]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/yahoo.py
Fix description of the modularization.
[youtubedl] / youtube_dl / extractor / yahoo.py
1 import datetime
2 import itertools
3 import json
4 import re
5
6 from .common import InfoExtractor, SearchInfoExtractor
7 from ..utils import (
8 compat_urllib_parse,
9
10 ExtractorError,
11 )
12
13 class YahooIE(InfoExtractor):
14 """Information extractor for screen.yahoo.com."""
15 _VALID_URL = r'http://screen\.yahoo\.com/.*?-(?P<id>\d*?)\.html'
16
17 def _real_extract(self, url):
18 mobj = re.match(self._VALID_URL, url)
19 if mobj is None:
20 raise ExtractorError(u'Invalid URL: %s' % url)
21 video_id = mobj.group('id')
22 webpage = self._download_webpage(url, video_id)
23 m_id = re.search(r'YUI\.namespace\("Media"\)\.CONTENT_ID = "(?P<new_id>.+?)";', webpage)
24
25 if m_id is None:
26 # TODO: Check which url parameters are required
27 info_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;lmsoverride=1;outputformat=mrss;cb=974419660;id=%s;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id
28 webpage = self._download_webpage(info_url, video_id, u'Downloading info webpage')
29 info_re = r'''<title><!\[CDATA\[(?P<title>.*?)\]\]></title>.*
30 <description><!\[CDATA\[(?P<description>.*?)\]\]></description>.*
31 <media:pubStart><!\[CDATA\[(?P<date>.*?)\ .*\]\]></media:pubStart>.*
32 <media:content\ medium="image"\ url="(?P<thumb>.*?)"\ name="LARGETHUMB"
33 '''
34 self.report_extraction(video_id)
35 m_info = re.search(info_re, webpage, re.VERBOSE|re.DOTALL)
36 if m_info is None:
37 raise ExtractorError(u'Unable to extract video info')
38 video_title = m_info.group('title')
39 video_description = m_info.group('description')
40 video_thumb = m_info.group('thumb')
41 video_date = m_info.group('date')
42 video_date = datetime.datetime.strptime(video_date, '%m/%d/%Y').strftime('%Y%m%d')
43
44 # TODO: Find a way to get mp4 videos
45 rest_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;element=stream;outputformat=mrss;id=%s;lmsoverride=1;bw=375;dynamicstream=1;cb=83521105;tech=flv,mp4;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id
46 webpage = self._download_webpage(rest_url, video_id, u'Downloading video url webpage')
47 m_rest = re.search(r'<media:content url="(?P<url>.*?)" path="(?P<path>.*?)"', webpage)
48 video_url = m_rest.group('url')
49 video_path = m_rest.group('path')
50 if m_rest is None:
51 raise ExtractorError(u'Unable to extract video url')
52
53 else: # We have to use a different method if another id is defined
54 long_id = m_id.group('new_id')
55 info_url = 'http://video.query.yahoo.com/v1/public/yql?q=SELECT%20*%20FROM%20yahoo.media.video.streams%20WHERE%20id%3D%22' + long_id + '%22%20AND%20format%3D%22mp4%2Cflv%22%20AND%20protocol%3D%22rtmp%2Chttp%22%20AND%20plrs%3D%2286Gj0vCaSzV_Iuf6hNylf2%22%20AND%20acctid%3D%22389%22%20AND%20plidl%3D%22%22%20AND%20pspid%3D%22792700001%22%20AND%20offnetwork%3D%22false%22%20AND%20site%3D%22ivy%22%20AND%20lang%3D%22en-US%22%20AND%20region%3D%22US%22%20AND%20override%3D%22none%22%3B&env=prod&format=json&callback=YUI.Env.JSONP.yui_3_8_1_1_1368368376830_335'
56 webpage = self._download_webpage(info_url, video_id, u'Downloading info json')
57 json_str = re.search(r'YUI.Env.JSONP.yui.*?\((.*?)\);', webpage).group(1)
58 info = json.loads(json_str)
59 res = info[u'query'][u'results'][u'mediaObj'][0]
60 stream = res[u'streams'][0]
61 video_path = stream[u'path']
62 video_url = stream[u'host']
63 meta = res[u'meta']
64 video_title = meta[u'title']
65 video_description = meta[u'description']
66 video_thumb = meta[u'thumbnail']
67 video_date = None # I can't find it
68
69 info_dict = {
70 'id': video_id,
71 'url': video_url,
72 'play_path': video_path,
73 'title':video_title,
74 'description': video_description,
75 'thumbnail': video_thumb,
76 'upload_date': video_date,
77 'ext': 'flv',
78 }
79 return info_dict
80
81 class YahooSearchIE(SearchInfoExtractor):
82 """Information Extractor for Yahoo! Video search queries."""
83
84 _MAX_RESULTS = 1000
85 IE_NAME = u'screen.yahoo:search'
86 _SEARCH_KEY = 'yvsearch'
87
88 def _get_n_results(self, query, n):
89 """Get a specified number of results for a query"""
90
91 res = {
92 '_type': 'playlist',
93 'id': query,
94 'entries': []
95 }
96 for pagenum in itertools.count(0):
97 result_url = u'http://video.search.yahoo.com/search/?p=%s&fr=screen&o=js&gs=0&b=%d' % (compat_urllib_parse.quote_plus(query), pagenum * 30)
98 webpage = self._download_webpage(result_url, query,
99 note='Downloading results page '+str(pagenum+1))
100 info = json.loads(webpage)
101 m = info[u'm']
102 results = info[u'results']
103
104 for (i, r) in enumerate(results):
105 if (pagenum * 30) +i >= n:
106 break
107 mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r)
108 e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
109 res['entries'].append(e)
110 if (pagenum * 30 +i >= n) or (m[u'last'] >= (m[u'total'] -1 )):
111 break
112
113 return res