[adn] fix extraction and add support for positioning styles(closes #20549)

This commit is contained in:
Remita Amine 2019-04-04 14:50:16 +01:00
parent 220828f2d6
commit b966740cf7

View file

@ -21,7 +21,6 @@ from ..utils import (
intlist_to_bytes, intlist_to_bytes,
long_to_bytes, long_to_bytes,
pkcs1pad, pkcs1pad,
srt_subtitles_timecode,
strip_or_none, strip_or_none,
urljoin, urljoin,
) )
@ -42,6 +41,18 @@ class ADNIE(InfoExtractor):
} }
_BASE_URL = 'http://animedigitalnetwork.fr' _BASE_URL = 'http://animedigitalnetwork.fr'
_RSA_KEY = (0xc35ae1e4356b65a73b551493da94b8cb443491c0aa092a357a5aee57ffc14dda85326f42d716e539a34542a0d3f363adf16c5ec222d713d5997194030ee2e4f0d1fb328c01a81cf6868c090d50de8e169c6b13d1675b9eeed1cbc51e1fffca9b38af07f37abd790924cd3bee59d0257cfda4fe5f3f0534877e21ce5821447d1b, 65537) _RSA_KEY = (0xc35ae1e4356b65a73b551493da94b8cb443491c0aa092a357a5aee57ffc14dda85326f42d716e539a34542a0d3f363adf16c5ec222d713d5997194030ee2e4f0d1fb328c01a81cf6868c090d50de8e169c6b13d1675b9eeed1cbc51e1fffca9b38af07f37abd790924cd3bee59d0257cfda4fe5f3f0534877e21ce5821447d1b, 65537)
_POS_ALIGN_MAP = {
'start': 1,
'end': 3,
}
_LINE_ALIGN_MAP = {
'middle': 8,
'end': 4,
}
@staticmethod
def _ass_subtitles_timecode(seconds):
return '%01d:%02d:%02d.%02d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 100)
def _get_subtitles(self, sub_path, video_id): def _get_subtitles(self, sub_path, video_id):
if not sub_path: if not sub_path:
@ -49,14 +60,14 @@ class ADNIE(InfoExtractor):
enc_subtitles = self._download_webpage( enc_subtitles = self._download_webpage(
urljoin(self._BASE_URL, sub_path), urljoin(self._BASE_URL, sub_path),
video_id, fatal=False) video_id, 'Downloading subtitles data', fatal=False)
if not enc_subtitles: if not enc_subtitles:
return None return None
# http://animedigitalnetwork.fr/components/com_vodvideo/videojs/adn-vjs.min.js # http://animedigitalnetwork.fr/components/com_vodvideo/videojs/adn-vjs.min.js
dec_subtitles = intlist_to_bytes(aes_cbc_decrypt( dec_subtitles = intlist_to_bytes(aes_cbc_decrypt(
bytes_to_intlist(compat_b64decode(enc_subtitles[24:])), bytes_to_intlist(compat_b64decode(enc_subtitles[24:])),
bytes_to_intlist(binascii.unhexlify(self._K + '9032ad7083106400')), bytes_to_intlist(binascii.unhexlify(self._K + '083db5aebd9353b4')),
bytes_to_intlist(compat_b64decode(enc_subtitles[:24])) bytes_to_intlist(compat_b64decode(enc_subtitles[:24]))
)) ))
subtitles_json = self._parse_json( subtitles_json = self._parse_json(
@ -67,23 +78,27 @@ class ADNIE(InfoExtractor):
subtitles = {} subtitles = {}
for sub_lang, sub in subtitles_json.items(): for sub_lang, sub in subtitles_json.items():
srt = '' ssa = '''[Script Info]
for num, current in enumerate(sub): ScriptType:V4.00
start, end, text = ( [V4 Styles]
Format:Name,Fontname,Fontsize,PrimaryColour,Bold,BorderStyle,Outline,Alignment,MarginL,MarginR,MarginV
Style:Default,Arial,18,16777215,-1,1,1,2,20,20,20
[Events]
Format:Marked,Start,End,Style,Text'''
for current in sub:
start, end, text, line_align, position_align = (
float_or_none(current.get('startTime')), float_or_none(current.get('startTime')),
float_or_none(current.get('endTime')), float_or_none(current.get('endTime')),
current.get('text')) current.get('text'), current.get('lineAlign'),
current.get('positionAlign'))
if start is None or end is None or text is None: if start is None or end is None or text is None:
continue continue
