Merge pull request #3268 from mtmail/wikipedia-file-path-warning

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

diff --git a/nominatim/clicmd/refresh.py b/nominatim/clicmd/refresh.py
index e696e7b6291d5df03d3480609cff8d8a7d8a5532..5e1b044e734336bf305f1c69c649bf18fda0fddc 100644 (file)
--- a/nominatim/clicmd/refresh.py
+++ b/nominatim/clicmd/refresh.py
@@ -1,10 +1,21 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# This file is part of Nominatim. (https://nominatim.org)
+#
+# Copyright (C) 2022 by the Nominatim developer community.
+# For a full list of authors see the git log.

 """
 Implementation of 'refresh' subcommand.
 """
 """
 Implementation of 'refresh' subcommand.
 """

+from typing import Tuple, Optional
+import argparse

 import logging
 from pathlib import Path
 
 import logging
 from pathlib import Path
 

+from nominatim.config import Configuration

 from nominatim.db.connection import connect
 from nominatim.db.connection import connect

+from nominatim.tokenizer.base import AbstractTokenizer
+from nominatim.clicmd.args import NominatimArgs

 
 # Do not repeat documentation of subcommand classes.
 # pylint: disable=C0111
 
 # Do not repeat documentation of subcommand classes.
 # pylint: disable=C0111


@@ -13,18 +24,37 @@ from nominatim.db.connection import connect
 
 LOG = logging.getLogger()
 
 
 LOG = logging.getLogger()
 

+def _parse_osm_object(obj: str) -> Tuple[str, int]:
+    """ Parse the given argument into a tuple of OSM type and ID.
+        Raises an ArgumentError if the format is not recognized.
+    """
+    if len(obj) < 2 or obj[0].lower() not in 'nrw' or not obj[1:].isdigit():
+        raise argparse.ArgumentTypeError("Cannot parse OSM ID. Expect format: [N|W|R]<id>.")
+
+    return (obj[0].upper(), int(obj[1:]))
+
+

 class UpdateRefresh:
     """\
     Recompute auxiliary data used by the indexing process.
 
 class UpdateRefresh:
     """\
     Recompute auxiliary data used by the indexing process.
 

-    These functions must not be run in parallel with other update commands.
+    This sub-commands updates various static data and functions in the database.
+    It usually needs to be run after changing various aspects of the
+    configuration. The configuration documentation will mention the exact
+    command to use in such case.
+
+    Warning: the 'update' command must not be run in parallel with other update
+             commands like 'replication' or 'add-data'.

     """
     """

+    def __init__(self) -> None:
+        self.tokenizer: Optional[AbstractTokenizer] = None

 
 

-    @staticmethod
-    def add_args(parser):
+    def add_args(self, parser: argparse.ArgumentParser) -> None:

         group = parser.add_argument_group('Data arguments')
         group.add_argument('--postcodes', action='store_true',
                            help='Update postcode centroid table')
         group = parser.add_argument_group('Data arguments')
         group.add_argument('--postcodes', action='store_true',
                            help='Update postcode centroid table')

+        group.add_argument('--word-tokens', action='store_true',
+                           help='Clean up search terms')

         group.add_argument('--word-counts', action='store_true',
                            help='Compute frequency of full-word search terms')
         group.add_argument('--address-levels', action='store_true',
         group.add_argument('--word-counts', action='store_true',
                            help='Compute frequency of full-word search terms')
         group.add_argument('--address-levels', action='store_true',


@@ -32,57 +62,89 @@ class UpdateRefresh:
         group.add_argument('--functions', action='store_true',
                            help='Update the PL/pgSQL functions in the database')
         group.add_argument('--wiki-data', action='store_true',
         group.add_argument('--functions', action='store_true',
                            help='Update the PL/pgSQL functions in the database')
         group.add_argument('--wiki-data', action='store_true',

-                           help='Update Wikipedia/data importance numbers.')
+                           help='Update Wikipedia/data importance numbers')
+        group.add_argument('--secondary-importance', action='store_true',
+                           help='Update secondary importance raster data')

         group.add_argument('--importance', action='store_true',
                            help='Recompute place importances (expensive!)')
         group.add_argument('--website', action='store_true',
                            help='Refresh the directory that serves the scripts for the web API')
         group.add_argument('--importance', action='store_true',
                            help='Recompute place importances (expensive!)')
         group.add_argument('--website', action='store_true',
                            help='Refresh the directory that serves the scripts for the web API')

+        group.add_argument('--data-object', action='append',
+                           type=_parse_osm_object, metavar='OBJECT',
+                           help='Mark the given OSM object as requiring an update'
+                                ' (format: [NWR]<id>)')
+        group.add_argument('--data-area', action='append',
+                           type=_parse_osm_object, metavar='OBJECT',
+                           help='Mark the area around the given OSM object as requiring an update'
+                                ' (format: [NWR]<id>)')
+

         group = parser.add_argument_group('Arguments for function refresh')
         group.add_argument('--no-diff-updates', action='store_false', dest='diffs',
                            help='Do not enable code for propagating updates')
         group.add_argument('--enable-debug-statements', action='store_true',
                            help='Enable debug warning statements in functions')
 
         group = parser.add_argument_group('Arguments for function refresh')
         group.add_argument('--no-diff-updates', action='store_false', dest='diffs',
                            help='Do not enable code for propagating updates')
         group.add_argument('--enable-debug-statements', action='store_true',
                            help='Enable debug warning statements in functions')
 

-    @staticmethod
-    def run(args):
+
+    def run(self, args: NominatimArgs) -> int: #pylint: disable=too-many-branches, too-many-statements

         from ..tools import refresh, postcodes
         from ..tools import refresh, postcodes

-        from ..tokenizer import factory as tokenizer_factory

         from ..indexer.indexer import Indexer
 
         from ..indexer.indexer import Indexer
 

-        tokenizer = tokenizer_factory.get_tokenizer_for_db(args.config)

 
         if args.postcodes:
 
         if args.postcodes:

-            LOG.warning("Update postcodes centroid")
-            postcodes.update_postcodes(args.config.get_libpq_dsn(),
-                                       args.project_dir, tokenizer)
-            indexer = Indexer(args.config.get_libpq_dsn(), tokenizer,
-                              args.threads or 1)
-            indexer.index_postcodes()
+            if postcodes.can_compute(args.config.get_libpq_dsn()):
+                LOG.warning("Update postcodes centroid")
+                tokenizer = self._get_tokenizer(args.config)
+                postcodes.update_postcodes(args.config.get_libpq_dsn(),
+                                           args.project_dir, tokenizer)
+                indexer = Indexer(args.config.get_libpq_dsn(), tokenizer,
+                                  args.threads or 1)
+                indexer.index_postcodes()
+            else:
+                LOG.error("The place table doesn't exist. "
+                          "Postcode updates on a frozen database is not possible.")
+
+        if args.word_tokens:
+            LOG.warning('Updating word tokens')
+            tokenizer = self._get_tokenizer(args.config)
+            tokenizer.update_word_tokens()

 
         if args.word_counts:
 
         if args.word_counts:

-            LOG.warning('Recompute frequency of full-word search terms')
-            refresh.recompute_word_counts(args.config.get_libpq_dsn(), args.sqllib_dir)
+            LOG.warning('Recompute word statistics')
+            self._get_tokenizer(args.config).update_statistics()

 
         if args.address_levels:
 
         if args.address_levels:

-            cfg = Path(args.config.ADDRESS_LEVEL_CONFIG)
-            LOG.warning('Updating address levels from %s', cfg)
+            LOG.warning('Updating address levels')
+            with connect(args.config.get_libpq_dsn()) as conn:
+                refresh.load_address_levels_from_config(conn, args.config)
+
+        # Attention: must come BEFORE functions
+        if args.secondary_importance:

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

-                refresh.load_address_levels_from_file(conn, cfg)
+                # If the table did not exist before, then the importance code
+                # needs to be enabled.
+                if not conn.table_exists('secondary_importance'):
+                    args.functions = True
+
+            LOG.warning('Import secondary importance raster data from %s', args.project_dir)
+            if refresh.import_secondary_importance(args.config.get_libpq_dsn(),
+                                                args.project_dir) > 0:
+                LOG.fatal('FATAL: Cannot update secondary importance raster data')
+                return 1

 
         if args.functions:
             LOG.warning('Create functions')
             with connect(args.config.get_libpq_dsn()) as conn:
                 refresh.create_functions(conn, args.config,
                                          args.diffs, args.enable_debug_statements)
 
         if args.functions:
             LOG.warning('Create functions')
             with connect(args.config.get_libpq_dsn()) as conn:
                 refresh.create_functions(conn, args.config,
                                          args.diffs, args.enable_debug_statements)

-                tokenizer.update_sql_functions(args.config)
+                self._get_tokenizer(args.config).update_sql_functions(args.config)

 
         if args.wiki_data:
             data_path = Path(args.config.WIKIPEDIA_DATA_PATH
                              or args.project_dir)
 
         if args.wiki_data:
             data_path = Path(args.config.WIKIPEDIA_DATA_PATH
                              or args.project_dir)

-            LOG.warning('Import wikipdia article importance from %s', data_path)
+            LOG.warning('Import wikipedia article importance from %s', data_path)

             if refresh.import_wikipedia_articles(args.config.get_libpq_dsn(),
                                                  data_path) > 0:
             if refresh.import_wikipedia_articles(args.config.get_libpq_dsn(),
                                                  data_path) > 0:

-                LOG.fatal('FATAL: Wikipedia importance dump file not found')
+                LOG.fatal('FATAL: Wikipedia importance file not found in %s', data_path)

                 return 1
 
         # Attention: importance MUST come after wiki data import.
                 return 1
 
         # Attention: importance MUST come after wiki data import.


@@ -94,6 +156,28 @@ class UpdateRefresh:
         if args.website:
             webdir = args.project_dir / 'website'
             LOG.warning('Setting up website directory at %s', webdir)
         if args.website:
             webdir = args.project_dir / 'website'
             LOG.warning('Setting up website directory at %s', webdir)

+            # This is a little bit hacky: call the tokenizer setup, so that
+            # the tokenizer directory gets repopulated as well, in case it
+            # wasn't there yet.
+            self._get_tokenizer(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:
                 refresh.setup_website(webdir, args.config, conn)

+
+        if args.data_object or args.data_area:
+            with connect(args.config.get_libpq_dsn()) as conn:
+                for obj in args.data_object or []:
+                    refresh.invalidate_osm_object(*obj, conn, recursive=False)
+                for obj in args.data_area or []:
+                    refresh.invalidate_osm_object(*obj, conn, recursive=True)
+                conn.commit()
+

         return 0
         return 0

+
+
+    def _get_tokenizer(self, config: Configuration) -> AbstractTokenizer:
+        if self.tokenizer is None:
+            from ..tokenizer import factory as tokenizer_factory
+
+            self.tokenizer = tokenizer_factory.get_tokenizer_for_db(config)
+
+        return self.tokenizer