]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/tvp.py
2abfb78307186a2007e7f375bccb7a26ff0713ca
[youtubedl] / youtube_dl / extractor / tvp.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..utils import (
8 determine_ext,
9 clean_html,
10 get_element_by_attribute,
11 ExtractorError,
12 )
13
14
15 class TVPIE(InfoExtractor):
16 IE_NAME = 'tvp'
17 IE_DESC = 'Telewizja Polska'
18 _VALID_URL = r'https?://[^/]+\.tvp\.(?:pl|info)/(?:(?!\d+/)[^/]+/)*(?P<id>\d+)'
19
20 _TESTS = [{
21 'url': 'http://vod.tvp.pl/194536/i-seria-odc-13',
22 'md5': '8aa518c15e5cc32dfe8db400dc921fbb',
23 'info_dict': {
24 'id': '194536',
25 'ext': 'mp4',
26 'title': 'Czas honoru, I seria – odc. 13',
27 'description': 'md5:76649d2014f65c99477be17f23a4dead',
28 },
29 }, {
30 'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176',
31 'md5': 'b0005b542e5b4de643a9690326ab1257',
32 'info_dict': {
33 'id': '17916176',
34 'ext': 'mp4',
35 'title': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata',
36 'description': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata',
37 },
38 }, {
39 # page id is not the same as video id(#7799)
40 'url': 'http://vod.tvp.pl/22704887/08122015-1500',
41 'md5': 'cf6a4705dfd1489aef8deb168d6ba742',
42 'info_dict': {
43 'id': '22680786',
44 'ext': 'mp4',
45 'title': 'Wiadomości, 08.12.2015, 15:00',
46 },
47 }, {
48 'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272',
49 'only_matching': True,
50 }, {
51 'url': 'http://wiadomosci.tvp.pl/25169746/24052016-1200',
52 'only_matching': True,
53 }, {
54 'url': 'http://krakow.tvp.pl/25511623/25lecie-mck-wyjatkowe-miejsce-na-mapie-krakowa',
55 'only_matching': True,
56 }, {
57 'url': 'http://teleexpress.tvp.pl/25522307/wierni-wzieli-udzial-w-procesjach',
58 'only_matching': True,
59 }, {
60 'url': 'http://sport.tvp.pl/25522165/krychowiak-uspokaja-w-sprawie-kontuzji-dwa-tygodnie-to-maksimum',
61 'only_matching': True,
62 }, {
63 'url': 'http://www.tvp.info/25511919/trwa-rewolucja-wladza-zdecydowala-sie-na-pogwalcenie-konstytucji',
64 'only_matching': True,
65 }]
66
67 def _real_extract(self, url):
68 page_id = self._match_id(url)
69 webpage = self._download_webpage(url, page_id)
70 video_id = self._search_regex([
71 r'<iframe[^>]+src="[^"]*?object_id=(\d+)',
72 "object_id\s*:\s*'(\d+)'"], webpage, 'video id')
73 return {
74 '_type': 'url_transparent',
75 'url': 'tvp:' + video_id,
76 'description': self._og_search_description(webpage, default=None),
77 'thumbnail': self._og_search_thumbnail(webpage),
78 'ie_key': 'TVPEmbed',
79 }
80
81
82 class TVPEmbedIE(InfoExtractor):
83 IE_NAME = 'tvp:embed'
84 IE_DESC = 'Telewizja Polska'
85 _VALID_URL = r'(?:tvp:|https?://[^/]+\.tvp\.(?:pl|info)/sess/tvplayer\.php\?.*?object_id=)(?P<id>\d+)'
86
87 _TESTS = [{
88 'url': 'http://www.tvp.pl/sess/tvplayer.php?object_id=22670268',
89 'md5': '8c9cd59d16edabf39331f93bf8a766c7',
90 'info_dict': {
91 'id': '22670268',
92 'ext': 'mp4',
93 'title': 'Panorama, 07.12.2015, 15:40',
94 },
95 }, {
96 'url': 'tvp:22670268',
97 'only_matching': True,
98 }]
99
100 def _real_extract(self, url):
101 video_id = self._match_id(url)
102
103 webpage = self._download_webpage(
104 'http://www.tvp.pl/sess/tvplayer.php?object_id=%s' % video_id, video_id)
105
106 error_massage = get_element_by_attribute('class', 'msg error', webpage)
107 if error_massage:
108 raise ExtractorError('%s said: %s' % (
109 self.IE_NAME, clean_html(error_massage)), expected=True)
110
111 title = self._search_regex(
112 r'name\s*:\s*([\'"])Title\1\s*,\s*value\s*:\s*\1(?P<title>.+?)\1',
113 webpage, 'title', group='title')
114 series_title = self._search_regex(
115 r'name\s*:\s*([\'"])SeriesTitle\1\s*,\s*value\s*:\s*\1(?P<series>.+?)\1',
116 webpage, 'series', group='series', default=None)
117 if series_title:
118 title = '%s, %s' % (series_title, title)
119
120 thumbnail = self._search_regex(
121 r"poster\s*:\s*'([^']+)'", webpage, 'thumbnail', default=None)
122
123 video_url = self._search_regex(
124 r'0:{src:([\'"])(?P<url>.*?)\1', webpage,
125 'formats', group='url', default=None)
126 if not video_url or 'material_niedostepny.mp4' in video_url:
127 video_url = self._download_json(
128 'http://www.tvp.pl/pub/stat/videofileinfo?video_id=%s' % video_id,
129 video_id)['video_url']
130
131 formats = []
132 video_url_base = self._search_regex(
133 r'(https?://.+?/video)(?:\.(?:ism|f4m|m3u8)|-\d+\.mp4)',
134 video_url, 'video base url', default=None)
135 if video_url_base:
136 # TODO: <Group> found instead of <AdaptationSet> in MPD manifest.
