X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;ds=inline;f=youtube_dl%2Fextractor%2Ftagesschau.py;h=136e18f96cadf7bd5701e32b0a3bc7c8767e324e;hb=794e5dcd7e24784c05e042e7e0655c584347f5c3;hp=636607db5a21658e3e98e6dabf09cc4b7fa94459;hpb=c356620ec14033969d53ca8a2dd432b79046ee20;p=youtube-dl.git
diff --git a/youtube_dl/extractor/tagesschau.py b/youtube_dl/extractor/tagesschau.py
index 636607db5..136e18f96 100644
--- a/youtube_dl/extractor/tagesschau.py
+++ b/youtube_dl/extractor/tagesschau.py
@@ -4,164 +4,303 @@ from __future__ import unicode_literals
import re
from .common import InfoExtractor
-from ..utils import parse_filesize, ExtractorError
+from ..utils import (
+ determine_ext,
+ js_to_json,
+ parse_iso8601,
+ parse_filesize,
+)
+
+
+class TagesschauPlayerIE(InfoExtractor):
+ IE_NAME = 'tagesschau:player'
+ _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/(?Paudio|video)/(?P=kind)-(?P\d+)~player(?:_[^/?#&]+)?\.html'
+
+ _TESTS = [{
+ 'url': 'http://www.tagesschau.de/multimedia/video/video-179517~player.html',
+ 'md5': '8d09548d5c15debad38bee3a4d15ca21',
+ 'info_dict': {
+ 'id': '179517',
+ 'ext': 'mp4',
+ 'title': 'Marie Kristin Boese, ARD Berlin, über den zukünftigen Kurs der AfD',
+ 'thumbnail': 're:^https?:.*\.jpg$',
+ 'formats': 'mincount:6',
+ },
+ }, {
+ 'url': 'https://www.tagesschau.de/multimedia/audio/audio-29417~player.html',
+ 'md5': '76e6eec6ebd40740671cf0a2c88617e5',
+ 'info_dict': {
+ 'id': '29417',
+ 'ext': 'mp3',
+ 'title': 'Trabi - Bye, bye Rennpappe',
+ 'thumbnail': 're:^https?:.*\.jpg$',
+ 'formats': 'mincount:2',
+ },
+ }, {
+ 'url': 'http://www.tagesschau.de/multimedia/audio/audio-29417~player_autoplay-true.html',
+ 'only_matching': True,
+ }]
+
+ _FORMATS = {
+ 'xs': {'quality': 0},
+ 's': {'width': 320, 'height': 180, 'quality': 1},
+ 'm': {'width': 512, 'height': 288, 'quality': 2},
+ 'l': {'width': 960, 'height': 540, 'quality': 3},
+ 'xl': {'width': 1280, 'height': 720, 'quality': 4},
+ 'xxl': {'quality': 5},
+ }
+
+ def _extract_via_api(self, kind, video_id):
+ info = self._download_json(
+ 'https://www.tagesschau.de/api/multimedia/{0}/{0}-{1}.json'.format(kind, video_id),
+ video_id)
+ title = info['headline']
+ formats = []
+ for media in info['mediadata']:
+ for format_id, format_url in media.items():
+ if determine_ext(format_url) == 'm3u8':
+ formats.extend(self._extract_m3u8_formats(
+ format_url, video_id, 'mp4',
+ entry_protocol='m3u8_native', m3u8_id='hls'))
+ else:
+ formats.append({
+ 'url': format_url,
+ 'format_id': format_id,
+ 'vcodec': 'none' if kind == 'audio' else None,
+ })
+ self._sort_formats(formats)
+ timestamp = parse_iso8601(info.get('date'))
+ return {
+ 'id': video_id,
+ 'title': title,
+ 'timestamp': timestamp,
+ 'formats': formats,
+ }
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+ video_id = mobj.group('id')
+
+ # kind = mobj.group('kind').lower()
+ # if kind == 'video':
+ # return self._extract_via_api(kind, video_id)
+
+ # JSON api does not provide some audio formats (e.g. ogg) thus
+ # extractiong audio via webpage
+
+ webpage = self._download_webpage(url, video_id)
+
+ title = self._og_search_title(webpage).strip()
+ formats = []
+
+ for media_json in re.findall(r'({src\s*:\s*["\']http[^}]+type\s*:[^}]+})', webpage):
+ media = self._parse_json(js_to_json(media_json), video_id, fatal=False)
+ if not media:
+ continue
