2 Command-line interface to the Nominatim functions for import, update,
3 database administration and querying.
9 from pathlib import Path
13 from .config import Configuration
14 from .tools.exec_utils import run_legacy_script, run_api_script
16 LOG = logging.getLogger()
18 def _num_system_cpus():
20 cpus = len(os.sched_getaffinity(0))
21 except NotImplementedError:
24 return cpus or os.cpu_count()
27 class CommandlineParser:
28 """ Wraps some of the common functions for parsing the command line
29 and setting up subcommands.
31 def __init__(self, prog, description):
32 self.parser = argparse.ArgumentParser(
34 description=description,
35 formatter_class=argparse.RawDescriptionHelpFormatter)
37 self.subs = self.parser.add_subparsers(title='available commands',
40 # Arguments added to every sub-command
41 self.default_args = argparse.ArgumentParser(add_help=False)
42 group = self.default_args.add_argument_group('Default arguments')
43 group.add_argument('-h', '--help', action='help',
44 help='Show this help message and exit')
45 group.add_argument('-q', '--quiet', action='store_const', const=0,
46 dest='verbose', default=1,
47 help='Print only error messages')
48 group.add_argument('-v', '--verbose', action='count', default=1,
49 help='Increase verboseness of output')
50 group.add_argument('--project-dir', metavar='DIR', default='.',
51 help='Base directory of the Nominatim installation (default:.)')
52 group.add_argument('-j', '--threads', metavar='NUM', type=int,
53 help='Number of parallel threads to use')
56 def add_subcommand(self, name, cmd):
57 """ Add a subcommand to the parser. The subcommand must be a class
58 with a function add_args() that adds the parameters for the
59 subcommand and a run() function that executes the command.
61 parser = self.subs.add_parser(name, parents=[self.default_args],
62 help=cmd.__doc__.split('\n', 1)[0],
63 description=cmd.__doc__,
64 formatter_class=argparse.RawDescriptionHelpFormatter,
66 parser.set_defaults(command=cmd)
69 def run(self, **kwargs):
70 """ Parse the command line arguments of the program and execute the
71 appropriate subcommand.
73 args = self.parser.parse_args(args=kwargs.get('cli_args'))
75 if args.subcommand is None:
76 self.parser.print_help()
79 for arg in ('module_dir', 'osm2pgsql_path', 'phplib_dir', 'data_dir', 'phpcgi_path'):
80 setattr(args, arg, Path(kwargs[arg]))
81 args.project_dir = Path(args.project_dir)
83 logging.basicConfig(stream=sys.stderr,
84 format='%(asctime)s: %(message)s',
85 datefmt='%Y-%m-%d %H:%M:%S',
86 level=max(4 - args.verbose, 1) * 10)
88 args.config = Configuration(args.project_dir, args.data_dir / 'settings')
90 return args.command.run(args)
92 ##### Subcommand classes
94 # Each class needs to implement two functions: add_args() adds the CLI parameters
95 # for the subfunction, run() executes the subcommand.
97 # The class documentation doubles as the help text for the command. The
98 # first line is also used in the summary when calling the program without
101 # No need to document the functions each time.
102 # pylint: disable=C0111
107 Create a new Nominatim database from an OSM file.
111 def add_args(parser):
112 group_name = parser.add_argument_group('Required arguments')
113 group = group_name.add_mutually_exclusive_group(required=True)
114 group.add_argument('--osm-file',
115 help='OSM file to be imported.')
