]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/googlesearch.py
debian/copyright: Fix lintian's dep5-copyright-license-name-not-unique.
[youtubedl] / youtube_dl / extractor / googlesearch.py
1 from __future__ import unicode_literals
2
3 import itertools
4 import re
5
6 from .common import SearchInfoExtractor
7 from ..compat import (
8 compat_urllib_parse,
9 )
10
11
12 class GoogleSearchIE(SearchInfoExtractor):
13 IE_DESC = 'Google Video search'
14 _MAX_RESULTS = 1000
15 IE_NAME = 'video.google:search'
16 _SEARCH_KEY = 'gvsearch'
17 _TEST = {
18 'url': 'gvsearch15:python language',
19 'info_dict': {
20 'id': 'python language',
21 'title': 'python language',
22 },
23 'playlist_count': 15,
24 }
25
26 def _get_n_results(self, query, n):
27 """Get a specified number of results for a query"""
28
29 entries = []
30 res = {
31 '_type': 'playlist',
32 'id': query,
33 'title': query,
34 }
35
36 for pagenum in itertools.count():
37 result_url = (
38 'http://www.google.com/search?tbm=vid&q=%s&start=%s&hl=en'
39 % (compat_urllib_parse.quote_plus(query), pagenum * 10))
40
41 webpage = self._download_webpage(
42 result_url, 'gvsearch:' + query,
43 note='Downloading result page ' + str(pagenum + 1))
44
45 for hit_idx, mobj in enumerate(re.finditer(
46 r'<h3 class="r"><a href="([^"]+)"', webpage)):
47
48 # Skip playlists
49 if not re.search(r'id="vidthumb%d"' % (hit_idx + 1), webpage):
50 continue
51
52 entries.append({
53 '_type': 'url',
54 'url': mobj.group(1)
55 })
56
57 if (len(entries) >= n) or not re.search(r'id="pnnext"', webpage):
58 res['entries'] = entries[:n]
59 return res