searxng/searx/engines/bing.py

128 lines
3.7 KiB
Python
Raw Normal View History

"""
Bing (Web)
@website https://www.bing.com
@provide-api yes (http://datamarket.azure.com/dataset/bing/search),
max. 5000 query/month
@using-api no (because of query limit)
@results HTML (using search portal)
@stable no (HTML can change)
@parse url, title, content
@todo publishedDate
"""
2014-09-01 12:38:59 +00:00
import re
2014-02-05 19:24:31 +00:00
from lxml import html
from searx import logger, utils
2015-01-25 19:14:37 +00:00
from searx.engines.xpath import extract_text
2016-11-30 17:43:03 +00:00
from searx.url_utils import urlencode
from searx.utils import match_language, gen_useragent, eval_xpath
2013-10-24 21:52:57 +00:00
logger = logger.getChild('bing engine')
2014-09-01 12:38:59 +00:00
# engine dependent config
categories = ['general']
2014-01-29 20:14:38 +00:00
paging = True
2014-01-31 03:35:23 +00:00
language_support = True
supported_languages_url = 'https://www.bing.com/account/general'
language_aliases = {'zh-CN': 'zh-CHS', 'zh-TW': 'zh-CHT', 'zh-HK': 'zh-CHT'}
2014-01-29 20:14:38 +00:00
2014-09-01 12:38:59 +00:00
# search-url
base_url = 'https://www.bing.com/'
search_string = 'search?{query}&first={offset}'
2013-10-24 21:52:57 +00:00
2014-09-02 15:13:44 +00:00
def _get_offset_from_pageno(pageno):
return (pageno - 1) * 10 + 1
2014-09-01 12:38:59 +00:00
# do search-request
2013-10-24 21:52:57 +00:00
def request(query, params):
offset = _get_offset_from_pageno(params.get('pageno', 0))
2014-09-01 12:38:59 +00:00
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')
2014-09-01 12:38:59 +00:00
search_path = search_string.format(
2016-11-14 14:52:29 +00:00
query=urlencode({'q': query}),
2014-01-29 20:14:38 +00:00
offset=offset)
2014-01-31 03:35:23 +00:00
2013-10-24 21:52:57 +00:00
params['url'] = base_url + search_path
2013-10-24 21:52:57 +00:00
return params
2014-09-01 12:38:59 +00:00
# get response from search-request
2013-10-24 21:52:57 +00:00
def response(resp):
results = []
result_len = 0
2014-09-01 12:38:59 +00:00
dom = html.fromstring(resp.text)
2014-09-01 12:38:59 +00:00
# parse results
for result in eval_xpath(dom, '//div[@class="sa_cc"]'):
link = eval_xpath(result, './/h3/a')[0]
2013-10-24 21:52:57 +00:00
url = link.attrib.get('href')
2015-01-25 19:14:37 +00:00
title = extract_text(link)
content = extract_text(eval_xpath(result, './/p'))
2013-10-24 23:37:48 +00:00
2014-09-01 12:38:59 +00:00
# append result
results.append({'url': url,
'title': title,
2014-09-01 12:38:59 +00:00
'content': content})
# parse results again if nothing is found yet
for result in eval_xpath(dom, '//li[@class="b_algo"]'):
link = eval_xpath(result, './/h2/a')[0]
2013-10-24 23:37:48 +00:00
url = link.attrib.get('href')
2015-01-25 19:14:37 +00:00
title = extract_text(link)
content = extract_text(eval_xpath(result, './/p'))
2014-09-01 12:38:59 +00:00
# append result
results.append({'url': url,
'title': title,
2014-09-01 12:38:59 +00:00
'content': content})
try:
2020-01-02 21:28:47 +00:00
result_len_container = "".join(eval_xpath(dom, '//span[@class="sb_count"]//text()'))
if "-" in result_len_container:
# Remove the part "from-to" for paginated request ...
result_len_container = result_len_container[result_len_container.find("-") * 2 + 2:]
result_len_container = re.sub('[^0-9]', '', result_len_container)
if len(result_len_container) > 0:
result_len = int(result_len_container)
except Exception as e:
logger.debug('result error :\n%s', e)
pass
2020-01-02 21:28:47 +00:00
if result_len and _get_offset_from_pageno(resp.search_params.get("pageno", 0)) > result_len:
return []
results.append({'number_of_results': result_len})
2013-10-24 21:52:57 +00:00
return results
# get supported languages from their site
def _fetch_supported_languages(resp):
lang_tags = set()
setmkt = re.compile('setmkt=([^&]*)')
dom = html.fromstring(resp.text)
lang_links = eval_xpath(dom, "//li/a[contains(@href, 'setmkt')]")
for a in lang_links:
href = eval_xpath(a, './@href')[0]
match = setmkt.search(href)
l_tag = match.groups()[0]
_lang, _nation = l_tag.split('-',1)
l_tag = _lang.lower() + '-' + _nation.upper()
lang_tags.add(l_tag)
return list(lang_tags)