srt += os.linesep.join( alignment = self._POS_ALIGN_MAP.get(position_align, 2) + self._LINE_ALIGN_MAP.get(line_align, 0)
( ssa += os.linesep + 'Dialogue:Marked=0,%s,%s,Default,%s%s' % (
'%d' % num, self._ass_subtitles_timecode(start),
'%s --> %s' % ( self._ass_subtitles_timecode(end),
srt_subtitles_timecode(start), '{\\a%d}' % alignment if alignment != 2 else '',
srt_subtitles_timecode(end)), text.replace('\n', '\\N').replace('<i>', '{\\i1}').replace('</i>', '{\\i0}'))
text,
os.linesep,
))
if sub_lang == 'vostf': if sub_lang == 'vostf':
sub_lang = 'fr' sub_lang = 'fr'
@ -91,8 +106,8 @@ class ADNIE(InfoExtractor):
'ext': 'json', 'ext': 'json',
'data': json.dumps(sub), 'data': json.dumps(sub),
}, { }, {
'ext': 'srt', 'ext': 'ssa',
'data': srt, 'data': ssa,
}]) }])
return subtitles return subtitles
@ -100,7 +115,15 @@ class ADNIE(InfoExtractor):
video_id = self._match_id(url) video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
player_config = self._parse_json(self._search_regex( player_config = self._parse_json(self._search_regex(
r'playerConfig\s*=\s*({.+});', webpage, 'player config'), video_id) r'playerConfig\s*=\s*({.+});', webpage,
'player config', default='{}'), video_id, fatal=False)
if not player_config:
config_url = urljoin(self._BASE_URL, self._search_regex(
r'(?:id="player"|class="[^"]*adn-player-container[^"]*")[^>]+data-url="([^"]+)"',
webpage, 'config url'))
player_config = self._download_json(
config_url, video_id,
'Downloading player config JSON metadata')['player']
video_info = {} video_info = {}
video_info_str = self._search_regex( video_info_str = self._search_regex(
@ -129,12 +152,15 @@ class ADNIE(InfoExtractor):
encrypted_message = long_to_bytes(pow(bytes_to_long(padded_message), e, n)) encrypted_message = long_to_bytes(pow(bytes_to_long(padded_message), e, n))
authorization = base64.b64encode(encrypted_message).decode() authorization = base64.b64encode(encrypted_message).decode()
links_data = self._download_json( links_data = self._download_json(
urljoin(self._BASE_URL, links_url), video_id, headers={ urljoin(self._BASE_URL, links_url), video_id,
'Downloading links JSON metadata', headers={
'Authorization': 'Bearer ' + authorization, 'Authorization': 'Bearer ' + authorization,
}) })
links = links_data.get('links') or {} links = links_data.get('links') or {}
metas = metas or links_data.get('meta') or {} metas = metas or links_data.get('meta') or {}
sub_path = (sub_path or links_data.get('subtitles')) + '&token=' + token sub_path = sub_path or links_data.get('subtitles') or \
'index.php?option=com_vodapi&task=subtitles.getJSON&format=json&id=' + video_id
sub_path += '&token=' + token
error = links_data.get('error') error = links_data.get('error')
title = metas.get('title') or video_info['title'] title = metas.get('title') or video_info['title']
@ -142,9 +168,11 @@ class ADNIE(InfoExtractor):
for format_id, qualities in links.items(): for format_id, qualities in links.items():
if not isinstance(qualities, dict): if not isinstance(qualities, dict):
continue continue
for load_balancer_url in qualities.values(): for quality, load_balancer_url in qualities.items():
load_balancer_data = self._download_json( load_balancer_data = self._download_json(
load_balancer_url, video_id, fatal=False) or {} load_balancer_url, video_id,
'Downloading %s %s JSON metadata' % (format_id, quality),
fatal=False) or {}
m3u8_url = load_balancer_data.get('location') m3u8_url = load_balancer_data.get('location')
if not m3u8_url: if not m3u8_url:
continue continue