1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import calendar
5 import re
6 import time
7
8 from .amp import AMPIE
9 from .common import InfoExtractor
10 from .youtube import YoutubeIE
11 from ..compat import compat_urlparse
12
13
14 class AbcNewsVideoIE(AMPIE):
15 IE_NAME = 'abcnews:video'
16 _VALID_URL = r'''(?x)
17 https?://
18 abcnews\.go\.com/
19 (?:
20 [^/]+/video/(?P<display_id>[0-9a-z-]+)-|
21 video/embed\?.*?\bid=
22 )
23 (?P<id>\d+)
24 '''
25
26 _TESTS = [{
27 'url': 'http://abcnews.go.com/ThisWeek/video/week-exclusive-irans-foreign-minister-zarif-20411932',
28 'info_dict': {
29 'id': '20411932',
30 'ext': 'mp4',
31 'display_id': 'week-exclusive-irans-foreign-minister-zarif',
32 'title': '\'This Week\' Exclusive: Iran\'s Foreign Minister Zarif',
33 'description': 'George Stephanopoulos goes one-on-one with Iranian Foreign Minister Dr. Javad Zarif.',
34 'duration': 180,
35 'thumbnail': r're:^https?://.*\.jpg$',
36 },
37 'params': {
38 # m3u8 download
39 'skip_download': True,
40 },
41 }, {
42 'url': 'http://abcnews.go.com/video/embed?id=46979033',
43 'only_matching': True,
44 }, {
45 'url': 'http://abcnews.go.com/2020/video/2020-husband-stands-teacher-jail-student-affairs-26119478',
46 'only_matching': True,
47 }]
48
49 def _real_extract(self, url):
50 mobj = re.match(self._VALID_URL, url)
51 display_id = mobj.group('display_id')
52 video_id = mobj.group('id')
53 info_dict = self._extract_feed_info(
54 'http://abcnews.go.com/video/itemfeed?id=%s' % video_id)
55 info_dict.update({
56 'id': video_id,
57 'display_id': display_id,
58 })
59 return info_dict
60
61
62 class AbcNewsIE(InfoExtractor):
63 IE_NAME = 'abcnews'
64 _VALID_URL = r'https?://abcnews\.go\.com/(?:[^/]+/)+(?P<display_id>[0-9a-z-]+)/story\?id=(?P<id>\d+)'
65
66 _TESTS = [{
67 'url': 'http://abcnews.go.com/Blotter/News/dramatic-video-rare-death-job-america/story?id=10498713#.UIhwosWHLjY',
68 'info_dict': {
69 'id': '10498713',
70 'ext': 'flv',
71 'display_id': 'dramatic-video-rare-death-job-america',
72 'title': 'Occupational Hazards',
73 'description': 'Nightline investigates the dangers that lurk at various jobs.',
74 'thumbnail': r're:^https?://.*\.jpg$',
75 'upload_date': '20100428',
76 'timestamp': 1272412800,
77 },
78 'add_ie': ['AbcNewsVideo'],
79 }, {
80 'url': 'http://abcnews.go.com/Entertainment/justin-timberlake-performs-stop-feeling-eurovision-2016/story?id=39125818',
81 'info_dict': {
82 'id': '39125818',
83 'ext': 'mp4',
84 'display_id': 'justin-timberlake-performs-stop-feeling-eurovision-2016',
85 'title': 'Justin Timberlake Drops Hints For Secret Single',
86 'description': 'Lara Spencer reports the buzziest stories of the day in "GMA" Pop News.',
87 'upload_date': '20160515',
88 'timestamp': 1463329500,
89 },
90 'params': {
91 # m3u8 download
92 'skip_download': True,
93 # The embedded YouTube video is blocked due to copyright issues
94 'playlist_items': '1',
95 },
96 'add_ie': ['AbcNewsVideo'],
97 }, {
98 'url': 'http://abcnews.go.com/Technology/exclusive-apple-ceo-tim-cook-iphone-cracking-software/story?id=37173343',
99 'only_matching': True,
100 }]
101
102 def _real_extract(self, url):
103 mobj = re.match(self._VALID_URL, url)
104 display_id = mobj.group('display_id')
105 video_id = mobj.group('id')
106
107 webpage = self._download_webpage(url, video_id)
108 video_url = self._search_regex(
109 r'window\.abcnvideo\.url\s*=\s*"([^"]+)"', webpage, 'video URL')
110 full_video_url = compat_urlparse.urljoin(url, video_url)
111
112 youtube_url = YoutubeIE._extract_url(webpage)
113
114 timestamp = None
115 date_str = self._html_search_regex(
116 r'<span[^>]+class="timestamp">([^<]+)</span>',
117 webpage, 'timestamp', fatal=False)
118 if date_str:
119 tz_offset = 0
120 if date_str.endswith(' ET'): # Eastern Time
121 tz_offset = -5
122 date_str = date_str[:-3]
123 date_formats = ['%b. %d, %Y', '%b %d, %Y, %I:%M %p']
124 for date_format in date_formats:
125 try:
126 timestamp = calendar.timegm(time.strptime(date_str.strip(), date_format))
127 except ValueError:
128 continue
129 if timestamp is not None:
130 timestamp -= tz_offset * 3600
131
132 entry = {
133 '_type': 'url_transparent',
134 'ie_key': AbcNewsVideoIE.ie_key(),
135 'url': full_video_url,
136 'id': video_id,
137 'display_id': display_id,
138 'timestamp': timestamp,
139 }
140
141 if youtube_url:
142 entries = [entry, self.url_result(youtube_url, ie=YoutubeIE.ie_key())]
143 return self.playlist_result(entries)
144
145 return entry
|
