULL;
FOR nodeidpos in 1..array_upper(waynodes, 1) LOOP
select osm_id, address, geometry
from place where osm_type = 'N' and osm_id = waynodes[nodeidpos]::BIGINT
and address is not NULL and address ? 'housenumber' limit 1 INTO nextnode;
--RAISE NOTICE 'Nextnode.place_id: %s', nextnode.place_id;
IF nextnode.osm_id IS NOT NULL THEN
--RAISE NOTICE 'place_id is not null';
IF nodeidpos > 1 and nodeidpos < array_upper(waynodes, 1) THEN
-- Make sure that the point is actually on the line. That might
-- be a bit paranoid but ensures that the algorithm still works
-- should osm2pgsql attempt to repair geometries.
splitline := split_line_on_node(linegeo, nextnode.geometry);
sectiongeo := ST_GeometryN(splitline, 1);
linegeo := ST_GeometryN(splitline, 2);
ELSE
sectiongeo = linegeo;
END IF;
endnumber := substring(nextnode.address->'housenumber','[0-9]+')::integer;
IF startnumber IS NOT NULL AND endnumber IS NOT NULL
AND startnumber != endnumber
AND ST_GeometryType(sectiongeo) = 'ST_LineString' THEN
IF (startnumber > endnumber) THEN
housenum := endnumber;
endnumber := startnumber;
startnumber := housenum;
sectiongeo := ST_Reverse(sectiongeo);
END IF;
-- determine postcode
postcode := coalesce(interpol_postcode,
prevnode.address->'postcode',
nextnode.address->'postcode',
postcode);
IF postcode is NULL THEN
SELECT placex.postcode FROM placex WHERE place_id = NEW.parent_place_id INTO postcode;
END IF;
IF postcode is NULL THEN
postcode := get_nearest_postcode(NEW.country_code, nextnode.geometry);
END IF;
IF NEW.startnumber IS NULL THEN
NEW.startnumber := startnumber;
NEW.endnumber := endnumber;
NEW.linegeo := sectiongeo;
NEW.postcode := upper(trim(postcode));
ELSE
insert into location_property_osmline
(linegeo, partition, osm_id, parent_place_id,
startnumber, endnumber, interpolationtype,
address, postcode, country_code,
geometry_sector, indexed_status)
values (sectiongeo, NEW.partition, NEW.osm_id, NEW.parent_place_id,
startnumber, endnumber, NEW.interpolationtype,
NEW.address, postcode,
NEW.country_code, NEW.geometry_sector, 0);
END IF;
END IF;
-- early break if we are out of line string,
-- might happen when a line string loops back on itself
IF ST_GeometryType(linegeo) != 'ST_LineString' THEN
RETURN NEW;
END IF;
startnumber := substring(nextnode.address->'housenumber','[0-9]+')::integer;
prevnode := nextnode;
END IF;
END LOOP;
END IF;
-- marking descendants for reparenting is not needed, because there are
-- actually no descendants for interpolation lines
RETURN NEW;
END;
$$
LANGUAGE plpgsql;

View File

@@ -0,0 +1,408 @@
-- Functions for term normalisation and access to the 'word' table.
CREATE OR REPLACE FUNCTION transliteration(text) RETURNS text
AS '{modulepath}/nominatim.so', 'transliteration'
LANGUAGE c IMMUTABLE STRICT;
CREATE OR REPLACE FUNCTION gettokenstring(text) RETURNS text
AS '{modulepath}/nominatim.so', 'gettokenstring'
LANGUAGE c IMMUTABLE STRICT;
CREATE OR REPLACE FUNCTION make_standard_name(name TEXT) RETURNS TEXT
AS $$
DECLARE
o TEXT;
BEGIN
o := public.gettokenstring(public.transliteration(name));
RETURN trim(substr(o,1,length(o)));
END;
$$
LANGUAGE plpgsql IMMUTABLE;
-- returns NULL if the word is too common
CREATE OR REPLACE FUNCTION getorcreate_word_id(lookup_word TEXT)
RETURNS INTEGER
AS $$
DECLARE
lookup_token TEXT;
return_word_id INTEGER;
count INTEGER;
BEGIN
lookup_token := trim(lookup_word);
SELECT min(word_id), max(search_name_count) FROM word
WHERE word_token = lookup_token and class is null and type is null
INTO return_word_id, count;
IF return_word_id IS NULL THEN
return_word_id := nextval('seq_word');
INSERT INTO word VALUES (return_word_id, lookup_token, null, null, null, null, 0);
ELSE
IF count > get_maxwordfreq() THEN
return_word_id := NULL;
END IF;
END IF;
RETURN return_word_id;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION getorcreate_housenumber_id(lookup_word TEXT)
RETURNS INTEGER
AS $$
DECLARE
lookup_token TEXT;
return_word_id INTEGER;
BEGIN
lookup_token := ' ' || trim(lookup_word);
SELECT min(word_id) FROM word
WHERE word_token = lookup_token and class='place' and type='house'
INTO return_word_id;
IF return_word_id IS NULL THEN
return_word_id := nextval('seq_word');
INSERT INTO word VALUES (return_word_id, lookup_token, null,
'place', 'house', null, 0);
END IF;
RETURN return_word_id;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION getorcreate_postcode_id(postcode TEXT)
RETURNS INTEGER
AS $$
DECLARE
lookup_token TEXT;
lookup_word TEXT;
return_word_id INTEGER;
BEGIN
lookup_word := upper(trim(postcode));
lookup_token := ' ' || make_standard_name(lookup_word);
SELECT min(word_id) FROM word
WHERE word_token = lookup_token and word = lookup_word
and class='place' and type='postcode'
INTO return_word_id;
IF return_word_id IS NULL THEN
return_word_id := nextval('seq_word');
INSERT INTO word VALUES (return_word_id, lookup_token, lookup_word,
'place', 'postcode', null, 0);
END IF;
RETURN return_word_id;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION getorcreate_country(lookup_word TEXT,
lookup_country_code varchar(2))
RETURNS INTEGER
AS $$
DECLARE
lookup_token TEXT;
return_word_id INTEGER;
BEGIN
lookup_token := ' '||trim(lookup_word);
SELECT min(word_id) FROM word
WHERE word_token = lookup_token and country_code=lookup_country_code
INTO return_word_id;
IF return_word_id IS NULL THEN
return_word_id := nextval('seq_word');
INSERT INTO word VALUES (return_word_id, lookup_token, null,
null, null, lookup_country_code, 0);
END IF;
RETURN return_word_id;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION getorcreate_amenity(lookup_word TEXT, normalized_word TEXT,
lookup_class text, lookup_type text)
RETURNS INTEGER
AS $$
DECLARE
lookup_token TEXT;
return_word_id INTEGER;
BEGIN
lookup_token := ' '||trim(lookup_word);
SELECT min(word_id) FROM word
WHERE word_token = lookup_token and word = normalized_word
and class = lookup_class and type = lookup_type
INTO return_word_id;
IF return_word_id IS NULL THEN
return_word_id := nextval('seq_word');
INSERT INTO word VALUES (return_word_id, lookup_token, normalized_word,
lookup_class, lookup_type, null, 0);
END IF;
RETURN return_word_id;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION getorcreate_amenityoperator(lookup_word TEXT,
normalized_word TEXT,
lookup_class text,
lookup_type text,
op text)
RETURNS INTEGER
AS $$
DECLARE
lookup_token TEXT;
return_word_id INTEGER;
BEGIN
lookup_token := ' '||trim(lookup_word);
SELECT min(word_id) FROM word
WHERE word_token = lookup_token and word = normalized_word
and class = lookup_class and type = lookup_type and operator = op
INTO return_word_id;
IF return_word_id IS NULL THEN
return_word_id := nextval('seq_word');
INSERT INTO word VALUES (return_word_id, lookup_token, normalized_word,
lookup_class, lookup_type, null, 0, op);
END IF;
RETURN return_word_id;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION getorcreate_name_id(lookup_word TEXT, src_word TEXT)
RETURNS INTEGER
AS $$
DECLARE
lookup_token TEXT;
nospace_lookup_token TEXT;
return_word_id INTEGER;
BEGIN
lookup_token := ' '||trim(lookup_word);
SELECT min(word_id) FROM word
WHERE word_token = lookup_token and class is null and type is null
INTO return_word_id;
IF return_word_id IS NULL THEN
return_word_id := nextval('seq_word');
INSERT INTO word VALUES (return_word_id, lookup_token, src_word,
null, null, null, 0);
END IF;
RETURN return_word_id;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION getorcreate_name_id(lookup_word TEXT)
RETURNS INTEGER
AS $$
DECLARE
BEGIN
RETURN getorcreate_name_id(lookup_word, '');
END;
$$
LANGUAGE plpgsql;
-- Normalize a string and lookup its word ids (partial words).
CREATE OR REPLACE FUNCTION addr_ids_from_name(lookup_word TEXT)
RETURNS INTEGER[]
AS $$
DECLARE
lookup_token TEXT;
return_word_id INTEGER[];
BEGIN
lookup_token := make_standard_name(lookup_word);
SELECT array_agg(word_id) FROM word
WHERE word_token = lookup_token and class is null and type is null
INTO return_word_id;
RETURN return_word_id;
END;
$$
LANGUAGE plpgsql STABLE;
-- Normalize a string and look up its name ids (full words).
CREATE OR REPLACE FUNCTION word_ids_from_name(lookup_word TEXT)
RETURNS INTEGER[]
AS $$
DECLARE
lookup_token TEXT;
return_word_ids INTEGER[];
BEGIN
lookup_token := ' '|| make_standard_name(lookup_word);
SELECT array_agg(word_id) FROM word
WHERE word_token = lookup_token and class is null and type is null
INTO return_word_ids;
RETURN return_word_ids;
END;
$$
LANGUAGE plpgsql STABLE STRICT;
CREATE OR REPLACE FUNCTION create_country(src HSTORE, country_code varchar(2))
RETURNS VOID
AS $$
DECLARE
s TEXT;
w INTEGER;
words TEXT[];
item RECORD;
j INTEGER;
BEGIN
FOR item IN SELECT (each(src)).* LOOP
s := make_standard_name(item.value);
w := getorcreate_country(s, country_code);
words := regexp_split_to_array(item.value, E'[,;()]');
IF array_upper(words, 1) != 1 THEN
FOR j IN 1..array_upper(words, 1) LOOP
s := make_standard_name(words[j]);
IF s != '' THEN
w := getorcreate_country(s, country_code);
END IF;
END LOOP;
END IF;
END LOOP;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION make_keywords(src HSTORE)
RETURNS INTEGER[]
AS $$
DECLARE
result INTEGER[];
s TEXT;
w INTEGER;
words TEXT[];
item RECORD;
j INTEGER;
BEGIN
result := '{}'::INTEGER[];
FOR item IN SELECT (each(src)).* LOOP
s := make_standard_name(item.value);
w := getorcreate_name_id(s, item.value);
IF not(ARRAY[w] <@ result) THEN
result := result || w;
END IF;
w := getorcreate_word_id(s);
IF w IS NOT NULL AND NOT (ARRAY[w] <@ result) THEN
result := result || w;
END IF;
words := string_to_array(s, ' ');
IF array_upper(words, 1) IS NOT NULL THEN
FOR j IN 1..array_upper(words, 1) LOOP
IF (words[j] != '') THEN
w = getorcreate_word_id(words[j]);
IF w IS NOT NULL AND NOT (ARRAY[w] <@ result) THEN
result := result || w;
END IF;
END IF;
END LOOP;
END IF;
words := regexp_split_to_array(item.value, E'[,;()]');
IF array_upper(words, 1) != 1 THEN
FOR j IN 1..array_upper(words, 1) LOOP
s := make_standard_name(words[j]);
IF s != '' THEN
w := getorcreate_word_id(s);
IF w IS NOT NULL AND NOT (ARRAY[w] <@ result) THEN
result := result || w;
END IF;
END IF;
END LOOP;
END IF;
s := regexp_replace(item.value, '市$', '');
IF s != item.value THEN
s := make_standard_name(s);
IF s != '' THEN
w := getorcreate_name_id(s, item.value);
IF NOT (ARRAY[w] <@ result) THEN
result := result || w;
END IF;
END IF;
END IF;
END LOOP;
RETURN result;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION make_keywords(src TEXT)
RETURNS INTEGER[]
AS $$
DECLARE
result INTEGER[];
s TEXT;
w INTEGER;
words TEXT[];
i INTEGER;
j INTEGER;
BEGIN
result := '{}'::INTEGER[];
s := make_standard_name(src);
w := getorcreate_name_id(s, src);
IF NOT (ARRAY[w] <@ result) THEN
result := result || w;
END IF;
w := getorcreate_word_id(s);
IF w IS NOT NULL AND NOT (ARRAY[w] <@ result) THEN
result := result || w;
END IF;
words := string_to_array(s, ' ');
IF array_upper(words, 1) IS NOT NULL THEN
FOR j IN 1..array_upper(words, 1) LOOP
IF (words[j] != '') THEN
w = getorcreate_word_id(words[j]);
IF w IS NOT NULL AND NOT (ARRAY[w] <@ result) THEN
result := result || w;
END IF;
END IF;
END LOOP;
END IF;
words := regexp_split_to_array(src, E'[,;()]');
IF array_upper(words, 1) != 1 THEN
FOR j IN 1..array_upper(words, 1) LOOP
s := make_standard_name(words[j]);
IF s != '' THEN
w := getorcreate_word_id(s);
IF w IS NOT NULL AND NOT (ARRAY[w] <@ result) THEN
result := result || w;
END IF;
END IF;
END LOOP;
END IF;
s := regexp_replace(src, '市$', '');
IF s != src THEN
s := make_standard_name(s);
IF s != '' THEN
w := getorcreate_name_id(s, src);
IF NOT (ARRAY[w] <@ result) THEN
result := result || w;
END IF;
END IF;
END IF;
RETURN result;
END;
$$
LANGUAGE plpgsql;

