diff --git a/searx/enginelib/traits.py b/searx/enginelib/traits.py index df7851594..ae27d46f1 100644 --- a/searx/enginelib/traits.py +++ b/searx/enginelib/traits.py @@ -13,7 +13,7 @@ used. from __future__ import annotations import json import dataclasses -from typing import Dict, Union, Callable, Optional, TYPE_CHECKING +from typing import Dict, Iterable, Union, Callable, Optional, TYPE_CHECKING from typing_extensions import Literal, Self from searx import locales @@ -81,7 +81,7 @@ class EngineTraits: """Data type, default is 'traits_v1'. """ - custom: Dict[str, Dict] = dataclasses.field(default_factory=dict) + custom: Dict[str, Union[Dict[str, Dict], Iterable[str]]] = dataclasses.field(default_factory=dict) """A place to store engine's custom traits, not related to the SearXNG core """ diff --git a/searx/engines/wikidata.py b/searx/engines/wikidata.py index 6ea77f092..b9de67ef9 100644 --- a/searx/engines/wikidata.py +++ b/searx/engines/wikidata.py @@ -18,7 +18,10 @@ from searx.data import WIKIDATA_UNITS from searx.network import post, get from searx.utils import searx_useragent, get_string_replaces_function from searx.external_urls import get_external_url, get_earth_coordinates_url, area_to_osm_zoom -from searx.engines.wikipedia import fetch_traits as _fetch_traits +from searx.engines.wikipedia import ( + fetch_wikimedia_traits, + get_wiki_params, +) from searx.enginelib.traits import EngineTraits if TYPE_CHECKING: @@ -165,17 +168,15 @@ def request(query, params): # wikidata does not support zh-classical (zh_Hans) / zh-TW, zh-HK and zh-CN # mapped to zh - sxng_lang = params['searxng_locale'].split('-')[0] - language = traits.get_language(sxng_lang, 'en') - - query, attributes = get_query(query, language) - logger.debug("request --> language %s // len(attributes): %s", language, len(attributes)) + eng_tag, _wiki_netloc = get_wiki_params(params['searxng_locale'], traits) + query, attributes = get_query(query, eng_tag) + logger.debug("request --> language %s // len(attributes): %s", eng_tag, len(attributes)) params['method'] = 'POST' params['url'] = SPARQL_ENDPOINT_URL params['data'] = {'query': query} params['headers'] = get_headers() - params['language'] = language + params['language'] = eng_tag params['attributes'] = attributes return params @@ -769,12 +770,16 @@ def init(engine_settings=None): # pylint: disable=unused-argument def fetch_traits(engine_traits: EngineTraits): - """Use languages evaluated from :py:obj:`wikipedia.fetch_traits - ` except zh-classical (zh_Hans) what - is not supported by wikidata.""" + """Uses languages evaluated from :py:obj:`wikipedia.fetch_wikimedia_traits + ` and removes - _fetch_traits(engine_traits) - # wikidata does not support zh-classical (zh_Hans) - engine_traits.languages.pop('zh_Hans') - # wikidata does not have net-locations for the languages + - ``traits.custom['wiki_netloc']``: wikidata does not have net-locations for + the languages and the list of all + + - ``traits.custom['WIKIPEDIA_LANGUAGES']``: not used in the wikipedia engine + + """ + + fetch_wikimedia_traits(engine_traits) engine_traits.custom['wiki_netloc'] = {} + engine_traits.custom['WIKIPEDIA_LANGUAGES'] = [] diff --git a/searx/engines/wikipedia.py b/searx/engines/wikipedia.py index 9d2d30afa..98b3d6f9e 100644 --- a/searx/engines/wikipedia.py +++ b/searx/engines/wikipedia.py @@ -5,10 +5,54 @@ are shared by other engines: - :ref:`wikidata engine` -The list of supported languages is fetched from the article linked by -:py:obj:`wikipedia_article_depth`. Unlike traditional search engines, wikipedia -does not support one Wikipedia for all the languages, but there is one Wikipedia -for every language (:py:obj:`fetch_traits`). +The list