From b63d645a52d663036b97e31a59818d79faa4da81 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?No=C3=A9mi=20V=C3=A1nyi?= Date: Sun, 6 Jan 2019 15:27:46 +0100 Subject: [PATCH] Revert "remove 'all' option from search languages" This reverts commit 4d1770398a6af8902e75c0bd885781584d39e796. --- searx/engines/archlinux.py | 4 +-- searx/engines/bing.py | 5 ++- searx/engines/bing_news.py | 5 ++- searx/engines/dailymotion.py | 5 ++- searx/engines/duckduckgo.py | 12 +++++-- searx/engines/duckduckgo_images.py | 8 +++-- searx/engines/faroo.py | 5 ++- searx/engines/gigablast.py | 9 +++-- searx/engines/google.py | 6 +++- searx/engines/google_news.py | 7 ++-- searx/engines/mediawiki.py | 5 ++- searx/engines/photon.py | 7 ++-- searx/engines/qwant.py | 5 +-- searx/engines/startpage.py | 5 +-- searx/engines/subtitleseeker.py | 2 +- searx/engines/swisscows.py | 8 +++-- searx/engines/twitter.py | 7 +++- searx/engines/wikidata.py | 5 ++- searx/engines/wikipedia.py | 3 ++ searx/engines/yacy.py | 4 ++- searx/engines/yahoo.py | 17 ++++++--- searx/engines/yahoo_news.py | 5 ++- searx/engines/youtube_api.py | 4 ++- searx/preferences.py | 1 + searx/search.py | 6 +--- searx/settings_robot.yml | 2 +- searx/templates/courgette/preferences.html | 1 + searx/templates/legacy/preferences.html | 1 + searx/templates/oscar/languages.html | 11 +++--- searx/templates/pix-art/preferences.html | 1 + searx/webapp.py | 17 +++++---- tests/unit/engines/test_archlinux.py | 2 +- tests/unit/engines/test_bing.py | 4 +++ tests/unit/engines/test_bing_news.py | 4 +++ tests/unit/engines/test_dailymotion.py | 4 +++ tests/unit/engines/test_duckduckgo_images.py | 2 +- tests/unit/engines/test_faroo.py | 4 +++ tests/unit/engines/test_gigablast.py | 7 +++- tests/unit/engines/test_google_news.py | 5 +++ tests/unit/engines/test_mediawiki.py | 4 +++ tests/unit/engines/test_qwant.py | 3 +- 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 | 2 ++ tests/unit/engines/test_wikipedia.py | 6 ++++ 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 | 15 +++----- 51 files changed, 245 insertions(+), 70 deletions(-) diff --git a/searx/engines/archlinux.py b/searx/engines/archlinux.py index fc08112a..dce862f5 100644 --- a/searx/engines/archlinux.py +++ b/searx/engines/archlinux.py @@ -36,7 +36,7 @@ def locale_to_lang_code(locale): # wikis for some languages were moved off from the main site, we need to make # requests to correct URLs to be able to get results in those languages lang_urls = { - 'en': { + 'all': { 'base': 'https://wiki.archlinux.org', 'search': '/index.php?title=Special:Search&offset={offset}&{query}' }, @@ -67,7 +67,7 @@ lang_urls = { def get_lang_urls(language): if language in lang_urls: return lang_urls[language] - return lang_urls['en'] + return lang_urls['all'] # Language names to build search requests for diff --git a/searx/engines/bing.py b/searx/engines/bing.py index 2da40619..742379c1 100644 --- a/searx/engines/bing.py +++ b/searx/engines/bing.py @@ -34,7 +34,10 @@ search_string = 'search?{query}&first={offset}' def request(query, params): offset = (params['pageno'] - 1) * 10 + 1 - lang = match_language(params['language'], supported_languages, language_aliases) + if params['language'] == 'all': + lang = 'EN' + else: + lang = match_language(params['language'], supported_languages, language_aliases) query = u'language:{} {}'.format(lang.split('-')[0].upper(), query.decode('utf-8')).encode('utf-8') diff --git a/searx/engines/bing_news.py b/searx/engines/bing_news.py index 39048a1f..669130c4 100644 --- a/searx/engines/bing_news.py +++ b/searx/engines/bing_news.py @@ -71,7 +71,10 @@ def request(query, params): offset = (params['pageno'] - 1) * 10 + 1 - language = match_language(params['language'], supported_languages, language_aliases) + if params['language'] == 'all': + language = 'en-US' + else: + language = match_language(params['language'], supported_languages, language_aliases) params['url'] = _get_url(query, language, offset, params['time_range']) diff --git a/searx/engines/dailymotion.py b/searx/engines/dailymotion.py index 8268b625..06a9c41f 100644 --- a/searx/engines/dailymotion.py +++ b/searx/engines/dailymotion.py @@ -33,7 +33,10 @@ supported_languages_url = 'https://api.dailymotion.com/languages' # do search-request def request(query, params): - locale = match_language(params['language'], supported_languages) + if params['language'] == 'all': + locale = 'en-US' + else: + locale = match_language(params['language'], supported_languages) params['url'] = search_url.format( query=urlencode({'search': query, 'localization': locale}), diff --git a/searx/engines/duckduckgo.py b/searx/engines/duckduckgo.py index 2c5dc50d..fb8f523a 100644 --- a/searx/engines/duckduckgo.py +++ b/searx/engines/duckduckgo.py @@ -54,6 +54,9 @@ content_xpath = './/a[@class="result__snippet"]' # match query's language to a region code that duckduckgo will accept def get_region_code(lang, lang_list=[]): + if lang == 'all': + return None + lang_code = match_language(lang, lang_list, language_aliases, 'wt-WT') lang_parts = lang_code.split('-') @@ -61,7 +64,6 @@ def get_region_code(lang, lang_list=[]): return lang_parts[1].lower() + '-' + lang_parts[0].lower() -# do search-request def request(query, params): if params['time_range'] and params['time_range'] not in time_range_dict: return params @@ -69,8 +71,12 @@ def request(query, params): offset = (params['pageno'] - 1) * 30 region_code = get_region_code(params['language'], supported_languages) - params['url'] = url.format( - query=urlencode({'q': query, 'kl': region_code}), offset=offset, dc_param=offset) + if region_code: + params['url'] = url.format( + query=urlencode({'q': query, 'kl': region_code}), offset=offset, dc_param=offset) + else: + params['url'] = url.format( + query=urlencode({'q': query}), offset=offset, dc_param=offset) if params['time_range'] in time_range_dict: params['url'] += time_range_url.format(range=time_range_dict[params['time_range']]) diff --git a/searx/engines/duckduckgo_images.py b/searx/engines/duckduckgo_images.py index 7905d0bc..8c46ec92 100644 --- a/searx/engines/duckduckgo_images.py +++ b/searx/engines/duckduckgo_images.py @@ -56,8 +56,12 @@ def request(query, params): safesearch = params['safesearch'] - 1 region_code = get_region_code(params['language'], lang_list=supported_languages) - params['url'] = images_url.format( - query=urlencode({'q': query, 'l': region_code}), offset=offset, safesearch=safesearch, vqd=vqd) + if region_code: + params['url'] = images_url.format( + query=urlencode({'q': query, 'l': region_code}), offset=offset, safesearch=safesearch, vqd=vqd) + else: + params['url'] = images_url.format( + query=urlencode({'q': query}), offset=offset, safesearch=safesearch, vqd=vqd) return params diff --git a/searx/engines/faroo.py b/searx/engines/faroo.py index 4e8b5674..a36ecf77 100644 --- a/searx/engines/faroo.py +++ b/searx/engines/faroo.py @@ -40,7 +40,10 @@ def request(query, params): offset = (params['pageno'] - 1) * number_of_results + 1 categorie = search_category.get(params['category'], 'web') - language = params['language'].split('-')[0] + if params['language'] == 'all': + language = 'en' + else: + language = params['language'].split('-')[0] # if language is not supported, put it in english if language != 'en' and\ diff --git a/searx/engines/gigablast.py b/searx/engines/gigablast.py index 9b9b9bd7..a6aa5d71 100644 --- a/searx/engines/gigablast.py +++ b/searx/engines/gigablast.py @@ -50,9 +50,12 @@ supported_languages_url = 'https://gigablast.com/search?