View File

@@ -0,0 +1,284 @@
CREATE OR REPLACE FUNCTION place_insert()
RETURNS TRIGGER
AS $$
DECLARE
i INTEGER;
existing RECORD;
existingplacex RECORD;
existingline RECORD;
existinggeometry GEOMETRY;
existingplace_id BIGINT;
result BOOLEAN;
partition INTEGER;
BEGIN
--DEBUG: RAISE WARNING '-----------------------------------------------------------------------------------';
--DEBUG: RAISE WARNING 'place_insert: % % % % %',NEW.osm_type,NEW.osm_id,NEW.class,NEW.type,st_area(NEW.geometry);
-- filter wrong tupels
IF ST_IsEmpty(NEW.geometry) OR NOT ST_IsValid(NEW.geometry) OR ST_X(ST_Centroid(NEW.geometry))::text in ('NaN','Infinity','-Infinity') OR ST_Y(ST_Centroid(NEW.geometry))::text in ('NaN','Infinity','-Infinity') THEN
INSERT INTO import_polygon_error (osm_type, osm_id, class, type, name, country_code, updated, errormessage, prevgeometry, newgeometry)
VALUES (NEW.osm_type, NEW.osm_id, NEW.class, NEW.type, NEW.name, NEW.address->'country', now(), ST_IsValidReason(NEW.geometry), null, NEW.geometry);
-- RAISE WARNING 'Invalid Geometry: % % % %',NEW.osm_type,NEW.osm_id,NEW.class,NEW.type;
RETURN null;
END IF;
-- decide, whether it is an osm interpolation line => insert intoosmline, or else just placex
IF NEW.class='place' and NEW.type='houses' and NEW.osm_type='W' and ST_GeometryType(NEW.geometry) = 'ST_LineString' THEN
-- Have we already done this place?
select * from place where osm_type = NEW.osm_type and osm_id = NEW.osm_id and class = NEW.class and type = NEW.type INTO existing;
-- Get the existing place_id
select * from location_property_osmline where osm_id = NEW.osm_id INTO existingline;
-- Handle a place changing type by removing the old data (this trigger is executed BEFORE INSERT of the NEW tupel)
IF existing.osm_type IS NULL THEN
DELETE FROM place where osm_type = NEW.osm_type and osm_id = NEW.osm_id and class = NEW.class;
END IF;
DELETE from import_polygon_error where osm_type = NEW.osm_type and osm_id = NEW.osm_id;
DELETE from import_polygon_delete where osm_type = NEW.osm_type and osm_id = NEW.osm_id;
-- update method for interpolation lines: delete all old interpolation lines with same osm_id (update on place) and insert the new one(s) (they can be split up, if they have > 2 nodes)
IF existingline.osm_id IS NOT NULL THEN
delete from location_property_osmline where osm_id = NEW.osm_id;
END IF;
-- for interpolations invalidate all nodes on the line
update placex p set indexed_status = 2
from planet_osm_ways w
where w.id = NEW.osm_id and p.osm_type = 'N' and p.osm_id = any(w.nodes);
INSERT INTO location_property_osmline (osm_id, address, linegeo)
VALUES (NEW.osm_id, NEW.address, NEW.geometry);
IF existing.osm_type IS NULL THEN
return NEW;
END IF;
IF coalesce(existing.address, ''::hstore) != coalesce(NEW.address, ''::hstore)
OR (coalesce(existing.extratags, ''::hstore) != coalesce(NEW.extratags, ''::hstore))
OR existing.geometry::text != NEW.geometry::text
THEN
update place set
name = NEW.name,
address = NEW.address,
extratags = NEW.extratags,
admin_level = NEW.admin_level,
geometry = NEW.geometry
where osm_type = NEW.osm_type and osm_id = NEW.osm_id and class = NEW.class and type = NEW.type;
END IF;
RETURN NULL;
ELSE -- insert to placex
-- Patch in additional country names
IF NEW.admin_level = 2 AND NEW.type = 'administrative'
AND NEW.address is not NULL AND NEW.address ? 'country' THEN
SELECT name FROM country_name WHERE country_code = lower(NEW.address->'country') INTO existing;
IF existing.name IS NOT NULL THEN
NEW.name = existing.name || NEW.name;
END IF;
END IF;
-- Have we already done this place?
select * from place where osm_type = NEW.osm_type and osm_id = NEW.osm_id and class = NEW.class and type = NEW.type INTO existing;
-- Get the existing place_id
select * from placex where osm_type = NEW.osm_type and osm_id = NEW.osm_id and class = NEW.class and type = NEW.type INTO existingplacex;
-- Handle a place changing type by removing the old data
-- My generated 'place' types are causing havok because they overlap with real keys
-- TODO: move them to their own special purpose key/class to avoid collisions
IF existing.osm_type IS NULL THEN
DELETE FROM place where osm_type = NEW.osm_type and osm_id = NEW.osm_id and class = NEW.class;
END IF;
--DEBUG: RAISE WARNING 'Existing: %',existing.osm_id;
--DEBUG: RAISE WARNING 'Existing PlaceX: %',existingplacex.place_id;
-- Log and discard
IF existing.geometry is not null AND st_isvalid(existing.geometry)
AND st_area(existing.geometry) > 0.02
AND ST_GeometryType(NEW.geometry) in ('ST_Polygon','ST_MultiPolygon')
AND st_area(NEW.geometry) < st_area(existing.geometry)*0.5
THEN
INSERT INTO import_polygon_error (osm_type, osm_id, class, type, name, country_code, updated, errormessage, prevgeometry, newgeometry)
VALUES (NEW.osm_type, NEW.osm_id, NEW.class, NEW.type, NEW.name, NEW.address->'country', now(),
'Area reduced from '||st_area(existing.geometry)||' to '||st_area(NEW.geometry), existing.geometry, NEW.geometry);
RETURN null;
END IF;
DELETE from import_polygon_error where osm_type = NEW.osm_type and osm_id = NEW.osm_id;
DELETE from import_polygon_delete where osm_type = NEW.osm_type and osm_id = NEW.osm_id;
-- To paraphrase, if there isn't an existing item, OR if the admin level has changed
IF existingplacex.osm_type IS NULL OR
(existingplacex.class = 'boundary' AND
((coalesce(existingplacex.admin_level, 15) != coalesce(NEW.admin_level, 15) AND existingplacex.type = 'administrative') OR
(existingplacex.type != NEW.type)))
THEN
IF existingplacex.osm_type IS NOT NULL THEN
-- sanity check: ignore admin_level changes on places with too many active children
-- or we end up reindexing entire countries because somebody accidentally deleted admin_level
--LIMIT INDEXING: SELECT count(*) FROM (SELECT 'a' FROM placex , place_addressline where address_place_id = existingplacex.place_id and placex.place_id = place_addressline.place_id and indexed_status = 0 and place_addressline.isaddress LIMIT 100001) sub INTO i;
--LIMIT INDEXING: IF i > 100000 THEN
--LIMIT INDEXING: RETURN null;
--LIMIT INDEXING: END IF;
END IF;
IF existing.osm_type IS NOT NULL THEN
-- pathological case caused by the triggerless copy into place during initial import
-- force delete even for large areas, it will be reinserted later
UPDATE place set geometry = ST_SetSRID(ST_Point(0,0), 4326) where osm_type = NEW.osm_type and osm_id = NEW.osm_id and class = NEW.class and type = NEW.type;
DELETE from place where osm_type = NEW.osm_type and osm_id = NEW.osm_id and class = NEW.class and type = NEW.type;
END IF;
-- No - process it as a new insertion (hopefully of low rank or it will be slow)
insert into placex (osm_type, osm_id, class, type, name,
admin_level, address, extratags, geometry)
values (NEW.osm_type, NEW.osm_id, NEW.class, NEW.type, NEW.name,
NEW.admin_level, NEW.address, NEW.extratags, NEW.geometry);
--DEBUG: RAISE WARNING 'insert done % % % % %',NEW.osm_type,NEW.osm_id,NEW.class,NEW.type,NEW.name;
RETURN NEW;
END IF;
-- Special case for polygon shape changes because they tend to be large and we can be a bit clever about how we handle them
IF existing.geometry::text != NEW.geometry::text
AND ST_GeometryType(existing.geometry) in ('ST_Polygon','ST_MultiPolygon')
AND ST_GeometryType(NEW.geometry) in ('ST_Polygon','ST_MultiPolygon')
THEN
-- Get the version of the geometry actually used (in placex table)
select geometry from placex where osm_type = NEW.osm_type and osm_id = NEW.osm_id and class = NEW.class and type = NEW.type into existinggeometry;
-- Performance limit
IF st_area(NEW.geometry) < 0.000000001 AND st_area(existinggeometry) < 1 THEN
-- re-index points that have moved in / out of the polygon, could be done as a single query but postgres gets the index usage wrong
update placex set indexed_status = 2 where indexed_status = 0
AND ST_Intersects(NEW.geometry, placex.geometry)
AND NOT ST_Intersects(existinggeometry, placex.geometry)
AND rank_search > existingplacex.rank_search AND (rank_search < 28 or name is not null);
update placex set indexed_status = 2 where indexed_status = 0
AND ST_Intersects(existinggeometry, placex.geometry)
AND NOT ST_Intersects(NEW.geometry, placex.geometry)
AND rank_search > existingplacex.rank_search AND (rank_search < 28 or name is not null);
END IF;
END IF;
IF coalesce(existing.name::text, '') != coalesce(NEW.name::text, '')
OR coalesce(existing.extratags::text, '') != coalesce(NEW.extratags::text, '')
OR coalesce(existing.address, ''::hstore) != coalesce(NEW.address, ''::hstore)
OR coalesce(existing.admin_level, 15) != coalesce(NEW.admin_level, 15)
OR existing.geometry::text != NEW.geometry::text
THEN
update place set
name = NEW.name,
address = NEW.address,
extratags = NEW.extratags,
admin_level = NEW.admin_level,
geometry = NEW.geometry
where osm_type = NEW.osm_type and osm_id = NEW.osm_id and class = NEW.class and type = NEW.type;
IF NEW.class in ('place','boundary') AND NEW.type in ('postcode','postal_code') THEN
IF NEW.address is NULL OR NOT NEW.address ? 'postcode' THEN
-- postcode was deleted, no longer retain in placex
DELETE FROM placex where place_id = existingplacex.place_id;
RETURN NULL;
END IF;
NEW.name := hstore('ref', NEW.address->'postcode');
END IF;
IF NEW.class in ('boundary')
AND ST_GeometryType(NEW.geometry) not in ('ST_Polygon','ST_MultiPolygon') THEN
DELETE FROM placex where place_id = existingplacex.place_id;
RETURN NULL;
END IF;
update placex set
name = NEW.name,
address = NEW.address,
parent_place_id = null,
extratags = NEW.extratags,
admin_level = NEW.admin_level,
indexed_status = 2,
geometry = NEW.geometry
where place_id = existingplacex.place_id;
-- if a node(=>house), which is part of a interpolation line, changes (e.g. the street attribute) => mark this line for reparenting
-- (already here, because interpolation lines are reindexed before nodes, so in the second call it would be too late)
IF NEW.osm_type='N'
and (coalesce(existing.address, ''::hstore) != coalesce(NEW.address, ''::hstore)
or existing.geometry::text != NEW.geometry::text)
THEN
result:= osmline_reinsert(NEW.osm_id, NEW.geometry);
END IF;
-- linked places should get potential new naming and addresses
IF existingplacex.linked_place_id is not NULL THEN
update placex x set
name = p.name,
extratags = p.extratags,
indexed_status = 2
from place p
where x.place_id = existingplacex.linked_place_id
and x.indexed_status = 0
and x.osm_type = p.osm_type
and x.osm_id = p.osm_id
and x.class = p.class;
END IF;
END IF;
-- Abort the add (we modified the existing place instead)
RETURN NULL;
END IF;
END;
$$ LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION place_delete()
RETURNS TRIGGER
AS $$
DECLARE
has_rank BOOLEAN;
BEGIN
--DEBUG: RAISE WARNING 'delete: % % % %',OLD.osm_type,OLD.osm_id,OLD.class,OLD.type;
-- deleting large polygons can have a massive effect on the system - require manual intervention to let them through
IF st_area(OLD.geometry) > 2 and st_isvalid(OLD.geometry) THEN
SELECT bool_or(not (rank_address = 0 or rank_address > 26)) as ranked FROM placex WHERE osm_type = OLD.osm_type and osm_id = OLD.osm_id and class = OLD.class and type = OLD.type INTO has_rank;
IF has_rank THEN
insert into import_polygon_delete (osm_type, osm_id, class, type) values (OLD.osm_type,OLD.osm_id,OLD.class,OLD.type);
RETURN NULL;
END IF;
END IF;
-- mark for delete
UPDATE placex set indexed_status = 100 where osm_type = OLD.osm_type and osm_id = OLD.osm_id and class = OLD.class and type = OLD.type;
-- interpolations are special
IF OLD.osm_type='W' and OLD.class = 'place' and OLD.type = 'houses' THEN
UPDATE location_property_osmline set indexed_status = 100 where osm_id = OLD.osm_id; -- osm_id = wayid (=old.osm_id)
END IF;
RETURN OLD;
END;
$$
LANGUAGE plpgsql;

