mirror of
https://github.com/searxng/searxng.git
synced 2024-11-23 04:27:46 +01:00
542f7d0d7b
In the past, some files were tested with the standard profile, others with a profile in which most of the messages were switched off ... some files were not checked at all. - ``PYLINT_SEARXNG_DISABLE_OPTION`` has been abolished - the distinction ``# lint: pylint`` is no longer necessary - the pylint tasks have been reduced from three to two 1. ./searx/engines -> lint engines with additional builtins 2. ./searx ./searxng_extra ./tests -> lint all other python files Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
67 lines
1.9 KiB
Python
Executable File
67 lines
1.9 KiB
Python
Executable File
#!/usr/bin/env python
|
|
# SPDX-License-Identifier: AGPL-3.0-or-later
|
|
"""Fetch units from :origin:`searx/engines/wikidata.py` engine.
|
|
|
|
Output file: :origin:`searx/data/wikidata_units.json` (:origin:`CI Update data
|
|
... <.github/workflows/data-update.yml>`).
|
|
|
|
"""
|
|
|
|
import json
|
|
import collections
|
|
|
|
# set path
|
|
from os.path import join
|
|
|
|
from searx import searx_dir
|
|
from searx.engines import wikidata, set_loggers
|
|
from searx.data import data_dir
|
|
|
|
DATA_FILE = data_dir / 'wikidata_units.json'
|
|
|
|
set_loggers(wikidata, 'wikidata')
|
|
|
|
# the response contains duplicate ?item with the different ?symbol
|
|
# "ORDER BY ?item DESC(?rank) ?symbol" provides a deterministic result
|
|
# even if a ?item has different ?symbol of the same rank.
|
|
# A deterministic result
|
|
# see:
|
|
# * https://www.wikidata.org/wiki/Help:Ranking
|
|
# * https://www.mediawiki.org/wiki/Wikibase/Indexing/RDF_Dump_Format ("Statement representation" section)
|
|
# * https://w.wiki/32BT
|
|
# see the result for https://www.wikidata.org/wiki/Q11582
|
|
# there are multiple symbols the same rank
|
|
SARQL_REQUEST = """
|
|
SELECT DISTINCT ?item ?symbol
|
|
WHERE
|
|
{
|
|
?item wdt:P31/wdt:P279 wd:Q47574 .
|
|
?item p:P5061 ?symbolP .
|
|
?symbolP ps:P5061 ?symbol ;
|
|
wikibase:rank ?rank .
|
|
FILTER(LANG(?symbol) = "en").
|
|
}
|
|
ORDER BY ?item DESC(?rank) ?symbol
|
|
"""
|
|
|
|
|
|
def get_data():
|
|
results = collections.OrderedDict()
|
|
response = wikidata.send_wikidata_query(SARQL_REQUEST)
|
|
for unit in response['results']['bindings']:
|
|
name = unit['item']['value'].replace('http://www.wikidata.org/entity/', '')
|
|
unit = unit['symbol']['value']
|
|
if name not in results:
|
|
# ignore duplicate: always use the first one
|
|
results[name] = unit
|
|
return results
|
|
|
|
|
|
def get_wikidata_units_filename():
|
|
return join(join(searx_dir, "data"), "")
|
|
|
|
|
|
if __name__ == '__main__':
|
|
with DATA_FILE.open('w', encoding="utf8") as f:
|
|
json.dump(get_data(), f, indent=4, sort_keys=True, ensure_ascii=False)
|