Merge remote-tracking branch 'upstream/master'

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

diff --git a/nominatim/clicmd/setup.py b/nominatim/clicmd/setup.py
index 066c296037153cc73f8fe126db4b6d79ec6750f5..878c88260602a691052fff87156115b1196410bd 100644 (file)
--- a/nominatim/clicmd/setup.py
+++ b/nominatim/clicmd/setup.py
@@ -52,10 +52,8 @@ class SetupAll:
 
     @staticmethod
     def run(args): # pylint: disable=too-many-statements
-        from ..tools import database_import
-        from ..tools import refresh
+        from ..tools import database_import, refresh, postcodes, freeze
         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():
@@ -100,26 +98,31 @@ class SetupAll:
         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:
-                database_import.truncate_data_tables(conn, args.config.MAX_WORD_FREQUENCY)
+                database_import.truncate_data_tables(conn)
 
             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)
 
         LOG.warning("Setting up tokenizer")
-        tokenizer = tokenizer_factory.create_tokenizer(args.config)
+        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')
-            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 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')
-            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)
 
@@ -128,11 +131,18 @@ class SetupAll:
             database_import.create_search_indices(conn, args.config,
                                                   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)
+            conn.commit()
+            if args.no_updates:
+                freeze.drop_update_tables(conn)
+        tokenizer.finalize_import(args.config)
+
 
         webdir = args.project_dir / 'website'
         LOG.warning('Setup website at %s', webdir)
-        refresh.setup_website(webdir, 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: