summaryrefslogtreecommitdiff
path: root/searxng_extra/update/update_currencies.py
blob: e51692e72a832c4a7ee7004a997a79da9c2c9ead (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
#!/usr/bin/env python
# lint: pylint
# SPDX-License-Identifier: AGPL-3.0-or-later

"""Fetch currencies from :origin:`searx/engines/wikidata.py` engine.

Output file: :origin:`searx/data/currencies.json` (:origin:`CI Update data ...
<.github/workflows/data-update.yml>`).

"""

# pylint: disable=invalid-name

import re
import unicodedata
import json

# set path
from os.path import join

from searx import searx_dir
from searx.locales import LOCALE_NAMES
from searx.engines import wikidata, set_loggers

set_loggers(wikidata, 'wikidata')

# ORDER BY (with all the query fields) is important to keep a deterministic result order
# so multiple invokation of this script doesn't change currencies.json
SARQL_REQUEST = """
SELECT DISTINCT ?iso4217 ?unit ?unicode ?label ?alias WHERE {
  ?item wdt:P498 ?iso4217; rdfs:label ?label.
  OPTIONAL { ?item skos:altLabel ?alias FILTER (LANG (?alias) = LANG(?label)). }
  OPTIONAL { ?item wdt:P5061 ?unit. }
  OPTIONAL { ?item wdt:P489 ?symbol.
             ?symbol wdt:P487 ?unicode. }
  MINUS { ?item wdt:P582 ?end_data . }                  # Ignore monney with an end date
  MINUS { ?item wdt:P31/wdt:P279* wd:Q15893266 . }      # Ignore "former entity" (obsolete currency)
  FILTER(LANG(?label) IN (%LANGUAGES_SPARQL%)).
}
ORDER BY ?iso4217 ?unit ?unicode ?label ?alias
"""

# ORDER BY (with all the query fields) is important to keep a deterministic result order
# so multiple invokation of this script doesn't change currencies.json
SPARQL_WIKIPEDIA_NAMES_REQUEST = """
SELECT DISTINCT ?iso4217 ?article_name WHERE {
  ?item wdt:P498 ?iso4217 .
  ?article schema:about ?item ;
           schema:name ?article_name ;
           schema:isPartOf [ wikibase:wikiGroup "wikipedia" ]
  MINUS { ?item wdt:P582 ?end_data . }                  # Ignore monney with an end date
  MINUS { ?item wdt:P31/wdt:P279* wd:Q15893266 . }      # Ignore "former entity" (obsolete currency)
  FILTER(LANG(?article_name) IN (%LANGUAGES_SPARQL%)).
}
ORDER BY ?iso4217 ?article_name
"""


LANGUAGES = LOCALE_NAMES.keys()
LANGUAGES_SPARQL = ', '.join(set(map(lambda l: repr(l.split('_')[0]), LANGUAGES)))


def remove_accents(name):
    return unicodedata.normalize('NFKD', name).lower()


def remove_extra(name):
    for c in ('(', ':'):
        if c in name:
            name = name.split(c)[0].strip()
    return name


def _normalize_name(name):
    name = re.sub(' +', ' ', remove_accents(name.lower()).replace('-', ' '))
    name = remove_extra(name)
    return name


def add_currency_name(db, name, iso4217, normalize_name=True):
    db_names = db['names']

    if normalize_name:
        name = _normalize_name(name)

    iso4217_set = db_names.setdefault(name, [])
    if iso4217 not in iso4217_set:
        iso4217_set.insert(0, iso4217)


def add_currency_label(db, label, iso4217, language):
    labels = db['iso4217'].setdefault(iso4217, {})
    labels[language] = label


def wikidata_request_result_iterator(request):
    result = wikidata.send_wikidata_query(request.replace('%LANGUAGES_SPARQL%', LANGUAGES_SPARQL))
    if result is not None:
        for r in result['results']['bindings']:
            yield r


def fetch_db():
    db = {
        'names': {},
        'iso4217': {},
    }

    for r in wikidata_request_result_iterator(SPARQL_WIKIPEDIA_NAMES_REQUEST):
        iso4217 = r['iso4217']['value']
        article_name = r['article_name']['value']
        article_lang = r['article_name']['xml:lang']
        add_currency_name(db, article_name, iso4217)
        add_currency_label(db, article_name, iso4217, article_lang)

    for r in wikidata_request_result_iterator(SARQL_REQUEST):
        iso4217 = r['iso4217']['value']
        if 'label' in r:
            label = r['label']['value']
            label_lang = r['label']['xml:lang']
            add_currency_name(db, label, iso4217)
            add_currency_label(db, label, iso4217, label_lang)

        if 'alias' in r:
            add_currency_name(db, r['alias']['value'], iso4217)

        if 'unicode' in r:
            add_currency_name(db, r['unicode']['value'], iso4217, normalize_name=False)

        if 'unit' in r:
            add_currency_name(db, r['unit']['value'], iso4217, normalize_name=False)

    # reduce memory usage:
    # replace lists with one item by the item.
    # see searx.search.processors.online_currency.name_to_iso4217
    for name in db['names']:
        if len(db['names'][name]) == 1:
            db['names'][name] = db['names'][name][0]

    return db


def get_filename():
    return join(join(searx_dir, "data"), "currencies.json")


def main():
    #
    db = fetch_db()
    # static
    add_currency_name(db, "euro", 'EUR')
    add_currency_name(db, "euros", 'EUR')
    add_currency_name(db, "dollar", 'USD')
    add_currency_name(db, "dollars", 'USD')
    add_currency_name(db, "peso", 'MXN')
    add_currency_name(db, "pesos", 'MXN')

    with open(get_filename(), 'w', encoding='utf8') as f:
        json.dump(db, f, ensure_ascii=False, indent=4)


if __name__ == '__main__':
    main()