Mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-08 04:02:40 +01:00
Code Issues Releases Activity

[puhutv] improve extraction

Browse Source 
- fix subtitles extraction
- transform HLS URLs to http URLs
- improve metadata extraction
This commit is contained in:
Remita Amine 2019-10-24 12:53:33 +01:00
parent 07154c7930
commit 162bcc68dc
1 changed files with 49 additions and 41 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

90
youtube_dl/extractor/puhutv.py
View File

@ -25,21 +25,21 @@ class PuhuTVIE(InfoExtractor):
_TESTS = [{ _TESTS = [{
# film # film
'url': 'https://puhutv.com/sut-kardesler-izle', 'url': 'https://puhutv.com/sut-kardesler-izle',
'md5': 'fbd8f2d8e7681f8bcd51b592475a6ae7', 'md5': 'a347470371d56e1585d1b2c8dab01c96',
'info_dict': { 'info_dict': {
'id': '5085', 'id': '5085',
'display_id': 'sut-kardesler', 'display_id': 'sut-kardesler',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Süt Kardeşler', 'title': 'Süt Kardeşler',
'description': 'md5:405fd024df916ca16731114eb18e511a', 'description': 'md5:ca09da25b7e57cbb5a9280d6e48d17aa',
'thumbnail': r're:^https?://.*\.jpg$', 'thumbnail': r're:^https?://.*\.jpg$',
'duration': 4832.44, 'duration': 4832.44,
'creator': 'Arzu Film', 'creator': 'Arzu Film',
'timestamp': 1469778212, 'timestamp': 1561062602,
'upload_date': '20160729', 'upload_date': '20190620',
'release_year': 1976, 'release_year': 1976,
'view_count': int, 'view_count': int,
'tags': ['Aile', 'Komedi', 'Klasikler'], 'tags': list,
}, },
}, { }, {
# episode, geo restricted, bypassable with --geo-verification-proxy # episode, geo restricted, bypassable with --geo-verification-proxy
@ -64,9 +64,10 @@ def _real_extract(self, url):
display_id)['data'] display_id)['data']
video_id = compat_str(info['id']) video_id = compat_str(info['id'])
title = info.get('name') or info['title']['name'] show = info.get('title') or {}
title = info.get('name') or show['name']
if info.get('display_name'): if info.get('display_name'):
title = '%s %s' % (title, info.get('display_name')) title = '%s %s' % (title, info['display_name'])
try: try:
videos = self._download_json( videos = self._download_json(
@ -78,17 +79,36 @@ def _real_extract(self, url):
self.raise_geo_restricted() self.raise_geo_restricted()
raise raise
urls = []
formats = [] formats = []
def add_http_from_hls(m3u8_f):
http_url = m3u8_f['url'].replace('/hls/', '/mp4/').replace('/chunklist.m3u8', '.mp4')
if http_url != m3u8_f['url']:
f = m3u8_f.copy()
f.update({
'format_id': f['format_id'].replace('hls', 'http'),
'protocol': 'http',
'url': http_url,
})
formats.append(f)
for video in videos['data']['videos']: for video in videos['data']['videos']:
media_url = url_or_none(video.get('url')) media_url = url_or_none(video.get('url'))
if not media_url: if not media_url or media_url in urls:
continue continue
urls.append(media_url)
playlist = video.get('is_playlist') playlist = video.get('is_playlist')
if video.get('stream_type') == 'hls' and playlist is True: if (video.get('stream_type') == 'hls' and playlist is True) or 'playlist.m3u8' in media_url:
formats.extend(self._extract_m3u8_formats( m3u8_formats = self._extract_m3u8_formats(
media_url, video_id, 'mp4', entry_protocol='m3u8_native', media_url, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=False)) m3u8_id='hls', fatal=False)
for m3u8_f in m3u8_formats:
formats.append(m3u8_f)
add_http_from_hls(m3u8_f)
continue continue
quality = int_or_none(video.get('quality')) quality = int_or_none(video.get('quality'))
f = { f = {
'url': media_url, 'url': media_url,
@ -96,34 +116,29 @@ def _real_extract(self, url):
'height': quality 'height': quality
} }
video_format = video.get('video_format') video_format = video.get('video_format')
if video_format == 'hls' and playlist is False: is_hls = (video_format == 'hls' or '/hls/' in media_url or '/chunklist.m3u8' in media_url) and playlist is False
