X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/46113edab215c2211a604c06245c16d5d4e57dcf..d1d8237037fb87a57bd1d073c3b9e642d4a75016:/youtube_dl/extractor/crunchyroll.py diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py index cc612d0..90a6430 100644 --- a/youtube_dl/extractor/crunchyroll.py +++ b/youtube_dl/extractor/crunchyroll.py @@ -5,33 +5,86 @@ import re import json import base64 import zlib -import xml.etree.ElementTree from hashlib import sha1 from math import pow, sqrt, floor -from .subtitles import SubtitlesInfoExtractor +from .common import InfoExtractor +from ..compat import ( + compat_etree_fromstring, + compat_urllib_parse_urlencode, + compat_urllib_request, + compat_urlparse, +) from ..utils import ( ExtractorError, - compat_urllib_parse, - compat_urllib_request, bytes_to_intlist, intlist_to_bytes, + int_or_none, + lowercase_escape, + remove_end, + sanitized_Request, unified_strdate, - clean_html, urlencode_postdata, + xpath_text, + extract_attributes, ) from ..aes import ( aes_cbc_decrypt, - inc, ) -from .common import InfoExtractor -class CrunchyrollIE(SubtitlesInfoExtractor): - _VALID_URL = r'https?://(?:(?Pwww|m)\.)?(?Pcrunchyroll\.com/(?:[^/]*/[^/?&]*?|media/\?id=)(?P[0-9]+))(?:[/?&]|$)' - _TEST = { +class CrunchyrollBaseIE(InfoExtractor): + _NETRC_MACHINE = 'crunchyroll' + + def _login(self): + (username, password) = self._get_login_info() + if username is None: + return + self.report_login() + login_url = 'https://www.crunchyroll.com/?a=formhandler' + data = urlencode_postdata({ + 'formname': 'RpcApiUser_Login', + 'name': username, + 'password': password, + }) + login_request = sanitized_Request(login_url, data) + login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') + self._download_webpage(login_request, None, False, 'Wrong login info') + + def _real_initialize(self): + self._login() + + def _download_webpage(self, url_or_request, *args, **kwargs): + request = (url_or_request if isinstance(url_or_request, compat_urllib_request.Request) + else sanitized_Request(url_or_request)) + # Accept-Language must be set explicitly to accept any language to avoid issues + # similar to https://github.com/rg3/youtube-dl/issues/6797. + # Along with IP address Crunchyroll uses Accept-Language to guess whether georestriction + # should be imposed or not (from what I can see it just takes the first language + # ignoring the priority and requires it to correspond the IP). By the way this causes + # Crunchyroll to not work in georestriction cases in some browsers that don't place + # the locale lang first in header. However allowing any language seems to workaround the issue. + request.add_header('Accept-Language', '*') + return super(CrunchyrollBaseIE, self)._download_webpage(request, *args, **kwargs) + + @staticmethod + def _add_skip_wall(url): + parsed_url = compat_urlparse.urlparse(url) + qs = compat_urlparse.parse_qs(parsed_url.query) + # Always force skip_wall to bypass maturity wall, namely 18+ confirmation message: + # > This content may be inappropriate for some people. + # > Are you sure you want to continue? + # since it's not disabled by default in crunchyroll account's settings. + # See https://github.com/rg3/youtube-dl/issues/7202. + qs['skip_wall'] = ['1'] + return compat_urlparse.urlunparse( + parsed_url._replace(query=compat_urllib_parse_urlencode(qs, True))) + + +class CrunchyrollIE(CrunchyrollBaseIE): + _VALID_URL = r'https?://(?:(?Pwww|m)\.)?(?Pcrunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P[0-9]+))(?:[/?