View File

@@ -0,0 +1,965 @@
-- Trigger functions for the placex table.
-- Find the parent road of a POI.
--
-- \returns Place ID of parent object or NULL if none
--
-- Copy data from linked items (POIs on ways, addr:street links, relations).
--
CREATE OR REPLACE FUNCTION find_parent_for_poi(poi_osm_type CHAR(1),
poi_osm_id BIGINT,
poi_partition SMALLINT,
bbox GEOMETRY,
addr_street TEXT,
addr_place TEXT,
fallback BOOL = true)
RETURNS BIGINT
AS $$
DECLARE
parent_place_id BIGINT DEFAULT NULL;
location RECORD;
parent RECORD;
BEGIN
--DEBUG: RAISE WARNING 'finding street for % %', poi_osm_type, poi_osm_id;
-- Is this object part of an associatedStreet relation?
FOR location IN
SELECT members FROM planet_osm_rels
WHERE parts @> ARRAY[poi_osm_id]
and members @> ARRAY[lower(poi_osm_type) || poi_osm_id]
and tags @> ARRAY['associatedStreet']
LOOP
FOR i IN 1..array_upper(location.members, 1) BY 2 LOOP
IF location.members[i+1] = 'street' THEN
--DEBUG: RAISE WARNING 'node in relation %',relation;
FOR parent IN
SELECT place_id from placex
WHERE osm_type = 'W' and osm_id = substring(location.members[i],2)::bigint
and name is not null
and rank_search between 26 and 27
LOOP
RETURN parent.place_id;
END LOOP;
END IF;
END LOOP;
END LOOP;
parent_place_id := find_parent_for_address(addr_street, addr_place,
poi_partition, bbox);
IF parent_place_id is not null THEN
RETURN parent_place_id;
END IF;
IF poi_osm_type = 'N' THEN
-- Is this node part of an interpolation?
FOR parent IN
SELECT q.parent_place_id
FROM location_property_osmline q, planet_osm_ways x
WHERE q.linegeo && bbox and x.id = q.osm_id
and poi_osm_id = any(x.nodes)
LIMIT 1
LOOP
--DEBUG: RAISE WARNING 'Get parent from interpolation: %', parent.parent_place_id;
RETURN parent.parent_place_id;
END LOOP;
-- Is this node part of any other way?
FOR location IN
SELECT p.place_id, p.osm_id, p.rank_search, p.address,
coalesce(p.centroid, ST_Centroid(p.geometry)) as centroid
FROM placex p, planet_osm_ways w
WHERE p.osm_type = 'W' and p.rank_search >= 26
and p.geometry && bbox
and w.id = p.osm_id and poi_osm_id = any(w.nodes)
LOOP
--DEBUG: RAISE WARNING 'Node is part of way % ', location.osm_id;
-- Way IS a road then we are on it - that must be our road
IF location.rank_search < 28 THEN
--DEBUG: RAISE WARNING 'node in way that is a street %',location;
return location.place_id;
END IF;
SELECT find_parent_for_poi('W', location.osm_id, poi_partition,
location.centroid,
location.address->'street',
location.address->'place',
false)
INTO parent_place_id;
IF parent_place_id is not null THEN
RETURN parent_place_id;
END IF;
END LOOP;
END IF;
IF fallback THEN
IF ST_Area(bbox) < 0.01 THEN
-- for smaller features get the nearest road
SELECT getNearestRoadPlaceId(poi_partition, bbox) INTO parent_place_id;
--DEBUG: RAISE WARNING 'Checked for nearest way (%)', parent_place_id;
ELSE
-- for larger features simply find the area with the largest rank that
-- contains the bbox, only use addressable features
FOR location IN
SELECT place_id FROM placex
WHERE bbox @ geometry AND _ST_Covers(geometry, ST_Centroid(bbox))
AND rank_address between 5 and 25
ORDER BY rank_address desc
LOOP
RETURN location.place_id;
END LOOP;
END IF;
END IF;
RETURN parent_place_id;
END;
$$
LANGUAGE plpgsql STABLE;
-- Try to find a linked place for the given object.
CREATE OR REPLACE FUNCTION find_linked_place(bnd placex)
RETURNS placex
AS $$
DECLARE
relation_members TEXT[];
rel_member RECORD;
linked_placex placex%ROWTYPE;
bnd_name TEXT;
BEGIN
IF bnd.rank_search >= 26 or bnd.rank_address = 0
or ST_GeometryType(bnd.geometry) NOT IN ('ST_Polygon','ST_MultiPolygon')
THEN
RETURN NULL;
END IF;
IF bnd.osm_type = 'R' THEN
-- see if we have any special relation members
SELECT members FROM planet_osm_rels WHERE id = bnd.osm_id INTO relation_members;
--DEBUG: RAISE WARNING 'Got relation members';
-- Search for relation members with role 'lable'.
IF relation_members IS NOT NULL THEN
FOR rel_member IN
SELECT get_rel_node_members(relation_members, ARRAY['label']) as member
LOOP
--DEBUG: RAISE WARNING 'Found label member %', rel_member.member;
FOR linked_placex IN
SELECT * from placex
WHERE osm_type = 'N' and osm_id = rel_member.member
and class = 'place'
LOOP
--DEBUG: RAISE WARNING 'Linked label member';
RETURN linked_placex;
END LOOP;
END LOOP;
END IF;
END IF;
IF bnd.name ? 'name' THEN
bnd_name := make_standard_name(bnd.name->'name');
IF bnd_name = '' THEN
bnd_name := NULL;
END IF;
END IF;
-- If extratags has a place tag, look for linked nodes by their place type.
-- Area and node still have to have the same name.
IF bnd.extratags ? 'place' and bnd_name is not null THEN
FOR linked_placex IN
SELECT * FROM placex
WHERE make_standard_name(name->'name') = bnd_name
AND placex.class = 'place' AND placex.type = bnd.extratags->'place'
AND placex.osm_type = 'N'
AND placex.rank_search < 26 -- needed to select the right index
AND _st_covers(bnd.geometry, placex.geometry)
LOOP
--DEBUG: RAISE WARNING 'Found type-matching place node %', linked_placex.osm_id;
RETURN linked_placex;
END LOOP;
END IF;
IF bnd.extratags ? 'wikidata' THEN
FOR linked_placex IN
SELECT * FROM placex
WHERE placex.class = 'place' AND placex.osm_type = 'N'
AND placex.extratags ? 'wikidata' -- needed to select right index
AND placex.extratags->'wikidata' = bnd.extratags->'wikidata'
AND placex.rank_search < 26
AND _st_covers(bnd.geometry, placex.geometry)
ORDER BY make_standard_name(name->'name') = bnd_name desc
LOOP
--DEBUG: RAISE WARNING 'Found wikidata-matching place node %', linked_placex.osm_id;
RETURN linked_placex;
END LOOP;
END IF;
-- Name searches can be done for ways as well as relations
IF bnd_name is not null THEN
--DEBUG: RAISE WARNING 'Looking for nodes with matching names';
FOR linked_placex IN
SELECT placex.* from placex
WHERE make_standard_name(name->'name') = bnd_name
AND ((bnd.rank_address > 0 and placex.rank_address = bnd.rank_address)
OR (bnd.rank_address = 0 and placex.rank_search = bnd.rank_search))
AND placex.osm_type = 'N'
AND placex.rank_search < 26 -- needed to select the right index
AND _st_covers(bnd.geometry, placex.geometry)
LOOP
--DEBUG: RAISE WARNING 'Found matching place node %', linked_placex.osm_id;
RETURN linked_placex;
END LOOP;
END IF;
RETURN NULL;
END;
$$
LANGUAGE plpgsql STABLE;
-- Insert address of a place into the place_addressline table.
--
-- \param obj_place_id Place_id of the place to compute the address for.
-- \param partition Partition number where the place is in.
-- \param maxrank Rank of the place. All address features must have
-- a search rank lower than the given rank.
-- \param address Address terms for the place.
-- \param geoemtry Geometry to which the address objects should be close.
--
-- \retval parent_place_id Place_id of the address object that is the direct
-- ancestor.
-- \retval postcode Postcode computed from the address. This is the
-- addr:postcode of one of the address objects. If
-- more than one of has a postcode, the highest ranking
-- one is used. May be NULL.
-- \retval nameaddress_vector Search terms for the address. This is the sum
-- of name terms of all address objects.
CREATE OR REPLACE FUNCTION insert_addresslines(obj_place_id BIGINT,
partition SMALLINT,
maxrank SMALLINT,
address HSTORE,
geometry GEOMETRY,
OUT parent_place_id BIGINT,
OUT postcode TEXT,
OUT nameaddress_vector INT[])
AS $$
DECLARE
current_rank_address INTEGER := 0;
location_distance FLOAT := 0;
location_parent GEOMETRY := NULL;
parent_place_id_rank SMALLINT := 0;
location_isaddress BOOLEAN;
address_havelevel BOOLEAN[];
location_keywords INT[];
location RECORD;
addr_item RECORD;
isin_tokens INT[];
isin TEXT[];
BEGIN
parent_place_id := 0;
nameaddress_vector := '{}'::int[];
isin_tokens := '{}'::int[];
---- convert address store to array of tokenids
IF address IS NOT NULL THEN
FOR addr_item IN SELECT * FROM each(address)
LOOP
IF addr_item.key IN ('city', 'tiger:county', 'state', 'suburb', 'province',
'district', 'region', 'county', 'municipality',
'hamlet', 'village', 'subdistrict', 'town',
'neighbourhood', 'quarter', 'parish')
THEN
isin_tokens := array_merge(isin_tokens,
word_ids_from_name(addr_item.value));
IF NOT %REVERSE-ONLY% THEN
nameaddress_vector := array_merge(nameaddress_vector,
addr_ids_from_name(addr_item.value));
END IF;
END IF;
END LOOP;
IF address ? 'is_in' THEN
-- is_in items need splitting
isin := regexp_split_to_array(address->'is_in', E'[;,]');
IF array_upper(isin, 1) IS NOT NULL THEN
FOR i IN 1..array_upper(isin, 1) LOOP
isin_tokens := array_merge(isin_tokens,
word_ids_from_name(isin[i]));
-- merge word into address vector
IF NOT %REVERSE-ONLY% THEN
nameaddress_vector := array_merge(nameaddress_vector,
addr_ids_from_name(isin[i]));
END IF;
END LOOP;
END IF;
END IF;
END IF;
IF NOT %REVERSE-ONLY% THEN
nameaddress_vector := array_merge(nameaddress_vector, isin_tokens);
END IF;
---- now compute the address terms
FOR i IN 1..28 LOOP
address_havelevel[i] := false;
END LOOP;
FOR location IN
SELECT * FROM getNearFeatures(partition, geometry, maxrank, isin_tokens)
LOOP
IF location.rank_address != current_rank_address THEN
current_rank_address := location.rank_address;
IF location.isguess THEN
location_distance := location.distance * 1.5;
ELSE
IF location.rank_address <= 12 THEN
-- for county and above, if we have an area consider that exact
-- (It would be nice to relax the constraint for places close to
-- the boundary but we'd need the exact geometry for that. Too
-- expensive.)
location_distance = 0;
ELSE
-- Below county level remain slightly fuzzy.
location_distance := location.distance * 0.5;
END IF;
END IF;
ELSE
CONTINUE WHEN location.keywords <@ location_keywords;
END IF;
IF location.distance < location_distance OR NOT location.isguess THEN
location_keywords := location.keywords;
location_isaddress := NOT address_havelevel[location.rank_address];
--DEBUG: RAISE WARNING 'should be address: %, is guess: %, rank: %', location_isaddress, location.isguess, location.rank_address;
IF location_isaddress AND location.isguess AND location_parent IS NOT NULL THEN
location_isaddress := ST_Contains(location_parent, location.centroid);
END IF;
--DEBUG: RAISE WARNING '% isaddress: %', location.place_id, location_isaddress;
-- Add it to the list of search terms
IF NOT %REVERSE-ONLY% THEN
nameaddress_vector := array_merge(nameaddress_vector,
location.keywords::integer[]);
END IF;
INSERT INTO place_addressline (place_id, address_place_id, fromarea,
isaddress, distance, cached_rank_address)
VALUES (obj_place_id, location.place_id, true,
location_isaddress, location.distance, location.rank_address);
IF location_isaddress THEN
-- add postcode if we have one
-- (If multiple postcodes are available, we end up with the highest ranking one.)
IF location.postcode is not null THEN
postcode = location.postcode;
END IF;