if is_hls:
format_id = 'hls' format_id = 'hls'
f['protocol'] = 'm3u8_native' f['protocol'] = 'm3u8_native'
elif video_format == 'mp4': elif video_format == 'mp4':
format_id = 'http' format_id = 'http'
else: else:
continue continue
if quality: if quality:
format_id += '-%sp' % quality format_id += '-%sp' % quality
f['format_id'] = format_id f['format_id'] = format_id
formats.append(f) formats.append(f)
if is_hls:
add_http_from_hls(f)
self._sort_formats(formats) self._sort_formats(formats)
description = try_get(
info, lambda x: x['title']['description'],
compat_str) or info.get('description')
timestamp = unified_timestamp(info.get('created_at'))
creator = try_get( creator = try_get(
info, lambda x: x['title']['producer']['name'], compat_str) show, lambda x: x['producer']['name'], compat_str)
duration = float_or_none( content = info.get('content') or {}
try_get(info, lambda x: x['content']['duration_in_ms'], int),
scale=1000)
view_count = try_get(info, lambda x: x['content']['watch_count'], int)
images = try_get( images = try_get(
info, lambda x: x['content']['images']['wide'], dict) or {} content, lambda x: x['images']['wide'], dict) or {}
thumbnails = [] thumbnails = []
for image_id, image_url in images.items(): for image_id, image_url in images.items():
if not isinstance(image_url, compat_str): if not isinstance(image_url, compat_str):
@ -137,14 +152,8 @@ def _real_extract(self, url):
}) })
thumbnails.append(t) thumbnails.append(t)
release_year = try_get(info, lambda x: x['title']['released_at'], int)
season_number = int_or_none(info.get('season_number'))
season_id = str_or_none(info.get('season_id'))
episode_number = int_or_none(info.get('episode_number'))
tags = [] tags = []
for genre in try_get(info, lambda x: x['title']['genres'], list) or []: for genre in show.get('genres') or []:
if not isinstance(genre, dict): if not isinstance(genre, dict):
continue continue
genre_name = genre.get('name') genre_name = genre.get('name')
@ -152,12 +161,11 @@ def _real_extract(self, url):
tags.append(genre_name) tags.append(genre_name)
subtitles = {} subtitles = {}
for subtitle in try_get( for subtitle in content.get('subtitles') or []:
info, lambda x: x['content']['subtitles'], list) or []:
if not isinstance(subtitle, dict): if not isinstance(subtitle, dict):
continue continue
lang = subtitle.get('language') lang = subtitle.get('language')
sub_url = url_or_none(subtitle.get('url')) sub_url = url_or_none(subtitle.get('url') or subtitle.get('file'))
if not lang or not isinstance(lang, compat_str) or not sub_url: if not lang or not isinstance(lang, compat_str) or not sub_url:
continue continue
subtitles[self._SUBTITLE_LANGS.get(lang, lang)] = [{ subtitles[self._SUBTITLE_LANGS.get(lang, lang)] = [{
@ -168,15 +176,15 @@ def _real_extract(self, url):
'id': video_id, 'id': video_id,
'display_id': display_id, 'display_id': display_id,
'title': title, 'title': title,
'description': description, 'description': info.get('description') or show.get('description'),
'season_id': season_id, 'season_id': str_or_none(info.get('season_id')),
'season_number': season_number, 'season_number': int_or_none(info.get('season_number')),
'episode_number': episode_number, 'episode_number': int_or_none(info.get('episode_number')),
'release_year': release_year, 'release_year': int_or_none(show.get('released_at')),
'timestamp': timestamp, 'timestamp': unified_timestamp(info.get('created_at')),
'creator': creator, 'creator': creator,
'view_count': view_count, 'view_count': int_or_none(content.get('watch_count')),
'duration': duration, 'duration': float_or_none(content.get('duration_in_ms'), 1000),
'tags': tags, 'tags': tags,
'subtitles': subtitles, 'subtitles': subtitles,
'thumbnails': thumbnails, 'thumbnails': thumbnails,