&]|$)' + _TESTS = [{ 'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513', - #'md5': 'b1639fd6ddfaa43788c85f6d1dddd412', 'info_dict': { 'id': '645513', 'ext': 'flv', @@ -46,7 +99,29 @@ class CrunchyrollIE(SubtitlesInfoExtractor): # rtmp 'skip_download': True, }, - } + }, { + 'url': 'http://www.crunchyroll.com/media-589804/culture-japan-1', + 'info_dict': { + 'id': '589804', + 'ext': 'flv', + 'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11', + 'description': 'md5:2fbc01f90b87e8e9137296f37b461c12', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'Danny Choo Network', + 'upload_date': '20120213', + }, + 'params': { + # rtmp + 'skip_download': True, + }, + }, { + 'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697', + 'only_matching': True, + }, { + # geo-restricted (US), 18+ maturity wall, non-premium available + 'url': 'http://www.crunchyroll.com/cosplay-complex-ova/episode-1-the-birth-of-the-cosplay-club-565617', + 'only_matching': True, + }] _FORMAT_IDS = { '360': ('60', '106'), @@ -55,29 +130,9 @@ class CrunchyrollIE(SubtitlesInfoExtractor): '1080': ('80', '108'), } - def _login(self): - (username, password) = self._get_login_info() - if username is None: - return - self.report_login() - login_url = 'https://www.crunchyroll.com/?a=formhandler' - data = urlencode_postdata({ - 'formname': 'RpcApiUser_Login', - 'name': username, - 'password': password, - }) - login_request = compat_urllib_request.Request(login_url, data) - login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - self._download_webpage(login_request, None, False, 'Wrong login info') - - - def _real_initialize(self): - self._login() - - def _decrypt_subtitles(self, data, iv, id): - data = bytes_to_intlist(data) - iv = bytes_to_intlist(iv) + data = bytes_to_intlist(base64.b64decode(data.encode('utf-8'))) + iv = bytes_to_intlist(base64.b64decode(iv.encode('utf-8'))) id = int(id) def obfuscate_key_aux(count, modulo, start): @@ -100,12 +155,7 @@ class CrunchyrollIE(SubtitlesInfoExtractor): return shaHash + [0] * 12 key = obfuscate_key(id) - class Counter: - __value = iv - def next_value(self): - temp = self.__value - self.__value = inc(self.__value) - return temp + decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv)) return zlib.decompress(decrypted_data) @@ -129,40 +179,40 @@ class CrunchyrollIE(SubtitlesInfoExtractor): return assvalue output = '[Script Info]\n' - output += 'Title: %s\n' % sub_root.attrib["title"] + output += 'Title: %s\n' % sub_root.attrib['title'] output += 'ScriptType: v4.00+\n' - output += 'WrapStyle: %s\n' % sub_root.attrib["wrap_style"] - output += 'PlayResX: %s\n' % sub_root.attrib["play_res_x"] - output += 'PlayResY: %s\n' % sub_root.attrib["play_res_y"] + output += 'WrapStyle: %s\n' % sub_root.attrib['wrap_style'] + output += 'PlayResX: %s\n' % sub_root.attrib['play_res_x'] + output += 'PlayResY: %s\n' % sub_root.attrib['play_res_y'] output += """ScaledBorderAndShadow: yes [V4+ Styles] Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding """ for style in sub_root.findall('./styles/style'): - output += 'Style: ' + style.attrib["name"] - output += ',' + style.attrib["font_name"] - output += ',' + style.attrib["font_size"] - output += ',' + style.attrib["primary_colour"] - output += ',' + style.attrib["secondary_colour"] - output += ',' + style.attrib["outline_colour"] - output += ',' + style.attrib["back_colour"] - output += ',' + ass_bool(style.attrib["bold"]) - output += ',' + ass_bool(style.attrib["italic"]) - output += ',' + ass_bool(style.attrib["underline"]) - output += ',' + ass_bool(style.attrib["strikeout"]) - output += ',' + style.attrib["scale_x"] - output += ',' + style.attrib["scale_y"] - output += ',' + style.attrib["spacing"] - output += ',' + style.attrib["angle"] - output += ',' + style.attrib["border_style"] - output += ',' + style.attrib["outline"] - output += ',' + style.attrib["shadow"] - output += ',' + style.attrib["alignment"] - output += ',' + style.attrib["margin_l"] - output += ',' + style.attrib["margin_r"] - output += ',' + style.attrib["margin_v"] - output += ',' + style.attrib["encoding"] + output += 'Style: ' + style.attrib['name'] + output += ',' + style.attrib['font_name'] + output += ',' + style.attrib['font_size'] + output += ',' + style.attrib['primary_colour'] + output += ',' + style.attrib['secondary_colour'] + output += ',' + style.attrib['outline_colour'] + output += ',' + style.attrib['back_colour'] + output += ',' + ass_bool(style.attrib['bold']) + output += ',' + ass_bool(style.attrib['italic']) + output += ',' + ass_bool(style.attrib['underline']) + output += ',' + ass_bool(style.attrib['strikeout']) + output += ',' + style.attrib['scale_x'] + output += ',' + style.attrib['scale_y'] + output += ',' + style.attrib['spacing'] + output += ',' + style.attrib['angle'] + output += ',' + style.attrib['border_style'] + output += ',' + style.attrib['outline'] + output += ',' + style.attrib['shadow'] + output += ',' + style.attrib['alignment'] + output += ',' + style.attrib['margin_l'] + output += ',' + style.attrib['margin_r'] + output += ',' + style.attrib['margin_v'] + output += ',' + style.attrib['encoding'] output += '\n' output += """ @@ -171,20 +221,48 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text """ for event in sub_root.findall('./events/event'): output += 'Dialogue: 0' - output += ',' + event.attrib["start"] - output += ',' + event.attrib["end"] - output += ',' + event.attrib["style"] - output += ',' + event.attrib["name"] - output += ',' + event.attrib["margin_l"] - output += ',' + event.attrib["margin_r"] - output += ',' + event.attrib["margin_v"] - output += ',' + event.attrib["effect"] - output += ',' + event.attrib["text"] + output += ',' + event.attrib['start'] + output += ',' + event.attrib['end'] + output += ',' + event.attrib['style'] + output += ',' + event.attrib['name'] + output += ',' + event.attrib['margin_l'] + output += ',' + event.attrib['margin_r'] + output += ',' + event.attrib['margin_v'] + output += ',' + event.attrib['effect'] + output += ',' + event.attrib['text'] output += '\n' return output - def _real_extract(self,url): + def _extract_subtitles(self, subtitle): + sub_root = compat_etree_fromstring(subtitle) + return [{ + 'ext': 'srt', + 'data': self._convert_subtitles_to_srt(sub_root), + }, { + 'ext': 'ass', + 'data': self._convert_subtitles_to_ass(sub_root), + }] + + def _get_subtitles(self, video_id, webpage): + subtitles = {} + for sub_id, sub_name in re.findall(r'\bssid=([0-9]+)"[^>]+?\btitle="([^"]+)', webpage): + sub_page = self._download_webpage( + 'http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id=' + sub_id, + video_id, note='Downloading subtitles for ' + sub_name) + id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False) + iv = self._search_regex(r'([^<]+)', sub_page, 'subtitle_iv', fatal=False) + data = self._search_regex(r'([^<]+)', sub_page, 'subtitle_data', fatal=False) + if not id or not iv or not data: + continue + subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8') + lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False) + if not lang_code: + continue + subtitles[lang_code] = self._extract_subtitles(subtitle) + return subtitles + + def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('video_id') @@ -194,8 +272,10 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text else: webpage_url = 'http://www.' + mobj.group('url') - webpage = self._download_webpage(webpage_url, video_id, 'Downloading webpage') - note_m = self._html_search_regex(r'
(.+?)
', webpage, 'trailer-notice', default='') + webpage = self._download_webpage(self._add_skip_wall(webpage_url), video_id, 'Downloading webpage') + note_m = self._html_search_regex( + r'
(.+?)
', + webpage, 'trailer-notice', default='') if note_m: raise ExtractorError(note_m) @@ -205,92 +285,117 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text if msg.get('type') == 'error': raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True) - video_title = self._html_search_regex(r']*>(.+?)', webpage, 'video_title', flags=re.DOTALL) + if 'To view this, please log in to verify you are 18 or older.' in webpage: + self.raise_login_required() + + video_title = self._html_search_regex( + r'(?s)]*>((?:(?!]+itemprop=["\']title["\'][^>]*>(?:(?!', + webpage, 'video_title') video_title = re.sub(r' {2,}', ' ', video_title) - video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='') - if not video_description: - video_description = None - video_upload_date = self._html_search_regex(r'
Availability for free users:(.+?)