address_havelevel[location.rank_address] := true;
-- add a hack against postcode ranks
IF NOT location.isguess
AND location.rank_address != 11 AND location.rank_address != 5
THEN
SELECT p.geometry FROM placex p
WHERE p.place_id = location.place_id INTO location_parent;
END IF;
IF location.rank_address > parent_place_id_rank THEN
parent_place_id = location.place_id;
parent_place_id_rank = location.rank_address;
END IF;
END IF;
END IF;
END LOOP;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION placex_insert()
RETURNS TRIGGER
AS $$
DECLARE
postcode TEXT;
result BOOLEAN;
is_area BOOLEAN;
country_code VARCHAR(2);
diameter FLOAT;
classtable TEXT;
BEGIN
--DEBUG: RAISE WARNING '% % % %',NEW.osm_type,NEW.osm_id,NEW.class,NEW.type;
NEW.place_id := nextval('seq_place');
NEW.indexed_status := 1; --STATUS_NEW
NEW.country_code := lower(get_country_code(NEW.geometry));
NEW.partition := get_partition(NEW.country_code);
NEW.geometry_sector := geometry_sector(NEW.partition, NEW.geometry);
IF NEW.osm_type = 'X' THEN
-- E'X'ternal records should already be in the right format so do nothing
ELSE
is_area := ST_GeometryType(NEW.geometry) IN ('ST_Polygon','ST_MultiPolygon');
IF NEW.class in ('place','boundary')
AND NEW.type in ('postcode','postal_code')
THEN
IF NEW.address IS NULL OR NOT NEW.address ? 'postcode' THEN
-- most likely just a part of a multipolygon postcode boundary, throw it away
RETURN NULL;
END IF;
NEW.name := hstore('ref', NEW.address->'postcode');
ELSEIF NEW.class = 'boundary' AND NOT is_area THEN
RETURN NULL;
ELSEIF NEW.class = 'boundary' AND NEW.type = 'administrative'
AND NEW.admin_level <= 4 AND NEW.osm_type = 'W'
THEN
RETURN NULL;
END IF;
SELECT * INTO NEW.rank_search, NEW.rank_address
FROM compute_place_rank(NEW.country_code,
CASE WHEN is_area THEN 'A' ELSE NEW.osm_type END,
NEW.class, NEW.type, NEW.admin_level,
(NEW.extratags->'capital') = 'yes',
NEW.address->'postcode');
-- a country code make no sense below rank 4 (country)
IF NEW.rank_search < 4 THEN
NEW.country_code := NULL;
END IF;
END IF;
--DEBUG: RAISE WARNING 'placex_insert:END: % % % %',NEW.osm_type,NEW.osm_id,NEW.class,NEW.type;
RETURN NEW; -- %DIFFUPDATES% The following is not needed until doing diff updates, and slows the main index process down
IF NEW.osm_type = 'N' and NEW.rank_search > 28 THEN
-- might be part of an interpolation
result := osmline_reinsert(NEW.osm_id, NEW.geometry);
ELSEIF NEW.rank_address > 0 THEN
IF (ST_GeometryType(NEW.geometry) in ('ST_Polygon','ST_MultiPolygon') AND ST_IsValid(NEW.geometry)) THEN
-- Performance: We just can't handle re-indexing for country level changes
IF st_area(NEW.geometry) < 1 THEN
-- mark items within the geometry for re-indexing
-- RAISE WARNING 'placex poly insert: % % % %',NEW.osm_type,NEW.osm_id,NEW.class,NEW.type;
-- work around bug in postgis, this may have been fixed in 2.0.0 (see http://trac.osgeo.org/postgis/ticket/547)
update placex set indexed_status = 2 where ST_Intersects(NEW.geometry, placex.geometry)
AND rank_search > NEW.rank_search and indexed_status = 0 and ST_geometrytype(placex.geometry) = 'ST_Point' and (rank_search < 28 or name is not null or (NEW.rank_search >= 16 and address ? 'place'));
update placex set indexed_status = 2 where ST_Intersects(NEW.geometry, placex.geometry)
AND rank_search > NEW.rank_search and indexed_status = 0 and ST_geometrytype(placex.geometry) != 'ST_Point' and (rank_search < 28 or name is not null or (NEW.rank_search >= 16 and address ? 'place'));
END IF;
ELSE
-- mark nearby items for re-indexing, where 'nearby' depends on the features rank_search and is a complete guess :(
diameter := update_place_diameter(NEW.rank_search);
IF diameter > 0 THEN
-- RAISE WARNING 'placex point insert: % % % % %',NEW.osm_type,NEW.osm_id,NEW.class,NEW.type,diameter;
IF NEW.rank_search >= 26 THEN
-- roads may cause reparenting for >27 rank places
update placex set indexed_status = 2 where indexed_status = 0 and rank_search > NEW.rank_search and ST_DWithin(placex.geometry, NEW.geometry, diameter);
-- reparenting also for OSM Interpolation Lines (and for Tiger?)
update location_property_osmline set indexed_status = 2 where indexed_status = 0 and ST_DWithin(location_property_osmline.linegeo, NEW.geometry, diameter);
ELSEIF NEW.rank_search >= 16 THEN
-- up to rank 16, street-less addresses may need reparenting
update placex set indexed_status = 2 where indexed_status = 0 and rank_search > NEW.rank_search and ST_DWithin(placex.geometry, NEW.geometry, diameter) and (rank_search < 28 or name is not null or address ? 'place');
ELSE
-- for all other places the search terms may change as well
update placex set indexed_status = 2 where indexed_status = 0 and rank_search > NEW.rank_search and ST_DWithin(placex.geometry, NEW.geometry, diameter) and (rank_search < 28 or name is not null);
END IF;
END IF;
END IF;
END IF;
-- add to tables for special search
-- Note: won't work on initial import because the classtype tables
-- do not yet exist. It won't hurt either.
classtable := 'place_classtype_' || NEW.class || '_' || NEW.type;
SELECT count(*)>0 FROM pg_tables WHERE tablename = classtable and schemaname = current_schema() INTO result;
IF result THEN
EXECUTE 'INSERT INTO ' || classtable::regclass || ' (place_id, centroid) VALUES ($1,$2)'
USING NEW.place_id, ST_Centroid(NEW.geometry);
END IF;
RETURN NEW;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION get_parent_address_level(geom GEOMETRY, in_level SMALLINT)
RETURNS SMALLINT
AS $$
DECLARE
address_rank SMALLINT;
BEGIN
IF in_level <= 3 or in_level > 15 THEN
address_rank := 3;
ELSE
SELECT rank_address INTO address_rank
FROM placex
WHERE osm_type = 'R' and class = 'boundary' and type = 'administrative'
and admin_level < in_level
and geometry && geom and ST_Covers(geometry, geom)
ORDER BY admin_level desc LIMIT 1;
END IF;
IF address_rank is NULL or address_rank <= 3 THEN
RETURN 3;
END IF;
RETURN address_rank;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION placex_update()
RETURNS TRIGGER
AS $$
DECLARE
i INTEGER;
location RECORD;
relation_members TEXT[];
centroid GEOMETRY;
parent_address_level SMALLINT;
addr_street TEXT;
addr_place TEXT;
name_vector INTEGER[];
nameaddress_vector INTEGER[];
linked_node_id BIGINT;
linked_importance FLOAT;
linked_wikipedia TEXT;
result BOOLEAN;
BEGIN
-- deferred delete
IF OLD.indexed_status = 100 THEN
--DEBUG: RAISE WARNING 'placex_update delete % %',NEW.osm_type,NEW.osm_id;
delete from placex where place_id = OLD.place_id;
RETURN NULL;
END IF;
IF NEW.indexed_status != 0 OR OLD.indexed_status = 0 THEN
RETURN NEW;
END IF;
--DEBUG: RAISE WARNING 'placex_update % % (%)',NEW.osm_type,NEW.osm_id,NEW.place_id;
NEW.indexed_date = now();
IF NOT %REVERSE-ONLY% THEN
DELETE from search_name WHERE place_id = NEW.place_id;
END IF;
result := deleteSearchName(NEW.partition, NEW.place_id);
DELETE FROM place_addressline WHERE place_id = NEW.place_id;
result := deleteRoad(NEW.partition, NEW.place_id);
result := deleteLocationArea(NEW.partition, NEW.place_id, NEW.rank_search);
UPDATE placex set linked_place_id = null, indexed_status = 2
where linked_place_id = NEW.place_id;
-- update not necessary for osmline, cause linked_place_id does not exist
IF NEW.linked_place_id is not null THEN
--DEBUG: RAISE WARNING 'place already linked to %', NEW.linked_place_id;
RETURN NEW;
END IF;
-- recompute the ranks, they might change when linking changes
SELECT * INTO NEW.rank_search, NEW.rank_address
FROM compute_place_rank(NEW.country_code,
CASE WHEN ST_GeometryType(NEW.geometry)
IN ('ST_Polygon','ST_MultiPolygon')
THEN 'A' ELSE NEW.osm_type END,
NEW.class, NEW.type, NEW.admin_level,
(NEW.extratags->'capital') = 'yes',
NEW.address->'postcode');
--DEBUG: RAISE WARNING 'Copy over address tags';
-- housenumber is a computed field, so start with an empty value
NEW.housenumber := NULL;
IF NEW.address is not NULL THEN
IF NEW.address ? 'conscriptionnumber' THEN
i := getorcreate_housenumber_id(make_standard_name(NEW.address->'conscriptionnumber'));
IF NEW.address ? 'streetnumber' THEN
i := getorcreate_housenumber_id(make_standard_name(NEW.address->'streetnumber'));
NEW.housenumber := (NEW.address->'conscriptionnumber') || '/' || (NEW.address->'streetnumber');
ELSE
NEW.housenumber := NEW.address->'conscriptionnumber';
END IF;
ELSEIF NEW.address ? 'streetnumber' THEN
NEW.housenumber := NEW.address->'streetnumber';
i := getorcreate_housenumber_id(make_standard_name(NEW.address->'streetnumber'));
ELSEIF NEW.address ? 'housenumber' THEN
NEW.housenumber := NEW.address->'housenumber';
i := getorcreate_housenumber_id(make_standard_name(NEW.housenumber));
END IF;
addr_street := NEW.address->'street';
addr_place := NEW.address->'place';
IF NEW.address ? 'postcode' and NEW.address->'postcode' not similar to '%(,|;)%' THEN
i := getorcreate_postcode_id(NEW.address->'postcode');
END IF;
END IF;
-- Speed up searches - just use the centroid of the feature
-- cheaper but less acurate
NEW.centroid := ST_PointOnSurface(NEW.geometry);
--DEBUG: RAISE WARNING 'Computing preliminary centroid at %',ST_AsText(NEW.centroid);
NEW.postcode := null;
-- recalculate country and partition
IF NEW.rank_search = 4 AND NEW.address is not NULL AND NEW.address ? 'country' THEN
-- for countries, believe the mapped country code,
-- so that we remain in the right partition if the boundaries
-- suddenly expand.
NEW.country_code := lower(NEW.address->'country');
NEW.partition := get_partition(lower(NEW.country_code));
IF NEW.partition = 0 THEN
NEW.country_code := lower(get_country_code(NEW.centroid));
NEW.partition := get_partition(NEW.country_code);
END IF;
ELSE
IF NEW.rank_search >= 4 THEN
NEW.country_code := lower(get_country_code(NEW.centroid));
ELSE
NEW.country_code := NULL;
END IF;
NEW.partition := get_partition(NEW.country_code);
END IF;
--DEBUG: RAISE WARNING 'Country updated: "%"', NEW.country_code;
-- waterway ways are linked when they are part of a relation and have the same class/type
IF NEW.osm_type = 'R' and NEW.class = 'waterway' THEN
FOR relation_members IN select members from planet_osm_rels r where r.id = NEW.osm_id and r.parts != array[]::bigint[]
LOOP
FOR i IN 1..array_upper(relation_members, 1) BY 2 LOOP
IF relation_members[i+1] in ('', 'main_stream', 'side_stream') AND substring(relation_members[i],1,1) = 'w' THEN
--DEBUG: RAISE WARNING 'waterway parent %, child %/%', NEW.osm_id, i, relation_members[i];
FOR linked_node_id IN SELECT place_id FROM placex
WHERE osm_type = 'W' and osm_id = substring(relation_members[i],2,200)::bigint
and class = NEW.class and type in ('river', 'stream', 'canal', 'drain', 'ditch')
and ( relation_members[i+1] != 'side_stream' or NEW.name->'name' = name->'name')
LOOP
UPDATE placex SET linked_place_id = NEW.place_id WHERE place_id = linked_node_id;
IF NOT %REVERSE-ONLY% THEN
DELETE FROM search_name WHERE place_id = linked_node_id;
