X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/1d3fd83f473663fce3e0a10303473a38d80cc3d0..ce7bc3aa7fcf5bc302342428c41df23fa8ce2887:/youtube_dl/extractor/crunchyroll.py diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py index 9c6cf00..3efdc8c 100644 --- a/youtube_dl/extractor/crunchyroll.py +++ b/youtube_dl/extractor/crunchyroll.py @@ -3,13 +3,13 @@ from __future__ import unicode_literals import re import json -import base64 import zlib from hashlib import sha1 from math import pow, sqrt, floor from .common import InfoExtractor from ..compat import ( + compat_b64decode, compat_etree_fromstring, compat_urllib_parse_urlencode, compat_urllib_request, @@ -38,11 +38,32 @@ class CrunchyrollBaseIE(InfoExtractor): _LOGIN_FORM = 'login_form' _NETRC_MACHINE = 'crunchyroll' + def _call_rpc_api(self, method, video_id, note=None, data=None): + data = data or {} + data['req'] = 'RpcApi' + method + data = compat_urllib_parse_urlencode(data).encode('utf-8') + return self._download_xml( + 'http://www.crunchyroll.com/xml/', + video_id, note, fatal=False, data=data, headers={ + 'Content-Type': 'application/x-www-form-urlencoded', + }) + def _login(self): (username, password) = self._get_login_info() if username is None: return + self._download_webpage( + 'https://www.crunchyroll.com/?a=formhandler', + None, 'Logging in', 'Wrong login info', + data=urlencode_postdata({ + 'formname': 'RpcApiUser_Login', + 'next_url': 'https://www.crunchyroll.com/acct/membership', + 'name': username, + 'password': password, + })) + + ''' login_page = self._download_webpage( self._LOGIN_URL, None, 'Downloading login page') @@ -86,6 +107,7 @@ class CrunchyrollBaseIE(InfoExtractor): raise ExtractorError('Unable to login: %s' % error, expected=True) raise ExtractorError('Unable to log in') + ''' def _real_initialize(self): self._login() @@ -171,14 +193,15 @@ class CrunchyrollIE(CrunchyrollBaseIE): 'info_dict': { 'id': '727589', 'ext': 'mp4', - 'title': "KONOSUBA -God's blessing on this wonderful world! 2 Episode 1 – Give Me Deliverance from this Judicial Injustice!", + 'title': "KONOSUBA -God's blessing on this wonderful world! 2 Episode 1 – Give Me Deliverance From This Judicial Injustice!", 'description': 'md5:cbcf05e528124b0f3a0a419fc805ea7d', 'thumbnail': r're:^https?://.*\.jpg$', 'uploader': 'Kadokawa Pictures Inc.', 'upload_date': '20170118', 'series': "KONOSUBA -God's blessing on this wonderful world!", + 'season': "KONOSUBA -God's blessing on this wonderful world! 2", 'season_number': 2, - 'episode': 'Give Me Deliverance from this Judicial Injustice!', + 'episode': 'Give Me Deliverance From This Judicial Injustice!', 'episode_number': 1, }, 'params': { @@ -222,6 +245,23 @@ class CrunchyrollIE(CrunchyrollBaseIE): # just test metadata extraction 'skip_download': True, }, + }, { + # A video with a vastly different season name compared to the series name + 'url': 'http://www.crunchyroll.com/nyarko-san-another-crawling-chaos/episode-1-test-590532', + 'info_dict': { + 'id': '590532', + 'ext': 'mp4', + 'title': 'Haiyoru! Nyaruani (ONA) Episode 1 – Test', + 'description': 'Mahiro and Nyaruko talk about official certification.', + 'uploader': 'TV TOKYO', + 'upload_date': '20120305', + 'series': 'Nyarko-san: Another Crawling Chaos', + 'season': 'Haiyoru! Nyaruani (ONA)', + }, + 'params': { + # Just test metadata extraction + 'skip_download': True, + }, }] _FORMAT_IDS = { @@ -232,8 +272,8 @@ class CrunchyrollIE(CrunchyrollBaseIE): } 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'))) + data = bytes_to_intlist(compat_b64decode(data)) + iv = bytes_to_intlist(compat_b64decode(iv)) id = int(id) def obfuscate_key_aux(count, modulo, start): @@ -347,15 +387,19 @@ 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'\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: + sub_doc = self._call_rpc_api( + 'Subtitle_GetXml', video_id, + 'Downloading subtitles for ' + sub_name, data={ + 'subtitle_script_id': sub_id, + }) + if sub_doc is None: continue - subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8') + sid = sub_doc.get('id') + iv = xpath_text(sub_doc, 'iv', 'subtitle iv') + data = xpath_text(sub_doc, 'data', 'subtitle data') + if not sid or not iv or not data: + continue + subtitle = self._decrypt_subtitles(data, iv, sid).decode('utf-8') lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False) if not lang_code: continue @@ -372,7 +416,9 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text else: webpage_url = 'http://www.' + mobj.group('url') - webpage = self._download_webpage(self._add_skip_wall(webpage_url), video_id, 'Downloading webpage') + webpage = self._download_webpage( + self._add_skip_wall(webpage_url), video_id, + headers=self.geo_verification_headers()) note_m = self._html_search_regex( r'
(.+?)
