Merge pull request #609 from return42/fix-bing-news

[fix] engine bing-news: replace the http:// by https://
This commit is contained in:
Alexandre Flament 2021-12-18 00:01:40 +01:00 committed by GitHub
commit 4da1e0026c
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -1,16 +1,27 @@
# SPDX-License-Identifier: AGPL-3.0-or-later # SPDX-License-Identifier: AGPL-3.0-or-later
""" # lint: pylint
Bing (News) """Bing (News)
""" """
from urllib.parse import (
urlencode,
urlparse,
parse_qsl,
quote,
)
from datetime import datetime from datetime import datetime
from dateutil import parser from dateutil import parser
from urllib.parse import urlencode, urlparse, parse_qsl
from lxml import etree from lxml import etree
from lxml.etree import XPath from lxml.etree import XPath
from searx.utils import match_language, eval_xpath_getindex from searx.utils import (
from searx.engines.bing import language_aliases match_language,
from searx.engines.bing import _fetch_supported_languages, supported_languages_url # NOQA # pylint: disable=unused-import eval_xpath_getindex
)
from searx.engines.bing import ( # pylint: disable=unused-import
language_aliases,
_fetch_supported_languages,
supported_languages_url,
)
# about # about
about = { about = {
@ -31,69 +42,71 @@ time_range_support = True
base_url = 'https://www.bing.com/' base_url = 'https://www.bing.com/'
search_string = 'news/search?{query}&first={offset}&format=RSS' search_string = 'news/search?{query}&first={offset}&format=RSS'
search_string_with_time = 'news/search?{query}&first={offset}&qft=interval%3d"{interval}"&format=RSS' search_string_with_time = 'news/search?{query}&first={offset}&qft=interval%3d"{interval}"&format=RSS'
time_range_dict = {'day': '7', time_range_dict = {
'day': '7',
'week': '8', 'week': '8',
'month': '9'} 'month': '9'
}
# remove click
def url_cleanup(url_string): def url_cleanup(url_string):
"""remove click"""
parsed_url = urlparse(url_string) parsed_url = urlparse(url_string)
if parsed_url.netloc == 'www.bing.com' and parsed_url.path == '/news/apiclick.aspx': if parsed_url.netloc == 'www.bing.com' and parsed_url.path == '/news/apiclick.aspx':
query = dict(parse_qsl(parsed_url.query)) query = dict(parse_qsl(parsed_url.query))
return query.get('url', None) url_string = query.get('url', None)
return url_string return url_string
# replace the http://*bing4.com/th?id=... by https://www.bing.com/th?id=...
def image_url_cleanup(url_string): def image_url_cleanup(url_string):
parsed_url = urlparse(url_string) """replace the http://*bing.com/th?id=... by https://www.bing.com/th?id=..."""
if parsed_url.netloc.endswith('bing4.com') and parsed_url.path == '/th':
query = dict(parse_qsl(parsed_url.query))
return "https://www.bing.com/th?id=" + query.get('id')
return url_string
parsed_url = urlparse(url_string)
if parsed_url.netloc.endswith('bing.com') and parsed_url.path == '/th':
query = dict(parse_qsl(parsed_url.query))
url_string = "https://www.bing.com/th?id=" + quote(query.get('id'))
return url_string
def _get_url(query, language, offset, time_range): def _get_url(query, language, offset, time_range):
if time_range in time_range_dict: if time_range in time_range_dict:
search_path = search_string_with_time.format( search_path = search_string_with_time.format(
query=urlencode({'q': query, 'setmkt': language}), query = urlencode({
'q': query,
'setmkt': language
}),
offset = offset, offset = offset,
interval=time_range_dict[time_range]) interval = time_range_dict[time_range]
)
else: else:
# e.g. setmkt=de-de&setlang=de # e.g. setmkt=de-de&setlang=de
search_path = search_string.format( search_path = search_string.format(
query=urlencode({'q': query, 'setmkt': language}), query = urlencode({
offset=offset) 'q': query,
'setmkt': language
}),
offset = offset
)
return base_url + search_path return base_url + search_path
# do search-request
def request(query, params): def request(query, params):
if params['time_range'] and params['time_range'] not in time_range_dict: if params['time_range'] and params['time_range'] not in time_range_dict:
return params return params
offset = (params['pageno'] - 1) * 10 + 1 offset = (params['pageno'] - 1) * 10 + 1
if params['language'] == 'all': if params['language'] == 'all':
language = 'en-US' language = 'en-US'
else: else:
language = match_language(params['language'], supported_languages, language_aliases) language = match_language(params['language'], supported_languages, language_aliases)
params['url'] = _get_url(query, language, offset, params['time_range']) params['url'] = _get_url(query, language, offset, params['time_range'])
return params return params
# get response from search-request
def response(resp): def response(resp):
results = [] results = []
rss = etree.fromstring(resp.content) rss = etree.fromstring(resp.content)
namespaces = rss.nsmap
ns = rss.nsmap
# parse results
for item in rss.xpath('./channel/item'): for item in rss.xpath('./channel/item'):
# url / title / content # url / title / content
url = url_cleanup(eval_xpath_getindex(item, './link/text()', 0, default=None)) url = url_cleanup(eval_xpath_getindex(item, './link/text()', 0, default=None))
@ -110,22 +123,26 @@ def response(resp):
publishedDate = datetime.now() publishedDate = datetime.now()
# thumbnail # thumbnail
thumbnail = eval_xpath_getindex(item, XPath('./News:Image/text()', namespaces=ns), 0, default=None) thumbnail = eval_xpath_getindex(
item, XPath('./News:Image/text()', namespaces=namespaces), 0, default=None)
if thumbnail is not None: if thumbnail is not None:
thumbnail = image_url_cleanup(thumbnail) thumbnail = image_url_cleanup(thumbnail)
# append result # append result
if thumbnail is not None: if thumbnail is not None:
results.append({'url': url, results.append({
'url': url,
'title': title, 'title': title,
'publishedDate': publishedDate, 'publishedDate': publishedDate,
'content': content, 'content': content,
'img_src': thumbnail}) 'img_src': thumbnail
})
else: else:
results.append({'url': url, results.append({
'url': url,
'title': title, 'title': title,
'publishedDate': publishedDate, 'publishedDate': publishedDate,
'content': content}) 'content': content
})
# return results
return results return results