1
0
mirror of https://github.com/searxng/searxng.git synced 2024-11-17 18:00:12 +01:00
searxng/searx/engines/wikipedia.py

89 lines
2.6 KiB
Python
Raw Normal View History

"""
Wikipedia (Web)
2020-09-08 07:05:21 +02:00
@website https://en.wikipedia.org/api/rest_v1/
@provide-api yes
@using-api yes
@results JSON
@stable yes
@parse url, infobox
"""
from urllib.parse import quote
from json import loads
from lxml.html import fromstring
2020-09-08 07:05:21 +02:00
from searx.utils import match_language, searx_useragent
# search-url
search_url = 'https://{language}.wikipedia.org/api/rest_v1/page/summary/{title}'
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]
2019-01-07 21:28:58 +01:00
if lang_pre == 'all' or lang_pre not in supported_languages and lang_pre not in language_aliases:
return 'en'
return match_language(lang, supported_languages, language_aliases).split('-')[0]
# do search-request
def request(query, params):
if query.islower():
2020-09-08 07:05:21 +02:00
query = query.title()
2020-09-08 07:05:21 +02:00
params['url'] = search_url.format(title=quote(query),
2016-11-30 18:43:03 +01:00
language=url_lang(params['language']))
2020-09-08 07:05:21 +02:00
params['headers']['User-Agent'] = searx_useragent()
return params
# get response from search-request
def response(resp):
2020-09-08 07:05:21 +02:00
if not resp.ok:
return []
2020-09-08 07:05:21 +02:00
results = []
api_result = loads(resp.text)
2020-09-08 07:05:21 +02:00
# skip disambiguation pages
if api_result['type'] != 'standard':
return []
2020-09-08 07:05:21 +02:00
title = api_result['title']
wikipedia_link = api_result['content_urls']['desktop']['page']
results.append({'url': wikipedia_link, 'title': title})
results.append({'infobox': title,
'id': wikipedia_link,
2020-09-08 07:05:21 +02:00
'content': api_result.get('extract', ''),
'img_src': api_result.get('thumbnail', {}).get('source'),
'urls': [{'title': 'Wikipedia', 'url': wikipedia_link}]})
return results
# get supported languages from their site
def _fetch_supported_languages(resp):
supported_languages = {}
dom = fromstring(resp.text)
tables = dom.xpath('//table[contains(@class,"sortable")]')
for table in tables:
# exclude header row
trs = table.xpath('.//tr')[1:]
for tr in trs:
td = tr.xpath('./td')
code = td[3].xpath('./a')[0].text
name = td[2].xpath('./a')[0].text
english_name = td[1].xpath('./a')[0].text
articles = int(td[4].xpath('./a/b')[0].text.replace(',', ''))
# exclude languages with too few articles
if articles >= 100:
supported_languages[code] = {"name": name, "english_name": english_name, "articles": articles}
return supported_languages