on master help sign in
Merge pull request #973 from MarcAbonce/languages

Remove 'all' option from search languages

commit49b845051f027cf9d29a20821968103a0f55c9d9 authorAdam Tauber <asciimoo@gmail.com> dateWed 6 Dec 2017, 14:34 (9 years ago) parentdf0b8ee4d17703
56 files changed+166−249
searx/engines/archlinux.py+3−3 searx/engines/bing.py+1−4 searx/engines/bing_news.py+1−4 searx/engines/dailymotion.py+1−4 searx/engines/duckduckgo.py+3−9 searx/engines/duckduckgo_images.py+2−6 searx/engines/faroo.py+1−4 searx/engines/gigablast.py+3−6 searx/engines/google.py+2−1 searx/engines/google_news.py+2−3 searx/engines/mediawiki.py+1−4 searx/engines/photon.py+3−4 searx/engines/qwant.py+11−12 searx/engines/startpage.py+2−3 searx/engines/subtitleseeker.py+1−1 searx/engines/swisscows.py+1−4 searx/engines/twitter.py+1−6 searx/engines/wikidata.py−4 searx/engines/wikipedia.py+1−1 searx/engines/yacy.py+1−3 searx/engines/yahoo.py+1−3 searx/engines/yahoo_news.py+1−4 searx/engines/youtube_api.py+1−3 searx/preferences.py−1 searx/query.py+19−15 searx/search.py+5−1 searx/settings.yml+1−1 searx/settings_robot.yml+1−1 searx/templates/courgette/preferences.html−1 searx/templates/legacy/preferences.html−1 searx/templates/oscar/languages.html+5−6 searx/templates/oscar/preferences.html+2−2 searx/templates/pix-art/preferences.html−1 searx/webapp.py+2−2 tests/unit/engines/test_archlinux.py+1−1 tests/unit/engines/test_bing.py+1−5 tests/unit/engines/test_bing_news.py−4 tests/unit/engines/test_dailymotion.py−4 tests/unit/engines/test_duckduckgo_images.py+1−1 tests/unit/engines/test_faroo.py+2−6 tests/unit/engines/test_gigablast.py+1−6 tests/unit/engines/test_google.py+1−1 tests/unit/engines/test_google_news.py+1−6 tests/unit/engines/test_mediawiki.py−4 tests/unit/engines/test_qwant.py+2−1 tests/unit/engines/test_startpage.py−4 tests/unit/engines/test_subtitleseeker.py−36 tests/unit/engines/test_swisscows.py−5 tests/unit/engines/test_twitter.py−6 tests/unit/engines/test_wikidata.py+3−3 tests/unit/engines/test_wikipedia.py−4 tests/unit/engines/test_yacy.py−5 tests/unit/engines/test_yahoo.py−7 tests/unit/engines/test_yahoo_news.py−7 tests/unit/test_preferences.py+10−5 tests/unit/test_query.py+64
searx/engines/archlinux.py+3−3
@@ -26,7 +26,7 @@ xpath_results = '//ul[@class="mw-search-results"]/li'
2626xpath_link = './/div[@class="mw-search-result-heading"]/a'
2727
2828
29−# cut 'en' from 'en_US', 'de' from 'de_CH', and so on
29+# cut 'en' from 'en-US', 'de' from 'de-CH', and so on
3030def locale_to_lang_code(locale):
3131 if locale.find('-') >= 0:
3232 locale = locale.split('-')[0]
@@ -36,7 +36,7 @@ def locale_to_lang_code(locale):
3636# wikis for some languages were moved off from the main site, we need to make
3737# requests to correct URLs to be able to get results in those languages
3838lang_urls = {
39− 'all': {
39+ 'en': {
4040 'base': 'https://wiki.archlinux.org',
4141 'search': '/index.php?title=Special:Search&offset={offset}&{query}'
4242 },
@@ -67,7 +67,7 @@ lang_urls = {
6767def get_lang_urls(language):
6868 if language in lang_urls:
6969 return lang_urls[language]
70− return lang_urls['all']
70+ return lang_urls['en']
7171
7272
7373# Language names to build search requests for
searx/engines/bing.py+1−4
@@ -32,10 +32,7 @@ search_string = 'search?{query}&first={offset}'
3232def request(query, params):
3333 offset = (params['pageno'] - 1) * 10 + 1
3434
35− if params['language'] != 'all':
36− lang = params['language'].split('-')[0].upper()
37− else:
38− lang = 'EN'
35+ lang = params['language'].split('-')[0].upper()
3936
4037 query = u'language:{} {}'.format(lang, query.decode('utf-8')).encode('utf-8')
4138
searx/engines/bing_news.py+1−4
@@ -71,10 +71,7 @@ def request(query, params):
7171
7272 offset = (params['pageno'] - 1) * 10 + 1
7373
74− if params['language'] == 'all':
75− language = 'en-US'
76− else:
77− language = params['language']
74+ language = params['language']
7875
7976 params['url'] = _get_url(query, language, offset, params['time_range'])
8077
searx/engines/dailymotion.py+1−4
@@ -32,10 +32,7 @@ supported_languages_url = 'https://api.dailymotion.com/languages'
3232
3333# do search-request
3434def request(query, params):
35− if params['language'] == 'all':
36− locale = 'en-US'
37− else:
38− locale = params['language']
35+ locale = params['language']
3936
4037 params['url'] = search_url.format(
4138 query=urlencode({'search': query, 'localization': locale}),
searx/engines/duckduckgo.py+3−9
@@ -44,9 +44,7 @@ content_xpath = './/a[@class="result__snippet"]'
4444# match query's language to a region code that duckduckgo will accept
4545def get_region_code(lang, lang_list=None):
4646 # custom fixes for languages
47− if lang == 'all':
48− region_code = None
49− elif lang[:2] == 'ja':
47+ if lang[:2] == 'ja':
5048 region_code = 'jp-jp'
5149 elif lang[:2] == 'sl':
5250 region_code = 'sl-sl'
@@ -82,12 +80,8 @@ def request(query, params):
8280 offset = (params['pageno'] - 1) * 30
8381
8482 region_code = get_region_code(params['language'])
85− if region_code:
86− params['url'] = url.format(
87− query=urlencode({'q': query, 'kl': region_code}), offset=offset, dc_param=offset)
88− else:
89− params['url'] = url.format(
90− query=urlencode({'q': query}), offset=offset, dc_param=offset)
83+ params['url'] = url.format(
84+ query=urlencode({'q': query, 'kl': region_code}), offset=offset, dc_param=offset)
9185
9286 if params['time_range'] in time_range_dict:
9387 params['url'] += time_range_url.format(range=time_range_dict[params['time_range']])
searx/engines/duckduckgo_images.py+2−6
@@ -53,12 +53,8 @@ def request(query, params):
5353 safesearch = params['safesearch'] - 1
5454
5555 region_code = get_region_code(params['language'], lang_list=supported_languages)
56− if region_code:
57− params['url'] = images_url.format(
58− query=urlencode({'q': query, 'l': region_code}), offset=offset, safesearch=safesearch, vqd=vqd)
59− else:
60− params['url'] = images_url.format(
61− query=urlencode({'q': query}), offset=offset, safesearch=safesearch, vqd=vqd)
56+ params['url'] = images_url.format(
57+ query=urlencode({'q': query, 'l': region_code}), offset=offset, safesearch=safesearch, vqd=vqd)
6258
6359 return params
6460
searx/engines/faroo.py+1−4
@@ -40,10 +40,7 @@ def request(query, params):
4040 offset = (params['pageno'] - 1) * number_of_results + 1
4141 categorie = search_category.get(params['category'], 'web')
4242
43− if params['language'] == 'all':
44− language = 'en'
45− else:
46− language = params['language'].split('_')[0]
43+ language = params['language'].split('-')[0]
4744
4845 # if language is not supported, put it in english
4946 if language != 'en' and\
searx/engines/gigablast.py+3−6
@@ -49,12 +49,9 @@ supported_languages_url = 'https://gigablast.com/search?&rxikd=1'
4949def request(query, params):
5050 offset = (params['pageno'] - 1) * number_of_results
5151
52− if params['language'] == 'all':
53− language = 'xx'
54− else:
55− language = params['language'].replace('-', '_').lower()
56− if language.split('-')[0] != 'zh':
57− language = language.split('-')[0]
52+ language = params['language'].replace('-', '_').lower()
53+ if language.split('-')[0] != 'zh':
54+ language = language.split('-')[0]
5855
5956 if params['safesearch'] >= 1:
6057 safesearch = 1
searx/engines/google.py+2−1
@@ -165,7 +165,8 @@ def extract_text_from_dom(result, xpath):
165165def request(query, params):
166166 offset = (params['pageno'] - 1) * 10
167167
168− if params['language'] == 'all':
168+ # temporary fix until a way of supporting en-US is found
169+ if params['language'] == 'en-US':
169170 params['language'] = 'en-GB'
170171
171172 if params['language'][:2] == 'jv':
searx/engines/google_news.py+2−3
@@ -50,9 +50,8 @@ def request(query, params):
5050 params['url'] = search_url.format(query=urlencode({'q': query}),
5151 search_options=urlencode(search_options))
5252
53− if params['language'] != 'all':
54− language_array = params['language'].lower().split('-')
55− params['url'] += '&lr=lang_' + language_array[0]
53+ language_array = params['language'].lower().split('-')
54+ params['url'] += '&lr=lang_' + language_array[0]
5655
5756 return params
5857
searx/engines/mediawiki.py+1−4
@@ -45,10 +45,7 @@ def request(query, params):
4545
4646 format_strings = list(Formatter().parse(base_url))
4747
48− if params['language'] == 'all':
49− language = 'en'
50− else:
51− language = params['language'].split('-')[0]
48+ language = params['language'].split('-')[0]
5249
5350 # format_string [('https://', 'language', '', None), ('.wikipedia.org/', None, None, None)]
5451 if any(x[1] == 'language' for x in format_strings):
searx/engines/photon.py+3−4
@@ -35,10 +35,9 @@ def request(query, params):
3535 search_string.format(query=urlencode({'q': query}),
3636 limit=number_of_results)
3737
38− if params['language'] != 'all':
39− language = params['language'].split('_')[0]
40− if language in supported_languages:
41− params['url'] = params['url'] + "&lang=" + language
38+ language = params['language'].split('-')[0]
39+ if language in supported_languages:
40+ params['url'] = params['url'] + "&lang=" + language
4241
4342 # using searx User-Agent
4443 params['headers']['User-Agent'] = searx_useragent()
searx/engines/qwant.py+11−12
@@ -44,18 +44,17 @@ def request(query, params):
4444 query=urlencode({'q': query}),
4545 offset=offset)
4646
47− # add language tag if specified
48− if params['language'] != 'all':
49− if params['language'] == 'no' or params['language'].startswith('no-'):
50− params['language'] = params['language'].replace('no', 'nb', 1)
51− if params['language'].find('-') < 0:
52− # tries to get a country code from language
53− for lang in supported_languages:
54− lc = lang.split('-')
55− if params['language'] == lc[0]:
56− params['language'] = lang
57− break
58− params['url'] += '&locale=' + params['language'].replace('-', '_').lower()
47+ # add language tag
48+ if params['language'] == 'no' or params['language'].startswith('no-'):
49+ params['language'] = params['language'].replace('no', 'nb', 1)
50+ if params['language'].find('-') < 0:
51+ # tries to get a country code from language
52+ for lang in supported_languages:
53+ lc = lang.split('-')
54+ if params['language'] == lc[0]:
55+ params['language'] = lang
56+ break
57+ params['url'] += '&locale=' + params['language'].replace('-', '_').lower()
5958
6059 return params
6160
searx/engines/startpage.py+2−3
@@ -45,9 +45,8 @@ def request(query, params):
4545 params['data'] = {'query': query,
4646 'startat': offset}
4747
48− # set language if specified
49− if params['language'] != 'all':
50− params['data']['with_language'] = ('lang_' + params['language'].split('-')[0])
48+ # set language
49+ params['data']['with_language'] = ('lang_' + params['language'].split('-')[0])
5150
5251 return params
5352
searx/engines/subtitleseeker.py+1−1
@@ -48,7 +48,7 @@ def response(resp):
4848 search_lang = 'Farsi'
4949 elif resp.search_params['language'] == 'pt-BR':
5050 search_lang = 'Brazilian'
51− elif resp.search_params['language'] != 'all':
51+ else:
5252 search_lang = [lc[3]
5353 for lc in language_codes
5454 if lc[0].split('-')[0] == resp.search_params['language'].split('-')[0]]
searx/engines/swisscows.py+1−4
@@ -35,10 +35,7 @@ regex_img_url_remove_start = re.compile(b'^https?://i\.swisscows\.ch/\?link=')
3535
3636# do search-request
3737def request(query, params):
38− if params['language'] == 'all':
39− ui_language = 'browser'
40− region = 'browser'
41− elif params['language'].split('-')[0] == 'no':
38+ if params['language'].split('-')[0] == 'no':
4239 region = 'nb-NO'
4340 else:
4441 region = params['language']
searx/engines/twitter.py+1−6
@@ -37,12 +37,7 @@ timestamp_xpath = './/span[contains(@class,"_timestamp")]'
3737# do search-request
3838def request(query, params):
3939 params['url'] = search_url + urlencode({'q': query})
40−
41− # set language if specified
42− if params['language'] != 'all':
43− params['cookies']['lang'] = params['language'].split('-')[0]
44− else:
45− params['cookies']['lang'] = 'en'
40+ params['cookies']['lang'] = params['language'].split('-')[0]
4641
4742 return params
4843
searx/engines/wikidata.py−4
@@ -57,8 +57,6 @@ calendar_name_xpath = './/sup[contains(@class,"wb-calendar-name")]'
5757
5858def request(query, params):
5959 language = params['language'].split('-')[0]
60− if language == 'all':
61− language = 'en'
6260
6361 params['url'] = url_search.format(
6462 query=urlencode({'label': query, 'language': language}))
@@ -71,8 +69,6 @@ def response(resp):
7169 wikidata_ids = html.xpath(wikidata_ids_xpath)
7270
7371 language = resp.search_params['language'].split('-')[0]
74− if language == 'all':
75− language = 'en'
7672
7773 # TODO: make requests asynchronous to avoid timeout when result_count > 1
7874 for wikidata_id in wikidata_ids[:result_count]:
searx/engines/wikipedia.py+1−1
@@ -31,7 +31,7 @@ supported_languages_url = 'https://meta.wikimedia.org/wiki/List_of_Wikipedias'
3131# set language in base_url
3232def url_lang(lang):
3333 lang = lang.split('-')[0]
34− if lang == 'all' or lang not in supported_languages:
34+ if lang not in supported_languages:
3535 language = 'en'
3636 else:
3737 language = lang
searx/engines/yacy.py+1−3
@@ -51,9 +51,7 @@ def request(query, params):
5151 limit=number_of_results,
5252 search_type=search_type)
5353
54− # add language tag if specified
55− if params['language'] != 'all':
56− params['url'] += '&lr=lang_' + params['language'].split('-')[0]
54+ params['url'] += '&lr=lang_' + params['language'].split('-')[0]
5755
5856 return params
5957
searx/engines/yahoo.py+1−3
@@ -71,9 +71,7 @@ def _get_url(query, offset, language, time_range):
7171
7272
7373def _get_language(params):
74− if params['language'] == 'all':
75− return 'en'
76− elif params['language'][:2] == 'zh':
74+ if params['language'][:2] == 'zh':
7775 if params['language'] == 'zh' or params['language'] == 'zh-CH':
7876 return 'szh'
7977 else:
searx/engines/yahoo_news.py+1−4
@@ -38,10 +38,7 @@ suggestion_xpath = '//div[contains(@class,"VerALSOTRY")]//a'
3838def request(query, params):
3939 offset = (params['pageno'] - 1) * 10 + 1
4040
41− if params['language'] == 'all':
42− language = 'en'
43− else:
44− language = params['language'].split('_')[0]
41+ language = params['language'].split('-')[0]
4542
4643 params['url'] = search_url.format(offset=offset,
4744 query=urlencode({'p': query}),
searx/engines/youtube_api.py+1−3
@@ -34,9 +34,7 @@ def request(query, params):
3434 params['url'] = search_url.format(query=urlencode({'q': query}),
3535 api_key=api_key)
3636
37− # add language tag if specified
38− if params['language'] != 'all':
39− params['url'] += '&relevanceLanguage=' + params['language'].split('-')[0]
37+ params['url'] += '&relevanceLanguage=' + params['language'].split('-')[0]
4038
4139 return params
4240
searx/preferences.py−1
@@ -12,7 +12,6 @@ if version[0] == '3':
1212
1313COOKIE_MAX_AGE = 60 * 60 * 24 * 365 * 5 # 5 years
1414LANGUAGE_CODES = [l[0] for l in languages]
15−LANGUAGE_CODES.append('all')
1615DISABLED = 0
1716ENABLED = 1
1817DOI_RESOLVERS = list(settings['doi_resolvers'])
searx/query.py+19−15
@@ -73,11 +73,6 @@ class RawTextQuery(object):
7373 if query_part[0] == ':':
7474 lang = query_part[1:].lower().replace('_', '-')
7575
76− # user may set a valid, yet not selectable language
77− if VALID_LANGUAGE_CODE.match(lang):
78− self.languages.append(lang)
79− parse_next = True
80−
8176 # check if any language-code is equal with
8277 # declared language-codes
8378 for lc in language_codes:
@@ -85,16 +80,25 @@ class RawTextQuery(object):
8580
8681 # if correct language-code is found
8782 # set it as new search-language
88− if lang == lang_id\
89− or lang_id.startswith(lang)\
90− or lang == lang_name\
91− or lang == english_name\
92− or lang.replace('-', ' ') == country:
93− parse_next = True
94− self.languages.append(lang_id)
95− # to ensure best match (first match is not necessarily the best one)
96− if lang == lang_id:
97− break
83+ if (lang == lang_id
84+ or lang == lang_name
85+ or lang == english_name
86+ or lang.replace('-', ' ') == country)\
87+ and lang not in self.languages:
88+ parse_next = True
89+ lang_parts = lang_id.split('-')
90+ if len(lang_parts) == 2:
91+ self.languages.append(lang_parts[0] + '-' + lang_parts[1].upper())
92+ else:
93+ self.languages.append(lang_id)
94+ # to ensure best match (first match is not necessarily the best one)
95+ if lang == lang_id:
96+ break
97+
98+ # user may set a valid, yet not selectable language
99+ if not self.languages and VALID_LANGUAGE_CODE.match(lang):
100+ self.languages.append(lang)
101+ parse_next = True
98102
99103 # this force a engine or category
100104 if query_part[0] == '!' or query_part[0] == '?':
searx/search.py+5−1
@@ -24,7 +24,7 @@ from flask_babel import gettext
2424import requests.exceptions
2525import searx.poolrequests as requests_lib
2626from searx.engines import (
27− categories, engines
27+ categories, engines, settings
2828)
2929from searx.answerers import ask
3030from searx.utils import gen_useragent
@@ -220,6 +220,10 @@ def get_search_query_from_webapp(preferences, form):
220220 else:
221221 query_lang = preferences.get_value('language')
222222
223+ # provides backwards compatibility for requests using old language default
224+ if query_lang == 'all':
225+ query_lang = settings['search']['language']
226+
223227 # check language
224228 if not VALID_LANGUAGE_CODE.match(query_lang):
225229 raise SearxParameterException('language', query_lang)
searx/settings.yml+1−1
@@ -5,7 +5,7 @@ general:
55search:
66 safe_search : 0 # Filter results. 0: None, 1: Moderate, 2: Strict
77 autocomplete : "" # Existing autocomplete backends: "dbpedia", "duckduckgo", "google", "startpage", "wikipedia" - leave blank to turn it off by default
8− language : "all"
8+ language : "en-US"
99
1010server:
1111 port : 8888
searx/settings_robot.yml+1−1
@@ -5,7 +5,7 @@ general:
55search:
66 safe_search : 0
77 autocomplete : ""
8− language: "all"
8+ language: "en-US"
99
1010server:
1111 port : 11111
searx/templates/courgette/preferences.html−1
@@ -13,7 +13,6 @@
1313 <legend>{{ _('Search language') }}</legend>
1414 <p>
1515 <select name='language'>
16− <option value="all" {% if current_language == 'all' %}selected="selected"{% endif %}>{{ _('Default language') }}</option>
1716 {% for lang_id,lang_name,country_name,english_name in language_codes | sort(attribute=1) %}
1817 <option value="{{ lang_id }}" {% if lang_id == current_language %}selected="selected"{% endif %}>{{ lang_name }} {% if country_name %}({{ country_name }}) {% endif %}- {{ lang_id }}</option>
1918 {% endfor %}
searx/templates/legacy/preferences.html−1
@@ -14,7 +14,6 @@
1414 <legend>{{ _('Search language') }}</legend>
1515 <p>
1616 <select name='language'>
17− <option value="all" {% if current_language == 'all' %}selected="selected"{% endif %}>{{ _('Default language') }}</option>
1817 {% for lang_id,lang_name,country_name,english_name in language_codes | sort(attribute=1) %}
1918 <option value="{{ lang_id }}" {% if lang_id == current_language %}selected="selected"{% endif %}>{{ lang_name }} {% if country_name %}({{ country_name }}) {% endif %}- {{ lang_id }}</option>
2019 {% endfor %}
searx/templates/oscar/languages.html+5−6
@@ -3,10 +3,9 @@
33{% else %}
44<select class="time_range custom-select form-control" id='language' name='language'>
55{% endif %}
6− <option value="all" {% if current_language == 'all' %}selected="selected"{% endif %}>{{ _('Default language') }}</option>
7− {% for lang_id,lang_name,country_name,english_name in language_codes | sort(attribute=1) %}
8− <option value="{{ lang_id }}" {% if lang_id == current_language %}selected="selected"{% endif %}>
9− {{ lang_name }} {% if country_name %}({{ country_name }}) {% endif %}- {{ lang_id }}
10− </option>
11− {% endfor %}
6+ {% for lang_id,lang_name,country_name,english_name in language_codes | sort(attribute=1) %}
7+ <option value="{{ lang_id }}" {% if lang_id == current_language %}selected="selected"{% endif %}>
8+ {{ lang_name }} {% if country_name %}({{ country_name }}) {% endif %}- {{ lang_id }}
9+ </option>
10+ {% endfor %}
1211</select>
searx/templates/oscar/preferences.html+2−2
@@ -187,7 +187,7 @@
187187 </td>
188188 <th>{{ search_engine.name }}</th>
189189 <td class="name">{{ shortcuts[search_engine.name] }}</td>
190− <td>{{ support_toggle(current_language == 'all' or current_language in search_engine.supported_languages or current_language.split('-')[0] in search_engine.supported_languages) }}</td>
190+ <td>{{ support_toggle(current_language in search_engine.supported_languages or current_language.split('-')[0] in search_engine.supported_languages) }}</td>
191191 <td>{{ support_toggle(search_engine.safesearch==True) }}</td>
192192 <td>{{ support_toggle(search_engine.time_range_support==True) }}</td>
193193 <td class="{{ 'danger' if stats[search_engine.name]['warn_time'] else '' }}">{{ 'N/A' if stats[search_engine.name].time==None else stats[search_engine.name].time }}</td>
@@ -197,7 +197,7 @@
197197 <td class="{{ 'danger' if stats[search_engine.name]['warn_time'] else '' }}">{{ 'N/A' if stats[search_engine.name].time==None else stats[search_engine.name].time }}</td>
198198 <td>{{ support_toggle(search_engine.time_range_support==True) }}</td>
199199 <td>{{ support_toggle(search_engine.safesearch==True) }}</td>
200− <td>{{ support_toggle(current_language == 'all' or current_language in search_engine.supported_languages or current_language.split('-')[0] in search_engine.supported_languages) }}</td>
200+ <td>{{ support_toggle(current_language in search_engine.supported_languages or current_language.split('-')[0] in search_engine.supported_languages) }}</td>
201201 <td>{{ shortcuts[search_engine.name] }}</td>
202202 <th>{{ search_engine.name }}</th>
203203 <td class="onoff-checkbox">
searx/templates/pix-art/preferences.html−1
@@ -9,7 +9,6 @@
99 <legend>{{ _('Search language') }}</legend>
1010 <p>
1111 <select name='language'>
12− <option value="all" {% if current_language == 'all' %}selected="selected"{% endif %}>{{ _('Default language') }}</option>
1312 {% for lang_id,lang_name,country_name,english_name in language_codes | sort(attribute=1) %}
1413 <option value="{{ lang_id }}" {% if lang_id == current_language %}selected="selected"{% endif %}>{{ lang_name }} {% if country_name %}({{ country_name }}) {% endif %}- {{ lang_id }}</option>
1514 {% endfor %}
searx/webapp.py+2−2
@@ -630,8 +630,8 @@ def autocompleter():
630630 if len(raw_results) <= 3 and completer:
631631 # get language from cookie
632632 language = request.preferences.get_value('language')
633− if not language or language == 'all':
634− language = 'en'
633+ if not language:
634+ language = settings['search']['language']
635635 else:
636636 language = language.split('-')[0]
637637 # run autocompletion
tests/unit/engines/test_archlinux.py+1−1
@@ -4,7 +4,7 @@ from searx.engines import archlinux
44from searx.testing import SearxTestCase
55
66domains = {
7− 'all': 'https://wiki.archlinux.org',
7+ 'en': 'https://wiki.archlinux.org',
88 'de': 'https://wiki.archlinux.de',
99 'fr': 'https://wiki.archlinux.fr',
1010 'ja': 'https://wiki.archlinuxjp.org',
tests/unit/engines/test_bing.py+1−5
@@ -10,17 +10,13 @@ class TestBingEngine(SearxTestCase):
1010 query = u'test_query'
1111 dicto = defaultdict(dict)
1212 dicto['pageno'] = 0
13− dicto['language'] = 'fr_FR'
13+ dicto['language'] = 'fr-FR'
1414 params = bing.request(query.encode('utf-8'), dicto)
1515 self.assertTrue('url' in params)
1616 self.assertTrue(query in params['url'])
1717 self.assertTrue('language%3AFR' in params['url'])
1818 self.assertTrue('bing.com' in params['url'])
1919
20− dicto['language'] = 'all'
21− params = bing.request(query.encode('utf-8'), dicto)
22− self.assertTrue('language' in params['url'])
23−
2420 def test_response(self):
2521 self.assertRaises(AttributeError, bing.response, None)
2622 self.assertRaises(AttributeError, bing.response, [])
tests/unit/engines/test_bing_news.py−4
@@ -19,10 +19,6 @@ class TestBingNewsEngine(SearxTestCase):
1919 self.assertIn('bing.com', params['url'])
2020 self.assertIn('fr', params['url'])
2121
22− dicto['language'] = 'all'
23− params = bing_news.request(query, dicto)
24− self.assertIn('en', params['url'])
25−
2622 def test_no_url_in_request_year_time_range(self):
2723 dicto = defaultdict(dict)
2824 query = 'test_query'
tests/unit/engines/test_dailymotion.py−4
@@ -18,10 +18,6 @@ class TestDailymotionEngine(SearxTestCase):
1818 self.assertTrue('dailymotion.com' in params['url'])
1919 self.assertTrue('fr' in params['url'])
2020
21− dicto['language'] = 'all'
22− params = dailymotion.request(query, dicto)
23− self.assertTrue('en' in params['url'])
24−
2521 def test_response(self):
2622 self.assertRaises(AttributeError, dailymotion.response, None)
2723 self.assertRaises(AttributeError, dailymotion.response, [])
tests/unit/engines/test_duckduckgo_images.py+1−1
@@ -15,7 +15,7 @@ class TestDuckduckgoImagesEngine(SearxTestCase):
1515 dicto['is_test'] = True
1616 dicto['pageno'] = 1
1717 dicto['safesearch'] = 0
18− dicto['language'] = 'all'
18+ dicto['language'] = 'en-US'
1919 params = duckduckgo_images.request(query, dicto)
2020 self.assertIn('url', params)
2121 self.assertIn(query, params['url'])
tests/unit/engines/test_faroo.py+2−6
@@ -11,7 +11,7 @@ class TestFarooEngine(SearxTestCase):
1111 query = 'test_query'
1212 dicto = defaultdict(dict)
1313 dicto['pageno'] = 1
14− dicto['language'] = 'fr_FR'
14+ dicto['language'] = 'fr-FR'
1515 dicto['category'] = 'general'
1616 params = faroo.request(query, dicto)
1717 self.assertIn('url', params)
@@ -20,11 +20,7 @@ class TestFarooEngine(SearxTestCase):
2020 self.assertIn('en', params['url'])
2121 self.assertIn('web', params['url'])
2222
23− dicto['language'] = 'all'
24− params = faroo.request(query, dicto)
25− self.assertIn('en', params['url'])
26−
27− dicto['language'] = 'de_DE'
23+ dicto['language'] = 'de-DE'
2824 params = faroo.request(query, dicto)
2925 self.assertIn('de', params['url'])
3026
tests/unit/engines/test_gigablast.py+1−6
@@ -11,16 +11,11 @@ class TestGigablastEngine(SearxTestCase):
1111 dicto = defaultdict(dict)
1212 dicto['pageno'] = 0
1313 dicto['safesearch'] = 0
14− dicto['language'] = 'all'
14+ dicto['language'] = 'en-US'
1515 params = gigablast.request(query, dicto)
1616 self.assertTrue('url' in params)
1717 self.assertTrue(query in params['url'])
1818 self.assertTrue('gigablast.com' in params['url'])
19− self.assertTrue('xx' in params['url'])
20−
21− dicto['language'] = 'en-US'
22− params = gigablast.request(query, dicto)
23− self.assertTrue('en' in params['url'])
2419 self.assertFalse('en-US' in params['url'])
2520
2621 def test_response(self):
tests/unit/engines/test_google.py+1−1
@@ -26,7 +26,7 @@ class TestGoogleEngine(SearxTestCase):
2626 self.assertIn('google.fr', params['url'])
2727 self.assertIn('fr', params['headers']['Accept-Language'])
2828
29− dicto['language'] = 'all'
29+ dicto['language'] = 'en-US'
3030 params = google.request(query, dicto)
3131 self.assertIn('google.co', params['url'])
3232 self.assertIn('en', params['headers']['Accept-Language'])
tests/unit/engines/test_google_news.py+1−6
@@ -12,18 +12,13 @@ class TestGoogleNewsEngine(SearxTestCase):
1212 query = 'test_query'
1313 dicto = defaultdict(dict)
1414 dicto['pageno'] = 1
15− dicto['language'] = 'fr_FR'
15+ dicto['language'] = 'fr-FR'
1616 dicto['time_range'] = 'w'
1717 params = google_news.request(query, dicto)
1818 self.assertIn('url', params)
1919 self.assertIn(query, params['url'])
2020 self.assertIn('fr', params['url'])
2121
22− dicto['language'] = 'all'
23− params = google_news.request(query, dicto)
24− self.assertIn('url', params)
25− self.assertNotIn('fr', params['url'])
26−
2722 def test_response(self):
2823 self.assertRaises(AttributeError, google_news.response, None)
2924 self.assertRaises(AttributeError, google_news.response, [])
tests/unit/engines/test_mediawiki.py−4
@@ -18,10 +18,6 @@ class TestMediawikiEngine(SearxTestCase):
1818 self.assertIn('wikipedia.org', params['url'])
1919 self.assertIn('fr', params['url'])
2020
21− dicto['language'] = 'all'
22− params = mediawiki.request(query, dicto)
23− self.assertIn('en', params['url'])
24−
2521 mediawiki.base_url = "http://test.url/"
2622 mediawiki.search_url = mediawiki.base_url +\
2723 'w/api.php?action=query'\
tests/unit/engines/test_qwant.py+2−1
@@ -19,10 +19,11 @@ class TestQwantEngine(SearxTestCase):
1919 self.assertIn('qwant.com', params['url'])
2020 self.assertIn('fr_fr', params['url'])
2121
22− dicto['language'] = 'all'
22+ dicto['language'] = 'en-US'
2323 qwant.categories = ['news']
2424 params = qwant.request(query, dicto)
2525 self.assertFalse('fr' in params['url'])
26+ self.assertIn('en_us', params['url'])
2627 self.assertIn('news', params['url'])
2728
2829 qwant.supported_languages = ['en', 'fr-FR', 'fr-CA']
tests/unit/engines/test_startpage.py−4
@@ -21,10 +21,6 @@ class TestStartpageEngine(SearxTestCase):
2121 self.assertIn('with_language', params['data'])
2222 self.assertIn('lang_fr', params['data']['with_language'])
2323
24− dicto['language'] = 'all'
25− params = startpage.request(query, dicto)
26− self.assertNotIn('with_language', params['data'])
27−
2824 def test_response(self):
2925 self.assertRaises(AttributeError, startpage.response, None)
3026 self.assertRaises(AttributeError, startpage.response, [])
tests/unit/engines/test_subtitleseeker.py−36
@@ -73,42 +73,6 @@ class TestSubtitleseekerEngine(SearxTestCase):
7373 results = subtitleseeker.response(response)
7474 self.assertEqual(results[0]['url'], 'http://this.is.the.url/Brazilian/')
7575
76− html = """
77− <div class="boxRows">
78− <div class="boxRowsInner" style="width:600px;">
79− <img src="http://static.subtitleseeker.com/images/movie.gif"
80− style="width:16px; height:16px;" class="icon">
81− <a href="http://this.is.the.url/"
82− class="blue" title="Title subtitle" >
83− This is the Title
84− </a>
85− </div>
86− <div class="boxRowsInner f12b red" style="width:70px;">
87− 1998
88− </div>
89− <div class="boxRowsInner grey-web f12" style="width:120px;">
90− <img src="http://static.subtitleseeker.com/images/basket_put.png"
91− style="width:16px; height:16px;" class="icon">
92− 1039 Subs
93− </div>
94− <div class="boxRowsInner grey-web f10" style="width:130px;">
95− <img src="http://static.subtitleseeker.com/images/arrow_refresh_small.png"
96− style="width:16px; height:16px;" class="icon">
97− 1 hours ago
98− </div>
99− <div class="clear"></div>
100− </div>
101− """
102− dicto['language'] = 'all'
103− response = mock.Mock(text=html, search_params=dicto)
104− results = subtitleseeker.response(response)
105− self.assertEqual(type(results), list)
106− self.assertEqual(len(results), 1)
107− self.assertEqual(results[0]['title'], 'This is the Title')
108− self.assertEqual(results[0]['url'], 'http://this.is.the.url/')
109− self.assertIn('1998', results[0]['content'])
110− self.assertIn('1039 Subs', results[0]['content'])
111−
11276 html = """
11377 <div class="boxRows">
11478 <div class="boxRowsInner" style="width:600px;">
tests/unit/engines/test_swisscows.py−5
@@ -18,11 +18,6 @@ class TestSwisscowsEngine(SearxTestCase):
1818 self.assertTrue('uiLanguage=de' in params['url'])
1919 self.assertTrue('region=de-DE' in params['url'])
2020
21− dicto['language'] = 'all'
22− params = swisscows.request(query, dicto)
23− self.assertTrue('uiLanguage=browser' in params['url'])
24− self.assertTrue('region=browser' in params['url'])
25−
2621 dicto['category'] = 'images'
2722 params = swisscows.request(query, dicto)
2823 self.assertIn('image', params['url'])
tests/unit/engines/test_twitter.py−6
@@ -20,12 +20,6 @@ class TestTwitterEngine(SearxTestCase):
2020 self.assertIn('lang', params['cookies'])
2121 self.assertIn('fr', params['cookies']['lang'])
2222
23− dicto['language'] = 'all'
24− params = twitter.request(query, dicto)
25− self.assertIn('cookies', params)
26− self.assertIn('lang', params['cookies'])
27− self.assertIn('en', params['cookies']['lang'])
28−
2923 def test_response(self):
3024 self.assertRaises(AttributeError, twitter.response, None)
3125 self.assertRaises(AttributeError, twitter.response, [])
tests/unit/engines/test_wikidata.py+3−3
@@ -11,14 +11,14 @@ class TestWikidataEngine(SearxTestCase):
1111 def test_request(self):
1212 query = 'test_query'
1313 dicto = defaultdict(dict)
14− dicto['language'] = 'all'
14+ dicto['language'] = 'en-US'
1515 params = wikidata.request(query, dicto)
1616 self.assertIn('url', params)
1717 self.assertIn(query, params['url'])
1818 self.assertIn('wikidata.org', params['url'])
1919 self.assertIn('en', params['url'])
2020
21− dicto['language'] = 'es_ES'
21+ dicto['language'] = 'es-ES'
2222 params = wikidata.request(query, dicto)
2323 self.assertIn(query, params['url'])
2424 self.assertIn('es', params['url'])
@@ -30,7 +30,7 @@ class TestWikidataEngine(SearxTestCase):
3030 self.assertRaises(AttributeError, wikidata.response, '')
3131 self.assertRaises(AttributeError, wikidata.response, '[]')
3232
33− response = mock.Mock(text='<html></html>', search_params={"language": "all"})
33+ response = mock.Mock(text='<html></html>', search_params={"language": "en"})
3434 self.assertEqual(wikidata.response(response), [])
3535
3636 def test_getDetail(self):
tests/unit/engines/test_wikipedia.py−4
@@ -25,10 +25,6 @@ class TestWikipediaEngine(SearxTestCase):
2525 self.assertIn('Test_Query', params['url'])
2626 self.assertNotIn('test_query', params['url'])
2727
28− dicto['language'] = 'all'
29− params = wikipedia.request(query, dicto)
30− self.assertIn('en', params['url'])
31−
3228 dicto['language'] = 'xx'
3329 params = wikipedia.request(query, dicto)
3430 self.assertIn('en', params['url'])
tests/unit/engines/test_yacy.py−5
@@ -17,11 +17,6 @@ class TestYacyEngine(SearxTestCase):
1717 self.assertIn('localhost', params['url'])
1818 self.assertIn('fr', params['url'])
1919
20− dicto['language'] = 'all'
21− params = yacy.request(query, dicto)
22− self.assertIn('url', params)
23− self.assertNotIn('lr=lang_', params['url'])
24−
2520 def test_response(self):
2621 self.assertRaises(AttributeError, yacy.response, None)
2722 self.assertRaises(AttributeError, yacy.response, [])
tests/unit/engines/test_yahoo.py−7
@@ -39,13 +39,6 @@ class TestYahooEngine(SearxTestCase):
3939 self.assertIn('sB', params['cookies'])
4040 self.assertIn('fr', params['cookies']['sB'])
4141
42− dicto['language'] = 'all'
43− params = yahoo.request(query, dicto)
44− self.assertIn('cookies', params)
45− self.assertIn('sB', params['cookies'])
46− self.assertIn('en', params['cookies']['sB'])
47− self.assertIn('en', params['url'])
48−
4942 def test_no_url_in_request_year_time_range(self):
5043 dicto = defaultdict(dict)
5144 query = 'test_query'
tests/unit/engines/test_yahoo_news.py−7
@@ -22,13 +22,6 @@ class TestYahooNewsEngine(SearxTestCase):
2222 self.assertIn('sB', params['cookies'])
2323 self.assertIn('fr', params['cookies']['sB'])
2424
25− dicto['language'] = 'all'
26− params = yahoo_news.request(query, dicto)
27− self.assertIn('cookies', params)
28− self.assertIn('sB', params['cookies'])
29− self.assertIn('en', params['cookies']['sB'])
30− self.assertIn('en', params['url'])
31−
3225 def test_sanitize_url(self):
3326 url = "test.url"
3427 self.assertEqual(url, yahoo_news.sanitize_url(url))
tests/unit/test_preferences.py+10−5
@@ -90,25 +90,30 @@ class TestSettings(SearxTestCase):
9090
9191 # search language settings
9292 def test_lang_setting_valid_choice(self):
93− setting = SearchLanguageSetting('all', choices=['all', 'de', 'en'])
93+ setting = SearchLanguageSetting('en', choices=['de', 'en'])
9494 setting.parse('de')
9595 self.assertEquals(setting.get_value(), 'de')
9696
9797 def test_lang_setting_invalid_choice(self):
98− setting = SearchLanguageSetting('all', choices=['all', 'de', 'en'])
98+ setting = SearchLanguageSetting('en', choices=['de', 'en'])
9999 setting.parse('xx')
100− self.assertEquals(setting.get_value(), 'all')
100+ self.assertEquals(setting.get_value(), 'en')
101101
102102 def test_lang_setting_old_cookie_choice(self):
103− setting = SearchLanguageSetting('all', choices=['all', 'es', 'es-ES'])
103+ setting = SearchLanguageSetting('en', choices=['en', 'es', 'es-ES'])
104104 setting.parse('es_XA')
105105 self.assertEquals(setting.get_value(), 'es')
106106
107107 def test_lang_setting_old_cookie_format(self):
108− setting = SearchLanguageSetting('all', choices=['all', 'es', 'es-ES'])
108+ setting = SearchLanguageSetting('en', choices=['en', 'es', 'es-ES'])
109109 setting.parse('es_ES')
110110 self.assertEquals(setting.get_value(), 'es-ES')
111111
112+ def test_lang_setting_old_default(self):
113+ setting = SearchLanguageSetting('en', choices=['en', 'es', 'de'])
114+ setting.parse('all')
115+ self.assertEquals(setting.get_value(), 'en')
116+
112117 # plugins settings
113118 def test_plugins_setting_all_default_enabled(self):
114119 plugin1 = PluginStub('plugin1', True)
tests/unit/test_query.pynew file+64
@@ -0,0 +1,64 @@
1+from searx.query import RawTextQuery
2+from searx.testing import SearxTestCase
3+
4+
5+class TestQuery(SearxTestCase):
6+
7+ def test_simple_query(self):
8+ query_text = 'the query'
9+ query = RawTextQuery(query_text, [])
10+ query.parse_query()
11+
12+ self.assertEquals(query.getFullQuery(), query_text)
13+ self.assertEquals(len(query.query_parts), 1)
14+ self.assertEquals(len(query.languages), 0)
15+ self.assertFalse(query.specific)
16+
17+ def test_language_code(self):
18+ language = 'es-ES'
19+ query_text = 'the query'
20+ full_query = ':' + language + ' ' + query_text
21+ query = RawTextQuery(full_query, [])
22+ query.parse_query()
23+
24+ self.assertEquals(query.getFullQuery(), full_query)
25+ self.assertEquals(len(query.query_parts), 3)
26+ self.assertEquals(len(query.languages), 1)
27+ self.assertIn(language, query.languages)
28+ self.assertFalse(query.specific)
29+
30+ def test_language_name(self):
31+ language = 'english'
32+ query_text = 'the query'
33+ full_query = ':' + language + ' ' + query_text
34+ query = RawTextQuery(full_query, [])
35+ query.parse_query()
36+
37+ self.assertEquals(query.getFullQuery(), full_query)
38+ self.assertEquals(len(query.query_parts), 3)
39+ self.assertIn('en', query.languages)
40+ self.assertFalse(query.specific)
41+
42+ def test_unlisted_language_code(self):
43+ language = 'all'
44+ query_text = 'the query'
45+ full_query = ':' + language + ' ' + query_text
46+ query = RawTextQuery(full_query, [])
47+ query.parse_query()
48+
49+ self.assertEquals(query.getFullQuery(), full_query)
50+ self.assertEquals(len(query.query_parts), 3)
51+ self.assertIn('all', query.languages)
52+ self.assertFalse(query.specific)
53+
54+ def test_invalid_language_code(self):
55+ language = 'not_a_language'
56+ query_text = 'the query'
57+ full_query = ':' + language + ' ' + query_text
58+ query = RawTextQuery(full_query, [])
59+ query.parse_query()
60+
61+ self.assertEquals(query.getFullQuery(), full_query)
62+ self.assertEquals(len(query.query_parts), 1)
63+ self.assertEquals(len(query.languages), 0)
64+ self.assertFalse(query.specific)