X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2Fextractor%2Fceskatelevize.py;h=65f6be62313dfc623cf1f9aa7adc52282872aade;hb=f05d0e73c6d38d86393f8f552fbfcdda80f37607;hp=92ae4ffcb09ae5adb99cad617f00fa910f4a6376;hpb=8cfb5bbf92bec632ae3bc08c3d87c0eb7c199b75;p=youtube-dl.git
diff --git a/youtube_dl/extractor/ceskatelevize.py b/youtube_dl/extractor/ceskatelevize.py
index 92ae4ffcb..65f6be623 100644
--- a/youtube_dl/extractor/ceskatelevize.py
+++ b/youtube_dl/extractor/ceskatelevize.py
@@ -2,43 +2,54 @@
from __future__ import unicode_literals
import re
-import json
-from urllib import unquote
from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
compat_urllib_request,
compat_urllib_parse,
- compat_urllib_parse_urlparse
+ compat_urllib_parse_urlparse,
+)
+from ..utils import (
+ ExtractorError,
+ float_or_none,
)
class CeskaTelevizeIE(InfoExtractor):
_VALID_URL = r'https?://www\.ceskatelevize\.cz/(porady|ivysilani)/(.+/)?(?P[^?#]+)'
- _TESTS = [{
- 'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/213512120230004-spanelska-chripka',
- 'info_dict': {
- 'id': '213512120230004',
- 'ext': 'flv',
- 'title': 'Prvnà republika: Å panÄlská chÅipka',
- 'duration': 3107.4,
- },
- 'params': {
- 'skip_download': True, # requires rtmpdump
+ _TESTS = [
+ {
+ 'url': 'http://www.ceskatelevize.cz/ivysilani/ivysilani/10441294653-hyde-park-civilizace/214411058091220',
+ 'info_dict': {
+ 'id': '214411058091220',
+ 'ext': 'mp4',
+ 'title': 'Hyde Park Civilizace',
+ 'description': 'VÄda a souÄasná civilizace. Interaktivnà poÅad - prostor pro vaÅ¡e otázky a komentáÅe',
+ 'thumbnail': 're:^https?://.*\.jpg',
+ 'duration': 3350,
+ },
+ 'params': {
+ # m3u8 download
+ 'skip_download': True,
+ },
},
- }, {
- 'url': 'http://www.ceskatelevize.cz/ivysilani/1030584952-tsatsiki-maminka-a-policajt',
- 'info_dict': {
- 'id': '20138143440',
- 'ext': 'flv',
- 'title': 'Tsatsiki, maminka a policajt',
- 'duration': 6754.1,
+ {
+ 'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/bonus/14716-zpevacka-z-duparny-bobina',
+ 'info_dict': {
+ 'id': '14716',
+ 'ext': 'mp4',
+ 'title': 'Prvnà republika: ZpÄvaÄka z Dupárny Bobina',
+ 'description': 'Sága mapujÃcà atmosféru prvnà republiky od r. 1918 do r. 1945.',
+ 'thumbnail': 're:^https?://.*\.jpg',
+ 'duration': 88.4,
+ },
+ 'params': {
+ # m3u8 download
+ 'skip_download': True,
+ },
},
- 'params': {
- 'skip_download': True, # requires rtmpdump
- },
- }]
+ ]
def _real_extract(self, url):
url = url.replace('/porady/', '/ivysilani/').replace('/video/', '')
@@ -48,6 +59,10 @@ class CeskaTelevizeIE(InfoExtractor):
webpage = self._download_webpage(url, video_id)
+ NOT_AVAILABLE_STRING = 'This content is not available at your territory due to limited copyright.'
