git.iko.soy/searx
on master help sign in
Merge pull request #88 from pointhi/engines
f36d1e2Adam Tauber
searx/engines/bing.py+39−6
compare with master
unifiedsplit
@@ -1,48 +1,81 @@
1+## Bing (Web)
2+#
3+# @website https://www.bing.com
4+# @provide-api yes (http://datamarket.azure.com/dataset/bing/search), max. 5000 query/month
5+#
6+# @using-api no (because of query limit)
7+# @results HTML (using search portal)
8+# @stable no (HTML can change)
9+# @parse url, title, content
10+#
11+# @todo publishedDate
12+
1from urllib import urlencode
13from urllib import urlencode
2from cgi import escape
14from cgi import escape
3from lxml import html
15from lxml import html
4
16
5−base_url = 'http://www.bing.com/'
17+# engine dependent config
6−search_string = 'search?{query}&first={offset}'
18+categories = ['general']
7paging = True
19paging = True
8language_support = True
20language_support = True
9
21
22+# search-url
23+base_url = 'https://www.bing.com/'
24+search_string = 'search?{query}&first={offset}'
10
25
26+# do search-request
11def request(query, params):
27def request(query, params):
12 offset = (params['pageno'] - 1) * 10 + 1
28 offset = (params['pageno'] - 1) * 10 + 1
29+
13 if params['language'] == 'all':
30 if params['language'] == 'all':
14 language = 'en-US'
31 language = 'en-US'
15 else:
32 else:
16 language = params['language'].replace('_', '-')
33 language = params['language'].replace('_', '-')
34+
17 search_path = search_string.format(
35 search_path = search_string.format(
18 query=urlencode({'q': query, 'setmkt': language}),
36 query=urlencode({'q': query, 'setmkt': language}),
19 offset=offset)
37 offset=offset)
20
38
21 params['cookies']['SRCHHPGUSR'] = \
39 params['cookies']['SRCHHPGUSR'] = \
22 'NEWWND=0&NRSLT=-1&SRCHLANG=' + language.split('-')[0]
40 'NEWWND=0&NRSLT=-1&SRCHLANG=' + language.split('-')[0]
23− #if params['category'] == 'images':
41+
24− # params['url'] = base_url + 'images/' + search_path
25 params['url'] = base_url + search_path
42 params['url'] = base_url + search_path
26 return params
43 return params
27
44
28
45
46+# get response from search-request
29def response(resp):
47def response(resp):
30 results = []
48 results = []
49+
31 dom = html.fromstring(resp.content)
50 dom = html.fromstring(resp.content)
51+
52+ # parse results
32 for result in dom.xpath('//div[@class="sa_cc"]'):
53 for result in dom.xpath('//div[@class="sa_cc"]'):
33 link = result.xpath('.//h3/a')[0]
54 link = result.xpath('.//h3/a')[0]
34 url = link.attrib.get('href')
55 url = link.attrib.get('href')
35 title = ' '.join(link.xpath('.//text()'))
56 title = ' '.join(link.xpath('.//text()'))
36 content = escape(' '.join(result.xpath('.//p//text()')))
57 content = escape(' '.join(result.xpath('.//p//text()')))
37− results.append({'url': url, 'title': title, 'content': content})
38
58
59+ # append result
60+ results.append({'url': url,
61+ 'title': title,
62+ 'content': content})
63+
64+ # return results if something is found
39 if results:
65 if results:
40 return results
66 return results
41
67
68+ # parse results again if nothing is found yet
42 for result in dom.xpath('//li[@class="b_algo"]'):
69 for result in dom.xpath('//li[@class="b_algo"]'):
43 link = result.xpath('.//h2/a')[0]
70 link = result.xpath('.//h2/a')[0]
44 url = link.attrib.get('href')
71 url = link.attrib.get('href')
45 title = ' '.join(link.xpath('.//text()'))
72 title = ' '.join(link.xpath('.//text()'))
46 content = escape(' '.join(result.xpath('.//p//text()')))
73 content = escape(' '.join(result.xpath('.//p//text()')))
74+
75+ # append result
47− results.append({'url': url, 'title': title, 'content': content})
76+ results.append({'url': url,
77+ 'title': title,
78+ 'content': content})
79+
80+ # return results
48 return results
81 return results
searx/engines/bing_images.pynew file+81
@@ -0,0 +1,81 @@
1+## Bing (Images)
2+#
3+# @website https://www.bing.com/images
4+# @provide-api yes (http://datamarket.azure.com/dataset/bing/search), max. 5000 query/month
5+#
6+# @using-api no (because of query limit)
7+# @results HTML (using search portal)
8+# @stable no (HTML can change)
9+# @parse url, title, img_src
10+#
11+# @todo currently there are up to 35 images receive per page, because bing does not parse count=10. limited response to 10 images
12+
13+from urllib import urlencode
14+from cgi import escape
15+from lxml import html
16+from yaml import load
17+import re
18+
19+# engine dependent config