', webpage, 'video_upload_date', fatal=False, flags=re.DOTALL) + video_description = self._html_search_regex( + r']*>\s*.+?\[media_id=%s\].+?"description"\s*:\s*"([^"]+)' % video_id, + webpage, 'description', default=None) + if video_description: + video_description = lowercase_escape(video_description.replace(r'\r\n', '\n')) + video_upload_date = self._html_search_regex( + [r'
Availability for free users:(.+?)
', r'
[^<>]+\s*(.+?\d{4})\s*
'], + webpage, 'video_upload_date', fatal=False, flags=re.DOTALL) if video_upload_date: video_upload_date = unified_strdate(video_upload_date) - video_uploader = self._html_search_regex(r'
\s*Publisher:(.+?)
', webpage, 'video_uploader', fatal=False, flags=re.DOTALL) - - playerdata_url = compat_urllib_parse.unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url')) - playerdata_req = compat_urllib_request.Request(playerdata_url) - playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url}) - playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded') - playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info') - - stream_id = self._search_regex(r'([^<]+)', playerdata, 'stream_id') - video_thumbnail = self._search_regex(r'([^<]+)', playerdata, 'thumbnail', fatal=False) - + video_uploader = self._html_search_regex( + r']+href="/publisher/[^"]+"[^>]*>([^<]+)', webpage, + 'video_uploader', fatal=False) + + available_fmts = [] + for a, fmt in re.findall(r'(]+token=["\']showmedia\.([0-9]{3,4})p["\'][^>]+>)', webpage): + attrs = extract_attributes(a) + href = attrs.get('href') + if href and '/freetrial' in href: + continue + available_fmts.append(fmt) + if not available_fmts: + for p in (r'token=["\']showmedia\.([0-9]{3,4})p"', r'showmedia\.([0-9]{3,4})p'): + available_fmts = re.findall(p, webpage) + if available_fmts: + break + video_encode_ids = [] formats = [] - for fmt in re.findall(r'\?p([0-9]{3,4})=1', webpage): + for fmt in available_fmts: stream_quality, stream_format = self._FORMAT_IDS[fmt] - video_format = fmt+'p' - streamdata_req = compat_urllib_request.Request('http://www.crunchyroll.com/xml/') - # urlencode doesn't work! - streamdata_req.data = 'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality='+stream_quality+'&media%5Fid='+stream_id+'&video%5Fformat='+stream_format + video_format = fmt + 'p' + streamdata_req = sanitized_Request( + 'http://www.crunchyroll.com/xml/?req=RpcApiVideoPlayer_GetStandardConfig&media_id=%s&video_format=%s&video_quality=%s' + % (video_id, stream_format, stream_quality), + compat_urllib_parse_urlencode({'current_page': url}).encode('utf-8')) streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded') - streamdata_req.add_header('Content-Length', str(len(streamdata_req.data))) streamdata = self._download_xml( streamdata_req, video_id, note='Downloading media info for %s' % video_format) - video_url = streamdata.find('.//host').text - video_play_path = streamdata.find('.//file').text - formats.append({ + stream_info = streamdata.find('./{default}preload/stream_info') + video_encode_id = xpath_text(stream_info, './video_encode_id') + if video_encode_id in video_encode_ids: + continue + video_encode_ids.append(video_encode_id) + video_url = xpath_text(stream_info, './host') + video_play_path = xpath_text(stream_info, './file') + if not video_url or not video_play_path: + continue + metadata = stream_info.find('./metadata') + format_info = { + 'format': video_format, + 'format_id': video_format, + 'height': int_or_none(xpath_text(metadata, './height')), + 'width': int_or_none(xpath_text(metadata, './width')), + } + + if '.fplive.net/' in video_url: + video_url = re.sub(r'^rtmpe?://', 'http://', video_url.strip()) + parsed_video_url = compat_urlparse.urlparse(video_url) + direct_video_url = compat_urlparse.urlunparse(parsed_video_url._replace( + netloc='v.lvlt.crcdn.net', + path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_play_path.split(':')[-1]))) + if