prepare 4.0.1 release

Don't penalize French 'bis' housenumbers

CMakeLists.txt

@@ -20,7 +20,7 @@ project(nominatim)
 
 set(NOMINATIM_VERSION_MAJOR 4)
 set(NOMINATIM_VERSION_MINOR 0)
-set(NOMINATIM_VERSION_PATCH 0)
+set(NOMINATIM_VERSION_PATCH 1)
 
 set(NOMINATIM_VERSION "${NOMINATIM_VERSION_MAJOR}.${NOMINATIM_VERSION_MINOR}.${NOMINATIM_VERSION_PATCH}")
 

ChangeLog

@@ -1,3 +1,12 @@
+4.0.1
+
+ * fix initialisation error in replication script
+ * ICU tokenizer: avoid any special characters in word tokens
+ * better error message when API php script does not exist
+ * fix quoting of house numbers in SQL queries
+ * small fixes and improvements in search query parsing
+ * add documentation for moving the database to a different machine
+
 4.0.0
 
  * refactor name token computation and introduce ICU tokenizer

docs/admin/Advanced-Installations.md

@@ -101,7 +101,7 @@ 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.
 
-## Importing Nominatim to an external PostgreSQL database
+## Using an external PostgreSQL database
 
 You can install Nominatim using a database that runs on a different server when
 you have physical access to the file system on the other server. Nominatim
@@ -109,6 +109,11 @@ uses a custom normalization library that needs to be made accessible to the
 PostgreSQL server. This section explains how to set up the normalization
 library.
 
+!!! note
+    The external module is only needed when using the legacy tokenizer.
+    If you have choosen the ICU tokenizer, then you can ignore this section
+    and follow the standard import documentation.
+
 ### Option 1: Compiling the library on the database server
 
 The most sure way to get a working library is to compile it on the database
@@ -167,3 +172,44 @@ NOMINATIM_DATABASE_MODULE_PATH="<directory on the database server where nominati
 
 Now change the `NOMINATIM_DATABASE_DSN` to point to your remote server and continue
 to follow the [standard instructions for importing](Import.md).
+
+
+## Moving the database to another machine
+
+For some configurations it may be useful to run the import on one machine, then
+move the database to another machine and run the Nominatim service from there.
+For example, you might want to use a large machine to be able to run the import
+quickly but only want a smaller machine for production because there is not so
+much load. Or you might want to do the import once and then replicate the
+database to many machines.
+
+The important thing to keep in mind when transferring the Nominatim installation
+is that you need to transfer the database _and the project directory_. Both
+parts are essential for your installation.
+
+The Nominatim database can be transferred using the `pg_dump`/`pg_restore` tool.
+Make sure to use the same version of PostgreSQL and PostGIS on source and
+target machine.
+
+!!! note
+    Before creating a dump of your Nominatim database, consider running
+    `nominatim freeze` first. Your database looses the ability to receive further
+    data updates but the resulting database is only about a third of the size
+    of a full database.
+
+Next install Nominatim on the target machine by following the standard installation
+instructions. Again make sure to use the same version as the source machine.
+
+You can now copy the project directory from the source machine to the new machine.
+If necessary, edit the `.env` file to point it to the restored database.
+Finally run
+
+    nominatim refresh --website
+
+to make sure that the local installation of Nominatim will be used.
+
+If you are using the legacy tokenizer you might also have to switch to the
+PostgreSQL module that was compiled on your target machine. If you get errors
+that PostgreSQL cannot find or access `nominatim.so` then copy the installed
+version into the `module` directory of your project directory. The installed
+copy can usually be found under `/usr/local/lib/nominatim/module/nominatim.so`.

lib-php/SearchDescription.php

@@ -257,7 +257,7 @@ class SearchDescription
         if (empty($this->aName)) {
             $this->bNameNeedsAddress = $bNeedsAddress;
         } else {
-            $this->bNameNeedsAddress |= $bNeedsAddress;
+            $this->bNameNeedsAddress &= $bNeedsAddress;
         }
         if ($bSearchable) {
             $this->aName[$iId] = $iId;
@@ -584,11 +584,11 @@ class SearchDescription
         // will be narrowed down by an address. Remember that with ordering
         // every single result has to be checked.
         if ($this->sHouseNumber && ($this->bRareName || !empty($this->aAddress) || $this->sPostcode)) {
-            $sHouseNumberRegex = '\\\\m'.$this->sHouseNumber.'\\\\M';
+            $sHouseNumberRegex = $oDB->getDBQuoted('\\\\m'.$this->sHouseNumber.'\\\\M');
 