20+categories = ['images']
21+paging = True
22+
23+# search-url
24+base_url = 'https://www.bing.com/'
25+search_string = 'images/search?{query}&count=10&first={offset}'
26+
27+# do search-request
28+def request(query, params):
29+ offset = (params['pageno'] - 1) * 10 + 1
30+
31+ # required for cookie
32+ language = 'en-US'
33+
34+ search_path = search_string.format(
35+ query=urlencode({'q': query}),
36+ offset=offset)
37+
38+ params['cookies']['SRCHHPGUSR'] = \
39+ 'NEWWND=0&NRSLT=-1&SRCHLANG=' + language.split('-')[0]
40+
41+ params['url'] = base_url + search_path
42+
43+ print(params['url'])
44+
45+ return params
46+
47+
48+# get response from search-request
49+def response(resp):
50+ results = []
51+
52+ dom = html.fromstring(resp.content)
53+
54+ # init regex for yaml-parsing
55+ p = re.compile( '({|,)([a-z]+):(")')
56+
57+ # parse results
58+ for result in dom.xpath('//div[@class="dg_u"]'):
59+ link = result.xpath('./a')[0]
60+
61+ # parse yaml-data (it is required to add a space, to make it parsable)
62+ yaml_data = load(p.sub( r'\1\2: \3', link.attrib.get('m')))
63+
64+ title = link.attrib.get('t1')
65+ #url = 'http://' + link.attrib.get('t3')
66+ url = yaml_data.get('surl')
67+ img_src = yaml_data.get('imgurl')
68+
69+ # append result
70+ results.append({'template': 'images.html',
71+ 'url': url,
72+ 'title': title,
73+ 'content': '',
74+ 'img_src': img_src})
75+
76+ # TODO stop parsing if 10 images are found
77+ if len(results) >= 10:
78+ break
79+
80+ # return results
81+ return results
searx/engines/bing_news.py+53−17
@@ -1,50 +1,86 @@
1+## Bing (News)
2+#
3+# @website https://www.bing.com/news
4+# @provide-api yes (http://datamarket.azure.com/dataset/bing/search), max. 5000 query/month
5+#
6+# @using-api no (because of query limit)
7+# @results HTML (using search portal)
8+# @stable no (HTML can change)
9+# @parse url, title, content, publishedDate
10+
1from urllib import urlencode
11from urllib import urlencode
2from cgi import escape
12from cgi import escape
3from lxml import html
13from lxml import html
14+from datetime import datetime, timedelta
15+from dateutil import parser
16+import re
4
17
18+# engine dependent config
5categories = ['news']
19categories = ['news']
6−
7−base_url = 'http://www.bing.com/'
8−search_string = 'news/search?{query}&first={offset}'
9paging = True
20paging = True
10language_support = True
21language_support = True
11
22
23+# search-url
24+base_url = 'https://www.bing.com/'
25+search_string = 'news/search?{query}&first={offset}'
12
26
27+# do search-request
13def request(query, params):
28def request(query, params):
14 offset = (params['pageno'] - 1) * 10 + 1
29 offset = (params['pageno'] - 1) * 10 + 1
30+
15 if params['language'] == 'all':
31 if params['language'] == 'all':
16 language = 'en-US'
32 language = 'en-US'
17 else:
33 else:
18 language = params['language'].replace('_', '-')
34 language = params['language'].replace('_', '-')
35+
19 search_path = search_string.format(
36 search_path = search_string.format(
20 query=urlencode({'q': query, 'setmkt': language}),
37 query=urlencode({'q': query, 'setmkt': language}),
21 offset=offset)
38 offset=offset)
22
39
23 params['cookies']['SRCHHPGUSR'] = \
40 params['cookies']['SRCHHPGUSR'] = \
24 'NEWWND=0&NRSLT=-1&SRCHLANG=' + language.split('-')[0]
41 'NEWWND=0&NRSLT=-1&SRCHLANG=' + language.split('-')[0]
25− #if params['category'] == 'images':
42+
26− # params['url'] = base_url + 'images/' + search_path
27 params['url'] = base_url + search_path
43 params['url'] = base_url + search_path
28 return params
44 return params
29
45
30
46
47+# get response from search-request
31def response(resp):
48def response(resp):
32 results = []
49 results = []
50+
33 dom = html.fromstring(resp.content)
51 dom = html.fromstring(resp.content)
52+
53+ # parse results
34− for result in dom.xpath('//div[@class="sa_cc"]'):
54+ for result in dom.xpath('//div[@class="sn_r"]'):
35− link = result.xpath('.//h3/a')[0]
55+ link = result.xpath('.//div[@class="newstitle"]/a')[0]
36 url = link.attrib.get('href')
56 url = link.attrib.get('href')
37 title = ' '.join(link.xpath('.//text()'))
57 title = ' '.join(link.xpath('.//text()'))
38− content = escape(' '.join(result.xpath('.//p//text()')))
58+ content = escape(' '.join(result.xpath('.//div[@class="sn_txt"]/div//span[@class="sn_snip"]//text()')))
39− results.append({'url': url, 'title': title, 'content': content})
59+
60+ # parse publishedDate