+ src = media.get('src')
+ if not src:
+ return
+ quality = media.get('quality')
+ kind = media.get('type', '').split('/')[0]
+ ext = determine_ext(src)
+ f = {
+ 'url': src,
+ 'format_id': '%s_%s' % (quality, ext) if quality else ext,
+ 'ext': ext,
+ 'vcodec': 'none' if kind == 'audio' else None,
+ }
+ f.update(self._FORMATS.get(quality, {}))
+ formats.append(f)
+
+ self._sort_formats(formats)
+
+ thumbnail = self._og_search_thumbnail(webpage)
+
+ return {
+ 'id': video_id,
+ 'title': title,
+ 'thumbnail': thumbnail,
+ 'formats': formats,
+ }
class TagesschauIE(InfoExtractor):
- _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/(?:sendung/(ts|tsg|tt|nm|bab/bab)|video/video|tsvorzwanzig)(?P-?[0-9]+)(?:~[-_a-zA-Z0-9]*)?\.html'
+ _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/(?P[^/]+/(?:[^/]+/)*?(?P[^/#?]+?(?:-?[0-9]+)?))(?:~_?[^/#?]+?)?\.html'
_TESTS = [{
'url': 'http://www.tagesschau.de/multimedia/video/video-102143.html',
- 'md5': '917a228bc7df7850783bc47979673a09',
+ 'md5': 'f7c27a0eff3bfe8c7727e65f8fe1b1e6',
'info_dict': {
- 'id': '102143',
+ 'id': 'video-102143',
'ext': 'mp4',
'title': 'Regierungsumbildung in Athen: Neue Minister in Griechenland vereidigt',
- 'description': 'md5:171feccd9d9b3dd54d05d501568f6359',
- 'thumbnail': 're:^http:.*\.jpg$',
+ 'description': '18.07.2015 20:10 Uhr',
+ 'thumbnail': 're:^https?:.*\.jpg$',
},
}, {
'url': 'http://www.tagesschau.de/multimedia/sendung/ts-5727.html',
'md5': '3c54c1f6243d279b706bde660ceec633',
'info_dict': {
- 'id': '5727',
+ 'id': 'ts-5727',
'ext': 'mp4',
- 'description': 'md5:695c01bfd98b7e313c501386327aea59',
'title': 'Sendung: tagesschau \t04.12.2014 20:00 Uhr',
- 'thumbnail': 're:^http:.*\.jpg$',
- }
+ 'description': 'md5:695c01bfd98b7e313c501386327aea59',
+ 'thumbnail': 're:^https?:.*\.jpg$',
+ },
}, {
- 'url': 'http://www.tagesschau.de/multimedia/sendung/tsg-3771.html',
- 'md5': '90757268b49ef56deae90c7b48928d58',
+ # exclusive audio
+ 'url': 'http://www.tagesschau.de/multimedia/audio/audio-29417.html',
+ 'md5': '76e6eec6ebd40740671cf0a2c88617e5',
'info_dict': {
- 'id': '3771',
- 'ext': 'mp4',
- 'description': None,
- 'title': 'Sendung: tagesschau (mit Gebärdensprache) \t14.07.2015 20:00 Uhr',
- 'thumbnail': 're:^http:.*\.jpg$',
- }
+ 'id': 'audio-29417',
+ 'ext': 'mp3',
+ 'title': 'Trabi - Bye, bye Rennpappe',
+ 'description': 'md5:8687dda862cbbe2cfb2df09b56341317',
+ 'thumbnail': 're:^https?:.*\.jpg$',
+ },
}, {
- 'url': 'http://www.tagesschau.de/multimedia/sendung/tt-3827.html',
- 'md5': '6e3ebdc75e8d67da966a8d06721eda71',
+ # audio in article
+ 'url': 'http://www.tagesschau.de/inland/bnd-303.html',
+ 'md5': 'e0916c623e85fc1d2b26b78f299d3958',
'info_dict': {
- 'id': '3827',
- 'ext': 'mp4',
- 'description': 'md5:d511d0e278b0ad341a95ad9ab992ce66',
- 'title': 'Sendung: tagesthemen \t14.07.2015 22:15 Uhr',
- 'thumbnail': 're:^http:.*\.jpg$',
- }
+ 'id': 'bnd-303',
+ 'ext': 'mp3',
+ 'title': 'Viele Baustellen für neuen BND-Chef',
+ 'description': 'md5:1e69a54be3e1255b2b07cdbce5bcd8b4',
+ 'thumbnail': 're:^https?:.*\.jpg$',
+ },
}, {
- 'url': 'http://www.tagesschau.de/multimedia/sendung/nm-3475.html',
- 'md5': '8a8875a568f0a5ae5ceef93c501a225f',
+ 'url': 'http://www.tagesschau.de/inland/afd-parteitag-135.html',
'info_dict': {
- 'id': '3475',
- 'ext': 'mp4',
- 'description': 'md5:ed149f5649cda3dac86813a9d777e131',
