git.iko.soy/searx
on master help sign in
Merge pull request #93 from dalf/master
1e99cf2Adam Tauber
searx/engines/bing_news.py+18−5
compare with master
unifiedsplit
@@ -56,10 +56,14 @@ def response(resp):
56 link = result.xpath('.//div[@class="newstitle"]/a')[0]
56 link = result.xpath('.//div[@class="newstitle"]/a')[0]
57 url = link.attrib.get('href')
57 url = link.attrib.get('href')
58 title = ' '.join(link.xpath('.//text()'))
58 title = ' '.join(link.xpath('.//text()'))
59− content = escape(' '.join(result.xpath('.//div[@class="sn_txt"]/div//span[@class="sn_snip"]//text()')))
59+ contentXPath = result.xpath('.//div[@class="sn_txt"]/div//span[@class="sn_snip"]//text()')
60−
60+ if contentXPath != None:
61+ content = escape(' '.join(contentXPath))
62+
61 # parse publishedDate
63 # parse publishedDate
62− publishedDate = escape(' '.join(result.xpath('.//div[@class="sn_txt"]/div//span[@class="sn_ST"]//span[@class="sn_tm"]//text()')))
64+ publishedDateXPath = result.xpath('.//div[@class="sn_txt"]/div//span[contains(@class,"sn_ST")]//span[contains(@class,"sn_tm")]//text()')
65+ if publishedDateXPath != None:
66+ publishedDate = escape(' '.join(publishedDateXPath))
63
67
64 if re.match("^[0-9]+ minute(s|) ago$", publishedDate):
68 if re.match("^[0-9]+ minute(s|) ago$", publishedDate):
65 timeNumbers = re.findall(r'\d+', publishedDate)
69 timeNumbers = re.findall(r'\d+', publishedDate)
@@ -74,9 +78,18 @@ def response(resp):
74 publishedDate = datetime.now()\
78 publishedDate = datetime.now()\
75 - timedelta(hours=int(timeNumbers[0]))\
79 - timedelta(hours=int(timeNumbers[0]))\
76 - timedelta(minutes=int(timeNumbers[1]))
80 - timedelta(minutes=int(timeNumbers[1]))
81+ elif re.match("^[0-9]+ day(s|) ago$", publishedDate):
82+ timeNumbers = re.findall(r'\d+', publishedDate)
83+ publishedDate = datetime.now()\
84+ - timedelta(days=int(timeNumbers[0]))
77 else:
85 else:
86+ try:
87+ # FIXME use params['language'] to parse either mm/dd or dd/mm
78− publishedDate = parser.parse(publishedDate)
88+ publishedDate = parser.parse(publishedDate, dayfirst=False)
79−
89+ except TypeError:
90+ # FIXME
91+ publishedDate = datetime.now()
92+
80 # append result
93 # append result
81 results.append({'url': url,
94 results.append({'url': url,
82 'title': title,
95 'title': title,
searx/engines/dailymotion.py+6−1
@@ -16,8 +16,8 @@ from lxml import html
16
16
17# engine dependent config
17# engine dependent config
18categories = ['videos']
18categories = ['videos']
19−locale = 'en_US'
20paging = True
19paging = True
20+language_support = True
21
21
22# search-url
22# search-url
23# see http://www.dailymotion.com/doc/api/obj-video.html
23# see http://www.dailymotion.com/doc/api/obj-video.html
@@ -26,6 +26,11 @@ search_url = 'https://api.dailymotion.com/videos?fields=title,description,durati
26
26
27# do search-request
27# do search-request
28def request(query, params):
28def request(query, params):
29+ if params['language'] == 'all':
30+ locale = 'en-US'
31+ else:
32+ locale = params['language']
33+
29 params['url'] = search_url.format(
34 params['url'] = search_url.format(
30 query=urlencode({'search': query, 'localization': locale}),
35 query=urlencode({'search': query, 'localization': locale}),
31 pageno=params['pageno'])
36 pageno=params['pageno'])
searx/engines/yahoo.py+5−3
@@ -40,9 +40,11 @@ def parse_url(url_string):
40 if endpos > -1:
40 if endpos > -1:
41 endpositions.append(endpos)
41 endpositions.append(endpos)
42
42
43+ if start==0 or len(endpositions) == 0:
44+ return url_string
45+ else:
43− end = min(endpositions)
46+ end = min(endpositions)
44−
45− return unquote(url_string[start:end])
47+ return unquote(url_string[start:end])
46
48
47
49
48# do search-request
50# do search-request