X-Git-Url: https://git.openstreetmap.org./nominatim.git/blobdiff_plain/e5206133628c0ab1cacd6c5a04a2a9a973bfc86c..a7a920a9a5b55bf4290b184f05898a8589c95b40:/nominatim/clicmd/index.py diff --git a/nominatim/clicmd/index.py b/nominatim/clicmd/index.py index 96a69396..86a504e7 100644 --- a/nominatim/clicmd/index.py +++ b/nominatim/clicmd/index.py @@ -1,32 +1,37 @@ +# 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 the 'index' subcommand. """ -import os +import argparse -from ..db import status -from ..db.connection import connect +import psutil + +from nominatim.db import status +from nominatim.db.connection import connect +from nominatim.clicmd.args import NominatimArgs # Do not repeat documentation of subcommand classes. # pylint: disable=C0111 # Using non-top-level imports to avoid eventually unused imports. # pylint: disable=E0012,C0415 -def _num_system_cpus(): - try: - cpus = len(os.sched_getaffinity(0)) - except NotImplementedError: - cpus = None - - return cpus or os.cpu_count() - class UpdateIndex: """\ Reindex all new and modified data. + + Indexing is the process of computing the address and search terms for + the places in the database. Every time data is added or changed, indexing + needs to be run. Imports and replication updates automatically take care + of indexing. For other cases, this function allows to run indexing manually. """ - @staticmethod - def add_args(parser): + def add_args(self, parser: argparse.ArgumentParser) -> None: group = parser.add_argument_group('Filter arguments') group.add_argument('--boundaries-only', action='store_true', help="""Index only administrative boundaries.""") @@ -37,17 +42,21 @@ class UpdateIndex: group.add_argument('--maxrank', '-R', type=int, metavar='RANK', default=30, help='Maximum/finishing rank') - @staticmethod - def run(args): + + def run(self, args: NominatimArgs) -> int: from ..indexer.indexer import Indexer + from ..tokenizer import factory as tokenizer_factory + + tokenizer = tokenizer_factory.get_tokenizer_for_db(args.config) - indexer = Indexer(args.config.get_libpq_dsn(), - args.threads or _num_system_cpus() or 1) + indexer = Indexer(args.config.get_libpq_dsn(), tokenizer, + args.threads or psutil.cpu_count() or 1) if not args.no_boundaries: indexer.index_boundaries(args.minrank, args.maxrank) if not args.boundaries_only: indexer.index_by_rank(args.minrank, args.maxrank) + indexer.index_postcodes() if not args.no_boundaries and not args.boundaries_only \ and args.minrank == 0 and args.maxrank == 30: