use make_keywords for place search terms also

[nominatim.git] / nominatim / clicmd / setup.py

diff --git a/nominatim/clicmd/setup.py b/nominatim/clicmd/setup.py
index fe7c8dc18de4ee650e7bbb42824c813f4e2d9753..3f59068618db15e95fe09588a570bc3529b6ba36 100644 (file)
--- a/nominatim/clicmd/setup.py
+++ b/nominatim/clicmd/setup.py
@@ -56,6 +56,7 @@ class SetupAll:
         from ..tools import refresh
         from ..indexer.indexer import Indexer
         from ..tools import postcodes
         from ..tools import refresh
         from ..indexer.indexer import Indexer
         from ..tools import postcodes

+        from ..tokenizer import factory as tokenizer_factory

 
         if args.osm_file and not Path(args.osm_file).is_file():
             LOG.fatal("OSM file '%s' does not exist.", args.osm_file)
 
         if args.osm_file and not Path(args.osm_file).is_file():
             LOG.fatal("OSM file '%s' does not exist.", args.osm_file)


@@ -67,12 +68,6 @@ class SetupAll:
                                                     args.no_partitions,
                                                     rouser=args.config.DATABASE_WEBUSER)
 
                                                     args.no_partitions,
                                                     rouser=args.config.DATABASE_WEBUSER)
 

-            LOG.warning('Installing database module')
-            with connect(args.config.get_libpq_dsn()) as conn:
-                database_import.install_module(args.module_dir, args.project_dir,
-                                               args.config.DATABASE_MODULE_PATH,
-                                               conn=conn)
-

             LOG.warning('Importing OSM data file')
             database_import.import_osm_data(Path(args.osm_file),
                                             args.osm2pgsql_options(0, 1),
             LOG.warning('Importing OSM data file')
             database_import.import_osm_data(Path(args.osm_file),
                                             args.osm2pgsql_options(0, 1),


@@ -81,22 +76,19 @@ class SetupAll:
 
             with connect(args.config.get_libpq_dsn()) as conn:
                 LOG.warning('Create functions (1st pass)')
 
             with connect(args.config.get_libpq_dsn()) as conn:
                 LOG.warning('Create functions (1st pass)')

-                refresh.create_functions(conn, args.config, args.sqllib_dir,
-                                         False, False)
+                refresh.create_functions(conn, args.config, False, False)

                 LOG.warning('Create tables')
                 LOG.warning('Create tables')

-                database_import.create_tables(conn, args.config, args.sqllib_dir,
+                database_import.create_tables(conn, args.config,

                                               reverse_only=args.reverse_only)
                 refresh.load_address_levels_from_file(conn, Path(args.config.ADDRESS_LEVEL_CONFIG))
                 LOG.warning('Create functions (2nd pass)')
                                               reverse_only=args.reverse_only)
                 refresh.load_address_levels_from_file(conn, Path(args.config.ADDRESS_LEVEL_CONFIG))
                 LOG.warning('Create functions (2nd pass)')

-                refresh.create_functions(conn, args.config, args.sqllib_dir,
-                                         False, False)
+                refresh.create_functions(conn, args.config, False, False)

                 LOG.warning('Create table triggers')
                 LOG.warning('Create table triggers')

-                database_import.create_table_triggers(conn, args.config, args.sqllib_dir)
+                database_import.create_table_triggers(conn, args.config)

                 LOG.warning('Create partition tables')
                 LOG.warning('Create partition tables')

-                database_import.create_partition_tables(conn, args.config, args.sqllib_dir)
+                database_import.create_partition_tables(conn, args.config)

                 LOG.warning('Create functions (3rd pass)')
                 LOG.warning('Create functions (3rd pass)')

-                refresh.create_functions(conn, args.config, args.sqllib_dir,
-                                         False, False)
+                refresh.create_functions(conn, args.config, False, False)

 
             LOG.warning('Importing wikipedia importance data')
             data_path = Path(args.config.WIKIPEDIA_DATA_PATH or args.project_dir)
 
             LOG.warning('Importing wikipedia importance data')
             data_path = Path(args.config.WIKIPEDIA_DATA_PATH or args.project_dir)


@@ -105,36 +97,50 @@ class SetupAll:
                 LOG.error('Wikipedia importance dump file not found. '
                           'Will be using default importances.')
 
                 LOG.error('Wikipedia importance dump file not found. '
                           'Will be using default importances.')
 

