youtube-dl/youtube_dl/extractor/tvp.py

143 lines
4.9 KiB
Python
Raw Normal View History

# coding: utf-8
2014-11-26 20:58:53 +09:00
from __future__ import unicode_literals
2014-12-04 10:54:25 +09:00
import re
from .common import InfoExtractor
2013-11-13 19:06:53 +09:00
class TVPIE(InfoExtractor):
IE_NAME = 'tvp'
IE_DESC = 'Telewizja Polska'
_VALID_URL = r'https?://[^/]+\.tvp\.(?:pl|info)/(?:(?!\d+/)[^/]+/)*(?P<id>\d+)'
2015-01-10 10:38:27 +09:00
_TESTS = [{
'url': 'http://vod.tvp.pl/194536/i-seria-odc-13',
2015-01-20 02:00:22 +09:00
'md5': '8aa518c15e5cc32dfe8db400dc921fbb',
2015-01-10 10:38:27 +09:00
'info_dict': {
'id': '194536',
'ext': 'mp4',
'title': 'Czas honoru, I seria odc. 13',
},
}, {
'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176',
2015-01-20 02:00:22 +09:00
'md5': 'c3b15ed1af288131115ff17a17c19dda',
2015-01-10 10:38:27 +09:00
'info_dict': {
'id': '17916176',
'ext': 'mp4',
'title': 'TVP Gorzów pokaże filmy studentów z podroży dookoła świata',
},
}, {
'url': 'http://vod.tvp.pl/seriale/obyczajowe/na-sygnale/sezon-2-27-/odc-39/17834272',
'only_matching': True,
}, {
'url': 'http://wiadomosci.tvp.pl/25169746/24052016-1200',
'only_matching': True,
}, {
'url': 'http://krakow.tvp.pl/25511623/25lecie-mck-wyjatkowe-miejsce-na-mapie-krakowa',
'only_matching': True,
}, {
'url': 'http://teleexpress.tvp.pl/25522307/wierni-wzieli-udzial-w-procesjach',
'only_matching': True,
}, {
'url': 'http://sport.tvp.pl/25522165/krychowiak-uspokaja-w-sprawie-kontuzji-dwa-tygodnie-to-maksimum',
'only_matching': True,
}, {
'url': 'http://www.tvp.info/25511919/trwa-rewolucja-wladza-zdecydowala-sie-na-pogwalcenie-konstytucji',
'only_matching': True,
2015-01-10 10:38:27 +09:00
}]
def _real_extract(self, url):
2015-01-10 10:38:27 +09:00
video_id = self._match_id(url)
2015-01-20 02:00:22 +09:00
2014-12-04 10:54:25 +09:00
webpage = self._download_webpage(
'http://www.tvp.pl/sess/tvplayer.php?object_id=%s' % video_id, video_id)
2015-01-10 10:38:27 +09:00
2015-01-20 02:00:22 +09:00
title = self._search_regex(
r'name\s*:\s*([\'"])Title\1\s*,\s*value\s*:\s*\1(?P<title>.+?)\1',
webpage, 'title', group='title')
series_title = self._search_regex(
r'name\s*:\s*([\'"])SeriesTitle\1\s*,\s*value\s*:\s*\1(?P<series>.+?)\1',
2014-12-04 10:54:25 +09:00
webpage, 'series', group='series', default=None)
2015-01-20 02:00:22 +09:00
if series_title:
title = '%s, %s' % (series_title, title)
thumbnail = self._search_regex(
r"poster\s*:\s*'([^']+)'", webpage, 'thumbnail', default=None)
2015-01-10 10:38:27 +09:00
2014-12-04 10:54:25 +09:00
video_url = self._search_regex(
r'0:{src:([\'"])(?P<url>.*?)\1', webpage, 'formats', group='url', default=None)
2015-01-20 02:00:22 +09:00
if not video_url:
2014-12-04 10:54:25 +09:00
video_url = self._download_json(
'http://www.tvp.pl/pub/stat/videofileinfo?video_id=%s' % video_id,
video_id)['video_url']
ext = video_url.rsplit('.', 1)[-1]
if ext != 'ism/manifest':
if '/' in ext:
ext = 'mp4'
2015-01-10 10:38:27 +09:00
formats = [{
'format_id': 'direct',
2014-12-04 10:54:25 +09:00
'url': video_url,
2015-01-10 10:38:27 +09:00
'ext': ext,
}]
2014-12-04 10:54:25 +09:00
else:
m3u8_url = re.sub('([^/]*)\.ism/manifest', r'\1.ism/\1.m3u8', video_url)
formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4')
2015-01-10 10:38:27 +09:00
self._sort_formats(formats)
return {
'id': video_id,
'title': title,
2015-01-20 02:00:22 +09:00
'thumbnail': thumbnail,
2015-01-10 10:38:27 +09:00
'formats': formats,
}
2014-12-04 13:14:09 +09:00
class TVPSeriesIE(InfoExtractor):
IE_NAME = 'tvp:series'
2014-12-04 13:14:09 +09:00
_VALID_URL = r'https?://vod\.tvp\.pl/(?:[^/]+/){2}(?P<id>[^/]+)/?$'
2015-01-10 10:38:27 +09:00
_TESTS = [{
'url': 'http://vod.tvp.pl/filmy-fabularne/filmy-za-darmo/ogniem-i-mieczem',
'info_dict': {
'title': 'Ogniem i mieczem',
'id': '4278026',
},
'playlist_count': 4,
}, {
'url': 'http://vod.tvp.pl/audycje/podroze/boso-przez-swiat',
'info_dict': {
'title': 'Boso przez świat',
'id': '9329207',
},
'playlist_count': 86,
}]
2014-12-04 13:14:09 +09:00
def _real_extract(self, url):
display_id = self._match_id(url)
2014-12-04 22:12:09 +09:00
webpage = self._download_webpage(url, display_id, tries=5)
2015-01-10 10:38:27 +09:00
2014-12-04 13:14:09 +09:00
title = self._html_search_regex(
2014-12-04 22:12:09 +09:00
r'(?s) id=[\'"]path[\'"]>(?:.*? / ){2}(.*?)</span>', webpage, 'series')
2014-12-04 13:14:09 +09:00
playlist_id = self._search_regex(r'nodeId:\s*(\d+)', webpage, 'playlist id')
2014-12-04 22:12:09 +09:00
playlist = self._download_webpage(
2014-12-04 13:14:09 +09:00
'http://vod.tvp.pl/vod/seriesAjax?type=series&nodeId=%s&recommend'
2015-01-10 10:38:27 +09:00
'edId=0&sort=&page=0&pageSize=10000' % playlist_id, display_id, tries=5,
note='Downloading playlist')
2014-12-04 13:14:09 +09:00
videos_paths = re.findall(
'(?s)class="shortTitle">.*?href="(/[^"]+)', playlist)
entries = [
self.url_result('http://vod.tvp.pl%s' % v_path, ie=TVPIE.ie_key())
2014-12-04 13:14:09 +09:00
for v_path in videos_paths]
2015-01-10 10:38:27 +09:00
2014-12-04 13:14:09 +09:00
return {
'_type': 'playlist',
'id': playlist_id,
'display_id': display_id,
'title': title,
'entries': entries,
}