61+ publishedDate = escape(' '.join(result.xpath('.//div[@class="sn_txt"]/div//span[@class="sn_ST"]//span[@class="sn_tm"]//text()')))
40
62
41− if results:
63+ if re.match("^[0-9]+ minute(s|) ago$", publishedDate):
42− return results
64+ timeNumbers = re.findall(r'\d+', publishedDate)
65+ publishedDate = datetime.now()\
66+ - timedelta(minutes=int(timeNumbers[0]))
67+ elif re.match("^[0-9]+ hour(s|) ago$", publishedDate):
68+ timeNumbers = re.findall(r'\d+', publishedDate)
69+ publishedDate = datetime.now()\
70+ - timedelta(hours=int(timeNumbers[0]))
71+ elif re.match("^[0-9]+ hour(s|), [0-9]+ minute(s|) ago$", publishedDate):
72+ timeNumbers = re.findall(r'\d+', publishedDate)
73+ publishedDate = datetime.now()\
74+ - timedelta(hours=int(timeNumbers[0]))\
75+ - timedelta(minutes=int(timeNumbers[1]))
76+ else:
77+ publishedDate = parser.parse(publishedDate)
43
78
44− for result in dom.xpath('//li[@class="b_algo"]'):
79+ # append result
45− link = result.xpath('.//h2/a')[0]
46− url = link.attrib.get('href')
47− title = ' '.join(link.xpath('.//text()'))
48− content = escape(' '.join(result.xpath('.//p//text()')))
49− results.append({'url': url, 'title': title, 'content': content})
80+ results.append({'url': url,
81+ 'title': title,
82+ 'publishedDate': publishedDate,
83+ 'content': content})
84+
85+ # return results
50 return results
86 return results
searx/engines/dailymotion.py+35−19
@@ -1,45 +1,61 @@
1+## Dailymotion (Videos)
2+#
3+# @website https://www.dailymotion.com
4+# @provide-api yes (http://www.dailymotion.com/developer)
5+#
6+# @using-api yes
7+# @results JSON
8+# @stable yes
9+# @parse url, title, thumbnail
10+#
11+# @todo set content-parameter with correct data
12+
1from urllib import urlencode
13from urllib import urlencode
2from json import loads
14from json import loads
3from lxml import html
15from lxml import html
4
16
17+# engine dependent config
5categories = ['videos']
18categories = ['videos']
6locale = 'en_US'
19locale = 'en_US'
20+paging = True
7
21
22+# search-url
8# see http://www.dailymotion.com/doc/api/obj-video.html
23# see http://www.dailymotion.com/doc/api/obj-video.html
9−search_url = 'https://api.dailymotion.com/videos?fields=title,description,duration,url,thumbnail_360_url&sort=relevance&limit=25&page={pageno}&{query}' # noqa
24+search_url = 'https://api.dailymotion.com/videos?fields=title,description,duration,url,thumbnail_360_url&sort=relevance&limit=5&page={pageno}&{query}' # noqa
10−
11−# TODO use video result template
12−content_tpl = '<a href="{0}" title="{0}" ><img src="{1}" /></a><br />'
13−
14−paging = True
15
25
16
26
27+# do search-request
17def request(query, params):
28def request(query, params):
18 params['url'] = search_url.format(
29 params['url'] = search_url.format(
19 query=urlencode({'search': query, 'localization': locale}),
30 query=urlencode({'search': query, 'localization': locale}),
20 pageno=params['pageno'])
31 pageno=params['pageno'])
32+
21 return params
33 return params
22
34
23
35
36+# get response from search-request
24def response(resp):
37def response(resp):
25 results = []
38 results = []
39+
26 search_res = loads(resp.text)
40 search_res = loads(resp.text)
41+
42+ # return empty array if there are no results
27 if not 'list' in search_res:
43 if not 'list' in search_res:
28− return results
44+ return []
45+
46+ # parse results
29 for res in search_res['list']:
47 for res in search_res['list']:
30 title = res['title']
48 title = res['title']
31 url = res['url']
49 url = res['url']
32− if res['thumbnail_360_url']:
50+ #content = res['description']
33− content = content_tpl.format(url, res['thumbnail_360_url'])
34− else:
35− content = ''
51+ content = ''
36− if res['description']:
52+ thumbnail = res['thumbnail_360_url']
37− description = text_content_from_html(res['description'])
38− content += description[:500]
39− results.append({'url': url, 'title': title, 'content': content})
40− return results
41
53
54+ results.append({'template': 'videos.html',
55+ 'url': url,
56+ 'title': title,
57+ 'content': content,
58+ 'thumbnail': thumbnail})
42
59
43−def text_content_from_html(html_string):
60+ # return results
44− desc_html = html.fragment_fromstring(html_string, create_parent=True)
61+ return results
45− return desc_html.text_content()
searx/engines/google.py+24−4
@@ -1,37 +1,57 @@
1−#!/usr/bin/env python
1+## Google (Web)
2+#
3+# @website https://www.google.com
4+# @provide-api yes (https://developers.google.com/web-search/docs/), deprecated!
5+#
6+# @using-api yes
7+# @results JSON
8+# @stable yes (but deprecated)
9+# @parse url, title, content
2
10
3from urllib import urlencode
11from urllib import urlencode
4from json import loads
12from json import loads
5
13
14+# engine dependent config
6categories = ['general']
15categories = ['general']
16+paging = True
17+language_support = True
7
18
19+# search-url
8url = 'https://ajax.googleapis.com/'
20url = 'https://ajax.googleapis.com/'
9search_url = url + 'ajax/services/search/web?v=2.0&start={offset}&rsz=large&safe=off&filter=off&{query}&hl={language}' # noqa
21search_url = url + 'ajax/services/search/web?v=2.0&start={offset}&rsz=large&safe=off&filter=off&{query}&hl={language}' # noqa
10
22
11−paging = True
12−language_support = True
13−
14
23
24+# do search-request
15def request(query, params):
25def request(query, params):
16 offset = (params['pageno'] - 1) * 8
26 offset = (params['pageno'] - 1) * 8
27+
17 language = 'en-US'
28 language = 'en-US'
18 if params['language'] != 'all':
29 if params['language'] != 'all':
19 language = params['language'].replace('_', '-')
30 language = params['language'].replace('_', '-')
31+
20 params['url'] = search_url.format(offset=offset,
32 params['url'] = search_url.format(offset=offset,
21 query=urlencode({'q': query}),
33 query=urlencode({'q': query}),
22 language=language)
34 language=language)
35+
23 return params
36 return params
24
37
25
38
39+# get response from search-request
26def response(resp):
40def response(resp):
27 results = []
41 results = []
42+
28 search_res = loads(resp.text)
43 search_res = loads(resp.text)
29
44
45+ # return empty array if there are no results
30 if not search_res.get('responseData', {}).get('results'):
46 if not search_res.get('responseData', {}).get('results'):
31 return []
47 return []
32
48
49+ # parse results
33 for result in search_res['responseData']['results']:
50 for result in search_res['responseData']['results']:
51+ # append result
34 results.append({'url': result['unescapedUrl'],
52 results.append({'url': result['unescapedUrl'],
35 'title': result['titleNoFormatting'],
53 'title': result['titleNoFormatting'],
36 'content': result['content']})
54 'content': result['content']})
55+
56+ # return results
37 return results
57 return results
searx/engines/google_images.py+26−5
@@ -1,37 +1,58 @@
1−#!/usr/bin/env python
1+## Google (Images)
2+#
3+# @website https://www.google.com
4+# @provide-api yes (https://developers.google.com/web-search/docs/), deprecated!
5+#
6+# @using-api yes
7+# @results JSON
8+# @stable yes (but deprecated)
9+# @parse url, title, img_src
2
10
3from urllib import urlencode
11from urllib import urlencode
4from json import loads
12from json import loads
5
13
14+# engine dependent config
6categories = ['images']
15categories = ['images']
16+paging = True
7
17
18+# search-url
8url = 'https://ajax.googleapis.com/'
19url = 'https://ajax.googleapis.com/'
9search_url = url + 'ajax/services/search/images?v=1.0&start={offset}&rsz=large&safe=off&filter=off&{query}' # noqa
20search_url = url + 'ajax/services/search/images?v=1.0&start={offset}&rsz=large&safe=off&filter=off&{query}' # noqa
10
21
11−paging = True
12
22
23+# do search-request
13def request(query, params):
24def request(query, params):
14 offset = (params['pageno'] - 1) * 8
25 offset = (params['pageno'] - 1) * 8
26+
15 params['url'] = search_url.format(query=urlencode({'q': query}),
27 params['url'] = search_url.format(query=urlencode({'q': query}),
16 offset=offset)
28 offset=offset)
29+
17 return params
30 return params
18
31
19
32
33+# get response from search-request
20def response(resp):
34def response(resp):
21 results = []
35 results = []
36+
22 search_res = loads(resp.text)
37 search_res = loads(resp.text)
23− if not search_res.get('responseData'):
38+
24− return []
39+ # return empty array if there are no results
25− if not search_res['responseData'].get('results'):
40+ if not search_res.get('responseData', {}).get('results'):
26 return []
41 return []
42+
43+ # parse results
27 for result in search_res['responseData']['results']:
44 for result in search_res['responseData']['results']:
28 href = result['originalContextUrl']
45 href = result['originalContextUrl']
29 title = result['title']
46 title = result['title']
30 if not result['url']:
47 if not result['url']:
31 continue
48 continue
49+
50+ # append result
32 results.append({'url': href,
51 results.append({'url': href,
33 'title': title,
52 'title': title,
34 'content': '',
53 'content': '',
35 'img_src': result['url'],
54 'img_src': result['url'],
36 'template': 'images.html'})
55 'template': 'images.html'})
56+
57+ # return results
37 return results
58 return results
searx/engines/google_news.py+25−6
@@ -1,43 +1,62 @@
1−#!/usr/bin/env python
1+## Google (News)
2+#
3+# @website https://www.google.com
4+# @provide-api yes (https://developers.google.com/web-search/docs/), deprecated!
5+#
6+# @using-api yes
7+# @results JSON
8+# @stable yes (but deprecated)
9+# @parse url, title, content, publishedDate
2
10
3from urllib import urlencode
11from urllib import urlencode
4from json import loads
12from json import loads
5from dateutil import parser
13from dateutil import parser
6
14
15+# search-url
7categories = ['news']
16categories = ['news']
17+paging = True
18+language_support = True
8
19
20+# engine dependent config
9url = 'https://ajax.googleapis.com/'
21url = 'https://ajax.googleapis.com/'
10search_url = url + 'ajax/services/search/news?v=2.0&start={offset}&rsz=large&safe=off&filter=off&{query}&hl={language}' # noqa
22search_url = url + 'ajax/services/search/news?v=2.0&start={offset}&rsz=large&safe=off&filter=off&{query}&hl={language}' # noqa
11
23
12−paging = True
13−language_support = True
14−
15
24
25+# do search-request
16def request(query, params):
26def request(query, params):
17 offset = (params['pageno'] - 1) * 8
27 offset = (params['pageno'] - 1) * 8
28+
18 language = 'en-US'
29 language = 'en-US'
19 if params['language'] != 'all':
30 if params['language'] != 'all':
20 language = params['language'].replace('_', '-')
31 language = params['language'].replace('_', '-')
32+
21 params['url'] = search_url.format(offset=offset,
33 params['url'] = search_url.format(offset=offset,
22 query=urlencode({'q': query}),
34 query=urlencode({'q': query}),
23 language=language)
35 language=language)
36+
24 return params
37 return params
25
38
26
39
40+# get response from search-request
27def response(resp):
41def response(resp):
28 results = []
42 results = []
43+
29 search_res = loads(resp.text)
44 search_res = loads(resp.text)
30
45
46+ # return empty array if there are no results
31 if not search_res.get('responseData', {}).get('results'):
47 if not search_res.get('responseData', {}).get('results'):
32 return []
48 return []
33
49
50+ # parse results
34 for result in search_res['responseData']['results']:
51 for result in search_res['responseData']['results']:
35−
52+ # parse publishedDate