+        if args.continue_at is None or args.continue_at == 'load-data':

             LOG.warning('Initialise tables')
             with connect(args.config.get_libpq_dsn()) as conn:
             LOG.warning('Initialise tables')
             with connect(args.config.get_libpq_dsn()) as conn:

-                database_import.truncate_data_tables(conn, args.config.MAX_WORD_FREQUENCY)
+                database_import.truncate_data_tables(conn)

 
 

-        if args.continue_at is None or args.continue_at == 'load-data':

             LOG.warning('Load data into placex table')
             database_import.load_data(args.config.get_libpq_dsn(),
             LOG.warning('Load data into placex table')
             database_import.load_data(args.config.get_libpq_dsn(),

-                                      args.data_dir,

                                       args.threads or psutil.cpu_count() or 1)
 
                                       args.threads or psutil.cpu_count() or 1)
 

+        LOG.warning("Setting up tokenizer")
+        if args.continue_at is None or args.continue_at == 'load-data':
+            # (re)initialise the tokenizer data
+            tokenizer = tokenizer_factory.create_tokenizer(args.config)
+        else:
+            # just load the tokenizer
+            tokenizer = tokenizer_factory.get_tokenizer_for_db(args.config)
+
+        if args.continue_at is None or args.continue_at == 'load-data':

             LOG.warning('Calculate postcodes')
             LOG.warning('Calculate postcodes')

-            postcodes.import_postcodes(args.config.get_libpq_dsn(), args.project_dir)
+            postcodes.update_postcodes(args.config.get_libpq_dsn(),
+                                       args.project_dir, tokenizer)

 
         if args.continue_at is None or args.continue_at in ('load-data', 'indexing'):
 
         if args.continue_at is None or args.continue_at in ('load-data', 'indexing'):

+            if args.continue_at is not None and args.continue_at != 'load-data':
+                with connect(args.config.get_libpq_dsn()) as conn:
+                    SetupAll._create_pending_index(conn, args.config.TABLESPACE_ADDRESS_INDEX)

             LOG.warning('Indexing places')
             LOG.warning('Indexing places')

-            indexer = Indexer(args.config.get_libpq_dsn(),
+            indexer = Indexer(args.config.get_libpq_dsn(), tokenizer,

                               args.threads or psutil.cpu_count() or 1)
             indexer.index_full(analyse=not args.index_noanalyse)
 
         LOG.warning('Post-process tables')
         with connect(args.config.get_libpq_dsn()) as conn:
             database_import.create_search_indices(conn, args.config,
                               args.threads or psutil.cpu_count() or 1)
             indexer.index_full(analyse=not args.index_noanalyse)
 
         LOG.warning('Post-process tables')
         with connect(args.config.get_libpq_dsn()) as conn:
             database_import.create_search_indices(conn, args.config,

-                                                  args.sqllib_dir,

                                                   drop=args.no_updates)
             LOG.warning('Create search index for default country names.')
                                                   drop=args.no_updates)
             LOG.warning('Create search index for default country names.')

-            database_import.create_country_names(conn, args.config)
+            database_import.create_country_names(conn, tokenizer,
+                                                 args.config.LANGUAGES)
+        tokenizer.finalize_import(args.config)

 
         webdir = args.project_dir / 'website'
         LOG.warning('Setup website at %s', webdir)
 
         webdir = args.project_dir / 'website'
         LOG.warning('Setup website at %s', webdir)

-        refresh.setup_website(webdir, args.phplib_dir, args.config)
+        with connect(args.config.get_libpq_dsn()) as conn:
+            refresh.setup_website(webdir, args.config, conn)

 
         with connect(args.config.get_libpq_dsn()) as conn:
             try:
 
         with connect(args.config.get_libpq_dsn()) as conn:
             try:


@@ -148,3 +154,25 @@ class SetupAll:
                                     '{0[0]}.{0[1]}.{0[2]}-{0[3]}'.format(NOMINATIM_VERSION))
 
         return 0
                                     '{0[0]}.{0[1]}.{0[2]}-{0[3]}'.format(NOMINATIM_VERSION))
 
         return 0

+
+
+    @staticmethod
+    def _create_pending_index(conn, tablespace):
+        """ Add a supporting index for finding places still to be indexed.
+
+            This index is normally created at the end of the import process
+            for later updates. When indexing was partially done, then this
+            index can greatly improve speed going through already indexed data.
+        """
+        if conn.index_exists('idx_placex_pendingsector'):
+            return
+
+        with conn.cursor() as cur:
+            LOG.warning('Creating support index')
+            if tablespace:
+                tablespace = 'TABLESPACE ' + tablespace
+            cur.execute("""CREATE INDEX idx_placex_pendingsector
+                           ON placex USING BTREE (rank_address,geometry_sector)
+                           {} WHERE indexed_status > 0
+                        """.format(tablespace))
+        conn.commit()