kamp/searxng
1
0
Fork 0
mirror of https://github.com/searxng/searxng.git synced 2024-11-19 02:40:11 +01:00
Code

[fix] startpage engine: fetch CAPTCHA & issues related to PR-695

Browse Source 
In case of CAPTCHA raise a SearxEngineCaptchaException and suspend for 7 days.
When get_sc_code() fails raise a SearxEngineResponseException and suspend for 7
days.

[1] https://github.com/searxng/searxng/pull/695

Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
This commit is contained in:
Markus Heiser 2022-01-09 16:05:25 +01:00
parent 2f4e567e90
commit 21e884f369
1 changed files with 21 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
searx/engines/startpage.py
View File

@ -18,6 +18,11 @@ from babel.localedata import locale_identifiers
from searx import network from searx import network
from searx.utils import extract_text, eval_xpath, match_language from searx.utils import extract_text, eval_xpath, match_language
from searx.exceptions import (
SearxEngineResponseException,
SearxEngineCaptchaException,
)
# about # about
about = { about = {
@ -54,6 +59,13 @@ sc_code_ts = 0
sc_code = '' sc_code = ''
def raise_captcha(resp):
if str(resp.url).startswith('https://www.startpage.com/sp/captcha'):
# suspend CAPTCHA for 7 days
raise SearxEngineCaptchaException(suspended_time=7 * 24 * 3600)
def get_sc_code(headers): def get_sc_code(headers):
"""Get an actual `sc` argument from startpage's home page. """Get an actual `sc` argument from startpage's home page.
@ -73,10 +85,17 @@ def get_sc_code(headers):
logger.debug("query new sc time-stamp ...") logger.debug("query new sc time-stamp ...")
resp = network.get(base_url, headers=headers) resp = network.get(base_url, headers=headers)
raise_captcha(resp)
dom = html.fromstring(resp.text) dom = html.fromstring(resp.text)
try:
# href --> '/?sc=adrKJMgF8xwp20' # href --> '/?sc=adrKJMgF8xwp20'
href = eval_xpath(dom, '//a[@class="footer-home__logo"]')[0].get('href') href = eval_xpath(dom, '//a[@class="footer-home__logo"]')[0].get('href')
except IndexError as exc:
# suspend startpage API --> https://github.com/searxng/searxng/pull/695
raise SearxEngineResponseException(
suspended_time=7 * 24 * 3600, message="PR-695: query new sc time-stamp failed!"
)
sc_code = href[5:] sc_code = href[5:]
sc_code_ts = time() sc_code_ts = time()