', webpage, 'trailer-notice', default='') @@ -424,65 +470,79 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text for fmt in available_fmts: stream_quality, stream_format = self._FORMAT_IDS[fmt] 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 = self._download_xml( - streamdata_req, video_id, - note='Downloading media info for %s' % video_format) - 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_file = xpath_text(stream_info, './file') - if not video_file: - continue - if video_file.startswith('http'): - formats.extend(self._extract_m3u8_formats( - video_file, video_id, 'mp4', entry_protocol='m3u8_native', - m3u8_id='hls', fatal=False)) - continue - - video_url = xpath_text(stream_info, './host') - if not video_url: - 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_file.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) + stream_infos = [] + streamdata = self._call_rpc_api( + 'VideoPlayer_GetStandardConfig', video_id, + 'Downloading media info for %s' % video_format, data={ + 'media_id': video_id, + 'video_format': stream_format, + 'video_quality': stream_quality, + 'current_page': url, + }) + if streamdata is not None: + stream_info = streamdata.find('./{default}preload/stream_info') + if stream_info is not None: + stream_infos.append(stream_info) + stream_info = self._call_rpc_api( + 'VideoEncode_GetStreamInfo', video_id, + 'Downloading stream info for %s' % video_format, data={ + 'media_id': video_id, + 'video_format': stream_format, + 'video_encode_quality': stream_quality, + }) + if stream_info is not None: + stream_infos.append(stream_info) + for stream_info in stream_infos: + 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) - format_info.update({ - 'url': video_url, - 'play_path': video_file, - 'ext': 'flv', - }) - formats.append(format_info) - self._sort_formats(formats) + video_file = xpath_text(stream_info, './file') + if not video_file: + continue + if video_file.startswith('http'): + formats.extend(self._extract_m3u8_formats( + video_file, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False)) + continue - metadata = self._download_xml( - 'http://www.crunchyroll.com/xml', video_id, - note='Downloading media info', query={ - 'req': 'RpcApiVideoPlayer_GetMediaMetadata', + video_url = xpath_text(stream_info, './host') + if not video_url: + continue + metadata = stream_info.find('./metadata') + format_info = { + 'format': 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_file.split(':')[-1]))) + if self._is_valid_url(direct_video_url, video_id, video_format): + format_info.update({ + 'format_id': 'http-' + video_format, + 'url': direct_video_url, + }) + formats.append(format_info) + continue + + format_info.update({ + 'format_id': 'rtmp-' + video_format, + 'url': video_url, + 'play_path': video_file, + 'ext': 'flv', + }) + formats.append(format_info) + self._sort_formats(formats, ('height', 'width', 'tbr', 'fps')) + + metadata = self._call_rpc_api( + 'VideoPlayer_GetMediaMetadata', video_id, + note='Downloading media info', data={ 'media_id': video_id, }) @@ -490,14 +550,15 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text # webpage provide more accurate data than series_title from XML series = self._html_search_regex( - r'id=["\']showmedia_about_episode_num[^>]+>\s*]+>([^<]+)', - webpage, 'series', default=xpath_text(metadata, 'series_title')) + r'(?s)]+\bid=["\']showmedia_about_episode_num[^>]+>(.+?)]+id=["\']showmedia_about_episode_num[^>]+>.+?\s*

\s*Season (\d+)', + r'(?s)]+id=["\']showmedia_about_episode_num[^>]+>.+?\s*

\s*Season (\d+)', webpage, 'season number', default=None)) return { @@ -508,6 +569,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text 'uploader': video_uploader, 'upload_date': video_upload_date, 'series': series, + 'season': season, 'season_number': season_number, 'episode': episode, 'episode_number': episode_number, @@ -545,7 +607,9 @@ class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE): def _real_extract(self, url): show_id = self._match_id(url) - webpage = self._download_webpage(self._add_skip_wall(url), show_id) + webpage = self._download_webpage( + self._add_skip_wall(url), show_id, + headers=self.geo_verification_headers()) title = self._html_search_regex( r'(?s)]*>\s*(.*?)', webpage, 'title')