36−# Mon, 10 Mar 2014 16:26:15 -0700
37 publishedDate = parser.parse(result['publishedDate'])
53 publishedDate = parser.parse(result['publishedDate'])
38
54
55+ # append result
39 results.append({'url': result['unescapedUrl'],
56 results.append({'url': result['unescapedUrl'],
40 'title': result['titleNoFormatting'],
57 'title': result['titleNoFormatting'],
41 'publishedDate': publishedDate,
58 'publishedDate': publishedDate,
42 'content': result['content']})
59 'content': result['content']})
60+
61+ # return results
43 return results
62 return results
searx/engines/vimeo.py+40−22
@@ -1,43 +1,58 @@
1+## Vimeo (Videos)
2+#
3+# @website https://vimeo.com/
4+# @provide-api yes (http://developer.vimeo.com/api), they have a maximum count of queries/hour
5+#
6+# @using-api no (TODO, rewrite to api)
7+# @results HTML (using search portal)
8+# @stable no (HTML can change)
9+# @parse url, title, publishedDate, thumbnail
10+#
11+# @todo rewrite to api
12+# @todo set content-parameter with correct data
13+
1from urllib import urlencode
14from urllib import urlencode
2from HTMLParser import HTMLParser
15from HTMLParser import HTMLParser
3from lxml import html
16from lxml import html
4from searx.engines.xpath import extract_text
17from searx.engines.xpath import extract_text
5from dateutil import parser
18from dateutil import parser
6
19
7−base_url = 'http://vimeo.com'
20+# engine dependent config
8−search_url = base_url + '/search?{query}'
21+categories = ['videos']
9−url_xpath = None
22+paging = True
10−content_xpath = None
11−title_xpath = None
12−results_xpath = ''
13−content_tpl = '<a href="{0}"> <img src="{2}"/> </a>'
14−publishedDate_xpath = './/p[@class="meta"]//attribute::datetime'
15
23
16−# the cookie set by vimeo contains all the following values,
24+# search-url
17−# but only __utma seems to be requiered
25+base_url = 'https://vimeo.com'
18−cookie = {
26+search_url = base_url + '/search/page:{pageno}?{query}'
19− #'vuid':'918282893.1027205400'
27+
20− # 'ab_bs':'%7B%223%22%3A279%7D'
28+# specific xpath variables
21− '__utma': '00000000.000#0000000.0000000000.0000000000.0000000000.0'
29+url_xpath = './a/@href'
22− # '__utmb':'18302654.1.10.1388942090'
30+content_xpath = './a/img/@src'
23− #, '__utmc':'18302654'
31+title_xpath = './a/div[@class="data"]/p[@class="title"]/text()'
24− #, '__utmz':'18#302654.1388942090.1.1.utmcsr=(direct)|utmccn=(direct)|utmcmd=(none)' # noqa
32+results_xpath = '//div[@id="browse_content"]/ol/li'
25− #, '__utml':'search'
33+publishedDate_xpath = './/p[@class="meta"]//attribute::datetime'
26−}
27
34
28
35
36+# do search-request
29def request(query, params):
37def request(query, params):
38+ params['url'] = search_url.format(pageno=params['pageno'] ,
30− params['url'] = search_url.format(query=urlencode({'q': query}))
39+ query=urlencode({'q': query}))
31− params['cookies'] = cookie
40+
41+ # TODO required?
42+ params['cookies']['__utma'] = '00000000.000#0000000.0000000000.0000000000.0000000000.0'
43+
32 return params
44 return params
33
45
34
46
47+# get response from search-request
35def response(resp):
48def response(resp):
36 results = []
49 results = []
50+
37 dom = html.fromstring(resp.text)
51 dom = html.fromstring(resp.text)
38
52
39 p = HTMLParser()
53 p = HTMLParser()
40
54
55+ # parse results
41 for result in dom.xpath(results_xpath):
56 for result in dom.xpath(results_xpath):
42 url = base_url + result.xpath(url_xpath)[0]
57 url = base_url + result.xpath(url_xpath)[0]
43 title = p.unescape(extract_text(result.xpath(title_xpath)))
58 title = p.unescape(extract_text(result.xpath(title_xpath)))
@@ -45,10 +60,13 @@ def response(resp):
45 publishedDate = parser.parse(extract_text(
60 publishedDate = parser.parse(extract_text(
46 result.xpath(publishedDate_xpath)[0]))
61 result.xpath(publishedDate_xpath)[0]))
47
62
63+ # append result
48 results.append({'url': url,
64 results.append({'url': url,
49 'title': title,
65 'title': title,
50− 'content': content_tpl.format(url, title, thumbnail),
66+ 'content': '',
51 'template': 'videos.html',
67 'template': 'videos.html',
52 'publishedDate': publishedDate,
68 'publishedDate': publishedDate,
53 'thumbnail': thumbnail})
69 'thumbnail': thumbnail})
70+
71+ # return results
54 return results
72 return results
searx/engines/yahoo.py+41−6
@@ -1,64 +1,99 @@
1−#!/usr/bin/env python
1+## Yahoo (Web)
2+#
3+# @website https://search.yahoo.com/web
4+# @provide-api yes (https://developer.yahoo.com/boss/search/), $0.80/1000 queries
