OwlBurst/Ponysearch
1
0
Fork 0
forked from Ponysearch/Ponysearch
Code Pull requests Activity

[fix] engine - duckduckgo_images / determination of vqd value incorrect

Browse source 
Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
This commit is contained in:
Markus Heiser 2023-09-05 20:25:13 +02:00 committed by Markus Heiser
parent 4fb26cd96d
commit 696c35d2c3
2 changed files with 10 additions and 18 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

16
searx/engines/duckduckgo.py
View file

@ -25,7 +25,6 @@ from searx.utils import (
from searx.network import get # see https://github.com/searxng/searxng/issues/762 from searx.network import get # see https://github.com/searxng/searxng/issues/762
from searx import redisdb from searx import redisdb
from searx.enginelib.traits import EngineTraits from searx.enginelib.traits import EngineTraits
from searx.exceptions import SearxEngineAPIException
if TYPE_CHECKING: if TYPE_CHECKING:
import logging import logging
@ -78,7 +77,7 @@ def cache_vqd(query, value):
c.set(key, value, ex=600) c.set(key, value, ex=600)
def get_vqd(query, headers): def get_vqd(query):
"""Returns the ``vqd`` that fits to the *query*. If there is no ``vqd`` cached """Returns the ``vqd`` that fits to the *query*. If there is no ``vqd`` cached
(:py:obj:`cache_vqd`) the query is sent to DDG to get a vqd value from the (:py:obj:`cache_vqd`) the query is sent to DDG to get a vqd value from the
response. response.
@ -94,13 +93,10 @@ def get_vqd(query, headers):
logger.debug("re-use cached vqd value: %s", value) logger.debug("re-use cached vqd value: %s", value)
return value return value
query_url = 'https://duckduckgo.com/?q={query}&atb=v290-5'.format(query=urlencode({'q': query})) query_url = 'https://lite.duckduckgo.com/lite/?{args}'.format(args=urlencode({'q': query}))
res = get(query_url, headers=headers) res = get(query_url)
content = res.text # type: ignore doc = lxml.html.fromstring(res.text)
if content.find('vqd=\"') == -1: value = doc.xpath("//input[@name='vqd']/@value")[0]
raise SearxEngineAPIException('Request failed')
value = content[content.find('vqd=\"') + 5 :]
value = value[: value.find('\'')]
logger.debug("new vqd value: %s", value) logger.debug("new vqd value: %s", value)
cache_vqd(query, value) cache_vqd(query, value)
return value return value
@ -240,7 +236,7 @@ def request(query, params):
params['data']['dc'] = offset + 1 params['data']['dc'] = offset + 1
# request needs a vqd argument # request needs a vqd argument
params['data']['vqd'] = get_vqd(query, params["headers"]) params['data']['vqd'] = get_vqd(query)
# initial page does not have additional data in the input form # initial page does not have additional data in the input form
if params['pageno'] > 1: if params['pageno'] > 1:

12
searx/engines/duckduckgo_images.py
View file

@ -50,7 +50,8 @@ def request(query, params):
'o': 'json', 'o': 'json',
# 'u': 'bing', # 'u': 'bing',
'l': eng_region, 'l': eng_region,
'vqd': get_vqd(query, params["headers"]), 'f': ',,,,,',
'vqd': get_vqd(query),
} }
if params['pageno'] > 1: if params['pageno'] > 1:
@ -59,7 +60,6 @@ def request(query, params):
params['cookies']['ad'] = eng_lang # zh_CN params['cookies']['ad'] = eng_lang # zh_CN
params['cookies']['ah'] = eng_region # "us-en,de-de" params['cookies']['ah'] = eng_region # "us-en,de-de"
params['cookies']['l'] = eng_region # "hk-tzh" params['cookies']['l'] = eng_region # "hk-tzh"
logger.debug("cookies: %s", params['cookies'])
safe_search = safesearch_cookies.get(params['safesearch']) safe_search = safesearch_cookies.get(params['safesearch'])
if safe_search is not None: if safe_search is not None:
@ -68,13 +68,9 @@ def request(query, params):
if safe_search is not None: if safe_search is not None:
args['p'] = safe_search # "-1", "1" args['p'] = safe_search # "-1", "1"
logger.debug("cookies: %s", params['cookies'])
args = urlencode(args) args = urlencode(args)
params['url'] = 'https://duckduckgo.com/i.js?{args}&f={f}'.format(args=args, f=',,,,,') params['url'] = 'https://duckduckgo.com/i.js?{args}'.format(args=args)
params['headers']['Accept'] = 'application/json, text/javascript, */*; q=0.01'
params['headers']['Referer'] = 'https://duckduckgo.com/'
params['headers']['X-Requested-With'] = 'XMLHttpRequest'
logger.debug("headers: %s", params['headers'])
return params return params