mirror of https://github.com/yt-dlp/yt-dlp.git
[RTP] Fix RTP Play support
See #4661 See https://github.com/ytdl-org/youtube-dl/pull/29824
This commit is contained in:
somini
parent
39bc699d2e
commit
9ae42f7ed0
|
|
@ -4,23 +4,29 @@ import re
|
|||
import urllib.parse
|
||||
|
||||
from .common import InfoExtractor
|
||||
from ..utils import js_to_json
|
||||
from ..utils import ExtractorError, determine_ext, join_nonempty, js_to_json
|
||||
|
||||
|
||||
def decode_b64_url(code):
|
||||
decoded_url = re.match(r'[^[]*\[([^]]*)\]', code).groups()[0]
|
||||
return base64.b64decode(
|
||||
urllib.parse.unquote(re.sub(r'[\s"\',]', '', decoded_url)),
|
||||
).decode('utf-8')
|
||||
|
||||
|
||||
class RTPIE(InfoExtractor):
|
||||
_VALID_URL = r'https?://(?:www\.)?rtp\.pt/play/p(?P<program_id>[0-9]+)/(?P<id>[^/?#]+)/?'
|
||||
_VALID_URL = r'https?://(?:(?:(?:www\.)?rtp\.pt/play/(?P<subarea>.*/)?p(?P<program_id>[0-9]+)/(?P<episode_id>e[0-9]+/)?)|(?:arquivos\.rtp\.pt/conteudos/))(?P<id>[^/?#]+)/?'
|
||||
_TESTS = [{
|
||||
'url': 'http://www.rtp.pt/play/p405/e174042/paixoes-cruzadas',
|
||||
'md5': 'e736ce0c665e459ddb818546220b4ef8',
|
||||
'url': 'https://www.rtp.pt/play/p9165/e562949/por-do-sol',
|
||||
'info_dict': {
|
||||
'id': 'e174042',
|
||||
'ext': 'mp3',
|
||||
'title': 'Paixões Cruzadas',
|
||||
'description': 'As paixões musicais de António Cartaxo e António Macedo',
|
||||
'id': 'por-do-sol',
|
||||
'ext': 'mp4',
|
||||
'title': 'Pôr do Sol Episódio 1 - de 16 Ago 2021',
|
||||
'description': 'Madalena Bourbon de Linhaça vive atormentada pelo segredo que esconde desde 1990. Matilde Bourbon de Linhaça sonha fugir com o seu amor proibido. O en',
|
||||
'thumbnail': r're:^https?://.*\.jpg',
|
||||
},
|
||||
}, {
|
||||
'url': 'http://www.rtp.pt/play/p831/a-quimica-das-coisas',
|
||||
'url': 'https://www.rtp.pt/play/p510/aleixo-fm',
|
||||
'only_matching': True,
|
||||
}]
|
||||
|
||||
|
|
@ -44,39 +50,60 @@ class RTPIE(InfoExtractor):
|
|||
video_id = self._match_id(url)
|
||||
|
||||
webpage = self._download_webpage(url, video_id)
|
||||
title = self._html_search_meta(
|
||||
'twitter:title', webpage, display_name='title', fatal=True)
|
||||
|
||||
f, config = self._search_regex(
|
||||
r'''(?sx)
|
||||
var\s+f\s*=\s*(?P<f>".*?"|{[^;]+?});\s*
|
||||
var\s+player1\s+=\s+new\s+RTPPlayer\s*\((?P<config>{(?:(?!\*/).)+?})\);(?!\s*\*/)
|
||||
''', webpage,
|
||||
'player config', group=('f', 'config'))
|
||||
# Remove comments from webpage source
|
||||
webpage = re.sub(r'(?s)/\*.*\*/', '', webpage)
|
||||
webpage = re.sub(r'(?m)(?:^|\s)//.*$', '', webpage)
|
||||
|
||||
f = self._parse_json(
|
||||
f, video_id,
|
||||
lambda data: self.__unobfuscate(data, video_id=video_id))
|
||||
config = self._parse_json(
|
||||
config, video_id,
|
||||
lambda data: self.__unobfuscate(data, video_id=video_id))
|
||||
title = self._html_search_regex(r'<title>(.+?)</title>', webpage, 'title', default='')
|
||||
# Replace irrelevant text in title
|
||||
title = title.replace(' - RTP Play - RTP', '') or self._html_search_meta('twitter:title', webpage)
|
||||
|
||||
formats = []
|
||||
if isinstance(f, dict):
|
||||
f_hls = f.get('hls')
|
||||
if f_hls is not None:
|
||||
formats.extend(self._extract_m3u8_formats(
|
||||
f_hls, video_id, 'mp4', 'm3u8_native', m3u8_id='hls'))
|
||||
if 'Este episódio não se encontra disponí' in title:
|
||||
raise ExtractorError('Episode unavailable', expected=True)
|
||||
|
||||
f_dash = f.get('dash')
|
||||
if f_dash is not None:
|
||||
formats.extend(self._extract_mpd_formats(f_dash, video_id, mpd_id='dash'))
|
||||
part = self._html_search_regex(r'section\-parts.*<span.*>(.+?)</span>.*</ul>', webpage, 'part', default=None)
|
||||
title = join_nonempty(title, part, delim=' ')
|
||||
|
||||
# Get JS object
|
||||
js_object = self._search_regex(r'(?s)RTPPlayer *\( *({.+?}) *\);', webpage, 'player config')
|
||||
json_string_for_config = ''
|
||||
full_url = None
|
||||
|
||||
# Verify JS object since it isn't pure JSON and maybe it needs some tuning
|
||||
for line in js_object.splitlines():
|
||||
stripped_line = line.strip()
|
||||
|
||||
# key == 'fileKey', then we found what we wanted
|
||||
if re.match(r'fileKey:', stripped_line):
|
||||
if re.match(r'fileKey: *""', stripped_line):
|
||||
raise ExtractorError('Episode not found (probably removed)', expected=True)
|
||||
url = decode_b64_url(stripped_line)
|
||||
if 'mp3' in url:
|
||||
full_url = 'https://cdn-ondemand.rtp.pt' + url
|
||||
else:
|
||||
full_url = f'https://streaming-vod.rtp.pt/dash{url}/manifest.mpd'
|
||||
|
||||
elif not stripped_line.startswith('//') and not re.match('file *:', stripped_line) and not re.match('.*extraSettings ?:', stripped_line):
|
||||
# Ignore commented lines, `extraSettings` and `f`. The latter seems to some random unrelated video.
|
||||
json_string_for_config += '\n' + line
|
||||
|
||||
if not full_url:
|
||||
raise ExtractorError('No valid media source found in page')
|
||||
|
||||
# Finally send pure JSON string for JSON parsing
|
||||
config = self._parse_json(json_string_for_config, video_id, js_to_json)
|
||||
full_url = full_url.replace('drm-dash', 'dash')
|
||||
ext = determine_ext(full_url)
|
||||
|
||||
if ext == 'mpd':
|
||||
# Download via mpd file
|
||||
formats = self._extract_mpd_formats(full_url, video_id)
|
||||
else:
|
||||
formats.append({
|
||||
'format_id': 'f',
|
||||
'url': f,
|
||||
'vcodec': 'none' if config.get('mediaType') == 'audio' else None,
|
||||
})
|
||||
formats = [{
|
||||
'url': full_url,
|
||||
'ext': ext,
|
||||
}]
|
||||
|
||||
subtitles = {}
|
||||
|
||||
|
|
|
|||
Loading…
Reference in New Issue