5+#
6+# @using-api no (because pricing)
7+# @results HTML (using search portal)
8+# @stable no (HTML can change)
9+# @parse url, title, content, suggestion
2
10
3from urllib import urlencode
11from urllib import urlencode
4from urlparse import unquote
12from urlparse import unquote
5from lxml import html
13from lxml import html
6from searx.engines.xpath import extract_text, extract_url
14from searx.engines.xpath import extract_text, extract_url
7
15
16+# engine dependent config
8categories = ['general']
17categories = ['general']
18+paging = True
19+language_support = True
20+
21+# search-url
9−search_url = 'http://search.yahoo.com/search?{query}&b={offset}'
22+search_url = 'https://search.yahoo.com/search?{query}&b={offset}&fl=1&vl=lang_{lang}'
23+
24+# specific xpath variables
10results_xpath = '//div[@class="res"]'
25results_xpath = '//div[@class="res"]'
11url_xpath = './/h3/a/@href'
26url_xpath = './/h3/a/@href'
12title_xpath = './/h3/a'
27title_xpath = './/h3/a'
13content_xpath = './/div[@class="abstr"]'
28content_xpath = './/div[@class="abstr"]'
14suggestion_xpath = '//div[@id="satat"]//a'
29suggestion_xpath = '//div[@id="satat"]//a'
15
30
16−paging = True
17−
18
31
32+# remove yahoo-specific tracking-url
19def parse_url(url_string):
33def parse_url(url_string):
20 endings = ['/RS', '/RK']
34 endings = ['/RS', '/RK']
21 endpositions = []
35 endpositions = []
22 start = url_string.find('http', url_string.find('/RU=')+1)
36 start = url_string.find('http', url_string.find('/RU=')+1)
37+
23 for ending in endings:
38 for ending in endings:
24 endpos = url_string.rfind(ending)
39 endpos = url_string.rfind(ending)
25 if endpos > -1:
40 if endpos > -1:
26 endpositions.append(endpos)
41 endpositions.append(endpos)
27
42
28 end = min(endpositions)
43 end = min(endpositions)
44+
29 return unquote(url_string[start:end])
45 return unquote(url_string[start:end])
30
46
31
47
48+# do search-request
32def request(query, params):
49def request(query, params):
33 offset = (params['pageno'] - 1) * 10 + 1
50 offset = (params['pageno'] - 1) * 10 + 1
51+
34 if params['language'] == 'all':
52 if params['language'] == 'all':
35 language = 'en'
53 language = 'en'
36 else:
54 else:
37 language = params['language'].split('_')[0]
55 language = params['language'].split('_')[0]
56+
38 params['url'] = search_url.format(offset=offset,
57 params['url'] = search_url.format(offset=offset,
39− query=urlencode({'p': query}))
58+ query=urlencode({'p': query}),
59+ lang=language)
60+
61+ # TODO required?
40 params['cookies']['sB'] = 'fl=1&vl=lang_{lang}&sh=1&rw=new&v=1'\
62 params['cookies']['sB'] = 'fl=1&vl=lang_{lang}&sh=1&rw=new&v=1'\
41 .format(lang=language)
63 .format(lang=language)
64+
42 return params
65 return params
43
66
44
67
68+# get response from search-request
45def response(resp):
69def response(resp):
46 results = []
70 results = []
71+
47 dom = html.fromstring(resp.text)
72 dom = html.fromstring(resp.text)
48
73
74+ # parse results
49 for result in dom.xpath(results_xpath):
75 for result in dom.xpath(results_xpath):
50 try:
76 try:
51 url = parse_url(extract_url(result.xpath(url_xpath), search_url))
77 url = parse_url(extract_url(result.xpath(url_xpath), search_url))
52 title = extract_text(result.xpath(title_xpath)[0])
78 title = extract_text(result.xpath(title_xpath)[0])
53 except:
79 except:
54 continue
80 continue
81+
55 content = extract_text(result.xpath(content_xpath)[0])
82 content = extract_text(result.xpath(content_xpath)[0])
56− results.append({'url': url, 'title': title, 'content': content})
57
83
84+ # append result
85+ results.append({'url': url,
86+ 'title': title,
87+ 'content': content})
88+
89+ # if no suggestion found, return results
58 if not suggestion_xpath:
90 if not suggestion_xpath:
59 return results
91 return results
60
92
93+ # parse suggestion
61 for suggestion in dom.xpath(suggestion_xpath):
94 for suggestion in dom.xpath(suggestion_xpath):
95+ # append suggestion
62 results.append({'suggestion': extract_text(suggestion)})
96 results.append({'suggestion': extract_text(suggestion)})
63
97
98+ # return results
64 return results
99 return results
searx/engines/yahoo_news.py+31−11
@@ -1,4 +1,12 @@
1−#!/usr/bin/env python
1+## Yahoo (News)
2+#
3+# @website https://news.yahoo.com
4+# @provide-api yes (https://developer.yahoo.com/boss/search/), $0.80/1000 queries
5+#
6+# @using-api no (because pricing)
7+# @results HTML (using search portal)