- 'title': 'Sendung: nachtmagazin \t15.07.2015 00:15 Uhr',
- 'thumbnail': 're:^http:.*\.jpg$',
- }
+ 'id': 'afd-parteitag-135',
+ 'title': 'Möchtegern-Underdog mit Machtanspruch',
+ },
+ 'playlist_count': 2,
+ }, {
+ 'url': 'http://www.tagesschau.de/multimedia/sendung/tsg-3771.html',
+ 'only_matching': True,
+ }, {
+ 'url': 'http://www.tagesschau.de/multimedia/sendung/tt-3827.html',
+ 'only_matching': True,
+ }, {
+ 'url': 'http://www.tagesschau.de/multimedia/sendung/nm-3475.html',
+ 'only_matching': True,
+ }, {
+ 'url': 'http://www.tagesschau.de/multimedia/sendung/weltspiegel-3167.html',
+ 'only_matching': True,
}, {
'url': 'http://www.tagesschau.de/multimedia/tsvorzwanzig-959.html',
- 'md5': 'be4d6f0421f2acd8abe25ea29f6f015b',
- 'info_dict': {
- 'id': '959',
- 'ext': 'mp4',
- 'description': None,
- 'title': 'Sendung: tagesschau vor 20 Jahren \t14.07.2015 22:45 Uhr',
- 'thumbnail': 're:^http:.*\.jpg$',
- }
+ 'only_matching': True,
}, {
'url': 'http://www.tagesschau.de/multimedia/sendung/bab/bab-3299~_bab-sendung-209.html',
- 'md5': '42e3757018d9908581481a80cc1806da',
- 'info_dict': {
- 'id': '3299',
- 'ext': 'mp4',
- 'description': None,
- 'title': 'Nach dem Referendum: Schaltgespräch nach Athen',
- 'thumbnail': 're:^http:.*\.jpg$',
- }
+ 'only_matching': True,
+ }, {
+ 'url': 'http://www.tagesschau.de/multimedia/video/video-102303~_bab-sendung-211.html',
+ 'only_matching': True,
+ }, {
+ 'url': 'http://www.tagesschau.de/100sekunden/index.html',
+ 'only_matching': True,
+ }, {
+ # playlist article with collapsing sections
+ 'url': 'http://www.tagesschau.de/wirtschaft/faq-freihandelszone-eu-usa-101.html',
+ 'only_matching': True,
}]
- _FORMATS = {
- 's': {'width': 256, 'height': 144, 'quality': 1},
- 'm': {'width': 512, 'height': 288, 'quality': 2},
- 'l': {'width': 960, 'height': 544, 'quality': 3},
- }
+ @classmethod
+ def suitable(cls, url):
+ return False if TagesschauPlayerIE.suitable(url) else super(TagesschauIE, cls).suitable(url)
+
+ def _extract_formats(self, download_text, media_kind):
+ links = re.finditer(
+ r'',
+ download_text)
+ formats = []
+ for l in links:
+ link_url = l.group('url')
+ if not link_url:
+ continue
+ format_id = self._search_regex(
+ r'.*/[^/.]+\.([^/]+)\.[^/.]+$', link_url, 'format ID',
+ default=determine_ext(link_url))
+ format = {
+ 'format_id': format_id,
+ 'url': l.group('url'),
+ 'format_name': l.group('name'),
+ }
+ title = l.group('title')
+ if title:
+ if media_kind.lower() == 'video':
+ m = re.match(
+ r'''(?x)
+ Video:\s*(?P[a-zA-Z0-9/._-]+)\s*&\#10;
+ (?P[0-9]+)x(?P[0-9]+)px&\#10;
+ (?P[0-9]+)kbps&\#10;
+ Audio:\s*(?P[0-9]+)kbps,\s*(?P[A-Za-z\.0-9]+)&\#10;
+ Größe:\s*(?P[0-9.,]+\s+[a-zA-Z]*B)''',
+ title)
+ if m:
+ format.update({
+ 'format_note': m.group('audio_desc'),
+ 'vcodec': m.group('vcodec'),
+ 'width': int(m.group('width')),
+ 'height': int(m.group('height')),
+ 'abr': int(m.group('abr')),
+ 'vbr': int(m.group('vbr')),
+ 'filesize_approx': parse_filesize(m.group('filesize_approx')),
+ })
+ else:
+ m = re.match(
+ r'(?P.+?)-Format\s*:\s*(?P\d+)kbps\s*,\s*(?P.+)',
+ title)
+ if m:
+ format.update({
+ 'format_note': '%s, %s' % (m.group('format'), m.group('note')),
+ 'vcodec': 'none',
+ 'abr': int(m.group('abr')),
+ })
+ formats.append(format)
+ self._sort_formats(formats)
+ return formats
def _real_extract(self, url):
- video_id = self._match_id(url)
