Ponysearch/searx/engines/duckduckgo.py

133 lines
3.7 KiB
Python
Raw Normal View History

"""
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
"""
2013-10-23 23:55:37 +02:00
from urllib import urlencode
2014-03-21 16:33:17 +01:00
from lxml.html import fromstring
from requests import get
from json import loads
2015-02-02 17:55:39 +01:00
from searx.engines.xpath import extract_text
2013-10-14 23:09:13 +02:00
# engine dependent config
categories = ['general']
paging = True
language_support = True
supported_languages_url = 'https://duckduckgo.com/d2030.js'
2016-07-18 16:15:37 +02:00
time_range_support = True
2013-10-14 23:09:13 +02:00
# search-url
url = 'https://duckduckgo.com/html?{query}&s={offset}'
2016-07-18 16:15:37 +02:00
time_range_url = '&df={range}'
time_range_dict = {'day': 'd',
'week': 'w',
'month': 'm'}
# specific xpath variables
result_xpath = '//div[@class="result results_links results_links_deep web-result "]' # noqa
url_xpath = './/a[@class="result__a"]/@href'
title_xpath = './/a[@class="result__a"]'
content_xpath = './/a[@class="result__snippet"]'
2014-03-29 16:38:45 +01:00
# do search-request
2013-10-14 23:09:13 +02:00
def request(query, params):
if params['time_range'] and params['time_range'] not in time_range_dict:
return params
2014-01-30 01:03:19 +01:00
offset = (params['pageno'] - 1) * 30
# custom fixes for languages
if params['language'] == 'all':
locale = None
elif params['language'][:2] == 'ja':
locale = 'jp-jp'
2016-12-14 06:51:15 +01:00
elif params['language'][:2] == 'sl':
locale = 'sl-sl'
elif params['language'] == 'zh-TW':
locale = 'tw-tzh'
elif params['language'] == 'zh-HK':
locale = 'hk-tzh'
elif params['language'][-2:] == 'SA':
2016-12-14 06:51:15 +01:00
locale = 'xa-' + params['language'].split('-')[0]
elif params['language'][-2:] == 'GB':
2016-12-14 06:51:15 +01:00
locale = 'uk-' + params['language'].split('-')[0]
else:
locale = params['language'].split('-')
if len(locale) == 2:
# country code goes first
locale = locale[1].lower() + '-' + locale[0].lower()
else:
# tries to get a country code from language
locale = locale[0].lower()
2016-10-30 03:04:01 +01:00
for lc in supported_languages:
lc = lc.split('-')
if locale == lc[0]:
locale = lc[1].lower() + '-' + lc[0].lower()
break
if locale:
params['url'] = url.format(
query=urlencode({'q': query, 'kl': locale}), offset=offset)
else:
params['url'] = url.format(
query=urlencode({'q': query}), offset=offset)
2016-07-26 00:22:05 +02:00
if params['time_range'] in time_range_dict:
2016-07-18 16:15:37 +02:00
params['url'] += time_range_url.format(range=time_range_dict[params['time_range']])
2013-10-14 23:09:13 +02:00
return params
# get response from search-request
2013-10-14 23:09:13 +02:00
def response(resp):
2013-10-15 19:11:43 +02:00
results = []
2014-03-21 16:33:17 +01:00
doc = fromstring(resp.text)
2014-09-02 18:12:42 +02:00
# parse results
2014-03-21 16:33:17 +01:00
for r in doc.xpath(result_xpath):
2014-03-21 18:17:13 +01:00
try:
res_url = r.xpath(url_xpath)[-1]
except:
continue
2014-03-21 16:33:17 +01:00
if not res_url:
2013-10-15 19:11:43 +02:00
continue
2015-02-02 17:55:39 +01:00
title = extract_text(r.xpath(title_xpath))
content = extract_text(r.xpath(content_xpath))
# append result
2014-03-21 16:33:17 +01:00
results.append({'title': title,
'content': content,
2015-09-07 23:13:04 +02:00
'url': res_url})
2014-03-21 16:33:17 +01:00
# return results
2013-10-15 19:11:43 +02:00
return results
# get supported languages from their site
def _fetch_supported_languages(resp):
# response is a js file with regions as an embedded object
response_page = resp.text
response_page = response_page[response_page.find('regions:{') + 8:]
response_page = response_page[:response_page.find('}') + 1]
regions_json = loads(response_page)
supported_languages = map((lambda x: x[3:] + '-' + x[:2].upper()), regions_json.keys())
return supported_languages