&rxikd=1' def request(query, params): offset = (params['pageno'] - 1) * number_of_results - language = params['language'].replace('-', '_').lower() - if language.split('-')[0] != 'zh': - language = language.split('-')[0] + if params['language'] == 'all': + language = 'xx' + else: + language = params['language'].replace('-', '_').lower() + if language.split('-')[0] != 'zh': + language = language.split('-')[0] if params['safesearch'] >= 1: safesearch = 1 diff --git a/searx/engines/google.py b/searx/engines/google.py index 7a4390fc..36ca7a11 100644 --- a/searx/engines/google.py +++ b/searx/engines/google.py @@ -166,7 +166,11 @@ def extract_text_from_dom(result, xpath): def request(query, params): offset = (params['pageno'] - 1) * 10 - language = match_language(params['language'], supported_languages, language_aliases) + if params['language'] == 'all' or params['language'] == 'en-US': + language = 'en-GB' + else: + language = match_language(params['language'], supported_languages, language_aliases) + language_array = language.split('-') if params['language'].find('-') > 0: country = params['language'].split('-')[1] diff --git a/searx/engines/google_news.py b/searx/engines/google_news.py index bf5995c8..9c837b45 100644 --- a/searx/engines/google_news.py +++ b/searx/engines/google_news.py @@ -51,9 +51,10 @@ def request(query, params): params['url'] = search_url.format(query=urlencode({'q': query}), search_options=urlencode(search_options)) - language = match_language(params['language'], supported_languages, language_aliases).split('-')[0] - if language: - params['url'] += '&lr=lang_' + language + if params['language'] != 'all': + language = match_language(params['language'], supported_languages, language_aliases).split('-')[0] + if language: + params['url'] += '&lr=lang_' + language return params diff --git a/searx/engines/mediawiki.py b/searx/engines/mediawiki.py index c7b05ffc..0607ac93 100644 --- a/searx/engines/mediawiki.py +++ b/searx/engines/mediawiki.py @@ -45,7 +45,10 @@ def request(query, params): format_strings = list(Formatter().parse(base_url)) - language = params['language'].split('-')[0] + if params['language'] == 'all': + language = 'en' + else: + language = params['language'].split('-')[0] # format_string [('https://', 'language', '', None), ('.wikipedia.org/', None, None, None)] if any(x[1] == 'language' for x in format_strings): diff --git a/searx/engines/photon.py b/searx/engines/photon.py index 24084195..15236f68 100644 --- a/searx/engines/photon.py +++ b/searx/engines/photon.py @@ -35,9 +35,10 @@ def request(query, params): search_string.format(query=urlencode({'q': query}), limit=number_of_results) - language = params['language'].split('-')[0] - if language in supported_languages: - params['url'] = params['url'] + "&lang=" + language + if params['language'] != 'all': + language = params['language'].split('_')[0] + if language in supported_languages: + params['url'] = params['url'] + "&lang=" + language # using searx User-Agent params['headers']['User-Agent'] = searx_useragent() diff --git a/searx/engines/qwant.py b/searx/engines/qwant.py index 1bef07cc..de12955c 100644 --- a/searx/engines/qwant.py +++ b/searx/engines/qwant.py @@ -46,8 +46,9 @@ def request(query, params): offset=offset) # add language tag - language = match_language(params['language'], supported_languages, language_aliases) - params['url'] += '&locale=' + language.replace('-', '_').lower() + if params['language'] != 'all': + language = match_language(params['language'], supported_languages, language_aliases) + params['url'] += '&locale=' + language.replace('-', '_').lower() return params diff --git a/searx/engines/startpage.py b/searx/engines/startpage.py index 55efdc88..6638f3d8 100644 --- a/searx/engines/startpage.py +++ b/searx/engines/startpage.py @@ -46,8 +46,9 @@ def request(query, params): params['data'] = {'query': query, 'startat': offset} - # set language - params['data']['with_language'] = ('lang_' + params['language'].split('-')[0]) + # set language if specified + if params['language'] != 'all': + params['data']['with_language'] = ('lang_' + params['language'].split('-')[0]) return params diff --git a/searx/engines/subtitleseeker.py b/searx/engines/subtitleseeker.py index 118504ff..2cbc991b 100644 --- a/searx/engines/subtitleseeker.py +++ b/searx/engines/subtitleseeker.py @@ -48,7 +48,7 @@ def response(resp): search_lang = 'Farsi' elif resp.search_params['language'] == 'pt-BR': search_lang = 'Brazilian' - else: + elif resp.search_params['language'] != 'all': search_lang = [lc[3] for lc in language_codes if lc[0].split('-')[0] == resp.search_params['language'].split('-')[0]] diff --git a/searx/engines/swisscows.py b/searx/engines/swisscows.py index 0001d56b..e451bcff 100644 --- a/searx/engines/swisscows.py +++ b/searx/engines/swisscows.py @@ -36,8 +36,12 @@ regex_img_url_remove_start = re.compile(b'^https?://i\.swisscows\.ch/\?link=') # do search-request def request(query, params): - region = match_language(params['language'], supported_languages, language_aliases) - ui_language = region.split('-')[0] + if params['language'] == 'all': + ui_language = 'browser' + region = 'browser' + else: + region = match_language(params['language'], supported_languages, language_aliases) + ui_language = region.split('-')[0] search_path = search_string.format( query=urlencode({'query': query, 'uiLanguage': ui_language, 'region': region}), diff --git a/searx/engines/twitter.py b/searx/engines/twitter.py index 8641167d..d2a8d208 100644 --- a/searx/engines/twitter.py +++ b/searx/engines/twitter.py @@ -37,7 +37,12 @@ timestamp_xpath = './/span[contains(@class,"_timestamp")]' # do search-request def request(query, params): params['url'] = search_url + urlencode({'q': query}) - params['cookies']['lang'] = params['language'].split('-')[0] + + # set language if specified + if params['language'] != 'all': + params['cookies']['lang'] = params['language'].split('-')[0] + else: + params['cookies']['lang'] = 'en' return params diff --git a/searx/engines/wikidata.py b/searx/engines/wikidata.py index 2485b652..03a58a31 100644 --- a/searx/engines/wikidata.py +++ b/searx/engines/wikidata.py @@ -68,7 +68,10 @@ def response(resp): html = fromstring(resp.text) search_results = html.xpath(wikidata_ids_xpath) - language = match_language(resp.search_params['language'], supported_languages, language_aliases).split('-')[0] + if resp.search_params['language'].split('-')[0] == 'all': + language = 'en' + else: + language = match_language(resp.search_params['language'], supported_languages, language_aliases).split('-')[0] # TODO: make requests asynchronous to avoid timeout when result_count > 1 for search_result in search_results[:result_count]: diff --git a/searx/engines/wikipedia.py b/searx/engines/wikipedia.py index 23f23db4..39aec1c8 100644 --- a/searx/engines/wikipedia.py +++ b/searx/engines/wikipedia.py @@ -31,6 +31,9 @@ supported_languages_url = 'https://meta.wikimedia.org/wiki/List_of_Wikipedias' # set language in base_url def url_lang(lang): + lang_pre = lang.split('-')[0] + if lang_pre == 'all' or lang_pre not in supported_languages: + return 'en' return match_language(lang, supported_languages, language_aliases).split('-')[0] diff --git a/searx/engines/yacy.py b/searx/engines/yacy.py index 7c307ce5..25bc8368 100644 --- a/searx/engines/yacy.py +++ b/searx/engines/yacy.py @@ -51,7 +51,9 @@ def request(query, params): limit=number_of_results, search_type=search_type) - params['url'] += '&lr=lang_' + params['language'].split('-')[0] + # add language tag if specified + if params['language'] != 'all': + params['url'] += '&lr=lang_' + params['language'].split('-')[0] return params diff --git a/searx/engines/yahoo.py b/searx/engines/yahoo.py index ba4cb6af..73b78bcf 100644 --- a/searx/engines/yahoo.py +++ b/searx/engines/yahoo.py @@ -73,16 +73,25 @@ def _get_url(query, offset, language, time_range): lang=language) +def _get_language(params): + if params['language'] == 'all': + return 'en' + + language = match_language(params['language'], supported_languages, language_aliases) + if language not in language_aliases.values(): + language = language.split('-')[0] + language = language.replace('-', '_').lower() + + return language + + # do search-request def request(query, params): if params['time_range'] and