summaryrefslogtreecommitdiff
path: root/searx/engines/bing.py
diff options
context:
space:
mode:
authorMarkus Heiser <markus.heiser@darmarit.de>2021-12-18 13:41:12 +0100
committerMarkus Heiser <markus.heiser@darmarit.de>2021-12-18 13:50:38 +0100
commit6b85607274b7a74cadb4707d430011650a087a5b (patch)
tree1beb1c0bc22b3a1c28274ef7189a921211fb4b0b /searx/engines/bing.py
parentb2177e5916c58b817665623ca107bb2467773e97 (diff)
downloadsearxng-6b85607274b7a74cadb4707d430011650a087a5b.tar.gz
searxng-6b85607274b7a74cadb4707d430011650a087a5b.zip
[fix] bing engine: fix paging support, show inital page.
Follow up queries for the pages needed to be fixed. - Split search-term in one for initial query and one for following queries. - Set some headers in HTTP requests, bing needs for paging support. - IMO //div[@class="sa_cc"] does no longer match in a bing response. Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
Diffstat (limited to 'searx/engines/bing.py')
-rw-r--r--searx/engines/bing.py34
1 files changed, 30 insertions, 4 deletions
diff --git a/searx/engines/bing.py b/searx/engines/bing.py
index a25f0212d..3917e54c1 100644
--- a/searx/engines/bing.py
+++ b/searx/engines/bing.py
@@ -1,6 +1,8 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
# lint: pylint
"""Bing (Web)
+
+- https://github.com/searx/searx/issues/2019#issuecomment-648227442
"""
import re
@@ -27,14 +29,27 @@ language_aliases = {'zh-CN': 'zh-CHS', 'zh-TW': 'zh-CHT', 'zh-HK': 'zh-CHT'}
# search-url
base_url = 'https://www.bing.com/'
-search_string = 'search?{query}&first={offset}'
+
+# initial query: https://www.bing.com/search?q=foo&search=&form=QBLH
+inital_query = 'search?{query}&search=&form=QBLH'
+
+# following queries: https://www.bing.com/search?q=foo&search=&first=11&FORM=PERE
+page_query = 'search?{query}&search=&first={offset}&FORM=PERE'
def _get_offset_from_pageno(pageno):
return (pageno - 1) * 10 + 1
def request(query, params):
- offset = _get_offset_from_pageno(params.get('pageno', 0))
+ offset = _get_offset_from_pageno(params.get('pageno', 1))
+
+ # logger.debug("params['pageno'] --> %s", params.get('pageno'))
+ # logger.debug(" offset --> %s", offset)
+
+ search_string = page_query
+ if offset == 1:
+ search_string = inital_query
+
if params['language'] == 'all':
lang = 'EN'
else:
@@ -49,10 +64,18 @@ def request(query, params):
search_path = search_string.format(
query = urlencode({'q': query}),
offset = offset)
- params['url'] = base_url + search_path
- return params
+ if offset > 1:
+ referer = base_url + inital_query.format(query = urlencode({'q': query}))
+ params['headers']['Referer'] = referer
+ logger.debug("headers.Referer --> %s", referer )
+ params['url'] = base_url + search_path
+ params['headers']['Accept-Language'] = "en-US,en;q=0.5"
+ params['headers']['Accept'] = (
+ 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8'
+ )
+ return params
def response(resp):
@@ -63,6 +86,9 @@ def response(resp):
for result in eval_xpath(dom, '//div[@class="sa_cc"]'):
+ # IMO //div[@class="sa_cc"] does no longer match
+ logger.debug('found //div[@class="sa_cc"] --> %s', result)
+
link = eval_xpath(result, './/h3/a')[0]
url = link.attrib.get('href')
title = extract_text(link)