of supported languages is :py:obj:`fetched ` from +the article linked by :py:obj:`list_of_wikipedias`. + +Unlike traditional search engines, wikipedia does not support one Wikipedia for +all languages, but there is one Wikipedia for each supported language. Some of +these Wikipedias have a LanguageConverter_ enabled +(:py:obj:`rest_v1_summary_url`). + +A LanguageConverter_ (LC) is a system based on language variants that +automatically converts the content of a page into a different variant. A variant +is mostly the same language in a different script. + +- `Wikipedias in multiple writing systems`_ +- `Automatic conversion between traditional and simplified Chinese characters`_ + +PR-2554_: + The Wikipedia link returned by the API is still the same in all cases + (`https://zh.wikipedia.org/wiki/出租車`_) but if your browser's + ``Accept-Language`` is set to any of ``zh``, ``zh-CN``, ``zh-TW``, ``zh-HK`` + or .. Wikipedia's LC automatically returns the desired script in their + web-page. + + - You can test the API here: https://reqbin.com/gesg2kvx + +.. _https://zh.wikipedia.org/wiki/出租車: + https://zh.wikipedia.org/wiki/%E5%87%BA%E7%A7%9F%E8%BB%8A + +To support Wikipedia's LanguageConverter_, a SearXNG request to Wikipedia uses +:py:obj:`get_wiki_params` and :py:obj:`wiki_lc_locale_variants' in the +:py:obj:`fetch_wikimedia_traits` function. + +To test in SearXNG, query for ``!wp 出租車`` with each of the available Chinese +options: + +- ``!wp 出租車 :zh`` should show 出租車 +- ``!wp 出租車 :zh-CN`` should show 出租车 +- ``!wp 出租車 :zh-TW`` should show 計程車 +- ``!wp 出租車 :zh-HK`` should show 的士 +- ``!wp 出租車 :zh-SG`` should show 德士 + +.. _LanguageConverter: + https://www.mediawiki.org/wiki/Writing_systems#LanguageConverter +.. _Wikipedias in multiple writing systems: + https://meta.wikimedia.org/wiki/Wikipedias_in_multiple_writing_systems +.. _Automatic conversion between traditional and simplified Chinese characters: + https://en.wikipedia.org/wiki/Chinese_Wikipedia#Automatic_conversion_between_traditional_and_simplified_Chinese_characters +.. _PR-2554: https://github.com/searx/searx/pull/2554 + """ import urllib.parse @@ -16,8 +60,9 @@ import babel from lxml import html +from searx import utils from searx import network -from searx.locales import language_tag +from searx import locales from searx.enginelib.traits import EngineTraits traits: EngineTraits @@ -33,6 +78,12 @@ about = { } send_accept_language_header = True +"""The HTTP ``Accept-Language`` header is needed for wikis where +LanguageConverter_ is enabled.""" + +list_of_wikipedias = 'https://meta.wikimedia.org/wiki/List_of_Wikipedias' +"""`List of all wikipedias `_ +""" wikipedia_article_depth = 'https://meta.wikimedia.org/wiki/Wikipedia_article_depth' """The *editing depth* of Wikipedia is one of several possible rough indicators @@ -41,29 +92,68 @@ are updated. The measurement of depth was introduced after some limitations of the classic measurement of article count were realized. """ -# example: https://zh-classical.wikipedia.org/api/rest_v1/page/summary/日 rest_v1_summary_url = 'https://{wiki_netloc}/api/rest_v1/page/summary/{title}' -"""`wikipedia rest_v1 summary API`_: The summary response includes an extract of -the first paragraph of the page in plain text and HTML as well as the type of -page. This is useful for page previews (fka. Hovercards, aka. Popups) on the web -and link previews in the apps. +""" +`wikipedia rest_v1 summary API`_: + The summary response includes an extract of the first paragraph of the page in + plain text and HTML as well as the type of page. This is useful for page + previews (fka. Hovercards, aka. Popups) on the web and link previews in the + apps. -.. _wikipedia