8+# @stable no (HTML can change)
9+# @parse url, title, content, publishedDate
2
10
3from urllib import urlencode
11from urllib import urlencode
4from lxml import html
12from lxml import html
@@ -8,8 +16,15 @@ from datetime import datetime, timedelta
8import re
16import re
9from dateutil import parser
17from dateutil import parser
10
18
19+# engine dependent config
11categories = ['news']
20categories = ['news']
21+paging = True
22+language_support = True
23+
24+# search-url
12−search_url = 'http://news.search.yahoo.com/search?{query}&b={offset}'
25+search_url = 'https://news.search.yahoo.com/search?{query}&b={offset}&fl=1&vl=lang_{lang}'
26+
27+# specific xpath variables
13results_xpath = '//div[@class="res"]'
28results_xpath = '//div[@class="res"]'
14url_xpath = './/h3/a/@href'
29url_xpath = './/h3/a/@href'
15title_xpath = './/h3/a'
30title_xpath = './/h3/a'
@@ -17,30 +32,39 @@ content_xpath = './/div[@class="abstr"]'
17publishedDate_xpath = './/span[@class="timestamp"]'
32publishedDate_xpath = './/span[@class="timestamp"]'
18suggestion_xpath = '//div[@id="satat"]//a'
33suggestion_xpath = '//div[@id="satat"]//a'
19
34
20−paging = True
21−
22
35
36+# do search-request
23def request(query, params):
37def request(query, params):
24 offset = (params['pageno'] - 1) * 10 + 1
38 offset = (params['pageno'] - 1) * 10 + 1
39+
25 if params['language'] == 'all':
40 if params['language'] == 'all':
26 language = 'en'
41 language = 'en'
27 else:
42 else:
28 language = params['language'].split('_')[0]
43 language = params['language'].split('_')[0]
44+
29 params['url'] = search_url.format(offset=offset,
45 params['url'] = search_url.format(offset=offset,
30− query=urlencode({'p': query}))
46+ query=urlencode({'p': query}),
47+ lang=language)
48+
49+ # TODO required?
31 params['cookies']['sB'] = 'fl=1&vl=lang_{lang}&sh=1&rw=new&v=1'\
50 params['cookies']['sB'] = 'fl=1&vl=lang_{lang}&sh=1&rw=new&v=1'\
32 .format(lang=language)
51 .format(lang=language)
33 return params
52 return params
34
53
35
54
55+# get response from search-request
36def response(resp):
56def response(resp):
37 results = []
57 results = []
58+
38 dom = html.fromstring(resp.text)
59 dom = html.fromstring(resp.text)
39
60
61+ # parse results
40 for result in dom.xpath(results_xpath):
62 for result in dom.xpath(results_xpath):
41 url = parse_url(extract_url(result.xpath(url_xpath), search_url))
63 url = parse_url(extract_url(result.xpath(url_xpath), search_url))
42 title = extract_text(result.xpath(title_xpath)[0])
64 title = extract_text(result.xpath(title_xpath)[0])
43 content = extract_text(result.xpath(content_xpath)[0])
65 content = extract_text(result.xpath(content_xpath)[0])
66+
67+ # parse publishedDate
44 publishedDate = extract_text(result.xpath(publishedDate_xpath)[0])
68 publishedDate = extract_text(result.xpath(publishedDate_xpath)[0])
45
69
46 if re.match("^[0-9]+ minute(s|) ago$", publishedDate):
70 if re.match("^[0-9]+ minute(s|) ago$", publishedDate):
@@ -58,15 +82,11 @@ def response(resp):
58 if publishedDate.year == 1900:
82 if publishedDate.year == 1900:
59 publishedDate = publishedDate.replace(year=datetime.now().year)
83 publishedDate = publishedDate.replace(year=datetime.now().year)
60
84
85+ # append result
61 results.append({'url': url,
86 results.append({'url': url,
62 'title': title,
87 'title': title,
63 'content': content,
88 'content': content,
64 'publishedDate': publishedDate})
89 'publishedDate': publishedDate})
65
90
66− if not suggestion_xpath:
67− return results
91+ # return results
68−
69− for suggestion in dom.xpath(suggestion_xpath):
70− results.append({'suggestion': extract_text(suggestion)})
71−
72 return results
92 return results
searx/settings.yml+6−5
@@ -20,6 +20,11 @@ engines:
20 locale : en-US
20 locale : en-US
21 shortcut : bi
21 shortcut : bi
22
22
23+ - name : bing images
24+ engine : bing_images
25+ locale : en-US
26+ shortcut : bii
27+
23 - name : bing news
28 - name : bing news
24 engine : bing_news
29 engine : bing_news
25 locale : en-US
30 locale : en-US
@@ -148,11 +153,7 @@ engines:
148
153
149 - name : vimeo
154 - name : vimeo
150 engine : vimeo
155 engine : vimeo
151− categories : videos
156+ locale : en-US
152− results_xpath : //div[@id="browse_content"]/ol/li
153− url_xpath : ./a/@href
154− title_xpath : ./a/div[@class="data"]/p[@class="title"]/text()
155− content_xpath : ./a/img/@src
156 shortcut : vm
157 shortcut : vm
157
158
158locales:
159locales: