Mirroring
/
yt-dlp
mirror of https://github.com/yt-dlp/yt-dlp
1
0
Fork 0
Code Issues Releases Wiki Activity

[ie/3sat] Fix extractor (#12403)

Browse Source 
Fix 241ace4f10

Closes #12391
Authored by: bashonly
release
bashonly 2 weeks ago committed by GitHub
parent 6ca23ffaa4
commit e7882b682b
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 142 additions and 119 deletions
Show Stats Download Patch File Download Diff File

31
yt_dlp/extractor/dreisat.py
Unescape Escape View File

@ -1,10 +1,24 @@
from .zdf import ZDFIE
from .zdf import ZDFBaseIE
class DreiSatIE(ZDFIE): # XXX: Do not subclass from concrete IE
class DreiSatIE(ZDFBaseIE):
IE_NAME = '3sat'
_VALID_URL = r'https?://(?:www\.)?3sat\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)\.html'
_TESTS = [{
'url': 'https://www.3sat.de/dokumentation/reise/traumziele-suedostasiens-die-philippinen-und-vietnam-102.html',
'info_dict': {
'id': '231124_traumziele_philippinen_und_vietnam_dokreise',
'ext': 'mp4',
'title': 'Traumziele Südostasiens (1/2): Die Philippinen und Vietnam',
'description': 'md5:26329ce5197775b596773b939354079d',
'duration': 2625.0,
'thumbnail': 'https://www.3sat.de/assets/traumziele-suedostasiens-die-philippinen-und-vietnam-100~2400x1350?cb=1699870351148',
'episode': 'Traumziele Südostasiens (1/2): Die Philippinen und Vietnam',
'episode_id': 'POS_cc7ff51c-98cf-4d12-b99d-f7a551de1c95',
'timestamp': 1738593000,
'upload_date': '20250203',
},
}, {
# Same as https://www.zdf.de/dokumentation/ab-18/10-wochen-sommer-102.html
'url': 'https://www.3sat.de/film/ab-18/10-wochen-sommer-108.html',
'md5': '0aff3e7bc72c8813f5e0fae333316a1d',
@ -17,6 +31,7 @@ class DreiSatIE(ZDFIE): # XXX: Do not subclass from concrete IE
'timestamp': 1608604200,
'upload_date': '20201222',
},
'skip': '410 Gone',
}, {
'url': 'https://www.3sat.de/gesellschaft/schweizweit/waidmannsheil-100.html',
'info_dict': {
@ -30,6 +45,7 @@ class DreiSatIE(ZDFIE): # XXX: Do not subclass from concrete IE
'params': {
'skip_download': True,
},
'skip': '404 Not Found',
}, {
# Same as https://www.zdf.de/filme/filme-sonstige/der-hauptmann-112.html
'url': 'https://www.3sat.de/film/spielfilm/der-hauptmann-100.html',
@ -39,3 +55,14 @@ class DreiSatIE(ZDFIE): # XXX: Do not subclass from concrete IE
'url': 'https://www.3sat.de/wissen/nano/nano-21-mai-2019-102.html',
'only_matching': True,
}]
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id, fatal=False)
if webpage:
player = self._extract_player(webpage, url, fatal=False)
if player:
return self._extract_regular(url, player, video_id)
return self._extract_mobile(video_id)

