Sarah Hoffmann
7bbfb4b972
drop search_rank column from search_name
...
This is no longer needed for the initial lookup.
2026-03-09 10:05:31 +01:00
Sarah Hoffmann
a16c698fb2
enforce presence of importance value in placex
2026-03-09 10:05:31 +01:00
Sarah Hoffmann
abd5cbada6
revert importing of housenumber interpolations
2026-03-04 16:29:24 +01:00
Sarah Hoffmann
c2d6821f2f
adapt interpolation handling to use separate place_interpolation table
2026-03-04 16:29:24 +01:00
James Addison
53e8334206
Indexing: fixup: add presence check for hstore name field
2026-02-16 00:01:57 +00:00
Sarah Hoffmann
fed64cda5a
Merge pull request #3957 from jayaddison/issue-2714/linked-places-default-language
...
Indexing: add default-language placename from linked places
2026-02-11 15:08:18 +01:00
Sarah Hoffmann
7a62c7d812
sanity check class names before inserting into classtype tables
...
The subsequent INSERT is done on an unqouted table name, making in
theory an SQL injection through an OSM value possible. In practise
this cannot happen because we check for the existance of the table.
During the creation of the classtype tables there is a sanity
check in place to disallow any table names that consist of anything
other than alphanumeric characters.
2026-02-10 11:38:26 +01:00
James Addison
e62811cf97
Indexing: invert boolean logic to factor-out empty ELSE clause
...
Relates-to commit fa2a789e27 .
2026-02-09 18:33:02 +00:00
James Addison
fa2a789e27
Indexing: manage the case where no default-language exists
...
Relates-to commit 6fee784c9f .
2026-02-05 20:48:01 +00:00
James Addison
6fee784c9f
Indexing: add default-language placename from linked places
2026-02-05 15:19:48 +00:00
James Addison
f07676a376
Fixup: add single-quote escaping within debug message
2026-01-28 01:27:53 +00:00
Sarah Hoffmann
e47601754a
do not attempt to delete old data for newly created placex entries
2026-01-07 17:08:28 +01:00
Sarah Hoffmann
ba1fc5a5b8
do not insert entries with empty name into search name
2026-01-07 11:27:55 +01:00
Sarah Hoffmann
d35a71c123
ensure correct indexed_status transitions
2026-01-07 11:12:35 +01:00
Sarah Hoffmann
e31862b7b5
make sure that importance is always set to a non-null value
...
Secondary importance might return invalid values in some cases.
2026-01-07 10:29:45 +01:00
Sarah Hoffmann
89821d01e0
reorganise layout of location_postcode table
...
Also renames the table as this will make it easier to migrate.
2025-12-30 15:20:46 +01:00
Sarah Hoffmann
0aa9eee3e7
remove special casing for postcodes in trigger code
2025-12-30 15:20:46 +01:00
Sarah Hoffmann
9cfef7a31a
prefer wikidata over name match when linking
2025-11-26 17:44:47 +01:00
Sarah Hoffmann
139678f367
fix linkage removal when nothing has changed
2025-11-26 17:03:19 +01:00
Sarah Hoffmann
a93113bc44
use extra place_entrance table
2025-10-23 17:25:20 +02:00
Sarah Hoffmann
1d1d80e1e3
adapt BDD tests for new address ranks
2025-09-11 11:56:39 +02:00
Emily Love Watson
d0ad65f696
Select all entrances for results in one query
2025-08-29 10:26:29 -05:00
Emily Love Watson
e916d27b7c
Update entrances when entrance nodes are updated
2025-08-29 10:26:29 -05:00
Emily Love Watson
823ad5d279
Update entrances schema
2025-08-29 10:26:29 -05:00
Emily Love Watson
048d571e46
Index and return entrance coordinates for indexed locations
2025-08-29 10:25:44 -05:00
Sarah Hoffmann
6270c90052
replace ST_Relate by shortcut functions
...
For some reason ST_Relate returns wrong results in the context of
the trigger on Debian Trixie. Works fine with the Postgis version
from postgresql.org.
2025-08-06 14:43:07 +02:00
Sarah Hoffmann
1111597db5
reduce precision of computed centroids to 7 digits
2025-07-29 21:25:14 +02:00
Sarah Hoffmann
4cbbe04f7f
align limits for deferring delete and reindexing on insert
...
Right now when a boundary with an area between 1 and 2 broke, it
was deleted but on reinsert afer repair, the addresses are not updated
resulting in inconsistent data.
2025-07-21 16:11:06 +02:00
Sarah Hoffmann
ec42fda1bd
do not add postcodes from postcode boundaries to address vector
...
Postcodes will be found through a special search, so we can save
the space.
2025-07-08 14:49:16 +02:00
Thomas Butz
d86d491f2e
Mark functions as PARALLEL SAFE
2025-03-13 10:53:11 +01:00
Sarah Hoffmann
ad50016c49
use line interpolation to create centroid for lines
...
ST_PointOnSurface always returns one of the vertices of a line.
This means that a two-point line will have the centroid at
one of the ends, which is less then ideal.
2024-11-03 22:06:36 +01:00
Sarah Hoffmann
350bb0e1cd
restrict invalidation of child objects on large street features
...
When streets become very large, it is more likely a mapping error.
So ignore such changes.
2024-06-19 14:23:29 +02:00
Sarah Hoffmann
77631f90fd
reindex postcodes that loose their parents
...
When the parent place of a postcode is deleted, it needs to
be reindexed to get a new parent. Otherwise displaying of
results is broken.
2024-05-04 12:33:26 +02:00
Sarah Hoffmann
d2bf986eae
unlisted place names need to go into the address
...
Housenumbers with unlisted place names should be handled like
any ither housenumbers with missing address terms.
2024-05-02 11:47:35 +02:00
Sarah Hoffmann
62b7670e0c
for postcodes use rank_search as base rank for finding addresses
...
The rank_address reflects the position in the address which is
usually lower than what one would expect for a postcode area.
2024-02-28 14:40:36 +01:00
Sarah Hoffmann
019a68a4bb
Merge pull request #3345 from lonvia/simplify-large-geometries
...
Simplify very large polygons that are not used in addresses
2024-02-28 12:06:49 +01:00
Sarah Hoffmann
36b1660121
add support for new middle table format of osm2pgsql
...
Functions are adapted according to the format detected from the
osm2pgsql property table.
2024-02-27 18:18:19 +01:00
Sarah Hoffmann
56201feb28
simplify very large polygons non used in addresses
...
Polygons with rank_address = 0 are only used in search and (rarely)
for reverse lookup. Geometries do not need to be precise for that
because topology does not matter. OSM has some very large polygons
of natural features with sizes of more than 10MB. Simplify these
polygons to keep the database and indexes smaller.
2024-02-27 10:16:18 +01:00
Sarah Hoffmann
4c19762e33
extratags should become null when empty
...
Removing the artifical entries in the extratags may lead to
an empty hstore. Set it to null in that case.
Fixes #3055 .
2024-02-08 10:21:48 +01:00
Sarah Hoffmann
9627352ee4
search postcodes for highway areas around the area
...
So far the code would only accept postcodes that are inside the area.
Fixes #3304 .
2024-01-26 18:14:11 +01:00
Sarah Hoffmann
a873f260cf
fix merging of linked names into unnamed boundaries
...
The NULL value of the boundaries' name field was erasing all
content when used in SQL operations.
2023-06-30 22:14:11 +02:00
Sarah Hoffmann
645ea5a057
use information from tokenizer to determine street vs. place address
...
So far the SQL logic used the information from the address field
to determine if an address is attached to a street or place.
This changes the logic to use the information provided in the
token_info. This allows sanitizers to enforce a certain parenting
without changing the visible address information.
2023-06-30 11:08:25 +02:00
Sarah Hoffmann
d574ceb598
restrict place rank inheritance to address items
...
Place tags must have no influence on street- or POI-level
objects.
2023-02-17 16:25:26 +01:00
Sarah Hoffmann
922352e215
do not assign postcodes to long linear features
...
This avoids a postcode in particular for waterway features and
long natural featues like ridges and valleys.
Fixes #2915 .
2022-12-10 14:53:08 +01:00
Sarah Hoffmann
4f05a03d13
handle associatedStreet relations with multiple streets
...
When a associatedStreet relation has multiple street members
always take the closest one. Avoid geometry operations for
the frequent case that there is only one street.
2022-11-16 17:25:51 +01:00
Sarah Hoffmann
abf349fb0d
simplify use of secondary importance
...
The values in the raster are already normalized between 0 and 2**16,
so a simple conversion to [0, 1] will do.
Check for existance of secondary_importance table statically when
creating the SQL function. For that to work importance tables need
to be created before the functions.
2022-10-01 11:01:49 +02:00
Tareq Al-Ahdal
0ab0f0ea44
Integrated OSM views into importance computation
2022-10-01 11:01:49 +02:00
Sarah Hoffmann
f017e1e9a1
make sure indexes are used
2022-09-25 14:09:45 +02:00
Sarah Hoffmann
33ba6896a8
further split up the big geometry index
...
Adds partial indexes for all geometry queries used during import.
A full index is not necessary anymore at that point. Still create
the index afterwards for use in queries.
Also adds documentation for all indexes on where they are used.
2022-09-21 16:21:41 +02:00
Sarah Hoffmann
aef014a47d
add indexes for lookup of addressable areas
...
The generic geometry index has become to slow for that purpose.
2022-09-18 16:57:12 +02:00