END IF;
END LOOP;
END IF;
END LOOP;
END LOOP;
--DEBUG: RAISE WARNING 'Waterway processed';
END IF;
NEW.importance := null;
SELECT wikipedia, importance
FROM compute_importance(NEW.extratags, NEW.country_code, NEW.osm_type, NEW.osm_id)
INTO NEW.wikipedia,NEW.importance;
--DEBUG: RAISE WARNING 'Importance computed from wikipedia: %', NEW.importance;
-- ---------------------------------------------------------------------------
-- For low level elements we inherit from our parent road
IF (NEW.rank_search > 27 OR (NEW.type = 'postcode' AND NEW.rank_search = 25)) THEN
--DEBUG: RAISE WARNING 'finding street for % %', NEW.osm_type, NEW.osm_id;
NEW.parent_place_id := null;
-- if we have a POI and there is no address information,
-- see if we can get it from a surrounding building
IF NEW.osm_type = 'N' AND addr_street IS NULL AND addr_place IS NULL
AND NEW.housenumber IS NULL THEN
FOR location IN
-- The additional && condition works around the misguided query
-- planner of postgis 3.0.
SELECT address from placex where ST_Covers(geometry, NEW.centroid)
and geometry && NEW.centroid
and (address ? 'housenumber' or address ? 'street' or address ? 'place')
and rank_search > 28 AND ST_GeometryType(geometry) in ('ST_Polygon','ST_MultiPolygon')
limit 1
LOOP
NEW.housenumber := location.address->'housenumber';
addr_street := location.address->'street';
addr_place := location.address->'place';
--DEBUG: RAISE WARNING 'Found surrounding building % %', location.osm_type, location.osm_id;
END LOOP;
END IF;
-- We have to find our parent road.
NEW.parent_place_id := find_parent_for_poi(NEW.osm_type, NEW.osm_id,
NEW.partition,
ST_Envelope(NEW.geometry),
addr_street, addr_place);
-- If we found the road take a shortcut here.
-- Otherwise fall back to the full address getting method below.
IF NEW.parent_place_id is not null THEN
-- Get the details of the parent road
SELECT p.country_code, p.postcode FROM placex p
WHERE p.place_id = NEW.parent_place_id INTO location;
NEW.country_code := location.country_code;
--DEBUG: RAISE WARNING 'Got parent details from search name';
-- determine postcode
IF NEW.address is not null AND NEW.address ? 'postcode' THEN
NEW.postcode = upper(trim(NEW.address->'postcode'));
ELSE
NEW.postcode := location.postcode;
END IF;
IF NEW.postcode is null THEN
NEW.postcode := get_nearest_postcode(NEW.country_code, NEW.geometry);
END IF;
-- If there is no name it isn't searchable, don't bother to create a search record
IF NEW.name is NULL THEN
--DEBUG: RAISE WARNING 'Not a searchable place % %', NEW.osm_type, NEW.osm_id;
return NEW;
END IF;
NEW.name := add_default_place_name(NEW.country_code, NEW.name);
name_vector := make_keywords(NEW.name);
-- Performance, it would be more acurate to do all the rest of the import
-- process but it takes too long
-- Just be happy with inheriting from parent road only
IF NEW.rank_search <= 25 and NEW.rank_address > 0 THEN
result := add_location(NEW.place_id, NEW.country_code, NEW.partition, name_vector, NEW.rank_search, NEW.rank_address, upper(trim(NEW.address->'postcode')), NEW.geometry);
--DEBUG: RAISE WARNING 'Place added to location table';
END IF;
result := insertSearchName(NEW.partition, NEW.place_id, name_vector,
NEW.rank_search, NEW.rank_address, NEW.geometry);
IF NOT %REVERSE-ONLY% THEN
-- Merge address from parent
SELECT array_merge(s.name_vector, s.nameaddress_vector)
INTO nameaddress_vector
FROM search_name s
WHERE s.place_id = NEW.parent_place_id;
INSERT INTO search_name (place_id, search_rank, address_rank,
importance, country_code, name_vector,
nameaddress_vector, centroid)
VALUES (NEW.place_id, NEW.rank_search, NEW.rank_address,
NEW.importance, NEW.country_code, name_vector,
nameaddress_vector, NEW.centroid);
--DEBUG: RAISE WARNING 'Place added to search table';
END IF;
return NEW;
END IF;
END IF;
-- ---------------------------------------------------------------------------
-- Full indexing
--DEBUG: RAISE WARNING 'Using full index mode for % %', NEW.osm_type, NEW.osm_id;
SELECT * INTO location FROM find_linked_place(NEW);
IF location.place_id is not null THEN
--DEBUG: RAISE WARNING 'Linked %', location;
-- Use the linked point as the centre point of the geometry,
-- but only if it is within the area of the boundary.
centroid := coalesce(location.centroid, ST_Centroid(location.geometry));
IF centroid is not NULL AND ST_Within(centroid, NEW.geometry) THEN
NEW.centroid := centroid;
END IF;
-- Use the address rank of the linked place, if it has one
parent_address_level := get_parent_address_level(NEW.geometry, NEW.admin_level);
--DEBUG: RAISE WARNING 'parent address: % rank address: %', parent_address_level, location.rank_address;
IF location.rank_address > parent_address_level
and location.rank_address < 26
THEN
NEW.rank_address := location.rank_address;
END IF;
-- merge in the label name
IF NOT location.name IS NULL THEN
NEW.name := location.name || NEW.name;
END IF;
-- merge in extra tags
NEW.extratags := hstore('linked_' || location.class, location.type)
|| coalesce(location.extratags, ''::hstore)
|| coalesce(NEW.extratags, ''::hstore);
-- mark the linked place (excludes from search results)
UPDATE placex set linked_place_id = NEW.place_id
WHERE place_id = location.place_id;
-- ensure that those places are not found anymore
IF NOT %REVERSE-ONLY% THEN
DELETE FROM search_name WHERE place_id = location.place_id;
END IF;
PERFORM deleteLocationArea(NEW.partition, location.place_id, NEW.rank_search);
SELECT wikipedia, importance
FROM compute_importance(location.extratags, NEW.country_code,
'N', location.osm_id)
INTO linked_wikipedia,linked_importance;
-- Use the maximum importance if one could be computed from the linked object.
IF linked_importance is not null AND
(NEW.importance is null or NEW.importance < linked_importance)
THEN
NEW.importance = linked_importance;
END IF;
END IF;
-- Initialise the name vector using our name
NEW.name := add_default_place_name(NEW.country_code, NEW.name);
name_vector := make_keywords(NEW.name);
-- make sure all names are in the word table
IF NEW.admin_level = 2
AND NEW.class = 'boundary' AND NEW.type = 'administrative'
AND NEW.country_code IS NOT NULL AND NEW.osm_type = 'R'
THEN
PERFORM create_country(NEW.name, lower(NEW.country_code));
--DEBUG: RAISE WARNING 'Country names updated';
END IF;
SELECT * FROM insert_addresslines(NEW.place_id, NEW.partition,
NEW.rank_search, NEW.address,
CASE WHEN NEW.rank_search >= 26
AND NEW.rank_search < 30
THEN NEW.geometry ELSE NEW.centroid END)
INTO NEW.parent_place_id, NEW.postcode, nameaddress_vector;
--DEBUG: RAISE WARNING 'RETURN insert_addresslines: %, %, %', NEW.parent_place_id, NEW.postcode, nameaddress_vector;
IF NEW.address is not null AND NEW.address ? 'postcode'
AND NEW.address->'postcode' not similar to '%(,|;)%' THEN
NEW.postcode := upper(trim(NEW.address->'postcode'));
END IF;
IF NEW.postcode is null AND NEW.rank_search > 8 THEN
NEW.postcode := get_nearest_postcode(NEW.country_code, NEW.geometry);
END IF;
-- if we have a name add this to the name search table
IF NEW.name IS NOT NULL THEN
IF NEW.rank_search <= 25 and NEW.rank_address > 0 THEN
result := add_location(NEW.place_id, NEW.country_code, NEW.partition, name_vector, NEW.rank_search, NEW.rank_address, upper(trim(NEW.address->'postcode')), NEW.geometry);
--DEBUG: RAISE WARNING 'added to location (full)';
END IF;
IF NEW.rank_search between 26 and 27 and NEW.class = 'highway' THEN
result := insertLocationRoad(NEW.partition, NEW.place_id, NEW.country_code, NEW.geometry);
--DEBUG: RAISE WARNING 'insert into road location table (full)';
END IF;
result := insertSearchName(NEW.partition, NEW.place_id, name_vector,
NEW.rank_search, NEW.rank_address, NEW.geometry);
--DEBUG: RAISE WARNING 'added to search name (full)';
IF NOT %REVERSE-ONLY% THEN
INSERT INTO search_name (place_id, search_rank, address_rank,
importance, country_code, name_vector,
nameaddress_vector, centroid)
VALUES (NEW.place_id, NEW.rank_search, NEW.rank_address,
NEW.importance, NEW.country_code, name_vector,
nameaddress_vector, NEW.centroid);
END IF;
END IF;
--DEBUG: RAISE WARNING 'place update % % finsihed.', NEW.osm_type, NEW.osm_id;
RETURN NEW;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION placex_delete()
RETURNS TRIGGER
AS $$
DECLARE
b BOOLEAN;
classtable TEXT;
BEGIN
-- RAISE WARNING 'placex_delete % %',OLD.osm_type,OLD.osm_id;
update placex set linked_place_id = null, indexed_status = 2 where linked_place_id = OLD.place_id and indexed_status = 0;
--DEBUG: RAISE WARNING 'placex_delete:01 % %',OLD.osm_type,OLD.osm_id;
update placex set linked_place_id = null where linked_place_id = OLD.place_id;
--DEBUG: RAISE WARNING 'placex_delete:02 % %',OLD.osm_type,OLD.osm_id;
IF OLD.rank_address < 30 THEN
-- mark everything linked to this place for re-indexing
--DEBUG: RAISE WARNING 'placex_delete:03 % %',OLD.osm_type,OLD.osm_id;
UPDATE placex set indexed_status = 2 from place_addressline where address_place_id = OLD.place_id
and placex.place_id = place_addressline.place_id and indexed_status = 0 and place_addressline.isaddress;
--DEBUG: RAISE WARNING 'placex_delete:04 % %',OLD.osm_type,OLD.osm_id;
DELETE FROM place_addressline where address_place_id = OLD.place_id;
--DEBUG: RAISE WARNING 'placex_delete:05 % %',OLD.osm_type,OLD.osm_id;
b := deleteRoad(OLD.partition, OLD.place_id);
--DEBUG: RAISE WARNING 'placex_delete:06 % %',OLD.osm_type,OLD.osm_id;
update placex set indexed_status = 2 where parent_place_id = OLD.place_id and indexed_status = 0;
--DEBUG: RAISE WARNING 'placex_delete:07 % %',OLD.osm_type,OLD.osm_id;
-- reparenting also for OSM Interpolation Lines (and for Tiger?)
update location_property_osmline set indexed_status = 2 where indexed_status = 0 and parent_place_id = OLD.place_id;
END IF;
--DEBUG: RAISE WARNING 'placex_delete:08 % %',OLD.osm_type,OLD.osm_id;
IF OLD.rank_address < 26 THEN
b := deleteLocationArea(OLD.partition, OLD.place_id, OLD.rank_search);
END IF;
--DEBUG: RAISE WARNING 'placex_delete:09 % %',OLD.osm_type,OLD.osm_id;
IF OLD.name is not null THEN
IF NOT %REVERSE-ONLY% THEN
DELETE from search_name WHERE place_id = OLD.place_id;
END IF;
b := deleteSearchName(OLD.partition, OLD.place_id);
END IF;
--DEBUG: RAISE WARNING 'placex_delete:10 % %',OLD.osm_type,OLD.osm_id;
DELETE FROM place_addressline where place_id = OLD.place_id;
--DEBUG: RAISE WARNING 'placex_delete:11 % %',OLD.osm_type,OLD.osm_id;
-- remove from tables for special search
classtable := 'place_classtype_' || OLD.class || '_' || OLD.type;
SELECT count(*)>0 FROM pg_tables WHERE tablename = classtable and schemaname = current_schema() INTO b;
IF b THEN
EXECUTE 'DELETE FROM ' || classtable::regclass || ' WHERE place_id = $1' USING OLD.place_id;
END IF;
--DEBUG: RAISE WARNING 'placex_delete:12 % %',OLD.osm_type,OLD.osm_id;
RETURN OLD;
END;
$$
LANGUAGE plpgsql;

View File

@@ -0,0 +1,40 @@
-- Trigger functions for location_postcode table.
-- Trigger for updates of location_postcode
--
-- Computes the parent object the postcode most likely refers to.
-- This will be the place that determines the address displayed when
-- searching for this postcode.
CREATE OR REPLACE FUNCTION postcode_update()
RETURNS TRIGGER
AS $$
DECLARE
partition SMALLINT;
location RECORD;
BEGIN
IF NEW.indexed_status != 0 OR OLD.indexed_status = 0 THEN
RETURN NEW;
END IF;
NEW.indexed_date = now();
partition := get_partition(NEW.country_code);
SELECT * FROM get_postcode_rank(NEW.country_code, NEW.postcode)
INTO NEW.rank_search, NEW.rank_address;
NEW.parent_place_id = 0;
FOR location IN
SELECT place_id
FROM getNearFeatures(partition, NEW.geometry, NEW.rank_search, '{}'::int[])
WHERE NOT isguess ORDER BY rank_address DESC LIMIT 1
LOOP
NEW.parent_place_id = location.place_id;
END LOOP;
RETURN NEW;
END;
$$
LANGUAGE plpgsql;

192
sql/functions/ranking.sql Normal file
View File

@@ -0,0 +1,192 @@
-- Functions related to search and address ranks
-- Return an approximate search radius according to the search rank.
CREATE OR REPLACE FUNCTION reverse_place_diameter(rank_search SMALLINT)
RETURNS FLOAT
AS $$
BEGIN
IF rank_search <= 4 THEN
RETURN 5.0;
ELSIF rank_search <= 8 THEN
RETURN 1.8;
ELSIF rank_search <= 12 THEN
RETURN 0.6;
ELSIF rank_search <= 17 THEN
RETURN 0.16;
ELSIF rank_search <= 18 THEN
RETURN 0.08;
ELSIF rank_search <= 19 THEN
RETURN 0.04;
END IF;
RETURN 0.02;
END;
$$
LANGUAGE plpgsql IMMUTABLE;
-- Return an approximate update radius according to the search rank.
CREATE OR REPLACE FUNCTION update_place_diameter(rank_search SMALLINT)
RETURNS FLOAT
AS $$
BEGIN
-- postcodes
IF rank_search = 11 or rank_search = 5 THEN
RETURN 0.05;
-- anything higher than city is effectively ignored (polygon required)
ELSIF rank_search < 16 THEN
RETURN 0;
ELSIF rank_search < 18 THEN
RETURN 0.1;
ELSIF rank_search < 20 THEN
RETURN 0.05;
ELSIF rank_search = 21 THEN
RETURN 0.001;
ELSIF rank_search < 24 THEN
RETURN 0.02;
ELSIF rank_search < 26 THEN
RETURN 0.002;
ELSIF rank_search < 28 THEN
RETURN 0.001;
END IF;
RETURN 0;
END;
$$
LANGUAGE plpgsql IMMUTABLE;
-- Guess a ranking for postcodes from country and postcode format.
CREATE OR REPLACE FUNCTION get_postcode_rank(country_code VARCHAR(2), postcode TEXT,
OUT rank_search SMALLINT,
OUT rank_address SMALLINT)
AS $$
DECLARE
part TEXT;
BEGIN
rank_search := 30;
rank_address := 30;
postcode := upper(postcode);
IF country_code = 'gb' THEN
IF postcode ~ '^([A-Z][A-Z]?[0-9][0-9A-Z]? [0-9][A-Z][A-Z])$' THEN
rank_search := 25;
rank_address := 5;
ELSEIF postcode ~ '^([A-Z][A-Z]?[0-9][0-9A-Z]? [0-9])$' THEN
rank_search := 23;
rank_address := 5;
ELSEIF postcode ~ '^([A-Z][A-Z]?[0-9][0-9A-Z])$' THEN
rank_search := 21;
rank_address := 5;
END IF;
ELSEIF country_code = 'sg' THEN
IF postcode ~ '^([0-9]{6})$' THEN
rank_search := 25;
rank_address := 11;
END IF;
ELSEIF country_code = 'de' THEN
IF postcode ~ '^([0-9]{5})$' THEN
rank_search := 21;
rank_address := 11;
END IF;
ELSE
-- Guess at the postcode format and coverage (!)
IF postcode ~ '^[A-Z0-9]{1,5}$' THEN -- Probably too short to be very local
rank_search := 21;
rank_address := 11;
ELSE
-- Does it look splitable into and area and local code?
part := substring(postcode from '^([- :A-Z0-9]+)([- :][A-Z0-9]+)$');
IF part IS NOT NULL THEN
rank_search := 25;
rank_address := 11;
ELSEIF postcode ~ '^[- :A-Z0-9]{6,}$' THEN
rank_search := 21;
rank_address := 11;
END IF;
END IF;
END IF;
END;
$$
LANGUAGE plpgsql IMMUTABLE;
-- Get standard search and address rank for an object.
--
-- \param country Two-letter country code where the object is in.
-- \param extended_type OSM type (N, W, R) or area type (A).
-- \param place_class Class (or tag key) of object.
-- \param place_type Type (or tag value) of object.
-- \param admin_level Value of admin_level tag.
-- \param is_major If true, boost search rank by one.
-- \param postcode Value of addr:postcode tag.
-- \param[out] search_rank Computed search rank.
-- \param[out] address_rank Computed address rank.
--
CREATE OR REPLACE FUNCTION compute_place_rank(country VARCHAR(2),
extended_type VARCHAR(1),
place_class TEXT, place_type TEXT,
admin_level SMALLINT,
is_major BOOLEAN,
postcode TEXT,
OUT search_rank SMALLINT,
OUT address_rank SMALLINT)
AS $$
DECLARE
classtype TEXT;
BEGIN
IF place_class in ('place','boundary')
and place_type in ('postcode','postal_code')
THEN
SELECT * INTO search_rank, address_rank
FROM get_postcode_rank(country, postcode);
IF NOT extended_type = 'A' THEN
address_rank := 0;
END IF;
ELSEIF extended_type = 'N' AND place_class = 'highway' THEN
search_rank = 30;
address_rank = 0;
ELSEIF place_class = 'landuse' AND extended_type != 'A' THEN
search_rank = 30;
address_rank = 0;
ELSE
IF place_class = 'boundary' and place_type = 'administrative' THEN
classtype = place_type || admin_level::TEXT;
ELSE
classtype = place_type;
END IF;
SELECT l.rank_search, l.rank_address INTO search_rank, address_rank
FROM address_levels l
WHERE (l.country_code = country or l.country_code is NULL)
AND l.class = place_class AND (l.type = classtype or l.type is NULL)
ORDER BY l.country_code, l.class, l.type LIMIT 1;
IF search_rank is NULL THEN
search_rank := 30;
END IF;
IF address_rank is NULL THEN
address_rank := 30;
END IF;
-- some postcorrections
IF place_class = 'waterway' AND extended_type = 'R' THEN
-- Slightly promote waterway relations so that they are processed
-- before their members.
search_rank := search_rank - 1;
END IF;
IF is_major THEN
search_rank := search_rank - 1;
END IF;
END IF;
END;
$$
LANGUAGE plpgsql IMMUTABLE;

505
sql/functions/utils.sql Normal file
View File

@@ -0,0 +1,505 @@
-- Assorted helper functions for the triggers.
CREATE OR REPLACE FUNCTION geometry_sector(partition INTEGER, place geometry)
RETURNS INTEGER
AS $$
DECLARE
NEWgeometry geometry;
BEGIN
-- RAISE WARNING '%',place;
NEWgeometry := ST_PointOnSurface(place);
RETURN (partition*1000000) + (500-ST_X(NEWgeometry)::integer)*1000 + (500-ST_Y(NEWgeometry)::integer);
END;
$$
LANGUAGE plpgsql IMMUTABLE;
CREATE OR REPLACE FUNCTION array_merge(a INTEGER[], b INTEGER[])
RETURNS INTEGER[]
AS $$
DECLARE
i INTEGER;
r INTEGER[];
BEGIN
IF array_upper(a, 1) IS NULL THEN
RETURN b;
END IF;
IF array_upper(b, 1) IS NULL THEN
RETURN a;
END IF;
r := a;
FOR i IN 1..array_upper(b, 1) LOOP
IF NOT (ARRAY[b[i]] <@ r) THEN
r := r || b[i];
END IF;
END LOOP;
RETURN r;
END;
$$
LANGUAGE plpgsql IMMUTABLE;
-- Return the node members with a given label from a relation member list
-- as a set.
--
-- \param members Member list in osm2pgsql middle format.
-- \param memberLabels Array of labels to accept.
--
-- \returns Set of OSM ids of nodes that are found.
--
CREATE OR REPLACE FUNCTION get_rel_node_members(members TEXT[],
memberLabels TEXT[])
RETURNS SETOF BIGINT
AS $$
DECLARE
i INTEGER;
BEGIN
FOR i IN 1..ARRAY_UPPER(members,1) BY 2 LOOP
IF members[i+1] = ANY(memberLabels)
AND upper(substring(members[i], 1, 1))::char(1) = 'N'
THEN
RETURN NEXT substring(members[i], 2)::bigint;
END IF;
END LOOP;
RETURN;
END;
$$
LANGUAGE plpgsql IMMUTABLE;
-- Copy 'name' to or from the default language.
--
-- \param country_code Country code of the object being named.
-- \param[inout] name List of names of the object.
--
-- If the country named by country_code has a single default language,
-- then a `name` tag is copied to `name:<country_code>` if this tag does
-- not yet exist and vice versa.
CREATE OR REPLACE FUNCTION add_default_place_name(country_code VARCHAR(2),
INOUT name HSTORE)
AS $$
DECLARE
default_language VARCHAR(10);
BEGIN
IF name is not null AND array_upper(akeys(name),1) > 1 THEN
default_language := get_country_language_code(country_code);
IF default_language IS NOT NULL THEN
IF name ? 'name' AND NOT name ? ('name:'||default_language) THEN
name := name || hstore(('name:'||default_language), (name -> 'name'));
ELSEIF name ? ('name:'||default_language) AND NOT name ? 'name' THEN
name := name || hstore('name', (name -> ('name:'||default_language)));
END IF;
END IF;
END IF;
END;
$$
LANGUAGE plpgsql IMMUTABLE;
-- Find the nearest artificial postcode for the given geometry.
-- TODO For areas there should not be more than two inside the geometry.
CREATE OR REPLACE FUNCTION get_nearest_postcode(country VARCHAR(2), geom GEOMETRY)
RETURNS TEXT
AS $$
DECLARE
outcode TEXT;
cnt INTEGER;
BEGIN
-- If the geometry is an area then only one postcode must be within
-- that area, otherwise consider the area as not having a postcode.
IF ST_GeometryType(geom) in ('ST_Polygon','ST_MultiPolygon') THEN
SELECT min(postcode), count(*) FROM
(SELECT postcode FROM location_postcode
WHERE ST_Contains(geom, location_postcode.geometry) LIMIT 2) sub
INTO outcode, cnt;
IF cnt = 1 THEN
RETURN outcode;
ELSE
RETURN null;
END IF;
END IF;
SELECT postcode FROM location_postcode
WHERE ST_DWithin(geom, location_postcode.geometry, 0.05)
AND location_postcode.country_code = country
ORDER BY ST_Distance(geom, location_postcode.geometry) LIMIT 1
INTO outcode;
RETURN outcode;
END;
$$
LANGUAGE plpgsql STABLE;
CREATE OR REPLACE FUNCTION get_country_code(place geometry)
RETURNS TEXT
AS $$
DECLARE
place_centre GEOMETRY;
nearcountry RECORD;
BEGIN
place_centre := ST_PointOnSurface(place);
-- RAISE WARNING 'get_country_code, start: %', ST_AsText(place_centre);
-- Try for a OSM polygon
FOR nearcountry IN
SELECT country_code from location_area_country
WHERE country_code is not null and st_covers(geometry, place_centre) limit 1
LOOP
RETURN nearcountry.country_code;
END LOOP;
-- RAISE WARNING 'osm fallback: %', ST_AsText(place_centre);
-- Try for OSM fallback data
-- The order is to deal with places like HongKong that are 'states' within another polygon
FOR nearcountry IN
SELECT country_code from country_osm_grid
WHERE st_covers(geometry, place_centre) order by area asc limit 1
LOOP
RETURN nearcountry.country_code;
END LOOP;
-- RAISE WARNING 'near osm fallback: %', ST_AsText(place_centre);
--
FOR nearcountry IN
SELECT country_code from country_osm_grid
WHERE st_dwithin(geometry, place_centre, 0.5)
ORDER BY st_distance(geometry, place_centre) asc, area asc limit 1
LOOP
RETURN nearcountry.country_code;
END LOOP;
RETURN NULL;
END;
$$
LANGUAGE plpgsql STABLE;
CREATE OR REPLACE FUNCTION get_country_language_code(search_country_code VARCHAR(2))
RETURNS TEXT
AS $$
DECLARE
nearcountry RECORD;
BEGIN
FOR nearcountry IN
SELECT distinct country_default_language_code from country_name
WHERE country_code = search_country_code limit 1
LOOP
RETURN lower(nearcountry.country_default_language_code);
END LOOP;
RETURN NULL;
END;
$$
LANGUAGE plpgsql STABLE;
CREATE OR REPLACE FUNCTION get_partition(in_country_code VARCHAR(10))
RETURNS INTEGER
AS $$
DECLARE
nearcountry RECORD;
BEGIN
FOR nearcountry IN
SELECT partition from country_name where country_code = in_country_code
LOOP
RETURN nearcountry.partition;
END LOOP;
RETURN 0;
END;
$$
LANGUAGE plpgsql STABLE;
-- Find the parent of an address with addr:street/addr:place tag.
--
-- \param street Value of addr:street or NULL if tag is missing.
-- \param place Value of addr:place or NULL if tag is missing.
-- \param partition Partition where to search the parent.
-- \param centroid Location of the address.
--
-- \return Place ID of the parent if one was found, NULL otherwise.
CREATE OR REPLACE FUNCTION find_parent_for_address(street TEXT, place TEXT,
partition SMALLINT,
centroid GEOMETRY)
RETURNS BIGINT
AS $$
DECLARE
parent_place_id BIGINT;
word_ids INTEGER[];
BEGIN
IF street is not null THEN
-- Check for addr:street attributes
-- Note that addr:street links can only be indexed, once the street itself is indexed
word_ids := word_ids_from_name(street);
IF word_ids is not null THEN
parent_place_id := getNearestNamedRoadPlaceId(partition, centroid, word_ids);
IF parent_place_id is not null THEN
--DEBUG: RAISE WARNING 'Get parent form addr:street: %', parent.place_id;
RETURN parent_place_id;
END IF;
END IF;
END IF;
-- Check for addr:place attributes.
IF place is not null THEN
word_ids := word_ids_from_name(place);
IF word_ids is not null THEN
parent_place_id := getNearestNamedPlacePlaceId(partition, centroid, word_ids);
IF parent_place_id is not null THEN
--DEBUG: RAISE WARNING 'Get parent form addr:place: %', parent.place_id;
RETURN parent_place_id;
END IF;
END IF;
END IF;
RETURN NULL;
END;
$$
LANGUAGE plpgsql STABLE;
CREATE OR REPLACE FUNCTION delete_location(OLD_place_id BIGINT)
RETURNS BOOLEAN
AS $$
DECLARE
BEGIN
DELETE FROM location_area where place_id = OLD_place_id;
-- TODO:location_area
RETURN true;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION add_location(place_id BIGINT, country_code varchar(2),
partition INTEGER, keywords INTEGER[],
rank_search INTEGER, rank_address INTEGER,
in_postcode TEXT, geometry GEOMETRY)
RETURNS BOOLEAN
AS $$
DECLARE
locationid INTEGER;
centroid GEOMETRY;
diameter FLOAT;
x BOOLEAN;
splitGeom RECORD;
secgeo GEOMETRY;
postcode TEXT;
BEGIN
IF rank_search > 25 THEN
RAISE EXCEPTION 'Adding location with rank > 25 (% rank %)', place_id, rank_search;
END IF;
x := deleteLocationArea(partition, place_id, rank_search);
-- add postcode only if it contains a single entry, i.e. ignore postcode lists
postcode := NULL;
IF in_postcode is not null AND in_postcode not similar to '%(,|;)%' THEN
postcode := upper(trim (in_postcode));
END IF;
IF ST_GeometryType(geometry) in ('ST_Polygon','ST_MultiPolygon') THEN
centroid := ST_Centroid(geometry);
FOR secgeo IN select split_geometry(geometry) AS geom LOOP
x := insertLocationAreaLarge(partition, place_id, country_code, keywords, rank_search, rank_address, false, postcode, centroid, secgeo);
END LOOP;
ELSE
diameter := 0.02;
IF rank_address = 0 THEN
diameter := 0.02;
ELSEIF rank_search <= 14 THEN
diameter := 1.2;
ELSEIF rank_search <= 15 THEN
diameter := 1;
ELSEIF rank_search <= 16 THEN
diameter := 0.5;
ELSEIF rank_search <= 17 THEN
diameter := 0.2;
ELSEIF rank_search <= 21 THEN
diameter := 0.05;
ELSEIF rank_search = 25 THEN
diameter := 0.005;
END IF;
-- RAISE WARNING 'adding % diameter %', place_id, diameter;
secgeo := ST_Buffer(geometry, diameter);
x := insertLocationAreaLarge(partition, place_id, country_code, keywords, rank_search, rank_address, true, postcode, ST_Centroid(geometry), secgeo);
END IF;
RETURN true;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION quad_split_geometry(geometry GEOMETRY, maxarea FLOAT,
maxdepth INTEGER)
RETURNS SETOF GEOMETRY
AS $$
DECLARE
xmin FLOAT;
ymin FLOAT;
xmax FLOAT;
ymax FLOAT;
xmid FLOAT;
ymid FLOAT;
secgeo GEOMETRY;
secbox GEOMETRY;
seg INTEGER;
geo RECORD;
area FLOAT;
remainingdepth INTEGER;
added INTEGER;
BEGIN
-- RAISE WARNING 'quad_split_geometry: maxarea=%, depth=%',maxarea,maxdepth;
IF (ST_GeometryType(geometry) not in ('ST_Polygon','ST_MultiPolygon') OR NOT ST_IsValid(geometry)) THEN
RETURN NEXT geometry;
RETURN;
END IF;
remainingdepth := maxdepth - 1;
area := ST_AREA(geometry);
IF remainingdepth < 1 OR area < maxarea THEN
RETURN NEXT geometry;
RETURN;
END IF;
xmin := st_xmin(geometry);
xmax := st_xmax(geometry);
ymin := st_ymin(geometry);
ymax := st_ymax(geometry);
secbox := ST_SetSRID(ST_MakeBox2D(ST_Point(ymin,xmin),ST_Point(ymax,xmax)),4326);
-- if the geometry completely covers the box don't bother to slice any more
IF ST_AREA(secbox) = area THEN
RETURN NEXT geometry;
RETURN;
END IF;
xmid := (xmin+xmax)/2;
ymid := (ymin+ymax)/2;
added := 0;
FOR seg IN 1..4 LOOP
IF seg = 1 THEN
secbox := ST_SetSRID(ST_MakeBox2D(ST_Point(xmin,ymin),ST_Point(xmid,ymid)),4326);
END IF;
IF seg = 2 THEN
secbox := ST_SetSRID(ST_MakeBox2D(ST_Point(xmin,ymid),ST_Point(xmid,ymax)),4326);
END IF;
IF seg = 3 THEN
secbox := ST_SetSRID(ST_MakeBox2D(ST_Point(xmid,ymin),ST_Point(xmax,ymid)),4326);
END IF;
IF seg = 4 THEN
secbox := ST_SetSRID(ST_MakeBox2D(ST_Point(xmid,ymid),ST_Point(xmax,ymax)),4326);
END IF;
IF st_intersects(geometry, secbox) THEN
secgeo := st_intersection(geometry, secbox);
IF NOT ST_IsEmpty(secgeo) AND ST_GeometryType(secgeo) in ('ST_Polygon','ST_MultiPolygon') THEN
FOR geo IN select quad_split_geometry(secgeo, maxarea, remainingdepth) as geom LOOP
IF NOT ST_IsEmpty(geo.geom) AND ST_GeometryType(geo.geom) in ('ST_Polygon','ST_MultiPolygon') THEN
added := added + 1;
RETURN NEXT geo.geom;
END IF;
END LOOP;
END IF;
END IF;
END LOOP;
RETURN;
END;
$$
LANGUAGE plpgsql IMMUTABLE;
CREATE OR REPLACE FUNCTION split_geometry(geometry GEOMETRY)
RETURNS SETOF GEOMETRY
AS $$
DECLARE
geo RECORD;
BEGIN
-- 10000000000 is ~~ 1x1 degree
FOR geo IN select quad_split_geometry(geometry, 0.01, 20) as geom LOOP
RETURN NEXT geo.geom;
END LOOP;
RETURN;
END;
$$
LANGUAGE plpgsql IMMUTABLE;
CREATE OR REPLACE FUNCTION place_force_delete(placeid BIGINT)
RETURNS BOOLEAN
AS $$
DECLARE
osmid BIGINT;
osmtype character(1);
pclass text;
ptype text;
BEGIN
SELECT osm_type, osm_id, class, type FROM placex WHERE place_id = placeid INTO osmtype, osmid, pclass, ptype;
DELETE FROM import_polygon_delete where osm_type = osmtype and osm_id = osmid and class = pclass and type = ptype;
DELETE FROM import_polygon_error where osm_type = osmtype and osm_id = osmid and class = pclass and type = ptype;
-- force delete from place/placex by making it a very small geometry
UPDATE place set geometry = ST_SetSRID(ST_Point(0,0), 4326) where osm_type = osmtype and osm_id = osmid and class = pclass and type = ptype;
DELETE FROM place where osm_type = osmtype and osm_id = osmid and class = pclass and type = ptype;
RETURN TRUE;
END;
$$
LANGUAGE plpgsql;
CREATE OR REPLACE FUNCTION place_force_update(placeid BIGINT)
RETURNS BOOLEAN
AS $$
DECLARE
placegeom GEOMETRY;
geom GEOMETRY;
diameter FLOAT;
rank INTEGER;
BEGIN
UPDATE placex SET indexed_status = 2 WHERE place_id = placeid;
SELECT geometry, rank_search FROM placex WHERE place_id = placeid INTO placegeom, rank;
IF placegeom IS NOT NULL AND ST_IsValid(placegeom) THEN
IF ST_GeometryType(placegeom) in ('ST_Polygon','ST_MultiPolygon') THEN
FOR geom IN select split_geometry(placegeom) FROM placex WHERE place_id = placeid LOOP
update placex set indexed_status = 2 where ST_Intersects(geom, placex.geometry)
AND rank_search > rank and indexed_status = 0 and ST_geometrytype(placex.geometry) = 'ST_Point' and (rank_search < 28 or name is not null or (rank >= 16 and address ? 'place'));
update placex set indexed_status = 2 where ST_Intersects(geom, placex.geometry)
AND rank_search > rank and indexed_status = 0 and ST_geometrytype(placex.geometry) != 'ST_Point' and (rank_search < 28 or name is not null or (rank >= 16 and address ? 'place'));
END LOOP;
ELSE
diameter := update_place_diameter(rank);
IF diameter > 0 THEN
IF rank >= 26 THEN
-- roads may cause reparenting for >27 rank places
update placex set indexed_status = 2 where indexed_status = 0 and rank_search > rank and ST_DWithin(placex.geometry, placegeom, diameter);
ELSEIF rank >= 16 THEN
-- up to rank 16, street-less addresses may need reparenting
update placex set indexed_status = 2 where indexed_status = 0 and rank_search > rank and ST_DWithin(placex.geometry, placegeom, diameter) and (rank_search < 28 or name is not null or address ? 'place');
ELSE
-- for all other places the search terms may change as well
update placex set indexed_status = 2 where indexed_status = 0 and rank_search > rank and ST_DWithin(placex.geometry, placegeom, diameter) and (rank_search < 28 or name is not null);
END IF;
END IF;
END IF;
RETURN TRUE;
END IF;
RETURN FALSE;
END;
$$
LANGUAGE plpgsql;