116 group.add_argument('--continue', dest='continue_at',
117 choices=['load-data', 'indexing', 'db-postprocess'],
118 help='Continue an import that was interrupted')
119 group = parser.add_argument_group('Optional arguments')
120 group.add_argument('--osm2pgsql-cache', metavar='SIZE', type=int,
121 help='Size of cache to be used by osm2pgsql (in MB)')
122 group.add_argument('--reverse-only', action='store_true',
123 help='Do not create tables and indexes for searching')
124 group.add_argument('--enable-debug-statements', action='store_true',
125 help='Include debug warning statements in SQL code')
126 group.add_argument('--no-partitions', action='store_true',
127 help="""Do not partition search indices
128 (speeds up import of single country extracts)""")
129 group.add_argument('--no-updates', action='store_true',
130 help="""Do not keep tables that are only needed for
131 updating the database later""")
132 group = parser.add_argument_group('Expert options')
133 group.add_argument('--ignore-errors', action='store_true',
134 help='Continue import even when errors in SQL are present')
135 group.add_argument('--index-noanalyse', action='store_true',
136 help='Do not perform analyse operations during index')
141 params = ['setup.php']
143 params.extend(('--all', '--osm-file', args.osm_file))
145 if args.continue_at == 'load-data':
146 params.append('--load-data')
147 if args.continue_at in ('load-data', 'indexing'):
148 params.append('--index')
149 params.extend(('--create-search-indices', '--create-country-names',
151 if args.osm2pgsql_cache:
152 params.extend(('--osm2pgsql-cache', args.osm2pgsql_cache))
153 if args.reverse_only:
154 params.append('--reverse-only')
155 if args.enable_debug_statements:
156 params.append('--enable-debug-statements')
157 if args.no_partitions:
158 params.append('--no-partitions')
160 params.append('--drop')
161 if args.ignore_errors:
162 params.append('--ignore-errors')
163 if args.index_noanalyse:
164 params.append('--index-noanalyse')
166 return run_legacy_script(*params, nominatim_env=args)
171 Make database read-only.
173 About half of data in the Nominatim database is kept only to be able to
174 keep the data up-to-date with new changes made in OpenStreetMap. This
175 command drops all this data and only keeps the part needed for geocoding
178 This command has the same effect as the `--no-updates` option for imports.
182 def add_args(parser):
187 return run_legacy_script('setup.php', '--drop', nominatim_env=args)
190 class SetupSpecialPhrases:
192 Maintain special phrases.
196 def add_args(parser):
197 group = parser.add_argument_group('Input arguments')
198 group.add_argument('--from-wiki', action='store_true',
199 help='Pull special phrases from the OSM wiki.')
200 group = parser.add_argument_group('Output arguments')
201 group.add_argument('-o', '--output', default='-',
202 help="""File to write the preprocessed phrases to.
203 If omitted, it will be written to stdout.""")
207 if args.output != '-':
208 raise NotImplementedError('Only output to stdout is currently implemented.')
209 return run_legacy_script('specialphrases.php', '--wiki-import', nominatim_env=args)
212 class UpdateReplication:
214 Update the database using an online replication service.
218 def add_args(parser):
219 group = parser.add_argument_group('Arguments for initialisation')
220 group.add_argument('--init', action='store_true',
221 help='Initialise the update process')
222 group.add_argument('--no-update-functions', dest='update_functions',
223 action='store_false',
224 help="""Do not update the trigger function to
225 support differential updates.""")
226 group = parser.add_argument_group('Arguments for updates')
227 group.add_argument('--check-for-updates', action='store_true',
228 help='Check if new updates are available and exit')
229 group.add_argument('--once', action='store_true',
230 help="""Download and apply updates only once. When
231 not set, updates are continuously applied""")
232 group.add_argument('--no-index', action='store_false', dest='do_index',
233 help="""Do not index the new data. Only applicable
234 together with --once""")
238 params = ['update.php']
240 params.append('--init-updates')
241 if not args.update_functions:
242 params.append('--no-update-functions')
243 elif args.check_for_updates:
244 params.append('--check-for-updates')
247 params.append('--import-osmosis')
249 params.append('--import-osmosis-all')
250 if not args.do_index:
251 params.append('--no-index')
253 return run_legacy_script(*params, nominatim_env=args)
258 Add additional data from a file or an online source.
260 Data is only imported, not indexed. You need to call `nominatim-update index`
261 to complete the process.