+ mobj = re.match(self._VALID_URL, url)
+ video_id = mobj.group('id') or mobj.group('path')
display_id = video_id.lstrip('-')
+
webpage = self._download_webpage(url, display_id)
- player_url = self._html_search_meta(
- 'twitter:player', webpage, 'player URL', default=None)
- if player_url:
- playerpage = self._download_webpage(
- player_url, display_id, 'Downloading player page')
-
- medias = re.findall(
- r'"(http://media.+?)", type:"video/(.+?)", quality:"(.+?)"',
- playerpage)
- formats = []
- for url, ext, res in medias:
- f = {
- 'format_id': res + '_' + ext,
- 'url': url,
- 'ext': ext,
- }
- f.update(self._FORMATS.get(res, {}))
- formats.append(f)
- thumbnail_fn = re.findall(r'"(/multimedia/.+?\.jpg)"', playerpage)[-1]
- title = self._og_search_title(webpage).strip()
- description = self._og_search_description(webpage).strip()
- else:
+ title = self._html_search_regex(
+ r']*class="headline"[^>]*>(.+?)',
+ webpage, 'title', default=None) or self._og_search_title(webpage)
+
+ DOWNLOAD_REGEX = r'(?s)Wir bieten dieses (?PVideo|Audio) in folgenden Formaten zum Download an:
\s*(?P.*?)
\s*'
+
+ webpage_type = self._og_search_property('type', webpage, default=None)
+ if webpage_type == 'website': # Article
+ entries = []
+ for num, (entry_title, media_kind, download_text) in enumerate(re.findall(
+ r'(?s)
]+class="infotext"[^>]*>\s*(?:]+>)?\s*(.+?).*?
.*?%s' % DOWNLOAD_REGEX,
+ webpage), 1):
+ entries.append({
+ 'id': '%s-%d' % (display_id, num),
+ 'title': '%s' % entry_title,
+ 'formats': self._extract_formats(download_text, media_kind),
+ })
+ if len(entries) > 1:
+ return self.playlist_result(entries, display_id, title)
+ formats = entries[0]['formats']
+ else: # Assume single video
download_text = self._search_regex(
- r'(?s)Wir bieten dieses Video in folgenden Formaten zum Download an:
\s*(.*?)
\s*',
- webpage, 'download links')
- links = re.finditer(
- r'
',
- download_text)
- formats = []
- for l in links:
- format_id = self._search_regex(
- r'.*/[^/.]+\.([^/]+)\.[^/.]+', l.group('url'), 'format ID')
- format = {
- 'format_id': format_id,
- 'url': l.group('url'),
- 'format_name': l.group('name'),
- }
- m = re.match(
- r'''(?x)
- Video:\s*(?P[a-zA-Z0-9/._-]+)\s*&\#10;
- (?P[0-9]+)x(?P[0-9]+)px&\#10;
- (?P[0-9]+)kbps&\#10;
- Audio:\s*(?P[0-9]+)kbps,\s*(?P[A-Za-z\.0-9]+)&\#10;
- Größe:\s*(?P[0-9.,]+\s+[a-zA-Z]*B)''',
- l.group('title'))
- if m:
- format.update({
- 'format_note': m.group('audio_desc'),
- 'vcodec': m.group('vcodec'),
- 'width': int(m.group('width')),
- 'height': int(m.group('height')),
- 'abr': int(m.group('abr')),
- 'vbr': int(m.group('vbr')),
- 'filesize_approx': parse_filesize(m.group('filesize_approx')),
- })
- formats.append(format)
- thumbnail_fn = self._search_regex(
- r'(?s)
(.*?)
',
- webpage, 'description', fatal=False, default=None)
- title = self._html_search_regex(
- r'(.*?)', webpage, 'title')
+ DOWNLOAD_REGEX, webpage, 'download links', group='links')
+ media_kind = self._search_regex(
+ DOWNLOAD_REGEX, webpage, 'media kind', default='Video', group='kind')
+ formats = self._extract_formats(download_text, media_kind)
+ thumbnail = self._og_search_thumbnail(webpage)
+ description = self._html_search_regex(
+ r'(?s)(.*?)
',
+ webpage, 'description', default=None)
self._sort_formats(formats)
- thumbnail = 'http://www.tagesschau.de' + thumbnail_fn
return {
'id': display_id,