add support index when continuing import at index phase

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

diff --git a/nominatim/clicmd/setup.py b/nominatim/clicmd/setup.py
index 71980739756c2e0f3a885689e3ad547e6ae1e6e7..fb7abdeca0a5021e0ab0eb3576ad1841af88728f 100644 (file)
--- a/nominatim/clicmd/setup.py
+++ b/nominatim/clicmd/setup.py
@@ -6,11 +6,10 @@ from pathlib import Path
 
 import psutil
 
-from ..tools.exec_utils import run_legacy_script
-from ..db.connection import connect
-from ..db import status, properties
-from ..version import NOMINATIM_VERSION
-from ..errors import UsageError
+from nominatim.db.connection import connect
+from nominatim.db import status, properties
+from nominatim.version import NOMINATIM_VERSION
+from nominatim.errors import UsageError
 
 # Do not repeat documentation of subcommand classes.
 # pylint: disable=C0111
@@ -39,11 +38,11 @@ class SetupAll:
         group.add_argument('--reverse-only', action='store_true',
                            help='Do not create tables and indexes for searching')
         group.add_argument('--no-partitions', action='store_true',
-                           help="""Do not partition search indices
-                                   (speeds up import of single country extracts)""")
+                           help=("Do not partition search indices "
+                                 "(speeds up import of single country extracts)"))
         group.add_argument('--no-updates', action='store_true',
-                           help="""Do not keep tables that are only needed for
-                                   updating the database later""")
+                           help="Do not keep tables that are only needed for "
+                                "updating the database later")
         group = parser.add_argument_group('Expert options')
         group.add_argument('--ignore-errors', action='store_true',
                            help='Continue import even when errors in SQL are present')
@@ -56,6 +55,7 @@ class SetupAll:
         from ..tools import database_import
         from ..tools import refresh
         from ..indexer.indexer import Indexer
+        from ..tools import postcodes
 
         if args.osm_file and not Path(args.osm_file).is_file():
             LOG.fatal("OSM file '%s' does not exist.", args.osm_file)
@@ -105,21 +105,23 @@ class SetupAll:
                 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:
                 database_import.truncate_data_tables(conn, args.config.MAX_WORD_FREQUENCY)
 
-        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(),
                                       args.data_dir,
                                       args.threads or psutil.cpu_count() or 1)
 
             LOG.warning('Calculate postcodes')
-            run_legacy_script('setup.php', '--calculate-postcodes',
-                              nominatim_env=args, throw_on_fail=not args.ignore_errors)
+            postcodes.import_postcodes(args.config.get_libpq_dsn(), args.project_dir)
 
         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(),
                               args.threads or psutil.cpu_count() or 1)
@@ -130,8 +132,8 @@ class SetupAll:
             database_import.create_search_indices(conn, args.config,
                                                   args.sqllib_dir,
                                                   drop=args.no_updates)
-        run_legacy_script('setup.php', '--create-country-names',
-                          nominatim_env=args, throw_on_fail=not args.ignore_errors)
+            LOG.warning('Create search index for default country names.')
+            database_import.create_country_names(conn, args.config)
 
         webdir = args.project_dir / 'website'
         LOG.warning('Setup website at %s', webdir)
@@ -149,3 +151,25 @@ class SetupAll:
                                     '{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()