265 def add_args(parser):
266 group_name = parser.add_argument_group('Source')
267 group = group_name.add_mutually_exclusive_group(required=True)
268 group.add_argument('--file', metavar='FILE',
269 help='Import data from an OSM file')
270 group.add_argument('--diff', metavar='FILE',
271 help='Import data from an OSM diff file')
272 group.add_argument('--node', metavar='ID', type=int,
273 help='Import a single node from the API')
274 group.add_argument('--way', metavar='ID', type=int,
275 help='Import a single way from the API')
276 group.add_argument('--relation', metavar='ID', type=int,
277 help='Import a single relation from the API')
278 group.add_argument('--tiger-data', metavar='DIR',
279 help='Add housenumbers from the US TIGER census database.')
280 group = parser.add_argument_group('Extra arguments')
281 group.add_argument('--use-main-api', action='store_true',
282 help='Use OSM API instead of Overpass to download objects')
287 os.environ['NOMINATIM_TIGER_DATA_PATH'] = args.tiger_data
288 return run_legacy_script('setup.php', '--import-tiger-data', nominatim_env=args)
290 params = ['update.php']
292 params.extend(('--import-file', args.file))
294 params.extend(('--import-diff', args.diff))
296 params.extend(('--import-node', args.node))
298 params.extend(('--import-way', args.way))
300 params.extend(('--import-relation', args.relation))
301 if args.use_main_api:
302 params.append('--use-main-api')
303 return run_legacy_script(*params, nominatim_env=args)
308 Reindex all new and modified data.
312 def add_args(parser):
313 group = parser.add_argument_group('Filter arguments')
314 group.add_argument('--boundaries-only', action='store_true',
315 help="""Index only administrative boundaries.""")
316 group.add_argument('--no-boundaries', action='store_true',
317 help="""Index everything except administrative boundaries.""")
318 group.add_argument('--minrank', '-r', type=int, metavar='RANK', default=0,
319 help='Minimum/starting rank')
320 group.add_argument('--maxrank', '-R', type=int, metavar='RANK', default=30,
321 help='Maximum/finishing rank')
325 from .indexer.indexer import Indexer
327 indexer = Indexer(args.config.get_libpq_dsn(),
328 args.threads or _num_system_cpus() or 1)
330 if not args.no_boundaries:
331 indexer.index_boundaries(args.minrank, args.maxrank)
332 if not args.boundaries_only:
333 indexer.index_by_rank(args.minrank, args.maxrank)
335 if not args.no_boundaries and not args.boundaries_only:
336 indexer.update_status_table()
343 Recompute auxiliary data used by the indexing process.
345 These functions must not be run in parallel with other update commands.
349 def add_args(parser):
350 group = parser.add_argument_group('Data arguments')
351 group.add_argument('--postcodes', action='store_true',
352 help='Update postcode centroid table')
353 group.add_argument('--word-counts', action='store_true',
354 help='Compute frequency of full-word search terms')
355 group.add_argument('--address-levels', action='store_true',
356 help='Reimport address level configuration')
357 group.add_argument('--functions', action='store_true',
358 help='Update the PL/pgSQL functions in the database')
359 group.add_argument('--wiki-data', action='store_true',
360 help='Update Wikipedia/data importance numbers.')
361 group.add_argument('--importance', action='store_true',
362 help='Recompute place importances (expensive!)')
363 group.add_argument('--website', action='store_true',
364 help='Refresh the directory that serves the scripts for the web API')
365 group = parser.add_argument_group('Arguments for function refresh')
366 group.add_argument('--no-diff-updates', action='store_false', dest='diffs',
367 help='Do not enable code for propagating updates')
368 group.add_argument('--enable-debug-statements', action='store_true',
369 help='Enable debug warning statements in functions')
373 import nominatim.tools.refresh
375 conn = psycopg2.connect(args.config.get_libpq_dsn())
378 LOG.warning("Update postcodes centroid")
379 nominatim.tools.refresh.update_postcodes(conn, args.data_dir)
381 LOG.warning('Recompute frequency of full-word search terms')
382 nominatim.tools.refresh.recompute_word_counts(conn, args.data_dir)
383 if args.address_levels:
384 cfg = Path(args.config.ADDRESS_LEVEL_CONFIG)
385 LOG.warning('Updating address levels from %s', cfg)
386 nominatim.tools.refresh.load_address_levels_from_file(conn, cfg)
388 params = ['setup.php', '--create-functions', '--create-partition-functions']
390 params.append('--enable-diff-updates')
391 if args.enable_debug_statements:
392 params.append('--enable-debug-statements')
393 run_legacy_script(*params, nominatim_env=args, throw_on_fail=True)
395 run_legacy_script('setup.php', '--import-wikipedia-articles',
396 nominatim_env=args, throw_on_fail=True)
397 # Attention: importance MUST come after wiki data import.
