rewrite duckduckgo engine and add comments

Browse Source
This commit is contained in:
Thomas Pointhuber 2014-09-02 17:14:57 +02:00
parent 3d61d9b930
commit e6e4de8ba0
2 changed files with 37 additions and 36 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

71
searx/engines/duckduckgo.py
View File

@ -1,24 +1,48 @@
## DuckDuckGo (Web)
#
# @website https://duckduckgo.com/
# @provide-api yes (https://duckduckgo.com/api), but not all results from search-site
#
# @using-api no
# @results HTML (using search portal)
# @stable no (HTML can change)
# @parse url, title, content
#
# @todo rewrite to api
# @todo language support
from urllib import urlencode from urllib import urlencode
from lxml.html import fromstring from lxml.html import fromstring
from searx.utils import html_to_text from searx.utils import html_to_text
url = 'https://duckduckgo.com/html?{query}&s={offset}' # engine dependent config
categories = ['general']
paging = True
locale = 'us-en' locale = 'us-en'
# search-url
url = 'https://duckduckgo.com/html?{query}&s={offset}'
# specific xpath variables
result_xpath = '//div[@class="results_links results_links_deep web-result"]' # noqa
url_xpath = './/a[@class="large"]/@href'
title_xpath = './/a[@class="large"]//text()'
content_xpath = './/div[@class="snippet"]//text()'
# do search-request
def request(query, params): def request(query, params):
offset = (params['pageno'] - 1) * 30 offset = (params['pageno'] - 1) * 30
q = urlencode({'q': query,
'l': locale}) params['url'] = url.format(
params['url'] = url.format(query=q, offset=offset) query=urlencode({'q': query, 'l': locale}),
offset=offset)
return params return params
# get response from search-request
def response(resp): def response(resp):
result_xpath = '//div[@class="results_links results_links_deep web-result"]' # noqa
url_xpath = './/a[@class="large"]/@href'
title_xpath = './/a[@class="large"]//text()'
content_xpath = './/div[@class="snippet"]//text()'
results = [] results = []
doc = fromstring(resp.text) doc = fromstring(resp.text)
@ -28,38 +52,17 @@ def response(resp):
res_url = r.xpath(url_xpath)[-1] res_url = r.xpath(url_xpath)[-1]
except: except:
continue continue
if not res_url: if not res_url:
continue continue
title = html_to_text(''.join(r.xpath(title_xpath))) title = html_to_text(''.join(r.xpath(title_xpath)))
content = html_to_text(''.join(r.xpath(content_xpath))) content = html_to_text(''.join(r.xpath(content_xpath)))
# append result
results.append({'title': title, results.append({'title': title,
'content': content, 'content': content,
'url': res_url}) 'url': res_url})
# return results
return results return results
#from json import loads
#search_url = url + 'd.js?{query}&p=1&s={offset}'
#
#paging = True
#
#
#def request(query, params):
# offset = (params['pageno'] - 1) * 30
# q = urlencode({'q': query,
# 'l': locale})
# params['url'] = search_url.format(query=q, offset=offset)
# return params
#
#
#def response(resp):
# results = []
# search_res = loads(resp.text[resp.text.find('[{'):-2])[:-1]
# for r in search_res:
# if not r.get('t'):
# continue
# results.append({'title': r['t'],
# 'content': html_to_text(r['a']),
# 'url': r['u']})
# return results

2
searx/settings.yml
View File

@ -37,7 +37,6 @@ engines:
- name : deviantart - name : deviantart
engine : deviantart engine : deviantart
categories : images
shortcut : da shortcut : da
timeout: 3.0 timeout: 3.0
@ -47,7 +46,6 @@ engines:
- name : duckduckgo - name : duckduckgo
engine : duckduckgo engine : duckduckgo
locale : en-us
shortcut : ddg shortcut : ddg
# down - website is under criminal investigation by the UK # down - website is under criminal investigation by the UK