+ if '%s
' % NOT_AVAILABLE_STRING in webpage:
+ raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
+
typ = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type')
episode_id = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id')
@@ -58,49 +73,83 @@ class CeskaTelevizeIE(InfoExtractor):
'requestSource': 'iVysilani',
}
- req = compat_urllib_request.Request('http://www.ceskatelevize.cz/ivysilani/ajax/get-playlist-url',
- data=compat_urllib_parse.urlencode(data))
+ req = compat_urllib_request.Request(
+ 'http://www.ceskatelevize.cz/ivysilani/ajax/get-client-playlist',
+ data=compat_urllib_parse.urlencode(data))
req.add_header('Content-type', 'application/x-www-form-urlencoded')
req.add_header('x-addr', '127.0.0.1')
req.add_header('X-Requested-With', 'XMLHttpRequest')
req.add_header('Referer', url)
- playlistpage = self._download_webpage(req, video_id)
+ playlistpage = self._download_json(req, video_id)
- req = compat_urllib_request.Request(unquote(json.loads(playlistpage)['url']))
+ playlist_url = playlistpage['url']
+ if playlist_url == 'error_region':
+ raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
+
+ req = compat_urllib_request.Request(compat_urllib_parse.unquote(playlist_url))
req.add_header('Referer', url)
- playlist = self._download_xml(req, video_id)
-
- formats = []
- for i in playlist.find('smilRoot/body'):
- if 'AD' not in i.attrib['id']:
- base_url = i.attrib['base']
- parsedurl = compat_urllib_parse_urlparse(base_url)
- duration = i.attrib['duration']
-
- for video in i.findall('video'):
- if video.attrib['label'] != 'AD':
- format_id = video.attrib['label']
- play_path = video.attrib['src']
- vbr = int(video.attrib['system-bitrate'])
-
- formats.append({
- 'format_id': format_id,
- 'url': base_url,
- 'vbr': vbr,
- 'play_path': play_path,
- 'app': parsedurl.path[1:] + '?' + parsedurl.query,
- 'rtmp_live': True,
- 'ext': 'flv',
- })
+ playlist = self._download_json(req, video_id)
+ item = playlist['playlist'][0]
+ formats = []
+ for format_id, stream_url in item['streamUrls'].items():
+ formats.extend(self._extract_m3u8_formats(stream_url, video_id, 'mp4'))
self._sort_formats(formats)
+ title = self._og_search_title(webpage)
+ description = self._og_search_description(webpage)
+ duration = float_or_none(item.get('duration'))
+ thumbnail = item.get('previewImageUrl')
+
+ subtitles = {}
+ subs = item.get('subtitles')
+ if subs:
+ subtitles = self.extract_subtitles(episode_id, subs)
+
return {
'id': episode_id,
- 'title': self._html_search_regex(r'(.+?) â iVysÃlánà â Äeská televize', webpage, 'title'),
- 'duration': float(duration),
+ 'title': title,
+ 'description': description,
+ 'thumbnail': thumbnail,
+ 'duration': duration,
'formats': formats,
+ 'subtitles': subtitles,
}
+
+ def _get_subtitles(self, episode_id, subs):
+ original_subtitles = self._download_webpage(
+ subs[0]['url'], episode_id, 'Downloading subtitles')
+ srt_subs = self._fix_subtitles(original_subtitles)
+ return {
+ 'cs': [{
+ 'ext': 'srt',
+ 'data': srt_subs,
+ }]
+ }
+
+ @staticmethod
+ def _fix_subtitles(subtitles):
+ """ Convert millisecond-based subtitles to SRT """
+
+ def _msectotimecode(msec):
+ """ Helper utility to convert milliseconds to timecode """
+ components = []
+ for divider in [1000, 60, 60, 100]:
+ components.append(msec % divider)
+ msec //= divider
+ return "{3:02}:{2:02}:{1:02},{0:03}".format(*components)
+
+ def _fix_subtitle(subtitle):
+ for line in subtitle.splitlines():
+ m = re.match(r"^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$", line)
+ if m:
+ yield m.group(1)
+ start, stop = (_msectotimecode(int(t)) for t in m.groups()[1:])
+ yield "{0} --> {1}".format(start, stop)
+ else:
+ yield line
+
+ return "\r\n".join(_fix_subtitle(subtitles))