rest_v1 summary API: https://en.wikipedia.org/api/rest_v1/#/Page%20content/get_page_summary__title_ +HTTP ``Accept-Language`` header (:py:obj:`send_accept_language_header`): + The desired language variant code for wikis where LanguageConverter_ is + enabled. + +.. _wikipedia rest_v1 summary API: + https://en.wikipedia.org/api/rest_v1/#/Page%20content/get_page_summary__title_ """ +wiki_lc_locale_variants = { + "zh": ( + "zh-CN", + "zh-HK", + "zh-MO", + "zh-MY", + "zh-SG", + "zh-TW", + ), + "zh-classical": ("zh-classical",), +} +"""Mapping rule of the LanguageConverter_ to map a language and its variants to +a Locale (used in the HTTP ``Accept-Language`` header). For example see `LC +Chinese`_. + +.. _LC Chinese: + https://meta.wikimedia.org/wiki/Wikipedias_in_multiple_writing_systems#Chinese +""" + +wikipedia_script_variants = { + "zh": ( + "zh_Hant", + "zh_Hans", + ) +} + + +def get_wiki_params(sxng_locale, eng_traits): + """Returns the Wikipedia language tag and the netloc that fits to the + ``sxng_locale``. To support LanguageConverter_ this function rates a locale + (region) higher than a language (compare :py:obj:`wiki_lc_locale_variants`). + + """ + eng_tag = eng_traits.get_region(sxng_locale, eng_traits.get_language(sxng_locale, 'en')) + wiki_netloc = eng_traits.custom['wiki_netloc'].get(eng_tag, 'en.wikipedia.org') + return eng_tag, wiki_netloc + def request(query, params): """Assemble a request (`wikipedia rest_v1 summary API`_).""" if query.islower(): query = query.title() - engine_language = traits.get_language(params['searxng_locale'], 'en') - wiki_netloc = traits.custom['wiki_netloc'].get(engine_language, 'https://en.wikipedia.org/wiki/') + _eng_tag, wiki_netloc = get_wiki_params(params['searxng_locale'], traits) title = urllib.parse.quote(query) - - # '!wikipedia 日 :zh-TW' --> https://zh-classical.wikipedia.org/ - # '!wikipedia 日 :zh' --> https://zh.wikipedia.org/ params['url'] = rest_v1_summary_url.format(wiki_netloc=wiki_netloc, title=title) params['raise_for_httperror'] = False @@ -93,7 +183,7 @@ def response(resp): network.raise_for_httperror(resp) api_result = resp.json() - title = api_result['title'] + title = utils.html_to_text(api_result.get('titles', {}).get('display') or api_result.get('title')) wikipedia_link = api_result['content_urls']['desktop']['page'] results.append({'url': wikipedia_link, 'title': title, 'content': api_result.get('description', '')}) @@ -116,44 +206,38 @@ def response(resp): # These Wikipedias use language codes that do not conform to the ISO 639 # standard (which is how wiki subdomains are chosen nowadays). -lang_map = { - 'be-tarask': 'bel', - 'ak': 'aka', - 'als': 'gsw', - 'bat-smg': 'sgs', - 'cbk-zam': 'cbk', - 'fiu-vro': 'vro', - 'map-bms': 'map', - 'nrm': 'nrf', - 'roa-rup': 'rup', - 'nds-nl': 'nds', - #'simple: – invented code used for the Simple English Wikipedia (not the official IETF code en-simple) - 'zh-min-nan': 'nan', - 'zh-yue': 'yue', - 'an': 'arg', - 'zh-classical': 'zh-Hant', # babel maps classical to zh-Hans (for whatever reason) -} - -unknown_langs = [ - 'an', # Aragonese - 'ba', # Bashkir - 'bar', # Bavarian - 'bcl', # Central Bicolano - 'be-tarask', # Belarusian variant / Belarusian is already covered by 'be' - 'bpy', # Bishnupriya Manipuri is unknown by babel - 'hif', # Fiji Hindi - 'ilo', # Ilokano - 'li', # Limburgish - 'sco', # Scots (sco) is not known by babel, Scottish Gaelic (gd) is known by babel - 'sh', # Serbo-Croatian - 'simple', # simple english is not know as a natural language different to english (babel) - 'vo', # Volapük - 'wa', # Walloon -] +lang_map = locales.LOCALE_BEST_MATCH.copy() +lang_map.update( + { + 'be-tarask': 'bel', + 'ak': 'aka', + 'als': 