Merge pull request #2312 from lonvia/icu-tokenizer

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

diff --git a/nominatim/clicmd/refresh.py b/nominatim/clicmd/refresh.py
index 4e2ad1dbe56aa108450cf3c4b1afa2523fce002a..e6e749121d24100d1393d6e59cc5d7c71bfa8218 100644 (file)
--- a/nominatim/clicmd/refresh.py
+++ b/nominatim/clicmd/refresh.py
@@ -4,8 +4,7 @@ Implementation of 'refresh' subcommand.
 import logging
 from pathlib import Path
 
-from ..db.connection import connect
-from ..tools.exec_utils import run_legacy_script
+from nominatim.db.connection import connect
 
 # Do not repeat documentation of subcommand classes.
 # pylint: disable=C0111
@@ -47,42 +46,48 @@ class UpdateRefresh:
     @staticmethod
     def run(args):
         from ..tools import refresh
+        from ..tokenizer import factory as tokenizer_factory
 
         if args.postcodes:
             LOG.warning("Update postcodes centroid")
-            conn = connect(args.config.get_libpq_dsn())
-            refresh.update_postcodes(conn, args.data_dir)
-            conn.close()
+            refresh.update_postcodes(args.config.get_libpq_dsn(), args.sqllib_dir)
 
         if args.word_counts:
             LOG.warning('Recompute frequency of full-word search terms')
-            conn = connect(args.config.get_libpq_dsn())
-            refresh.recompute_word_counts(conn, args.data_dir)
-            conn.close()
+            refresh.recompute_word_counts(args.config.get_libpq_dsn(), args.sqllib_dir)
 
         if args.address_levels:
             cfg = Path(args.config.ADDRESS_LEVEL_CONFIG)
             LOG.warning('Updating address levels from %s', cfg)
-            conn = connect(args.config.get_libpq_dsn())
-            refresh.load_address_levels_from_file(conn, cfg)
-            conn.close()
+            with connect(args.config.get_libpq_dsn()) as conn:
+                refresh.load_address_levels_from_file(conn, cfg)
 
         if args.functions:
             LOG.warning('Create functions')
-            conn = connect(args.config.get_libpq_dsn())
-            refresh.create_functions(conn, args.config, args.data_dir,
-                                     args.diffs, args.enable_debug_statements)
-            conn.close()
+            with connect(args.config.get_libpq_dsn()) as conn:
+                refresh.create_functions(conn, args.config,
+                                         args.diffs, args.enable_debug_statements)
+                tokenizer = tokenizer_factory.get_tokenizer_for_db(args.config)
+                tokenizer.update_sql_functions(args.config)
 
         if args.wiki_data:
-            run_legacy_script('setup.php', '--import-wikipedia-articles',
-                              nominatim_env=args, throw_on_fail=True)
+            data_path = Path(args.config.WIKIPEDIA_DATA_PATH
+                             or args.project_dir)
+            LOG.warning('Import wikipdia article importance from %s', data_path)
+            if refresh.import_wikipedia_articles(args.config.get_libpq_dsn(),
+                                                 data_path) > 0:
+                LOG.fatal('FATAL: Wikipedia importance dump file not found')
+                return 1
+
         # Attention: importance MUST come after wiki data import.
         if args.importance:
-            run_legacy_script('update.php', '--recompute-importance',
-                              nominatim_env=args, throw_on_fail=True)
+            LOG.warning('Update importance values for database')
+            with connect(args.config.get_libpq_dsn()) as conn:
+                refresh.recompute_importance(conn)
+
         if args.website:
-            run_legacy_script('setup.php', '--setup-website',
-                              nominatim_env=args, throw_on_fail=True)
+            webdir = args.project_dir / 'website'
+            LOG.warning('Setting up website directory at %s', webdir)
+            refresh.setup_website(webdir, args.config)
 
         return 0