View File

@@ -1,27 +1,26 @@
-- Indices used only during search and update.
-- These indices are created only after the indexing process is done.
CREATE INDEX CONCURRENTLY idx_word_word_id on word USING BTREE (word_id) {ts:search-index};
CREATE INDEX idx_word_word_id on word USING BTREE (word_id) {ts:search-index};
CREATE INDEX CONCURRENTLY idx_place_addressline_address_place_id on place_addressline USING BTREE (address_place_id) {ts:search-index};
CREATE INDEX idx_place_addressline_address_place_id on place_addressline USING BTREE (address_place_id) {ts:search-index};
DROP INDEX CONCURRENTLY IF EXISTS idx_placex_rank_search;
CREATE INDEX CONCURRENTLY idx_placex_rank_search ON placex USING BTREE (rank_search) {ts:search-index};
CREATE INDEX CONCURRENTLY idx_placex_rank_address ON placex USING BTREE (rank_address) {ts:search-index};
CREATE INDEX CONCURRENTLY idx_placex_pendingsector ON placex USING BTREE (rank_search,geometry_sector) {ts:address-index} where indexed_status > 0;
CREATE INDEX CONCURRENTLY idx_placex_parent_place_id ON placex USING BTREE (parent_place_id) {ts:search-index} where parent_place_id IS NOT NULL;
DROP INDEX IF EXISTS idx_placex_rank_search;
CREATE INDEX idx_placex_rank_search ON placex USING BTREE (rank_search) {ts:search-index};
CREATE INDEX idx_placex_rank_address ON placex USING BTREE (rank_address) {ts:search-index};
CREATE INDEX idx_placex_parent_place_id ON placex USING BTREE (parent_place_id) {ts:search-index} where parent_place_id IS NOT NULL;
CREATE INDEX CONCURRENTLY idx_placex_geometry_reverse_lookupPoint
CREATE INDEX idx_placex_geometry_reverse_lookupPoint
ON placex USING gist (geometry) {ts:search-index}
WHERE (name is not null or housenumber is not null or rank_address between 26 and 27)
AND class not in ('railway','tunnel','bridge','man_made')
AND rank_address >= 26 AND indexed_status = 0 AND linked_place_id is null;
CREATE INDEX CONCURRENTLY idx_placex_geometry_reverse_lookupPolygon
CREATE INDEX idx_placex_geometry_reverse_lookupPolygon
ON placex USING gist (geometry) {ts:search-index}
WHERE St_GeometryType(geometry) in ('ST_Polygon', 'ST_MultiPolygon')
AND rank_address between 4 and 25 AND type != 'postcode'
AND name is not null AND indexed_status = 0 AND linked_place_id is null;
CREATE INDEX CONCURRENTLY idx_placex_geometry_reverse_placeNode
CREATE INDEX idx_placex_geometry_reverse_placeNode
ON placex USING gist (geometry) {ts:search-index}
WHERE osm_type = 'N' AND rank_search between 5 and 25
AND class = 'place' AND type != 'postcode'
@@ -29,14 +28,8 @@ CREATE INDEX CONCURRENTLY idx_placex_geometry_reverse_placeNode
GRANT SELECT ON table country_osm_grid to "{www-user}";
CREATE INDEX CONCURRENTLY idx_location_area_country_place_id ON location_area_country USING BTREE (place_id) {ts:address-index};
CREATE INDEX idx_osmline_parent_place_id ON location_property_osmline USING BTREE (parent_place_id) {ts:search-index};
CREATE INDEX idx_osmline_parent_osm_id ON location_property_osmline USING BTREE (osm_id) {ts:search-index};
CREATE INDEX CONCURRENTLY idx_osmline_parent_place_id ON location_property_osmline USING BTREE (parent_place_id) {ts:search-index};
CREATE INDEX CONCURRENTLY idx_osmline_parent_osm_id ON location_property_osmline USING BTREE (osm_id) {ts:search-index};
DROP INDEX CONCURRENTLY IF EXISTS place_id_idx;
CREATE UNIQUE INDEX CONCURRENTLY idx_place_osm_unique on place using btree(osm_id,osm_type,class,type) {ts:address-index};
CREATE UNIQUE INDEX CONCURRENTLY idx_postcode_id ON location_postcode USING BTREE (place_id) {ts:search-index};
CREATE INDEX CONCURRENTLY idx_postcode_postcode ON location_postcode USING BTREE (postcode) {ts:search-index};
CREATE UNIQUE INDEX idx_postcode_id ON location_postcode USING BTREE (place_id) {ts:search-index};
CREATE INDEX idx_postcode_postcode ON location_postcode USING BTREE (postcode) {ts:search-index};

Some files were not shown because too many files have changed in this diff Show More