230
yt_dlp/extractor/zdf.py
Unescape Escape View File

@ -137,6 +137,116 @@ class ZDFBaseIE(InfoExtractor):
group='json'),
video_id)
def _extract_entry(self, url, player, content, video_id):
title = content.get('title') or content['teaserHeadline']
t = content['mainVideoContent']['http://zdf.de/rels/target']
ptmd_path = traverse_obj(t, (
(('streams', 'default'), None),
('http://zdf.de/rels/streams/ptmd', 'http://zdf.de/rels/streams/ptmd-template'),
), get_all=False)
if not ptmd_path:
raise ExtractorError('Could not extract ptmd_path')
info = self._extract_ptmd(
urljoin(url, ptmd_path.replace('{playerId}', 'android_native_5')), video_id, player['apiToken'], url)
thumbnails = []
layouts = try_get(
content, lambda x: x['teaserImageRef']['layouts'], dict)
if layouts:
for layout_key, layout_url in layouts.items():
layout_url = url_or_none(layout_url)
if not layout_url:
continue
thumbnail = {
'url': layout_url,
'format_id': layout_key,
}
mobj = re.search(r'(?P<width>\d+)x(?P<height>\d+)', layout_key)
if mobj:
thumbnail.update({
'width': int(mobj.group('width')),
'height': int(mobj.group('height')),
})
thumbnails.append(thumbnail)
chapter_marks = t.get('streamAnchorTag') or []
chapter_marks.append({'anchorOffset': int_or_none(t.get('duration'))})
chapters = [{
'start_time': chap.get('anchorOffset'),
'end_time': next_chap.get('anchorOffset'),
'title': chap.get('anchorLabel'),
} for chap, next_chap in zip(chapter_marks, chapter_marks[1:])]
return merge_dicts(info, {
'title': title,
'description': content.get('leadParagraph') or content.get('teasertext'),
'duration': int_or_none(t.get('duration')),
'timestamp': unified_timestamp(content.get('editorialDate')),
'thumbnails': thumbnails,
'chapters': chapters or None,
'episode': title,
**traverse_obj(content, ('programmeItem', 0, 'http://zdf.de/rels/target', {
'series_id': ('http://zdf.de/rels/cmdm/series', 'seriesUuid', {str}),
'series': ('http://zdf.de/rels/cmdm/series', 'seriesTitle', {str}),
'season': ('http://zdf.de/rels/cmdm/season', 'seasonTitle', {str}),
'season_number': ('http://zdf.de/rels/cmdm/season', 'seasonNumber', {int_or_none}),
'season_id': ('http://zdf.de/rels/cmdm/season', 'seasonUuid', {str}),
'episode_number': ('episodeNumber', {int_or_none}),
'episode_id': ('contentId', {str}),
})),
})
def _extract_regular(self, url, player, video_id, query=None):
player_url = player['content']
content = self._call_api(
update_url_query(player_url, query),
video_id, 'content', player['apiToken'], url)
return self._extract_entry(player_url, player, content, video_id)
def _extract_mobile(self, video_id):
video = self._download_v2_doc(video_id)
formats = []
formitaeten = try_get(video, lambda x: x['document']['formitaeten'], list)
document = formitaeten and video['document']
if formitaeten:
title = document['titel']
content_id = document['basename']
format_urls = set()
for f in formitaeten or []:
self._extract_format(content_id, formats, format_urls, f)
thumbnails = []
teaser_bild = document.get('teaserBild')
if isinstance(teaser_bild, dict):
for thumbnail_key, thumbnail in teaser_bild.items():
thumbnail_url = try_get(
thumbnail, lambda x: x['url'], str)
if thumbnail_url:
thumbnails.append({
'url': thumbnail_url,
'id': thumbnail_key,
'width': int_or_none(thumbnail.get('width')),
'height': int_or_none(thumbnail.get('height')),
})
return {
'id': content_id,
'title': title,
'description': document.get('beschreibung'),
'duration': int_or_none(document.get('length')),
'timestamp': unified_timestamp(document.get('date')) or unified_timestamp(
try_get(video, lambda x: x['meta']['editorialDate'], str)),
'thumbnails': thumbnails,
'subtitles': self._extract_subtitles(document),
'formats': formats,
}
class ZDFIE(ZDFBaseIE):
_VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)\.html'
@ -306,121 +416,6 @@ class ZDFIE(ZDFBaseIE):
},
}]
def _extract_entry(self, url, player, content, video_id):
title = content.get('title') or content['teaserHeadline']
t = content['mainVideoContent']['http://zdf.de/rels/target']
ptmd_path = traverse_obj(t, (
(('streams', 'default'), None),
('http://zdf.de/rels/streams/ptmd', 'http://zdf.de/rels/streams/ptmd-template'),
), get_all=False)
if not ptmd_path:
raise ExtractorError('Could not extract ptmd_path')
info = self._extract_ptmd(
urljoin(url, ptmd_path.replace('{playerId}', 'android_native_5')), video_id, player['apiToken'], url)
thumbnails = []
layouts = try_get(
content, lambda x: x['teaserImageRef']['layouts'], dict)
if layouts:
for layout_key, layout_url in layouts.items():
layout_url = url_or_none(layout_url)
if not layout_url:
continue
thumbnail = {
'url': layout_url,
'format_id': layout_key,
}
mobj = re.search(r'(?P<width>\d+)x(?P<height>\d+)', layout_key)
if mobj:
thumbnail.update({
'width': int(mobj.group('width')),
'height': int(mobj.group('height')),
})
thumbnails.append(thumbnail)
chapter_marks = t.get('streamAnchorTag') or []
chapter_marks.append({'anchorOffset': int_or_none(t.get('duration'))})
chapters = [{
'start_time': chap.get('anchorOffset'),
'end_time': next_chap.get('anchorOffset'),
'title': chap.get('anchorLabel'),
} for chap, next_chap in zip(chapter_marks, chapter_marks[1:])]
return merge_dicts(info, {
'title': title,
'description': content.get('leadParagraph') or content.get('teasertext'),
'duration': int_or_none(t.get('duration')),
'timestamp': unified_timestamp(content.get('editorialDate')),
'thumbnails': thumbnails,
'chapters': chapters or None,
'episode': title,
**traverse_obj(content, ('programmeItem', 0, 'http://zdf.de/rels/target', {
'series_id': ('http://zdf.de/rels/cmdm/series', 'seriesUuid', {str}),
'series': ('http://zdf.de/rels/cmdm/series', 'seriesTitle', {str}),
'season': ('http://zdf.de/rels/cmdm/season', 'seasonTitle', {str}),
'season_number': ('http://zdf.de/rels/cmdm/season', 'seasonNumber', {int_or_none}),
'season_id': ('http://zdf.de/rels/cmdm/season', 'seasonUuid', {str}),
'episode_number': ('episodeNumber', {int_or_none}),
'episode_id': ('contentId', {str}),
})),
})
def _extract_regular(self, url, player, video_id):
player_url = player['content']
try:
content = self._call_api(
update_url_query(player_url, {'profile': 'player-3'}),
video_id, 'content', player['apiToken'], url)
except ExtractorError as e:
self.report_warning(f'{video_id}: {e.orig_msg}; retrying with v2 profile')
content = self._call_api(
player_url, video_id, 'content', player['apiToken'], url)
return self._extract_entry(player_url, player, content, video_id)
def _extract_mobile(self, video_id):
video = self._download_v2_doc(video_id)
formats = []
formitaeten = try_get(video, lambda x: x['document']['formitaeten'], list)
document = formitaeten and video['document']
if formitaeten:
title = document['titel']
content_id = document['basename']
format_urls = set()
for f in formitaeten or []:
self._extract_format(content_id, formats, format_urls, f)
thumbnails = []
teaser_bild = document.get('teaserBild')
if isinstance(teaser_bild, dict):
for thumbnail_key, thumbnail in teaser_bild.items():
thumbnail_url = try_get(
thumbnail, lambda x: x['url'], str)
if thumbnail_url:
thumbnails.append({
'url': thumbnail_url,
'id': thumbnail_key,
'width': int_or_none(thumbnail.get('width')),
'height': int_or_none(thumbnail.get('height')),
})
return {
'id': content_id,
'title': title,
'description': document.get('beschreibung'),
'duration': int_or_none(document.get('length')),
'timestamp': unified_timestamp(document.get('date')) or unified_timestamp(
try_get(video, lambda x: x['meta']['editorialDate'], str)),
'thumbnails': thumbnails,
'subtitles': self._extract_subtitles(document),
'formats': formats,
}
def _real_extract(self, url):
video_id = self._match_id(url)
@ -428,7 +423,7 @@ class ZDFIE(ZDFBaseIE):
if webpage:
player = self._extract_player(webpage, url, fatal=False)
if player:
return self._extract_regular(url, player, video_id)
return self._extract_regular(url, player, video_id, query={'profile': 'player-3'})
return self._extract_mobile(video_id)
@ -474,7 +469,8 @@ class ZDFChannelIE(ZDFBaseIE):
'title': ('titel', {str}),
'description': ('beschreibung', {str}),
'duration': ('length', {float_or_none}),
# TODO: seasonNumber and episodeNumber can be extracted but need to also be in ZDFIE
'season_number': ('seasonNumber', {int_or_none}),
'episode_number': ('episodeNumber', {int_or_none}),
}))
def _entries(self, data, document_id):

Write Preview
Loading…
Cancel
Save