'gsw', + 'bat-smg': 'sgs', + 'cbk-zam': 'cbk', + 'fiu-vro': 'vro', + 'map-bms': 'map', + 'no': 'nb-NO', + 'nrm': 'nrf', + 'roa-rup': 'rup', + 'nds-nl': 'nds', + #'simple: – invented code used for the Simple English Wikipedia (not the official IETF code en-simple) + 'zh-min-nan': 'nan', + 'zh-yue': 'yue', + 'an': 'arg', + } +) def fetch_traits(engine_traits: EngineTraits): - """Fetch languages from Wikipedia. + fetch_wikimedia_traits(engine_traits) + print("WIKIPEDIA_LANGUAGES: %s" % len(engine_traits.custom['WIKIPEDIA_LANGUAGES'])) + + +def fetch_wikimedia_traits(engine_traits: EngineTraits): + """Fetch languages from Wikipedia. Not all languages from the + :py:obj:`list_of_wikipedias` are supported by SearXNG locales, only those + known from :py:obj:`searx.locales.LOCALE_NAMES` or those with a minimal + :py:obj:`editing depth `. The location of the Wikipedia address of a language is mapped in a :py:obj:`custom field ` @@ -169,15 +253,21 @@ def fetch_traits(engine_traits: EngineTraits): "zh": "zh.wikipedia.org", "zh-classical": "zh-classical.wikipedia.org" } - """ - + # pylint: disable=too-many-branches engine_traits.custom['wiki_netloc'] = {} + engine_traits.custom['WIKIPEDIA_LANGUAGES'] = [] - # insert alias to map from a region like zh-CN to a language zh_Hans - engine_traits.languages['zh_Hans'] = 'zh' + # insert alias to map from a script or region to a wikipedia variant - resp = network.get(wikipedia_article_depth) + for eng_tag, sxng_tag_list in wikipedia_script_variants.items(): + for sxng_tag in sxng_tag_list: + engine_traits.languages[sxng_tag] = eng_tag + for eng_tag, sxng_tag_list in wiki_lc_locale_variants.items(): + for sxng_tag in sxng_tag_list: + engine_traits.regions[sxng_tag] = eng_tag + + resp = network.get(list_of_wikipedias) if not resp.ok: print("ERROR: response from Wikipedia is not OK.") @@ -189,30 +279,31 @@ def fetch_traits(engine_traits: EngineTraits): continue cols = [c.text_content().strip() for c in cols] - depth = float(cols[3].replace('-', '0').replace(',', '')) + depth = float(cols[11].replace('-', '0').replace(',', '')) articles = int(cols[4].replace(',', '').replace(',', '')) - if articles < 10000: - # exclude languages with too few articles - continue - - if int(depth) < 20: - # Rough indicator of a Wikipedia’s quality, showing how frequently - # its articles are updated. - continue - - eng_tag = cols[2] - wiki_url = row.xpath('./td[3]/a/@href')[0] + eng_tag = cols[3] + wiki_url = row.xpath('./td[4]/a/@href')[0] wiki_url = urllib.parse.urlparse(wiki_url) - if eng_tag in unknown_langs: - continue - try: - sxng_tag = language_tag(babel.Locale.parse(lang_map.get(eng_tag, eng_tag), sep='-')) + sxng_tag = locales.language_tag(babel.Locale.parse(lang_map.get(eng_tag, eng_tag), sep='-')) except babel.UnknownLocaleError: - print("ERROR: %s [%s] is unknown by babel" % (cols[0], eng_tag)) + # print("ERROR: %s [%s] is unknown by babel" % (cols[0], eng_tag)) continue + finally: + engine_traits.custom['WIKIPEDIA_LANGUAGES'].append(eng_tag) + + if sxng_tag not in locales.LOCALE_NAMES: + + if articles < 10000: + # exclude languages with too few articles + continue + + if int(depth) < 20: + # Rough indicator of a Wikipedia’s quality, showing how + # frequently its articles are updated. + continue conflict = engine_traits.languages.get(sxng_tag) if conflict: @@ -222,3 +313,5 @@ def fetch_traits(engine_traits: EngineTraits): engine_traits.languages[sxng_tag] = eng_tag engine_traits.custom['wiki_netloc'][eng_tag] = wiki_url.netloc + + engine_traits.custom['WIKIPEDIA_LANGUAGES'].sort() diff --git a/searx/search/processors/online.py b/searx/search/processors/online.py index 697533d8c..7b2ec856b 100644 --- a/searx/search/processors/online.py +++ b/searx/search/processors/online.py @@ -75,6 +75,7 @@ class OnlineProcessor(EngineProcessor): ) params['headers']['Accept-Language'] = ac_lang + self.logger.debug('HTTP Accept-Language: %s', params['headers'].get('Accept-Language', '')) return params def _send_http_request(self, params): diff --git a/searxng_extra/update/update_engine_descriptions.py b/searxng_extra/update/update_engine_descriptions.py index 66bc303db..737b99775 100755 --- a/searxng_extra/update/update_engine_descriptions.py +++ b/searxng_extra/update/update_engine_descriptions.py @@ -70,7 +70,7 @@ SKIP_ENGINE_SOURCE = [ ] LANGUAGES = LOCALE_NAMES.keys() -WIKIPEDIA_LANGUAGES = {'language': 'wikipedia_language'} +WIKIPEDIA_LANGUAGES = {} # {'': ''} LANGUAGES_SPARQL = '' IDS = None @@ -103,7 +103,7 @@ def update_description(engine_name, lang, description, source, replace=True): def get_wikipedia_summary(lang, pageid): - params = {'language': lang.replace('_', '-'), 'headers': {}} + params = {'searxng_locale': lang.replace('_', '-'), 'headers': {}} searx.engines.engines['wikipedia'].request(pageid, params) try: response = searx.network.get(params['url'], headers=params['headers'], timeout=10) @@ -154,11 +154,25 @@ def get_website_description(url, lang1, lang2=None): def initialize(): - global IDS, WIKIPEDIA_LANGUAGES, LANGUAGES_SPARQL + global IDS, LANGUAGES_SPARQL searx.search.initialize() wikipedia_engine = searx.engines.engines['wikipedia'] - WIKIPEDIA_LANGUAGES = {language: wikipedia_engine.url_lang(language.replace('_', '-')) for language in LANGUAGES} - WIKIPEDIA_LANGUAGES['nb_NO'] = 'no' + + locale2lang = {'nl-BE': 'nl'} + for sxng_ui_lang in LANGUAGES: + + sxng_ui_alias = locale2lang.get(sxng_ui_lang, sxng_ui_lang) + wiki_lang = None + + if sxng_ui_alias in wikipedia_engine.traits.custom['WIKIPEDIA_LANGUAGES']: + wiki_lang = sxng_ui_alias + if not wiki_lang: + wiki_lang = wikipedia_engine.traits.get_language(sxng_ui_alias) + if not wiki_lang: + print(f"WIKIPEDIA_LANGUAGES missing {sxng_ui_lang}") + continue + WIKIPEDIA_LANGUAGES[sxng_ui_lang] = wiki_lang + LANGUAGES_SPARQL = ', '.join(f"'{l}'" for l in set(WIKIPEDIA_LANGUAGES.values())) for engine_name, engine in searx.engines.engines.items(): descriptions[engine_name] = {} @@ -170,6 +184,7 @@ def initialize(): def fetch_wikidata_descriptions(): + print('Fetching wikidata descriptions') searx.network.set_timeout_for_thread(60) result = wikidata.send_wikidata_query( SPARQL_DESCRIPTION.replace('%IDS%', IDS).replace('%LANGUAGES_SPARQL%', LANGUAGES_SPARQL) @@ -178,14 +193,17 @@ def fetch_wikidata_descriptions(): for binding in result['results']['bindings']: wikidata_id = binding['item']['value'].replace('http://www.wikidata.org/entity/', '') wikidata_lang = binding['itemDescription']['xml:lang'] - description = binding['itemDescription']['value'] + desc = binding['itemDescription']['value'] for engine_name in wd_to_engine_name[wikidata_id]: for lang in LANGUAGES: - if WIKIPEDIA_LANGUAGES[lang] == wikidata_lang: - update_description(engine_name, lang, description, 'wikidata') + if WIKIPEDIA_LANGUAGES[lang] != wikidata_lang: + continue + print(f" engine: {engine_name} / wikidata_lang: {wikidata_lang} / len(desc): {len(desc)}") + update_description(engine_name, lang, desc, 'wikidata') def fetch_wikipedia_descriptions(): + print('Fetching wikipedia descriptions') result = wikidata.send_wikidata_query( SPARQL_WIKIPEDIA_ARTICLE.replace('%IDS%', IDS).replace('%LANGUAGES_SPARQL%', LANGUAGES_SPARQL) ) @@ -196,9 +214,13 @@ def fetch_wikipedia_descriptions(): pageid = binding['name']['value'] for engine_name in wd_to_engine_name[wikidata_id]: for lang in LANGUAGES: - if WIKIPEDIA_LANGUAGES[lang] == wikidata_lang: - description = get_wikipedia_summary(lang, pageid) - update_description(engine_name, lang, description, 'wikipedia') + if WIKIPEDIA_LANGUAGES[lang] != wikidata_lang: + continue + desc = get_wikipedia_summary(lang, pageid) + if not desc: + continue + print(f" engine: {engine_name} / wikidata_lang: {wikidata_lang} / len(desc): {len(desc)}") + update_description(engine_name, lang, desc, 'wikipedia') def normalize_url(url): @@ -209,41 +231,60 @@ def normalize_url(url): def fetch_website_description(engine_name, website): + print(f"- fetch website descr: {engine_name} / {website}") default_lang, default_description = get_website_description(website, None, None) + if default_lang is None or default_description is None: # the front page can't be fetched: skip this engine return - wikipedia_languages_r = {V: K for K, V in WIKIPEDIA_LANGUAGES.items()} + # to specify an order in where the most common languages are in front of the + # language list .. languages = ['en', 'es', 'pt', 'ru', 'tr', 'fr'] languages = languages + [l for l in LANGUAGES if l not in languages] previous_matched_lang = None previous_count = 0 + for lang in languages: - if lang not in descriptions[engine_name]: - fetched_lang, desc = get_website_description(website, lang, WIKIPEDIA_LANGUAGES[lang]) - if fetched_lang is None or desc is None: - continue - matched_lang = match_locale(fetched_lang, LANGUAGES, fallback=None) - if matched_lang is None: - fetched_wikipedia_lang = match_locale(fetched_lang, WIKIPEDIA_LANGUAGES.values(), fallback=None) - matched_lang = wikipedia_languages_r.get(fetched_wikipedia_lang) - if matched_lang is not None: - update_description(engine_name, matched_lang, desc, website, replace=False) - # check if desc changed with the different lang values - if matched_lang == previous_matched_lang: - previous_count += 1 - if previous_count == 6: - # the website has returned the same description for 6 different languages in Accept-Language header - # stop now - break - else: - previous_matched_lang = matched_lang - previous_count = 0 + + if lang in descriptions[engine_name]: + continue + + fetched_lang, desc = get_website_description(website, lang, WIKIPEDIA_LANGUAGES[lang]) + if fetched_lang is None or desc is None: + continue + + # check if desc changed with the different lang values + + if fetched_lang == previous_matched_lang: + previous_count += 1 + if previous_count == 6: + # the website has returned the same description for 6 different languages in Accept-Language header + # stop now + break + else: + previous_matched_lang = fetched_lang + previous_count = 0 + + # Don't trust in the value of fetched_lang, some websites return + # for some inappropriate values, by example bing-images:: + # + # requested lang: zh-Hans-CN / fetched lang: ceb / desc: 查看根据您的兴趣量身定制的提要 + # + # The lang ceb is "Cebuano" but the description is given in zh-Hans-CN + + print( + f" engine: {engine_name:20} / requested lang:{lang:7}" + f" / fetched lang: {fetched_lang:7} / len(desc): {len(desc)}" + ) + + matched_lang = match_locale(fetched_lang, LANGUAGES, fallback=lang) + update_description(engine_name, matched_lang, desc, website, replace=False) def fetch_website_descriptions(): + print('Fetching website descriptions') for engine_name, engine in searx.engines.engines.items(): website = getattr(engine, "about", {}).get('website') if website is None and hasattr(engine, "search_url"): @@ -289,11 +330,8 @@ def get_output(): def main(): initialize() - print('Fetching wikidata descriptions') fetch_wikidata_descriptions() - print('Fetching wikipedia descriptions') fetch_wikipedia_descriptions() - print('Fetching website descriptions') fetch_website_descriptions() output = get_output()