             // Housenumbers on streets and places.
             $sChildHnr = 'SELECT * FROM placex WHERE parent_place_id = search_name.place_id';
-            $sChildHnr .= "    AND housenumber ~* E'".$sHouseNumberRegex."'";
+            $sChildHnr .= '    AND housenumber ~* E'.$sHouseNumberRegex;
             // Interpolations on streets and places.
             if (preg_match('/^[0-9]+$/', $this->sHouseNumber)) {
                 $sIpolHnr = 'SELECT * FROM location_property_osmline ';
@@ -601,7 +601,7 @@ class SearchDescription
             }
             // Housenumbers on the object iteself for unlisted places.
             $sSelfHnr = 'SELECT * FROM placex WHERE place_id = search_name.place_id';
-            $sSelfHnr .= "    AND housenumber ~* E'".$sHouseNumberRegex."'";
+            $sSelfHnr .= '    AND housenumber ~* E'.$sHouseNumberRegex;
 
             $sSql = '(CASE WHEN address_rank = 30 THEN EXISTS('.$sSelfHnr.') ';
             $sSql .= ' ELSE EXISTS('.$sChildHnr.') ';
@@ -739,9 +739,9 @@ class SearchDescription
             return $aResults;
         }
 
-        $sHouseNumberRegex = '\\\\m'.$this->sHouseNumber.'\\\\M';
+        $sHouseNumberRegex = $oDB->getDBQuoted('\\\\m'.$this->sHouseNumber.'\\\\M');
         $sSQL = 'SELECT place_id FROM placex WHERE';
-        $sSQL .= "  housenumber ~* E'".$sHouseNumberRegex."'";
+        $sSQL .= '  housenumber ~* E'.$sHouseNumberRegex;
         $sSQL .= ' AND ('.join(' OR ', $aIDCondition).')';
         $sSQL .= $this->oContext->excludeSQL(' AND place_id');
 

lib-php/TokenHousenumber.php

@@ -58,7 +58,7 @@ class HouseNumber
         // up of numbers, add a penalty
         $iSearchCost = 1;
         if (preg_match('/\\d/', $this->sToken) === 0
-            || preg_match_all('/[^0-9]/', $this->sToken, $aMatches) > 2) {
+            || preg_match_all('/[^0-9 ]/', $this->sToken, $aMatches) > 3) {
             $iSearchCost += strlen($this->sToken) - 1;
         }
         if (!$oSearch->hasOperator(\Nominatim\Operator::NONE)) {

nominatim/clicmd/api.py

@@ -4,6 +4,7 @@ Subcommand definitions for API calls from the command line.
 import logging
 
 from nominatim.tools.exec_utils import run_api_script
+from nominatim.errors import UsageError
 
 # Do not repeat documentation of subcommand classes.
 # pylint: disable=C0111
@@ -53,6 +54,18 @@ def _add_api_output_arguments(parser):
                              "Parameter is difference tolerance in degrees."))
 
 
+def _run_api(endpoint, args, params):
+    script_file = args.project_dir / 'website' / (endpoint + '.php')
+
+    if not script_file.exists():
+        LOG.error("Cannot find API script file.\n\n"
+                  "Make sure to run 'nominatim' from the project directory \n"
+                  "or use the option --project-dir.")
+        raise UsageError("API script not found.")
+
+    return run_api_script(endpoint, args.project_dir,
+                          phpcgi_bin=args.phpcgi_path, params=params)
+
 class APISearch:
     """\
     Execute a search query.
@@ -114,8 +127,7 @@ class APISearch:
         if not args.dedupe:
             params['dedupe'] = '0'
 
-        return run_api_script('search', args.project_dir,
-                              phpcgi_bin=args.phpcgi_path, params=params)
+        return _run_api('search', args, params)
 
 class APIReverse:
     """\
@@ -158,8 +170,7 @@ class APIReverse:
         if args.polygon_threshold:
             params['polygon_threshold'] = args.polygon_threshold
 
-        return run_api_script('reverse', args.project_dir,
-                              phpcgi_bin=args.phpcgi_path, params=params)
+        return _run_api('reverse', args, params)
 
 
 class APILookup:
@@ -198,8 +209,7 @@ class APILookup:
         if args.polygon_threshold:
             params['polygon_threshold'] = args.polygon_threshold
 
-        return run_api_script('lookup', args.project_dir,
-                              phpcgi_bin=args.phpcgi_path, params=params)
+        return _run_api('lookup', args, params)
 
 
 class APIDetails:
@@ -249,8 +259,7 @@ class APIDetails:
         for name, _ in DETAILS_SWITCHES:
             params[name] = '1' if getattr(args, name) else '0'
 
-        return run_api_script('details', args.project_dir,
-                              phpcgi_bin=args.phpcgi_path, params=params)
+        return _run_api('details', args, params)
 
 
 class APIStatus:
@@ -271,6 +280,4 @@ class APIStatus:
 
