forked from zaclys/searxng
[mod] bing_news: use async API & upgrade to data_type: traits_v1
Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
This commit is contained in:
Markus Heiser
parent
ff80e7637e
commit
15eaf0f15f
|
|
@ -1,20 +1,30 @@
|
||||||
# SPDX-License-Identifier: AGPL-3.0-or-later
|
# SPDX-License-Identifier: AGPL-3.0-or-later
|
||||||
# lint: pylint
|
# lint: pylint
|
||||||
"""Bing (News)
|
"""Bing-News: description see :py:obj:`searx.engines.bing`.
|
||||||
"""
|
"""
|
||||||
|
|
||||||
from urllib.parse import (
|
# pylint: disable=invalid-name
|
||||||
urlencode,
|
|
||||||
urlparse,
|
from typing import TYPE_CHECKING
|
||||||
parse_qsl,
|
import uuid
|
||||||
quote,
|
from urllib.parse import urlencode
|
||||||
)
|
|
||||||
from datetime import datetime
|
from lxml import html
|
||||||
from dateutil import parser
|
|
||||||
from lxml import etree
|
|
||||||
from lxml.etree import XPath
|
|
||||||
from searx.utils import match_language, eval_xpath_getindex
|
|
||||||
from searx.enginelib.traits import EngineTraits
|
from searx.enginelib.traits import EngineTraits
|
||||||
|
from searx.engines.bing import (
|
||||||
|
set_bing_cookies,
|
||||||
|
_fetch_traits,
|
||||||
|
)
|
||||||
|
from searx.engines.bing import send_accept_language_header # pylint: disable=unused-import
|
||||||
|
|
||||||
|
if TYPE_CHECKING:
|
||||||
|
import logging
|
||||||
|
|
||||||
|
logger: logging.Logger
|
||||||
|
|
||||||
|
traits: EngineTraits
|
||||||
|
|
||||||
|
|
||||||
# about
|
# about
|
||||||
about = {
|
about = {
|
||||||
|
|
@ -30,126 +40,111 @@ about = {
|
||||||
categories = ['news']
|
categories = ['news']
|
||||||
paging = True
|
paging = True
|
||||||
time_range_support = True
|
time_range_support = True
|
||||||
send_accept_language_header = True
|
time_map = {
|
||||||
|
'day': '4',
|
||||||
|
'week': '8',
|
||||||
|
'month': '9',
|
||||||
|
}
|
||||||
|
"""A string '4' means *last hour*. We use *last hour* for ``day`` here since the
|
||||||
|
difference of *last day* and *last week* in the result list is just marginally.
|
||||||
|
"""
|
||||||
|
|
||||||
# search-url
|
base_url = 'https://www.bing.com/news/infinitescrollajax'
|
||||||
base_url = 'https://www.bing.com/'
|
"""Bing (News) search URL"""
|
||||||
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'}
|
|
||||||
|
|
||||||
|
bing_traits_url = 'https://learn.microsoft.com/en-us/bing/search-apis/bing-news-search/reference/market-codes'
|
||||||
|
"""Bing (News) search API description"""
|
||||||
|
|
||||||
def url_cleanup(url_string):
|
mkt_alias = {
|
||||||
"""remove click"""
|
'zh': 'en-WW',
|
||||||
|
'zh-CN': 'en-WW',
|
||||||
parsed_url = urlparse(url_string)
|
}
|
||||||
if parsed_url.netloc == 'www.bing.com' and parsed_url.path == '/news/apiclick.aspx':
|
"""Bing News has an official market code 'zh-CN' but we won't get a result with
|
||||||
query = dict(parse_qsl(parsed_url.query))
|
this market code. For 'zh' and 'zh-CN' we better use the *Worldwide aggregate*
|
||||||
url_string = query.get('url', None)
|
market code (en-WW).
|
||||||
return url_string
|
"""
|
||||||
|
|
||||||
|
|
||||||
def image_url_cleanup(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(
|
|
||||||
# fmt: off
|
|
||||||
query = urlencode({
|
|
||||||
'q': query,
|
|
||||||
'setmkt': language
|
|
||||||
}),
|
|
||||||
offset = offset,
|
|
||||||
interval = time_range_dict[time_range]
|
|
||||||
# fmt: on
|
|
||||||
)
|
|
||||||
else:
|
|
||||||
# e.g. setmkt=de-de&setlang=de
|
|
||||||
search_path = search_string.format(
|
|
||||||
# fmt: off
|
|
||||||
query = urlencode({
|
|
||||||
'q': query,
|
|
||||||
'setmkt': language
|
|
||||||
}),
|
|
||||||
offset = offset
|
|
||||||
# fmt: on
|
|
||||||
)
|
|
||||||
return base_url + search_path
|
|
||||||
|
|
||||||
|
|
||||||
def request(query, params):
|
def request(query, params):
|
||||||
|
"""Assemble a Bing-News request."""
|
||||||
|
|
||||||
if params['time_range'] and params['time_range'] not in time_range_dict:
|
sxng_locale = params['searxng_locale']
|
||||||
return params
|
engine_region = traits.get_region(mkt_alias.get(sxng_locale, sxng_locale), traits.all_locale)
|
||||||
|
engine_language = traits.get_language(sxng_locale, 'en')
|
||||||
|
|
||||||
offset = (params['pageno'] - 1) * 10 + 1
|
SID = uuid.uuid1().hex.upper()
|
||||||
if params['language'] == 'all':
|
set_bing_cookies(params, engine_language, engine_region, SID)
|
||||||
language = 'en-US'
|
|
||||||
else:
|
# build URL query
|
||||||
language = match_language(params['language'], supported_languages, language_aliases)
|
#
|
||||||
params['url'] = _get_url(query, language, offset, params['time_range'])
|
# example: https://www.bing.com/news/infinitescrollajax?q=london&first=1
|
||||||
|
|
||||||
|
query_params = {
|
||||||
|
# fmt: off
|
||||||
|
'q': query,
|
||||||
|
'InfiniteScroll': 1,
|
||||||
|
# to simplify the page count lets use the default of 10 images per page
|
||||||
|
'first' : (int(params.get('pageno', 1)) - 1) * 10 + 1,
|
||||||
|
# fmt: on
|
||||||
|
}
|
||||||
|
|
||||||
|
if params['time_range']:
|
||||||
|
# qft=interval:"7"
|
||||||
|
query_params['qft'] = 'qft=interval="%s"' % time_map.get(params['time_range'], '9')
|
||||||
|
|
||||||
|
params['url'] = base_url + '?' + urlencode(query_params)
|
||||||
|
|
||||||
return params
|
return params
|
||||||
|
|
||||||
|
|
||||||
def response(resp):
|
def response(resp):
|
||||||
|
"""Get response from Bing-Video"""
|
||||||
results = []
|
results = []
|
||||||
rss = etree.fromstring(resp.content)
|
|
||||||
namespaces = rss.nsmap
|
|
||||||
|
|
||||||
for item in rss.xpath('./channel/item'):
|
if not resp.ok or not resp.text:
|
||||||
# url / title / content
|
return results
|
||||||
url = url_cleanup(eval_xpath_getindex(item, './link/text()', 0, default=None))
|
|
||||||
title = eval_xpath_getindex(item, './title/text()', 0, default=url)
|
|
||||||
content = eval_xpath_getindex(item, './description/text()', 0, default='')
|
|
||||||
|
|
||||||
# publishedDate
|
dom = html.fromstring(resp.text)
|
||||||
publishedDate = eval_xpath_getindex(item, './pubDate/text()', 0, default=None)
|
|
||||||
try:
|
|
||||||
publishedDate = parser.parse(publishedDate, dayfirst=False)
|
|
||||||
except TypeError:
|
|
||||||
publishedDate = datetime.now()
|
|
||||||
except ValueError:
|
|
||||||
publishedDate = datetime.now()
|
|
||||||
|
|
||||||
# thumbnail
|
for newsitem in dom.xpath('//div[contains(@class, "newsitem")]'):
|
||||||
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
|
url = newsitem.xpath('./@url')[0]
|
||||||
if thumbnail is not None:
|
title = ' '.join(newsitem.xpath('.//div[@class="caption"]//a[@class="title"]/text()')).strip()
|
||||||
results.append(
|
content = ' '.join(newsitem.xpath('.//div[@class="snippet"]/text()')).strip()
|
||||||
{'url': url, 'title': title, 'publishedDate': publishedDate, 'content': content, 'img_src': thumbnail}
|
thumbnail = None
|
||||||
)
|
author = newsitem.xpath('./@data-author')[0]
|
||||||
else:
|
metadata = ' '.join(newsitem.xpath('.//div[@class="source"]/span/text()')).strip()
|
||||||
results.append({'url': url, 'title': title, 'publishedDate': publishedDate, 'content': content})
|
|
||||||
|
img_src = newsitem.xpath('.//a[@class="imagelink"]//img/@src')
|
||||||
|
if img_src:
|
||||||
|
thumbnail = 'https://www.bing.com/' + img_src[0]
|
||||||
|
|
||||||
|
results.append(
|
||||||
|
{
|
||||||
|
'url': url,
|
||||||
|
'title': title,
|
||||||
|
'content': content,
|
||||||
|
'img_src': thumbnail,
|
||||||
|
'author': author,
|
||||||
|
'metadata': metadata,
|
||||||
|
}
|
||||||
|
)
|
||||||
|
|
||||||
return results
|
return results
|
||||||
|
|
||||||
|
|
||||||
def fetch_traits(engine_traits: EngineTraits):
|
def fetch_traits(engine_traits: EngineTraits):
|
||||||
"""Fetch languages and regions from Bing-News."""
|
"""Fetch languages and regions from Bing-News.
|
||||||
# pylint: disable=import-outside-toplevel
|
|
||||||
|
|
||||||
from searx.engines.bing import _fetch_traits
|
The :py:obj:`description <searx.engines.bing_news.bing_traits_url>` of the
|
||||||
|
first table says *"query parameter when calling the Video Search API."*
|
||||||
|
.. thats why I use the 4. table "News Category API markets" for the
|
||||||
|
``xpath_market_codes``.
|
||||||
|
|
||||||
url = 'https://learn.microsoft.com/en-us/bing/search-apis/bing-news-search/reference/market-codes'
|
"""
|
||||||
|
|
||||||
# The description of the first table says "query parameter when calling the
|
|
||||||
# Video Search API." .. thats why I use the 4. table "News Category API markets"
|
|
||||||
xpath_market_codes = '//table[4]/tbody/tr/td[3]'
|
xpath_market_codes = '//table[4]/tbody/tr/td[3]'
|
||||||
|
|
||||||
# xpath_country_codes = '//table[2]/tbody/tr/td[2]'
|
# xpath_country_codes = '//table[2]/tbody/tr/td[2]'
|
||||||
xpath_language_codes = '//table[3]/tbody/tr/td[2]'
|
xpath_language_codes = '//table[3]/tbody/tr/td[2]'
|
||||||
|
|
||||||
_fetch_traits(engine_traits, url, xpath_language_codes, xpath_market_codes)
|
_fetch_traits(engine_traits, bing_traits_url, xpath_language_codes, xpath_market_codes)
|
||||||
|
|
|
||||||
Loading…
Reference in New Issue