on master help sign in
Merge pull request #2602 from MarcAbonce/fix-bing-fetch-languages

Fix fetch_languages for Bing

commitf77983e174ec45913f32482210e36c3cb94cf810 authorAlexandre Flament <alex@al-f.net> dateMon 1 Mar 2021, 09:06 (6 years ago) parent5f6ac3ad6681fd
3 files changed+8−5
searx/engines/__init__.py+6−2 searx/engines/wikipedia.py+1−1 utils/fetch_languages.py+1−2
searx/engines/__init__.py+6−2
@@ -27,7 +27,7 @@ from searx import settings
2727from searx import logger
2828from searx.data import ENGINES_LANGUAGES
2929from searx.poolrequests import get, get_proxy_cycles
30−from searx.utils import load_module, match_language, get_engine_from_settings
30+from searx.utils import load_module, match_language, get_engine_from_settings, gen_useragent
3131
3232
3333logger = logger.getChild('engines')
@@ -131,8 +131,12 @@ def load_engine(engine_data):
131131
132132 # assign language fetching method if auxiliary method exists
133133 if hasattr(engine, '_fetch_supported_languages'):
134+ headers = {
135+ 'User-Agent': gen_useragent(),
136+ 'Accept-Language': 'ja-JP,ja;q=0.8,en-US;q=0.5,en;q=0.3', # bing needs a non-English language
137+ }
134138 setattr(engine, 'fetch_supported_languages',
135− lambda: engine._fetch_supported_languages(get(engine.supported_languages_url)))
139+ lambda: engine._fetch_supported_languages(get(engine.supported_languages_url, headers=headers)))
136140
137141 engine.stats = {
138142 'sent_search_count': 0, # sent search
searx/engines/wikipedia.py+1−1
@@ -106,6 +106,6 @@ def _fetch_supported_languages(resp):
106106 articles = int(td[4].xpath('./a/b')[0].text.replace(',', ''))
107107 # exclude languages with too few articles
108108 if articles >= 100:
109− supported_languages[code] = {"name": name, "english_name": english_name, "articles": articles}
109+ supported_languages[code] = {"name": name, "english_name": english_name}
110110
111111 return supported_languages
utils/fetch_languages.py+1−2
@@ -2,8 +2,7 @@
22
33# This script generates languages.py from intersecting each engine's supported languages.
44#
5−# Output files (engines_languages.json and languages.py)
6−# are written in current directory to avoid overwriting in case something goes wrong.
5+# Output files: searx/data/engines_languages.json and searx/languages.py
76
87import json
98from pathlib import Path