     @staticmethod
     def run(args):
-        return run_api_script('status', args.project_dir,
-                              phpcgi_bin=args.phpcgi_path,
-                              params=dict(format=args.format))
+        return _run_api('status', args, dict(format=args.format))

nominatim/clicmd/replication.py

@@ -136,6 +136,7 @@ class UpdateReplication:
             recheck_interval = args.config.get_int('REPLICATION_RECHECK_INTERVAL')
 
         tokenizer = tokenizer_factory.get_tokenizer_for_db(args.config)
+        indexer = Indexer(args.config.get_libpq_dsn(), tokenizer, args.threads or 1)
 
         while True:
             with connect(args.config.get_libpq_dsn()) as conn:
@@ -148,8 +149,6 @@ class UpdateReplication:
 
             if state is not replication.UpdateState.NO_CHANGES and args.do_index:
                 index_start = dt.datetime.now(dt.timezone.utc)
-                indexer = Indexer(args.config.get_libpq_dsn(), tokenizer,
-                                  args.threads or 1)
                 indexer.index_full(analyse=False)
 
                 with connect(args.config.get_libpq_dsn()) as conn:

nominatim/version.py

@@ -10,7 +10,7 @@ Version information for Nominatim.
 # and must always be increased when there is a change to the database or code
 # that requires a migration.
 # Released versions always have a database patch level of 0.
-NOMINATIM_VERSION = (4, 0, 0, 0)
+NOMINATIM_VERSION = (4, 0, 1, 0)
 
 POSTGRESQL_REQUIRED_VERSION = (9, 5)
 POSTGIS_REQUIRED_VERSION = (2, 2)

settings/icu_tokenizer.yaml

@@ -21,8 +21,8 @@ transliteration:
     - !include icu-rules/extended-unicode-to-asccii.yaml
     - ":: Ascii ()"
     - ":: NFD ()"
-    - "[^[:Ascii:]] >"
     - ":: lower ()"
+    - "[^a-z0-9[:Space:]] >"
     - ":: NFC ()"
 sanitizers:
     - step: split-name-list

test/python/test_cli.py

@@ -113,23 +113,36 @@ class TestCli:
         assert func.called == 1
 
 
-    @pytest.mark.parametrize("params", [('search', '--query', 'new'),
-                                        ('reverse', '--lat', '0', '--lon', '0'),
-                                        ('lookup', '--id', 'N1'),
-                                        ('details', '--node', '1'),
-                                        ('details', '--way', '1'),
-                                        ('details', '--relation', '1'),
-                                        ('details', '--place_id', '10001'),
-                                        ('status',)])
-    def test_api_commands_simple(self, mock_func_factory, params):
+@pytest.mark.parametrize("params", [('search', '--query', 'new'),
+                                    ('reverse', '--lat', '0', '--lon', '0'),
+                                    ('lookup', '--id', 'N1'),
+                                    ('details', '--node', '1'),
+                                    ('details', '--way', '1'),
+                                    ('details', '--relation', '1'),
+                                    ('details', '--place_id', '10001'),
+                                    ('status',)])
+class TestCliApiCall:
+
+    @pytest.fixture(autouse=True)
+    def setup_cli_call(self, cli_call):
+        self.call_nominatim = cli_call
+
+    def test_api_commands_simple(self, mock_func_factory, params, tmp_path):
+        (tmp_path / 'website').mkdir()
+        (tmp_path / 'website' / (params[0] + '.php')).write_text('')
         mock_run_api = mock_func_factory(nominatim.clicmd.api, 'run_api_script')
 
-        assert self.call_nominatim(*params) == 0
+        assert self.call_nominatim(*params, '--project-dir', str(tmp_path)) == 0
 
         assert mock_run_api.called == 1
         assert mock_run_api.last_args[0] == params[0]
 
 
+    def test_bad_project_idr(self, mock_func_factory, params):
+        mock_run_api = mock_func_factory(nominatim.clicmd.api, 'run_api_script')
+
+        assert self.call_nominatim(*params) == 1
+
 
 class TestCliWithDb:
 

test/python/test_cli_replication.py

@@ -106,6 +106,16 @@ class TestCliReplication:
         assert str(update_mock.last_args[1]['osm2pgsql']) == '/secret/osm2pgsql'
 
 
+    @pytest.mark.parametrize("update_interval", [60, 3600])
+    def test_replication_catchup(self, monkeypatch, index_mock, update_interval, placex_table):
+        monkeypatch.setenv('NOMINATIM_REPLICATION_UPDATE_INTERVAL', str(update_interval))
+        states = [nominatim.tools.replication.UpdateState.NO_CHANGES]
+        monkeypatch.setattr(nominatim.tools.replication, 'update',
+                            lambda *args, **kwargs: states.pop())
+
+        assert self.call_nominatim('--catch-up') == 0
+
+
     def test_replication_update_custom_threads(self, update_mock):
         assert self.call_nominatim('--once', '--no-index', '--threads', '4') == 0