diff options
| author | Johannes Schauer <josch@debian.org> | 2018-02-20 13:49:55 +0100 |
|---|---|---|
| committer | Johannes Schauer <josch@debian.org> | 2018-02-20 13:52:45 +0100 |
| commit | f6da658cf1d976fdd7847593ba96857e2c14fc8b (patch) | |
| tree | a548673c4b66d77d8e9f9f12169ac3fc09c400fc /searx/engines/google.py | |
| parent | 1619b7943f59fd6784ff70afdd604570f9bebc52 (diff) | |
New upstream version 0.14.0+dfsg1
Diffstat (limited to 'searx/engines/google.py')
| -rw-r--r-- | searx/engines/google.py | 9 |
1 files changed, 8 insertions, 1 deletions
diff --git a/searx/engines/google.py b/searx/engines/google.py index de2717d..0a86783 100644 --- a/searx/engines/google.py +++ b/searx/engines/google.py @@ -9,6 +9,7 @@ # @parse url, title, content, suggestion import re +from flask_babel import gettext from lxml import html, etree from searx.engines.xpath import extract_text, extract_url from searx import logger @@ -164,7 +165,8 @@ def extract_text_from_dom(result, xpath): def request(query, params): offset = (params['pageno'] - 1) * 10 - if params['language'] == 'all': + # temporary fix until a way of supporting en-US is found + if params['language'] == 'en-US': params['language'] = 'en-GB' if params['language'][:2] == 'jv': @@ -185,6 +187,8 @@ def request(query, params): else: google_hostname = default_hostname + # original format: ID=3e2b6616cee08557:TM=5556667580:C=r:IP=4.1.12.5-:S=23ASdf0soFgF2d34dfgf-_22JJOmHdfgg + params['cookies']['GOOGLE_ABUSE_EXEMPTION'] = 'x' params['url'] = search_url.format(offset=offset, query=urlencode({'q': query}), hostname=google_hostname, @@ -209,6 +213,9 @@ def response(resp): if resp_url.netloc == 'sorry.google.com' or resp_url.path == '/sorry/IndexRedirect': raise RuntimeWarning('sorry.google.com') + if resp_url.path.startswith('/sorry'): + raise RuntimeWarning(gettext('CAPTCHA required')) + # which hostname ? google_hostname = resp.search_params.get('google_hostname') google_url = "https://" + google_hostname |