self._is_valid_url(direct_video_url, video_id, video_format): + format_info.update({ + 'url': direct_video_url, + }) + formats.append(format_info) + continue + + format_info.update({ 'url': video_url, 'play_path': video_play_path, 'ext': 'flv', - 'format': video_format, - 'format_id': video_format, + }) + formats.append(format_info) + self._sort_formats(formats) + + metadata = self._download_xml( + 'http://www.crunchyroll.com/xml', video_id, + note='Downloading media info', query={ + 'req': 'RpcApiVideoPlayer_GetMediaMetadata', + 'media_id': video_id, }) - subtitles = {} - sub_format = self._downloader.params.get('subtitlesformat', 'srt') - for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage): - sub_page = self._download_webpage('http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id='+sub_id,\ - video_id, note='Downloading subtitles for '+sub_name) - id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False) - iv = self._search_regex(r'([^<]+)', sub_page, 'subtitle_iv', fatal=False) - data = self._search_regex(r'([^<]+)', sub_page, 'subtitle_data', fatal=False) - if not id or not iv or not data: - continue - id = int(id) - iv = base64.b64decode(iv) - data = base64.b64decode(data) - - subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8') - lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False) - if not lang_code: - continue - sub_root = xml.etree.ElementTree.fromstring(subtitle) - if not sub_root: - subtitles[lang_code] = '' - if sub_format == 'ass': - subtitles[lang_code] = self._convert_subtitles_to_ass(sub_root) - else: - subtitles[lang_code] = self._convert_subtitles_to_srt(sub_root) - - if self._downloader.params.get('listsubtitles', False): - self._list_available_subtitles(video_id, subtitles) - return + subtitles = self.extract_subtitles(video_id, webpage) return { - 'id': video_id, - 'title': video_title, + 'id': video_id, + 'title': video_title, 'description': video_description, - 'thumbnail': video_thumbnail, - 'uploader': video_uploader, + 'thumbnail': xpath_text(metadata, 'episode_image_url'), + 'uploader': video_uploader, 'upload_date': video_upload_date, - 'subtitles': subtitles, - 'formats': formats, + 'series': xpath_text(metadata, 'series_title'), + 'episode': xpath_text(metadata, 'episode_title'), + 'episode_number': int_or_none(xpath_text(metadata, 'episode_number')), + 'subtitles': subtitles, + 'formats': formats, } -class CrunchyrollShowPlaylistIE(InfoExtractor): - IE_NAME = "crunchyroll:playlist" - _VALID_URL = r'https?://(?:(?Pwww|m)\.)?(?Pcrunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P[\w\-]+))/?$' +class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE): + IE_NAME = 'crunchyroll:playlist' + _VALID_URL = r'https?://(?:(?Pwww|m)\.)?(?Pcrunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P[\w\-]+))/?(?:\?|$)' _TESTS = [{ 'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi', @@ -299,12 +404,25 @@ class CrunchyrollShowPlaylistIE(InfoExtractor): 'title': 'A Bridge to the Starry Skies - Hoshizora e Kakaru Hashi' }, 'playlist_count': 13, + }, { + # geo-restricted (US), 18+ maturity wall, non-premium available + 'url': 'http://www.crunchyroll.com/cosplay-complex-ova', + 'info_dict': { + 'id': 'cosplay-complex-ova', + 'title': 'Cosplay Complex OVA' + }, + 'playlist_count': 3, + 'skip': 'Georestricted', + }, { + # geo-restricted (US), 18+ maturity wall, non-premium will be available since 2015.11.14 + 'url': 'http://www.crunchyroll.com/ladies-versus-butlers?skip_wall=1', + 'only_matching': True, }] def _real_extract(self, url): show_id = self._match_id(url) - webpage = self._download_webpage(url, show_id) + webpage = self._download_webpage(self._add_skip_wall(url), show_id) title = self._html_search_regex( r'(?s)]*>\s*(.*?)', webpage, 'title')