399 run_legacy_script('update.php', '--recompute-importance',
400 nominatim_env=args, throw_on_fail=True)
402 run_legacy_script('setup.php', '--setup-website',
403 nominatim_env=args, throw_on_fail=True)
410 class AdminCheckDatabase:
412 Check that the database is complete and operational.
416 def add_args(parser):
421 return run_legacy_script('check_import_finished.php', nominatim_env=args)
426 Warm database caches for search and reverse queries.
430 def add_args(parser):
431 group = parser.add_argument_group('Target arguments')
432 group.add_argument('--search-only', action='store_const', dest='target',
434 help="Only pre-warm tables for search queries")
435 group.add_argument('--reverse-only', action='store_const', dest='target',
437 help="Only pre-warm tables for reverse queries")
441 params = ['warm.php']
442 if args.target == 'reverse':
443 params.append('--reverse-only')
444 if args.target == 'search':
445 params.append('--search-only')
446 return run_legacy_script(*params, nominatim_env=args)
451 Export addresses as CSV file from the database.
455 def add_args(parser):
456 group = parser.add_argument_group('Output arguments')
457 group.add_argument('--output-type', default='street',
458 choices=('continent', 'country', 'state', 'county',
459 'city', 'suburb', 'street', 'path'),
460 help='Type of places to output (default: street)')
461 group.add_argument('--output-format',
462 default='street;suburb;city;county;state;country',
463 help="""Semicolon-separated list of address types
464 (see --output-type). Multiple ranks can be
465 merged into one column by simply using a
466 comma-separated list.""")
467 group.add_argument('--output-all-postcodes', action='store_true',
468 help="""List all postcodes for address instead of
469 just the most likely one""")
470 group.add_argument('--language',
471 help="""Preferred language for output
472 (use local name, if omitted)""")
473 group = parser.add_argument_group('Filter arguments')
474 group.add_argument('--restrict-to-country', metavar='COUNTRY_CODE',
475 help='Export only objects within country')
476 group.add_argument('--restrict-to-osm-node', metavar='ID', type=int,
477 help='Export only children of this OSM node')
478 group.add_argument('--restrict-to-osm-way', metavar='ID', type=int,
479 help='Export only children of this OSM way')
480 group.add_argument('--restrict-to-osm-relation', metavar='ID', type=int,
481 help='Export only children of this OSM relation')
486 params = ['export.php',
487 '--output-type', args.output_type,
488 '--output-format', args.output_format]
489 if args.output_all_postcodes:
490 params.append('--output-all-postcodes')
492 params.extend(('--language', args.language))
493 if args.restrict_to_country:
494 params.extend(('--restrict-to-country', args.restrict_to_country))
495 if args.restrict_to_osm_node:
496 params.extend(('--restrict-to-osm-node', args.restrict_to_osm_node))
497 if args.restrict_to_osm_way:
498 params.extend(('--restrict-to-osm-way', args.restrict_to_osm_way))
499 if args.restrict_to_osm_relation:
500 params.extend(('--restrict-to-osm-relation', args.restrict_to_osm_relation))
502 return run_legacy_script(*params, nominatim_env=args)
505 ('street', 'housenumber and street'),
506 ('city', 'city, town or village'),
507 ('county', 'county'),
509 ('country', 'country'),
510 ('postalcode', 'postcode')
514 ('addressdetails', 'Include a breakdown of the address into elements.'),
515 ('extratags', """Include additional information if available
516 (e.g. wikipedia link, opening hours)."""),
517 ('namedetails', 'Include a list of alternative names.')
