7cb0a7418a
Merge in livestreamer's functionality of fetching Crunchyroll video information from their mobile API, which contains an independent set of streams from the desktop API. Grab the additional formats, process them, and sort them accordingly.
615 lines
25 KiB
Python
615 lines
25 KiB
Python
# encoding: utf-8
|
||
from __future__ import unicode_literals
|
||
|
||
import random
|
||
import re
|
||
import string
|
||
import json
|
||
import base64
|
||
import zlib
|
||
|
||
from hashlib import sha1
|
||
from math import pow, sqrt, floor
|
||
from .common import InfoExtractor
|
||
from ..compat import (
|
||
compat_etree_fromstring,
|
||
compat_urllib_parse_urlencode,
|
||
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,
|
||
extract_attributes,
|
||
)
|
||
from ..aes import (
|
||
aes_cbc_decrypt,
|
||
)
|
||
|
||
|
||
class CrunchyrollBaseIE(InfoExtractor):
|
||
_NETRC_MACHINE = 'crunchyroll'
|
||
|
||
def __init__(self, *args, **kwargs):
|
||
super(CrunchyrollBaseIE, self).__init__(*args, **kwargs)
|
||
self.api_session_id = None
|
||
self.api_session_auth = None
|
||
self.api_device_id = ''.join(random.sample(string.ascii_letters + string.digits, 32))
|
||
|
||
def _login(self):
|
||
(username, password) = self._get_login_info()
|
||
if username is None:
|
||
return
|
||
|
||
# Log into main website
|
||
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')
|
||
|
||
# Start session with mobile API
|
||
res = self._api_call(
|
||
'start_session',
|
||
{
|
||
'device_id': self.api_device_id,
|
||
'device_type': 'com.crunchyroll.iphone',
|
||
'access_token': 'QWjz212GspMHH9h'
|
||
},
|
||
video_id=None,
|
||
note='Starting session with mobile API',
|
||
errnote='Could not start session with mobile API',
|
||
# If mobile API fails, we can always fall back on the regular website
|
||
fatal=False
|
||
)
|
||
|
||
if res is not None:
|
||
self.api_session_id = res.get('data', {'session_id': None}).get('session_id')
|
||
|
||
# Log into mobile API
|
||
res = self._api_call(
|
||
'login',
|
||
{
|
||
'account': username,
|
||
'password': password
|
||
},
|
||
video_id=None,
|
||
note='Logging in to mobile API',
|
||
errnote='Could not log into mobile API',
|
||
# If mobile API fails, we can always fall back on the regular website
|
||
fatal=False
|
||
)
|
||
|
||
if res is not None:
|
||
self.api_session_auth = res.get('data', {'auth': None}).get('auth')
|
||
|
||
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)
|
||
|
||
def _api_call(self, entrypoint, params, video_id, *args, **kwargs):
|
||
'''Makes a call against the api.
|
||
|
||
:param entrypoint: API method to call.
|
||
:param params: parameters to include in the request data.
|
||
:param schema: schema to use to validate the data
|
||
|
||
Source adapted from:
|
||
https://github.com/chrippa/livestreamer/blob/develop/src/livestreamer/plugins/crunchyroll.py
|
||
Copyright (c) 2011-2015, Christopher Rosell
|
||
License: https://github.com/chrippa/livestreamer/blob/develop/LICENSE
|
||
'''
|
||
|
||
url = 'https://api.crunchyroll.com/{0}.0.json'.format(entrypoint)
|
||
|
||
# Default params
|
||
params = dict(params)
|
||
params.update({
|
||
'version': '2313.8',
|
||
'locale': 'enUS',
|
||
})
|
||
|
||
if self.api_session_id:
|
||
params["session_id"] = self.api_session_id
|
||
|
||
# Headers
|
||
headers = {
|
||
'Host': 'api.crunchyroll.com',
|
||
'Accept-Encoding': 'gzip, deflzate',
|
||
'Accept': '*/*',
|
||
'Content-Type': 'application/x-www-form-urlencoded'
|
||
}
|
||
headers['User-Agent'] = 'Mozilla/5.0 (iPhone; iPhone OS 8.3.0; en_US)'
|
||
|
||
return self._download_json(url, video_id, query=params, headers=headers, *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?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)'
|
||
_TESTS = [{
|
||
'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
|
||
'info_dict': {
|
||
'id': '645513',
|
||
'ext': 'mp4',
|
||
'title': 'Wanna be the Strongest in the World Episode 1 – An Idol-Wrestler is Born!',
|
||
'description': 'md5:2d17137920c64f2f49981a7797d275ef',
|
||
'thumbnail': 'http://img1.ak.crunchyroll.com/i/spire1-tmb/20c6b5e10f1a47b10516877d3c039cae1380951166_full.jpg',
|
||
'uploader': 'Yomiuri Telecasting Corporation (YTV)',
|
||
'upload_date': '20131013',
|
||
'url': 're:(?!.*&)',
|
||
},
|
||
'params': {
|
||
# m3u8 download
|
||
'skip_download': True,
|
||
},
|
||
}, {
|
||
'url': 'http://www.crunchyroll.com/media-589804/culture-japan-1',
|
||
'info_dict': {
|
||
'id': '589804',
|
||
'ext': 'flv',
|
||
'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11',
|
||
'description': 'md5:2fbc01f90b87e8e9137296f37b461c12',
|
||
'thumbnail': 're:^https?://.*\.jpg$',
|
||
'uploader': 'Danny Choo Network',
|
||
'upload_date': '20120213',
|
||
},
|
||
'params': {
|
||
# rtmp
|
||
'skip_download': True,
|
||
},
|
||
}, {
|
||
'url': 'http://www.crunchyroll.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,
|
||
}, {
|
||
# 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'),
|
||
'480': ('61', '106'),
|
||
'720': ('62', '106'),
|
||
'1080': ('80', '108'),
|
||
}
|
||
|
||
# Crunchyroll does not give us bitrate data for the RTMP sources,
|
||
# so by default self._sort_formats() will put the HLS streams first.
|
||
# However, the HLS streams are actually a lower bitrate than their
|
||
# RTMP counterparts.
|
||
_FORMAT_QUALITY = {
|
||
'ultralow': 2,
|
||
'low': 3,
|
||
'360p': 4,
|
||
'mid': 5,
|
||
'480p': 6,
|
||
'high': 7,
|
||
'720p': 8,
|
||
'ultra': 9,
|
||
'1080p': 10
|
||
}
|
||
|
||
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')))
|
||
id = int(id)
|
||
|
||
def obfuscate_key_aux(count, modulo, start):
|
||
output = list(start)
|
||
for _ in range(count):
|
||
output.append(output[-1] + output[-2])
|
||
# cut off start values
|
||
output = output[2:]
|
||
output = list(map(lambda x: x % modulo + 33, output))
|
||
return output
|
||
|
||
def obfuscate_key(key):
|
||
num1 = int(floor(pow(2, 25) * sqrt(6.9)))
|
||
num2 = (num1 ^ key) << 5
|
||
num3 = key ^ num1
|
||
num4 = num3 ^ (num3 >> 3) ^ num2
|
||
prefix = intlist_to_bytes(obfuscate_key_aux(20, 97, (1, 2)))
|
||
shaHash = bytes_to_intlist(sha1(prefix + str(num4).encode('ascii')).digest())
|
||
# Extend 160 Bit hash to 256 Bit
|
||
return shaHash + [0] * 12
|
||
|
||
key = obfuscate_key(id)
|
||
|
||
decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv))
|
||
return zlib.decompress(decrypted_data)
|
||
|
||
def _convert_subtitles_to_srt(self, sub_root):
|
||
output = ''
|
||
|
||
for i, event in enumerate(sub_root.findall('./events/event'), 1):
|
||
start = event.attrib['start'].replace('.', ',')
|
||
end = event.attrib['end'].replace('.', ',')
|
||
text = event.attrib['text'].replace('\\N', '\n')
|
||
output += '%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
|
||
return output
|
||
|
||
def _convert_subtitles_to_ass(self, sub_root):
|
||
output = ''
|
||
|
||
def ass_bool(strvalue):
|
||
assvalue = '0'
|
||
if strvalue == '1':
|
||
assvalue = '-1'
|
||
return assvalue
|
||
|
||
output = '[Script Info]\n'
|
||
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 += """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 += '\n'
|
||
|
||
output += """
|
||
[Events]
|
||
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 += '\n'
|
||
|
||
return output
|
||
|
||
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'<iv>([^<]+)', sub_page, 'subtitle_iv', fatal=False)
|
||
data = self._search_regex(r'<data>([^<]+)', 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')
|
||
|
||
# Fetch mobile webpage
|
||
if mobj.group('prefix') == 'm':
|
||
mobile_webpage = self._download_webpage(url, video_id, 'Downloading mobile webpage')
|
||
webpage_url = self._search_regex(r'<link rel="canonical" href="([^"]+)" />', mobile_webpage, 'webpage_url')
|
||
else:
|
||
webpage_url = 'http://www.' + mobj.group('url')
|
||
|
||
webpage = self._download_webpage(self._add_skip_wall(webpage_url), video_id, 'Downloading webpage')
|
||
note_m = self._html_search_regex(
|
||
r'<div class="showmedia-trailer-notice">(.+?)</div>',
|
||
webpage, 'trailer-notice', default='')
|
||
if note_m:
|
||
raise ExtractorError(note_m)
|
||
|
||
mobj = re.search(r'Page\.messaging_box_controller\.addItems\(\[(?P<msg>{.+?})\]\)', webpage)
|
||
if mobj:
|
||
msg = json.loads(mobj.group('msg'))
|
||
if msg.get('type') == 'error':
|
||
raise ExtractorError('crunchyroll returned error: %s' % msg['message_body'], expected=True)
|
||
|
||
if 'To view this, please log in to verify you are 18 or older.' in webpage:
|
||
self.raise_login_required()
|
||
|
||
# Extract title, description, and other metadata
|
||
# from the mobile pagedlist
|
||
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'<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)
|
||
if video_upload_date:
|
||
video_upload_date = unified_strdate(video_upload_date)
|
||
video_uploader = self._html_search_regex(
|
||
r'<a[^>]+href="/publisher/[^"]+"[^>]*>([^<]+)</a>', webpage,
|
||
'video_uploader', fatal=False)
|
||
|
||
# Extract the available RTMP formats
|
||
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 = []
|
||
# Process the RTMP formats
|
||
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': 'rtmp-' + video_format,
|
||
'height': int_or_none(xpath_text(metadata, './height')),
|
||
'width': int_or_none(xpath_text(metadata, './width')),
|
||
'quality': self._FORMAT_QUALITY[video_format]
|
||
}
|
||
|
||
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)
|
||
continue
|
||
|
||
format_info.update({
|
||
'url': video_url,
|
||
'play_path': video_file,
|
||
'ext': 'flv',
|
||
})
|
||
formats.append(format_info)
|
||
|
||
# Fetch HLS formats from the API
|
||
api_info = self._api_call('info', {
|
||
'media_id': video_id,
|
||
'fields': 'media.stream_data'
|
||
}, video_id)
|
||
|
||
available_fmts = (api_info
|
||
.get('data', {'stream_data': {'streams': []}})
|
||
.get('stream_data', {'streams': []})
|
||
.get('streams', []))
|
||
for fmt in available_fmts:
|
||
# Crunchyroll has a mysterious format with quality None, that is
|
||
# lower resolution than even "low" quality
|
||
if fmt['quality'] is None:
|
||
fmt['quality'] = 'ultralow'
|
||
# Ignore the "adaptive" format, which is just a single HLS URL
|
||
# that duplicates all the others
|
||
elif fmt['quality'] == 'adaptive':
|
||
continue
|
||
|
||
m3u8_formats = self._extract_m3u8_formats(
|
||
fmt.get('url'),
|
||
video_id,
|
||
)
|
||
|
||
for m3u8_fmt in m3u8_formats:
|
||
m3u8_fmt['quality'] = self._FORMAT_QUALITY[fmt['quality']]
|
||
|
||
if m3u8_fmt['format_id']:
|
||
m3u8_fmt['format_id'] = 'hls-%s-%s' % (fmt['quality'], m3u8_fmt['format_id'])
|
||
else:
|
||
m3u8_fmt['format_id'] = 'hls-%s' % (fmt['quality'])
|
||
|
||
if fmt.get('height') is not None:
|
||
m3u8_fmt['height'] = int_or_none(fmt['height'])
|
||
if fmt.get('width') is not None:
|
||
m3u8_fmt['width'] = int_or_none(fmt['width'])
|
||
if fmt.get('bitrate') is not None:
|
||
m3u8_fmt['tbr'] = int_or_none(fmt['bitrate'])
|
||
|
||
formats.extend(m3u8_formats)
|
||
|
||
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)
|
||
|
||
return {
|
||
'id': video_id,
|
||
'title': video_title,
|
||
'description': video_description,
|
||
'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'
|
||
_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 = [{
|
||
'url': 'http://www.crunchyroll.com/a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
|
||
'info_dict': {
|
||
'id': 'a-bridge-to-the-starry-skies-hoshizora-e-kakaru-hashi',
|
||
'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(self._add_skip_wall(url), show_id)
|
||
title = self._html_search_regex(
|
||
r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>',
|
||
webpage, 'title')
|
||
episode_paths = re.findall(
|
||
r'(?s)<li id="showview_videos_media_[0-9]+"[^>]+>.*?<a href="([^"]+)"',
|
||
webpage)
|
||
entries = [
|
||
self.url_result('http://www.crunchyroll.com' + ep, 'Crunchyroll')
|
||
for ep in episode_paths
|
||
]
|
||
entries.reverse()
|
||
|
||
return {
|
||
'_type': 'playlist',
|
||
'id': show_id,
|
||
'title': title,
|
||
'entries': entries,
|
||
}
|