X-Git-Url: http://git.bitcoin.ninja/index.cgi?p=youtube-dl;a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fcrunchyroll.py;h=c38fd095a3ea95facf214e6eb350fc71a20b39a6;hp=f8ce10111df8078638d9c56b837f354b2345ad73;hb=dcdb292fddc82ae11f4c0b647815a45c88a6b6d5;hpb=769078755318896fa1a6c5c8aba6f76d6aeddf78 diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py index f8ce10111..c38fd095a 100644 --- a/youtube_dl/extractor/crunchyroll.py +++ b/youtube_dl/extractor/crunchyroll.py @@ -1,18 +1,17 @@ -# encoding: utf-8 +# coding: utf-8 from __future__ import unicode_literals 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_urllib_parse, - compat_urllib_parse_unquote, + compat_etree_fromstring, + compat_urllib_parse_urlencode, compat_urllib_request, compat_urlparse, ) @@ -21,10 +20,13 @@ from ..utils import ( bytes_to_intlist, intlist_to_bytes, int_or_none, + lowercase_escape, remove_end, + sanitized_Request, unified_strdate, urlencode_postdata, xpath_text, + extract_attributes, ) from ..aes import ( aes_cbc_decrypt, @@ -32,29 +34,65 @@ from ..aes import ( class CrunchyrollBaseIE(InfoExtractor): + _LOGIN_URL = 'https://www.crunchyroll.com/login' + _LOGIN_FORM = 'login_form' _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_page = self._download_webpage( + self._LOGIN_URL, None, 'Downloading login page') + + def is_logged(webpage): + return 'Redirecting' in webpage + + # Already logged in + if is_logged(login_page): + return + + login_form_str = self._search_regex( + r'(?P<form><form[^>]+?id=(["\'])%s\2[^>]*>)' % self._LOGIN_FORM, + login_page, 'login form', group='form') + + post_url = extract_attributes(login_form_str).get('action') + if not post_url: + post_url = self._LOGIN_URL + elif not post_url.startswith('http'): + post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url) + + login_form = self._form_hidden_inputs(self._LOGIN_FORM, login_page) + + login_form.update({ + 'login_form[name]': username, + 'login_form[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') + + response = self._download_webpage( + post_url, None, 'Logging in', 'Wrong login info', + data=urlencode_postdata(login_form), + headers={'Content-Type': 'application/x-www-form-urlencoded'}) + + # Successful login + if is_logged(response): + return + + error = self._html_search_regex( + '(?s)<ul[^>]+class=["\']messages["\'][^>]*>(.+?)</ul>', + response, 'error message', default=None) + if error: + raise ExtractorError('Unable to login: %s' % error, expected=True) + + raise ExtractorError('Unable to log in') 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): + def _download_webpage(self, url_or_request, *args, **kwargs): request = (url_or_request if isinstance(url_or_request, compat_urllib_request.Request) - else compat_urllib_request.Request(url_or_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 @@ -63,8 +101,7 @@ class CrunchyrollBaseIE(InfoExtractor): # 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) + return super(CrunchyrollBaseIE, self)._download_webpage(request, *args, **kwargs) @staticmethod def _add_skip_wall(url): @@ -77,7 +114,7 @@ class CrunchyrollBaseIE(InfoExtractor): # 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))) + parsed_url._replace(query=compat_urllib_parse_urlencode(qs, True))) class CrunchyrollIE(CrunchyrollBaseIE): @@ -104,7 +141,7 @@ class CrunchyrollIE(CrunchyrollBaseIE): '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$', 'uploader': 'Danny Choo Network', 'upload_date': '20120213', @@ -113,6 +150,21 @@ class CrunchyrollIE(CrunchyrollBaseIE): # rtmp 'skip_download': True, }, + }, { + 'url': 'http://www.crunchyroll.com/rezero-starting-life-in-another-world-/episode-5-the-morning-of-our-promise-is-still-distant-702409', + 'info_dict': { + 'id': '702409', + 'ext': 'mp4', + 'title': 'Re:ZERO -Starting Life in Another World- Episode 5 – The Morning of Our Promise Is Still Distant', + 'description': 'md5:97664de1ab24bbf77a9c01918cb7dca9', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'TV TOKYO', + 'upload_date': '20160508', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }, { 'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697', 'only_matching': True, @@ -178,40 +230,40 @@ class CrunchyrollIE(CrunchyrollBaseIE): 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 += """ @@ -220,21 +272,21 @@ 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), @@ -287,11 +339,15 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text 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'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL) + video_title = self._html_search_regex( + r'(?s)<h1[^>]*>((?:(?!<h1).)*?<span[^>]+itemprop=["\']title["\'][^>]*>(?:(?!<h1).)+?)</h1>', + 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_description = self._html_search_regex( + r'<script[^>]*>\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'<div>Availability for free users:(.+?)</div>', r'<div>[^<>]+<span>\s*(.+?\d{4})\s*</span></div>'], webpage, 'video_upload_date', fatal=False, flags=re.DOTALL) @@ -301,30 +357,49 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text r'<a[^>]+href="/publisher/[^"]+"[^>]*>([^<]+)</a>', 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.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'<media_id>([^<]+)', playerdata, 'stream_id') - video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False) - + available_fmts = [] + for a, fmt in re.findall(r'(<a[^>]+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'showmedia\.([0-9]{3,4})p', webpage): + for fmt in available_fmts: 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')) + % (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_url = stream_info.find('./host').text - video_play_path = stream_info.find('./file').text + 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, @@ -338,7 +413,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text 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]))) + 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, @@ -348,10 +423,18 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text format_info.update({ 'url': video_url, - 'play_path': video_play_path, + 'play_path': video_file, 'ext': 'flv', }) 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 = self.extract_subtitles(video_id, webpage) @@ -359,16 +442,19 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text 'id': video_id, 'title': video_title, 'description': video_description, - 'thumbnail': video_thumbnail, + 'thumbnail': xpath_text(metadata, 'episode_image_url'), 'uploader': video_uploader, 'upload_date': video_upload_date, + '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(CrunchyrollBaseIE): - IE_NAME = "crunchyroll:playlist" + IE_NAME = 'crunchyroll:playlist' _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?!(?:news|anime-news|library|forum|launchcalendar|lineup|store|comics|freetrial|login))(?P<id>[\w\-]+))/?(?:\?|$)' _TESTS = [{