521 ('addressdetails', 'Include a breakdown of the address into elements.'),
522 ('keywords', 'Include a list of name keywords and address keywords.'),
523 ('linkedplaces', 'Include a details of places that are linked with this one.'),
524 ('hierarchy', 'Include details of places lower in the address hierarchy.'),
525 ('group_hierarchy', 'Group the places by type.'),
526 ('polygon_geojson', 'Include geometry of result.')
529 def _add_api_output_arguments(parser):
530 group = parser.add_argument_group('Output arguments')
531 group.add_argument('--format', default='jsonv2',
532 choices=['xml', 'json', 'jsonv2', 'geojson', 'geocodejson'],
533 help='Format of result')
534 for name, desc in EXTRADATA_PARAMS:
535 group.add_argument('--' + name, action='store_true', help=desc)
537 group.add_argument('--lang', '--accept-language', metavar='LANGS',
538 help='Preferred language order for presenting search results')
539 group.add_argument('--polygon-output',
540 choices=['geojson', 'kml', 'svg', 'text'],
541 help='Output geometry of results as a GeoJSON, KML, SVG or WKT.')
542 group.add_argument('--polygon-threshold', type=float, metavar='TOLERANCE',
543 help="""Simplify output geometry.
544 Parameter is difference tolerance in degrees.""")
549 Execute API search query.
553 def add_args(parser):
554 group = parser.add_argument_group('Query arguments')
555 group.add_argument('--query',
556 help='Free-form query string')
557 for name, desc in STRUCTURED_QUERY:
558 group.add_argument('--' + name, help='Structured query: ' + desc)
560 _add_api_output_arguments(parser)
562 group = parser.add_argument_group('Result limitation')
563 group.add_argument('--countrycodes', metavar='CC,..',
564 help='Limit search results to one or more countries.')
565 group.add_argument('--exclude_place_ids', metavar='ID,..',
566 help='List of search object to be excluded')
567 group.add_argument('--limit', type=int,
568 help='Limit the number of returned results')
569 group.add_argument('--viewbox', metavar='X1,Y1,X2,Y2',
570 help='Preferred area to find search results')
571 group.add_argument('--bounded', action='store_true',
572 help='Strictly restrict results to viewbox area')
574 group = parser.add_argument_group('Other arguments')
575 group.add_argument('--no-dedupe', action='store_false', dest='dedupe',
576 help='Do not remove duplicates from the result list')
582 params = dict(q=args.query)
584 params = {k : getattr(args, k) for k, _ in STRUCTURED_QUERY if getattr(args, k)}
586 for param, _ in EXTRADATA_PARAMS:
587 if getattr(args, param):
589 for param in ('format', 'countrycodes', 'exclude_place_ids', 'limit', 'viewbox'):
590 if getattr(args, param):
591 params[param] = getattr(args, param)
593 params['accept-language'] = args.lang
594 if args.polygon_output:
595 params['polygon_' + args.polygon_output] = '1'
596 if args.polygon_threshold:
597 params['polygon_threshold'] = args.polygon_threshold
599 params['bounded'] = '1'
601 params['dedupe'] = '0'
603 return run_api_script('search', args.project_dir,
604 phpcgi_bin=args.phpcgi_path, params=params)
608 Execute API reverse query.
612 def add_args(parser):
613 group = parser.add_argument_group('Query arguments')
614 group.add_argument('--lat', type=float, required=True,
615 help='Latitude of coordinate to look up (in WGS84)')
616 group.add_argument('--lon', type=float, required=True,
617 help='Longitude of coordinate to look up (in WGS84)')
618 group.add_argument('--zoom', type=int,
619 help='Level of detail required for the address')
621 _add_api_output_arguments(parser)
626 params = dict(lat=args.lat, lon=args.lon)
627 if args.zoom is not None:
628 params['zoom'] = args.zoom
630 for param, _ in EXTRADATA_PARAMS:
631 if getattr(args, param):
634 params['format'] = args.format
636 params['accept-language'] = args.lang
637 if args.polygon_output:
638 params['polygon_' + args.polygon_output] = '1'
639 if args.polygon_threshold:
640 params['polygon_threshold'] = args.polygon_threshold
642 return run_api_script('reverse', args.project_dir,
643 phpcgi_bin=args.phpcgi_path, params=params)
