commit b966740cf786d3755c4849b12c9559d4bfe8ec15
parent 220828f2d6c0147ddcf5c16ff856d8df56332eb7
Author: Remita Amine <remitamine@gmail.com>
Date:   Thu,  4 Apr 2019 14:50:16 +0100

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

Diffstat:
Myoutube_dl/extractor/adn.py | 74+++++++++++++++++++++++++++++++++++++++++++++++++++-----------------------
1 file changed, 51 insertions(+), 23 deletions(-)

diff --git a/youtube_dl/extractor/adn.py b/youtube_dl/extractor/adn.py @@ -21,7 +21,6 @@ from ..utils import ( intlist_to_bytes, long_to_bytes, pkcs1pad, - srt_subtitles_timecode, strip_or_none, urljoin, ) @@ -42,6 +41,18 @@ class ADNIE(InfoExtractor): } _BASE_URL = 'http://animedigitalnetwork.fr' _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): if not sub_path: @@ -49,14 +60,14 @@ class ADNIE(InfoExtractor): enc_subtitles = self._download_webpage( urljoin(self._BASE_URL, sub_path), - video_id, fatal=False) + video_id, 'Downloading subtitles data', fatal=False) if not enc_subtitles: return None # http://animedigitalnetwork.fr/components/com_vodvideo/videojs/adn-vjs.min.js dec_subtitles = intlist_to_bytes(aes_cbc_decrypt( 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])) )) subtitles_json = self._parse_json( @@ -67,23 +78,27 @@ class ADNIE(InfoExtractor): subtitles = {} for sub_lang, sub in subtitles_json.items(): - srt = '' - for num, current in enumerate(sub): - start, end, text = ( + ssa = '''[Script Info] +ScriptType:V4.00 +[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('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: continue - srt += os.linesep.join( - ( - '%d' % num, - '%s --> %s' % ( - srt_subtitles_timecode(start), - srt_subtitles_timecode(end)), - text, - os.linesep, - )) + 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' % ( + self._ass_subtitles_timecode(start), + self._ass_subtitles_timecode(end), + '{\\a%d}' % alignment if alignment != 2 else '', + text.replace('\n', '\\N').replace('<i>', '{\\i1}').replace('</i>', '{\\i0}')) if sub_lang == 'vostf': sub_lang = 'fr' @@ -91,8 +106,8 @@ class ADNIE(InfoExtractor): 'ext': 'json', 'data': json.dumps(sub), }, { - 'ext': 'srt', - 'data': srt, + 'ext': 'ssa', + 'data': ssa, }]) return subtitles @@ -100,7 +115,15 @@ class ADNIE(InfoExtractor): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) 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_str = self._search_regex( @@ -129,12 +152,15 @@ class ADNIE(InfoExtractor): encrypted_message = long_to_bytes(pow(bytes_to_long(padded_message), e, n)) authorization = base64.b64encode(encrypted_message).decode() 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, }) links = links_data.get('links') 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') title = metas.get('title') or video_info['title'] @@ -142,9 +168,11 @@ class ADNIE(InfoExtractor): for format_id, qualities in links.items(): if not isinstance(qualities, dict): continue - for load_balancer_url in qualities.values(): + for quality, load_balancer_url in qualities.items(): 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') if not m3u8_url: continue