137 # It's not mentioned in MPEG-DASH standard. Figure that out.
138 # formats.extend(self._extract_mpd_formats(
139 # video_url_base + '.ism/video.mpd',
140 # video_id, mpd_id='dash', fatal=False))
141 formats.extend(self._extract_f4m_formats(
142 video_url_base + '.ism/video.f4m',
143 video_id, f4m_id='hds', fatal=False))
144 m3u8_formats = self._extract_m3u8_formats(
145 video_url_base + '.ism/video.m3u8', video_id,
146 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
147 self._sort_formats(m3u8_formats)
148 m3u8_formats = list(filter(
149 lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
150 m3u8_formats))
151 formats.extend(m3u8_formats)
152 for i, m3u8_format in enumerate(m3u8_formats, 2):
153 http_url = '%s-%d.mp4' % (video_url_base, i)
154 if self._is_valid_url(http_url, video_id):
155 f = m3u8_format.copy()
156 f.update({
157 'url': http_url,
158 'format_id': f['format_id'].replace('hls', 'http'),
159 'protocol': 'http',
160 })
161 formats.append(f)
162 else:
163 formats = [{
164 'format_id': 'direct',
165 'url': video_url,
166 'ext': determine_ext(video_url, 'mp4'),
167 }]
168
169 self._sort_formats(formats)
170
171 return {
172 'id': video_id,
173 'title': title,
174 'thumbnail': thumbnail,
175 'formats': formats,
176 }
177
178
179 class TVPSeriesIE(InfoExtractor):
180 IE_NAME = 'tvp:series'
181 _VALID_URL = r'https?://vod\.tvp\.pl/(?:[^/]+/){2}(?P<id>[^/]+)/?$'
182
183 _TESTS = [{
184 'url': 'http://vod.tvp.pl/filmy-fabularne/filmy-za-darmo/ogniem-i-mieczem',
185 'info_dict': {
186 'title': 'Ogniem i mieczem',
187 'id': '4278026',
188 },
189 'playlist_count': 4,
190 }, {
191 'url': 'http://vod.tvp.pl/audycje/podroze/boso-przez-swiat',
192 'info_dict': {
193 'title': 'Boso przez świat',
194 'id': '9329207',
195 },
196 'playlist_count': 86,
197 }]
198
199 def _real_extract(self, url):
200 display_id = self._match_id(url)
201 webpage = self._download_webpage(url, display_id, tries=5)
202
203 title = self._html_search_regex(
204 r'(?s) id=[\'"]path[\'"]>(?:.*? / ){2}(.*?)</span>', webpage, 'series')
205 playlist_id = self._search_regex(r'nodeId:\s*(\d+)', webpage, 'playlist id')
206 playlist = self._download_webpage(
207 'http://vod.tvp.pl/vod/seriesAjax?type=series&nodeId=%s&recommend'
208 'edId=0&sort=&page=0&pageSize=10000' % playlist_id, display_id, tries=5,
209 note='Downloading playlist')
210
211 videos_paths = re.findall(
212 '(?s)class="shortTitle">.*?href="(/[^"]+)', playlist)
213 entries = [
214 self.url_result('http://vod.tvp.pl%s' % v_path, ie=TVPIE.ie_key())
215 for v_path in videos_paths]
216
217 return {
218 '_type': 'playlist',
219 'id': playlist_id,
220 'display_id': display_id,
221 'title': title,
222 'entries': entries,
223 }