params['time_range'] not in time_range_dict: return params offset = (params['pageno'] - 1) * 10 + 1 - language = match_language(params['language'], supported_languages, language_aliases) - if language not in language_aliases.values(): - language = language.split('-')[0] - language = language.replace('-', '_').lower() + language = _get_language(params) params['url'] = _get_url(query, offset, language, params['time_range']) diff --git a/searx/engines/yahoo_news.py b/searx/engines/yahoo_news.py index b61384d0..9f6a4159 100644 --- a/searx/engines/yahoo_news.py +++ b/searx/engines/yahoo_news.py @@ -41,7 +41,10 @@ suggestion_xpath = '//div[contains(@class,"VerALSOTRY")]//a' def request(query, params): offset = (params['pageno'] - 1) * 10 + 1 - language = match_language(params['language'], supported_languages, language_aliases).split('-')[0] + if params['language'] == 'all': + language = 'en' + else: + language = match_language(params['language'], supported_languages, language_aliases).split('-')[0] params['url'] = search_url.format(offset=offset, query=urlencode({'p': query}), diff --git a/searx/engines/youtube_api.py b/searx/engines/youtube_api.py index f8bc353f..6de18aa2 100644 --- a/searx/engines/youtube_api.py +++ b/searx/engines/youtube_api.py @@ -34,7 +34,9 @@ def request(query, params): params['url'] = search_url.format(query=urlencode({'q': query}), api_key=api_key) - params['url'] += '&relevanceLanguage=' + params['language'].split('-')[0] + # add language tag if specified + if params['language'] != 'all': + params['url'] += '&relevanceLanguage=' + params['language'].split('-')[0] return params diff --git a/searx/preferences.py b/searx/preferences.py index ed2cc402..acaa67f6 100644 --- a/searx/preferences.py +++ b/searx/preferences.py @@ -12,6 +12,7 @@ if version[0] == '3': COOKIE_MAX_AGE = 60 * 60 * 24 * 365 * 5 # 5 years LANGUAGE_CODES = [l[0] for l in languages] +LANGUAGE_CODES.append('all') DISABLED = 0 ENABLED = 1 DOI_RESOLVERS = list(settings['doi_resolvers']) diff --git a/searx/search.py b/searx/search.py index 950a49f2..2b7151e6 100644 --- a/searx/search.py +++ b/searx/search.py @@ -24,7 +24,7 @@ from flask_babel import gettext import requests.exceptions import searx.poolrequests as requests_lib from searx.engines import ( - categories, engines, settings + categories, engines ) from searx.answerers import ask from searx.utils import gen_useragent @@ -221,10 +221,6 @@ def get_search_query_from_webapp(preferences, form): else: query_lang = preferences.get_value('language') - # provides backwards compatibility for requests using old language default - if query_lang == 'all': - query_lang = settings['search']['language'] - # check language if not VALID_LANGUAGE_CODE.match(query_lang): raise SearxParameterException('language', query_lang) diff --git a/searx/settings_robot.yml b/searx/settings_robot.yml index e5f163ae..070a0edb 100644 --- a/searx/settings_robot.yml +++ b/searx/settings_robot.yml @@ -5,7 +5,7 @@ general: search: safe_search : 0 autocomplete : "" - language: "en-US" + language: "all" server: port : 11111 diff --git a/searx/templates/courgette/preferences.html b/searx/templates/courgette/preferences.html index 61f52147..56a6e020 100644 --- a/searx/templates/courgette/preferences.html +++ b/searx/templates/courgette/preferences.html @@ -13,6 +13,7 @@ {{ _('Search language') }}

+ {% for lang_id,lang_name,country_name,english_name in language_codes | sort(attribute=1) %} {% endfor %} diff --git a/searx/templates/oscar/languages.html b/searx/templates/oscar/languages.html index 369abca8..53ade43b 100644 --- a/searx/templates/oscar/languages.html +++ b/searx/templates/oscar/languages.html @@ -3,9 +3,10 @@ {% else %} diff --git a/searx/templates/pix-art/preferences.html b/searx/templates/pix-art/preferences.html index 0b2bb670..05876ded 100644 --- a/searx/templates/pix-art/preferences.html +++ b/searx/templates/pix-art/preferences.html @@ -9,6 +9,7 @@ {{ _('Search language') }}