forked from Ponysearch/Ponysearch
[fix] engine bing-news: replace the http:// by https://
BTW: add bing_news to the pylint process Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
This commit is contained in:
Markus Heiser
parent
c351993fdc
commit
f41734a543
1 changed files with 62 additions and 45 deletions
|
|
@ -1,16 +1,27 @@
|
|||
# SPDX-License-Identifier: AGPL-3.0-or-later
|
||||
"""
|
||||
Bing (News)
|
||||
# lint: pylint
|
||||
"""Bing (News)
|
||||
"""
|
||||
|
||||
from urllib.parse import (
|
||||
urlencode,
|
||||
urlparse,
|
||||
parse_qsl,
|
||||
quote,
|
||||
)
|
||||
from datetime import datetime
|
||||
from dateutil import parser
|
||||
from urllib.parse import urlencode, urlparse, parse_qsl
|
||||
from lxml import etree
|
||||
from lxml.etree import XPath
|
||||
from searx.utils import match_language, eval_xpath_getindex
|
||||
from searx.engines.bing import language_aliases
|
||||
from searx.engines.bing import _fetch_supported_languages, supported_languages_url # NOQA # pylint: disable=unused-import
|
||||
from searx.utils import (
|
||||
match_language,
|
||||
eval_xpath_getindex
|
||||
)
|
||||
from searx.engines.bing import ( # pylint: disable=unused-import
|
||||
language_aliases,
|
||||
_fetch_supported_languages,
|
||||
supported_languages_url,
|
||||
)
|
||||
|
||||
# about
|
||||
about = {
|
||||
|
|
@ -31,69 +42,71 @@ time_range_support = True
|
|||
base_url = 'https://www.bing.com/'
|
||||
search_string = 'news/search?{query}&first={offset}&format=RSS'
|
||||
search_string_with_time = 'news/search?{query}&first={offset}&qft=interval%3d"{interval}"&format=RSS'
|
||||
time_range_dict = {'day': '7',
|
||||
'week': '8',
|
||||
'month': '9'}
|
||||
time_range_dict = {
|
||||
'day': '7',
|
||||
'week': '8',
|
||||
'month': '9'
|
||||
}
|
||||
|
||||
|
||||
# remove click
|
||||
def url_cleanup(url_string):
|
||||
"""remove click"""
|
||||
|
||||
parsed_url = urlparse(url_string)
|
||||
if parsed_url.netloc == 'www.bing.com' and parsed_url.path == '/news/apiclick.aspx':
|
||||
query = dict(parse_qsl(parsed_url.query))
|
||||
return query.get('url', None)
|
||||
url_string = query.get('url', None)
|
||||
return url_string
|
||||
|
||||
|
||||
# replace the http://*bing4.com/th?id=... by https://www.bing.com/th?id=...
|
||||
def image_url_cleanup(url_string):
|
||||
parsed_url = urlparse(url_string)
|
||||
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
|
||||
"""replace the http://*bing.com/th?id=... by https://www.bing.com/th?id=..."""
|
||||
|
||||
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):
|
||||
if time_range in time_range_dict:
|
||||
search_path = search_string_with_time.format(
|
||||
query=urlencode({'q': query, 'setmkt': language}),
|
||||
offset=offset,
|
||||
interval=time_range_dict[time_range])
|
||||
query = urlencode({
|
||||
'q': query,
|
||||
'setmkt': language
|
||||
}),
|
||||
offset = offset,
|
||||
interval = time_range_dict[time_range]
|
||||
)
|
||||
else:
|
||||
# e.g. setmkt=de-de&setlang=de
|
||||
search_path = search_string.format(
|
||||
query=urlencode({'q': query, 'setmkt': language}),
|
||||
offset=offset)
|
||||
query = urlencode({
|
||||
'q': query,
|
||||
'setmkt': language
|
||||
}),
|
||||
offset = offset
|
||||
)
|
||||
return base_url + search_path
|
||||
|
||||
|
||||
# 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
|
||||
|
||||
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'])
|
||||
|
||||
return params
|
||||
|
||||
|
||||
# get response from search-request
|
||||
def response(resp):
|
||||
|
||||
results = []
|
||||
|
||||
rss = etree.fromstring(resp.content)
|
||||
namespaces = rss.nsmap
|
||||
|
||||
ns = rss.nsmap
|
||||
|
||||
# parse results
|
||||
for item in rss.xpath('./channel/item'):
|
||||
# url / title / content
|
||||
url = url_cleanup(eval_xpath_getindex(item, './link/text()', 0, default=None))
|
||||
|
|
@ -110,22 +123,26 @@ def response(resp):
|
|||
publishedDate = datetime.now()
|
||||
|
||||
# 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:
|
||||
thumbnail = image_url_cleanup(thumbnail)
|
||||
|
||||
# append result
|
||||
if thumbnail is not None:
|
||||
results.append({'url': url,
|
||||
'title': title,
|
||||
'publishedDate': publishedDate,
|
||||
'content': content,
|
||||
'img_src': thumbnail})
|
||||
results.append({
|
||||
'url': url,
|
||||
'title': title,
|
||||
'publishedDate': publishedDate,
|
||||
'content': content,
|
||||
'img_src': thumbnail
|
||||
})
|
||||
else:
|
||||
results.append({'url': url,
|
||||
'title': title,
|
||||
'publishedDate': publishedDate,
|
||||
'content': content})
|
||||
results.append({
|
||||
'url': url,
|
||||
'title': title,
|
||||
'publishedDate': publishedDate,
|
||||
'content': content
|
||||
})
|
||||
|
||||
# return results
|
||||
return results
|
||||
|
|
|
|||
Loading…
Reference in a new issue