648 Execute API reverse query.
652 def add_args(parser):
653 group = parser.add_argument_group('Query arguments')
654 group.add_argument('--id', metavar='OSMID',
655 action='append', required=True, dest='ids',
656 help='OSM id to lookup in format <NRW><id> (may be repeated)')
658 _add_api_output_arguments(parser)
663 params = dict(osm_ids=','.join(args.ids))
665 for param, _ in EXTRADATA_PARAMS:
666 if getattr(args, param):
669 params['format'] = args.format
671 params['accept-language'] = args.lang
672 if args.polygon_output:
673 params['polygon_' + args.polygon_output] = '1'
674 if args.polygon_threshold:
675 params['polygon_threshold'] = args.polygon_threshold
677 return run_api_script('lookup', args.project_dir,
678 phpcgi_bin=args.phpcgi_path, params=params)
683 Execute API lookup query.
687 def add_args(parser):
688 group = parser.add_argument_group('Query arguments')
689 objs = group.add_mutually_exclusive_group(required=True)
690 objs.add_argument('--node', '-n', type=int,
691 help="Look up the OSM node with the given ID.")
692 objs.add_argument('--way', '-w', type=int,
693 help="Look up the OSM way with the given ID.")
694 objs.add_argument('--relation', '-r', type=int,
695 help="Look up the OSM relation with the given ID.")
696 objs.add_argument('--place_id', '-p', type=int,
697 help='Database internal identifier of the OSM object to look up.')
698 group.add_argument('--class', dest='object_class',
699 help="""Class type to disambiguated multiple entries
700 of the same object.""")
702 group = parser.add_argument_group('Output arguments')
703 for name, desc in DETAILS_SWITCHES:
704 group.add_argument('--' + name, action='store_true', help=desc)
705 group.add_argument('--lang', '--accept-language', metavar='LANGS',
706 help='Preferred language order for presenting search results')
711 params = dict(osmtype='N', osmid=args.node)
713 params = dict(osmtype='W', osmid=args.node)
715 params = dict(osmtype='R', osmid=args.node)
717 params = dict(place_id=args.place_id)
718 if args.object_class:
719 params['class'] = args.object_class
720 for name, _ in DETAILS_SWITCHES:
721 params[name] = '1' if getattr(args, name) else '0'
723 return run_api_script('details', args.project_dir,
724 phpcgi_bin=args.phpcgi_path, params=params)
729 Execute API status query.
733 def add_args(parser):
734 group = parser.add_argument_group('API parameters')
735 group.add_argument('--format', default='text', choices=['text', 'json'],
736 help='Format of result')
740 return run_api_script('status', args.project_dir,
741 phpcgi_bin=args.phpcgi_path,
742 params=dict(format=args.format))
745 def nominatim(**kwargs):
747 Command-line tools for importing, updating, administrating and
748 querying the Nominatim database.
750 parser = CommandlineParser('nominatim', nominatim.__doc__)
752 parser.add_subcommand('import', SetupAll)
753 parser.add_subcommand('freeze', SetupFreeze)
754 parser.add_subcommand('replication', UpdateReplication)
756 parser.add_subcommand('check-database', AdminCheckDatabase)
757 parser.add_subcommand('warm', AdminWarm)
759 parser.add_subcommand('special-phrases', SetupSpecialPhrases)
761 parser.add_subcommand('add-data', UpdateAddData)
762 parser.add_subcommand('index', UpdateIndex)
763 parser.add_subcommand('refresh', UpdateRefresh)
765 parser.add_subcommand('export', QueryExport)
767 if kwargs.get('phpcgi_path'):
768 parser.add_subcommand('search', APISearch)
769 parser.add_subcommand('reverse', APIReverse)
770 parser.add_subcommand('lookup', APILookup)
771 parser.add_subcommand('details', APIDetails)
772 parser.add_subcommand('status', APIStatus)
774 parser.parser.epilog = 'php-cgi not found. Query commands not available.'
776 return parser.run(**kwargs)