mirror of
https://github.com/searx/searx
synced 2024-11-04 17:56:48 +01:00
fix startpage engine and add comments
* add language support * remove not required code * improve google-ad detection (no false detection anymore, I hope) * other improvements
This commit is contained in:
parent
a46bbb4042
commit
678a80f043
@ -1,47 +1,79 @@
|
|||||||
|
## Startpage (Web)
|
||||||
|
#
|
||||||
|
# @website https://startpage.com
|
||||||
|
# @provide-api no (nothing found)
|
||||||
|
#
|
||||||
|
# @using-api no
|
||||||
|
# @results HTML
|
||||||
|
# @stable no (HTML can change)
|
||||||
|
# @parse url, title, content
|
||||||
|
#
|
||||||
|
# @todo paging
|
||||||
|
|
||||||
from urllib import urlencode
|
from urllib import urlencode
|
||||||
from lxml import html
|
from lxml import html
|
||||||
from cgi import escape
|
from cgi import escape
|
||||||
|
import re
|
||||||
|
|
||||||
base_url = None
|
# engine dependent config
|
||||||
search_url = None
|
categories = ['general']
|
||||||
|
# there is a mechanism to block "bot" search (probably the parameter qid), require storing of qid's between mulitble search-calls
|
||||||
|
#paging = False
|
||||||
|
language_support = True
|
||||||
|
|
||||||
# TODO paging
|
# search-url
|
||||||
paging = False
|
base_url = 'https://startpage.com/'
|
||||||
# TODO complete list of country mapping
|
search_url = base_url + 'do/search'
|
||||||
country_map = {'en_US': 'eng',
|
|
||||||
'en_UK': 'uk',
|
# specific xpath variables
|
||||||
'nl_NL': 'ned'}
|
# ads xpath //div[@id="results"]/div[@id="sponsored"]//div[@class="result"]
|
||||||
|
# not ads: div[@class="result"] are the direct childs of div[@id="results"]
|
||||||
|
results_xpath = '//div[@class="result"]'
|
||||||
|
link_xpath = './/h3/a'
|
||||||
|
|
||||||
|
|
||||||
|
# do search-request
|
||||||
def request(query, params):
|
def request(query, params):
|
||||||
|
offset = (params['pageno'] - 1) * 10
|
||||||
query = urlencode({'q': query})[2:]
|
query = urlencode({'q': query})[2:]
|
||||||
|
|
||||||
params['url'] = search_url
|
params['url'] = search_url
|
||||||
params['method'] = 'POST'
|
params['method'] = 'POST'
|
||||||
params['data'] = {'query': query,
|
params['data'] = {'query': query,
|
||||||
'startat': (params['pageno'] - 1) * 10} # offset
|
'startat': offset}
|
||||||
country = country_map.get(params['language'], 'eng')
|
|
||||||
params['cookies']['preferences'] = \
|
# set language if specified
|
||||||
'lang_homepageEEEs/air/{country}/N1NsslEEE1N1Nfont_sizeEEEmediumN1Nrecent_results_filterEEE1N1Nlanguage_uiEEEenglishN1Ndisable_open_in_new_windowEEE0N1Ncolor_schemeEEEnewN1Nnum_of_resultsEEE10N1N'.format(country=country) # noqa
|
if params['language'] != 'all':
|
||||||
|
params['data']['with_language'] = 'lang_' + params['language'].split('_')[0]
|
||||||
|
|
||||||
return params
|
return params
|
||||||
|
|
||||||
|
|
||||||
|
# get response from search-request
|
||||||
def response(resp):
|
def response(resp):
|
||||||
results = []
|
results = []
|
||||||
|
|
||||||
dom = html.fromstring(resp.content)
|
dom = html.fromstring(resp.content)
|
||||||
# ads xpath //div[@id="results"]/div[@id="sponsored"]//div[@class="result"]
|
|
||||||
# not ads: div[@class="result"] are the direct childs of div[@id="results"]
|
# parse results
|
||||||
for result in dom.xpath('//div[@class="result"]'):
|
for result in dom.xpath(results_xpath):
|
||||||
link = result.xpath('.//h3/a')[0]
|
link = result.xpath(link_xpath)[0]
|
||||||
url = link.attrib.get('href')
|
url = link.attrib.get('href')
|
||||||
if url.startswith('http://www.google.')\
|
|
||||||
or url.startswith('https://www.google.'):
|
|
||||||
continue
|
|
||||||
title = escape(link.text_content())
|
title = escape(link.text_content())
|
||||||
|
|
||||||
content = ''
|
# block google-ad url's
|
||||||
|
if re.match("^http(s|)://www.google.[a-z]+/aclk.*$", url):
|
||||||
|
continue
|
||||||
|
|
||||||
if result.xpath('./p[@class="desc"]'):
|
if result.xpath('./p[@class="desc"]'):
|
||||||
content = escape(result.xpath('./p[@class="desc"]')[0].text_content())
|
content = escape(result.xpath('./p[@class="desc"]')[0].text_content())
|
||||||
|
else:
|
||||||
|
content = ''
|
||||||
|
|
||||||
results.append({'url': url, 'title': title, 'content': content})
|
# append result
|
||||||
|
results.append({'url': url,
|
||||||
|
'title': title,
|
||||||
|
'content': content})
|
||||||
|
|
||||||
|
# return results
|
||||||
return results
|
return results
|
||||||
|
@ -94,8 +94,6 @@ engines:
|
|||||||
|
|
||||||
- name : startpage
|
- name : startpage
|
||||||
engine : startpage
|
engine : startpage
|
||||||
base_url : 'https://startpage.com/'
|
|
||||||
search_url : 'https://startpage.com/do/search'
|
|
||||||
shortcut : sp
|
shortcut : sp
|
||||||
|
|
||||||
# +30% page load time
|
# +30% page load time
|
||||||
|
Loading…
Reference in New Issue
Block a user