X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/38bb9b1b0a044cabaf5691553815e334cd2e9213..aebf9c8f14c9f5d45f8c6aa523ee135fdfd421b2:/youtube_dl/extractor/crunchyroll.py diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py index d1b6d73..c7032ff 100644 --- a/youtube_dl/extractor/crunchyroll.py +++ b/youtube_dl/extractor/crunchyroll.py @@ -5,31 +5,85 @@ import re import json import base64 import zlib -import xml.etree.ElementTree from hashlib import sha1 from math import pow, sqrt, floor from .common import InfoExtractor from ..compat import ( + compat_etree_fromstring, compat_urllib_parse, compat_urllib_parse_unquote, compat_urllib_request, + compat_urlparse, ) from ..utils import ( ExtractorError, bytes_to_intlist, intlist_to_bytes, + int_or_none, + lowercase_escape, + remove_end, + sanitized_Request, unified_strdate, urlencode_postdata, + xpath_text, ) from ..aes import ( aes_cbc_decrypt, ) -class CrunchyrollIE(InfoExtractor): - _VALID_URL = r'https?://(?:(?Pwww|m)\.)?(?Pcrunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P[0-9]+))(?:[/?&]|$)' +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, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None): + 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, video_id, note, errnote, fatal, tries, timeout, encoding) + + @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', 'info_dict': { @@ -52,7 +106,7 @@ class CrunchyrollIE(InfoExtractor): 'id': '589804', 'ext': 'flv', 'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11', - 'description': 'md5:fe2743efedb49d279552926d0bd0cd9e', + 'description': 'md5:2fbc01f90b87e8e9137296f37b461c12', 'thumbnail': 're:^https?://.*\.jpg$', 'uploader': 'Danny Choo Network', 'upload_date': '20120213', @@ -61,10 +115,13 @@ class CrunchyrollIE(InfoExtractor): # 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 = { @@ -74,24 +131,6 @@ class CrunchyrollIE(InfoExtractor): '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(base64.b64decode(data.encode('utf-8'))) iv = bytes_to_intlist(base64.b64decode(iv.encode('utf-8'))) @@ -141,40 +180,40 @@ class CrunchyrollIE(InfoExtractor): 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 += """ @@ -183,21 +222,21 @@ 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 _extract_subtitles(self, subtitle): - sub_root = xml.etree.ElementTree.fromstring(subtitle) + sub_root = compat_etree_fromstring(subtitle) return [{ 'ext': 'srt', 'data': self._convert_subtitles_to_srt(sub_root), @@ -208,7 +247,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text def _get_subtitles(self, video_id, webpage): subtitles = {} - for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage): + 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) @@ -234,8 +273,10 @@ 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) @@ -245,18 +286,29 @@ 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) + video_uploader = self._html_search_regex( + r']+href="/publisher/[^"]+"[^>]*>([^<]+)', webpage, + 'video_uploader', fatal=False) 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 = sanitized_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') @@ -268,7 +320,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text for fmt in re.findall(r'showmedia\.([0-9]{3,4})p', webpage): stream_quality, stream_format = self._FORMAT_IDS[fmt] video_format = fmt + 'p' - streamdata_req = compat_urllib_request.Request( + streamdata_req = sanitized_Request( 'http://www.crunchyroll.com/xml/?req=RpcApiVideoPlayer_GetStandardConfig&media_id=%s&video_format=%s&video_quality=%s' % (stream_id, stream_format, stream_quality), compat_urllib_parse.urlencode({'current_page': url}).encode('utf-8')) @@ -277,15 +329,37 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text streamdata_req, video_id, note='Downloading media info for %s' % video_format) stream_info = streamdata.find('./{default}preload/stream_info') - video_url = stream_info.find('./host').text - video_play_path = stream_info.find('./file').text - formats.append({ + 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) subtitles = self.extract_subtitles(video_id, webpage) @@ -301,9 +375,9 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text } -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', @@ -312,12 +386,25 @@ 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')