Merge pull request #2273 from dstftw/crunchyroll

[crunchyroll] Add support for mobile URLs and use unicode literals
This commit is contained in:
Philipp Hagemeister 2014-01-29 20:15:38 -08:00
commit 3effa7ceaa

View file

@ -1,4 +1,6 @@
# encoding: utf-8 # encoding: utf-8
from __future__ import unicode_literals
import re, base64, zlib import re, base64, zlib
from hashlib import sha1 from hashlib import sha1
from math import pow, sqrt, floor from math import pow, sqrt, floor
@ -18,29 +20,29 @@ from ..aes import (
) )
class CrunchyrollIE(InfoExtractor): class CrunchyrollIE(InfoExtractor):
_VALID_URL = r'(?:https?://)?(?:www\.)?(?P<url>crunchyroll\.com/[^/]*/[^/?&]*?(?P<video_id>[0-9]+))(?:[/?&]|$)' _VALID_URL = r'(?:https?://)?(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?:[^/]*/[^/?&]*?|media/\?id=)(?P<video_id>[0-9]+))(?:[/?&]|$)'
_TESTS = [{ _TESTS = [{
u'url': u'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513', 'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
u'file': u'645513.flv', 'file': '645513.flv',
#u'md5': u'b1639fd6ddfaa43788c85f6d1dddd412', #'md5': 'b1639fd6ddfaa43788c85f6d1dddd412',
u'info_dict': { 'info_dict': {
u'title': u'Wanna be the Strongest in the World Episode 1 An Idol-Wrestler is Born!', 'title': 'Wanna be the Strongest in the World Episode 1 An Idol-Wrestler is Born!',
u'description': u'md5:2d17137920c64f2f49981a7797d275ef', 'description': 'md5:2d17137920c64f2f49981a7797d275ef',
u'thumbnail': u'http://img1.ak.crunchyroll.com/i/spire1-tmb/20c6b5e10f1a47b10516877d3c039cae1380951166_full.jpg', 'thumbnail': 'http://img1.ak.crunchyroll.com/i/spire1-tmb/20c6b5e10f1a47b10516877d3c039cae1380951166_full.jpg',
u'uploader': u'Yomiuri Telecasting Corporation (YTV)', 'uploader': 'Yomiuri Telecasting Corporation (YTV)',
u'upload_date': u'20131013', 'upload_date': '20131013',
}, },
u'params': { 'params': {
# rtmp # rtmp
u'skip_download': True, 'skip_download': True,
}, },
}] }]
_FORMAT_IDS = { _FORMAT_IDS = {
u'360': (u'60', u'106'), '360': ('60', '106'),
u'480': (u'61', u'106'), '480': ('61', '106'),
u'720': (u'62', u'106'), '720': ('62', '106'),
u'1080': (u'80', u'108'), '1080': ('80', '108'),
} }
def _decrypt_subtitles(self, data, iv, id): def _decrypt_subtitles(self, data, iv, id):
@ -63,7 +65,7 @@ class CrunchyrollIE(InfoExtractor):
num3 = key ^ num1 num3 = key ^ num1
num4 = num3 ^ (num3 >> 3) ^ num2 num4 = num3 ^ (num3 >> 3) ^ num2
prefix = intlist_to_bytes(obfuscate_key_aux(20, 97, (1, 2))) prefix = intlist_to_bytes(obfuscate_key_aux(20, 97, (1, 2)))
shaHash = bytes_to_intlist(sha1(prefix + str(num4).encode(u'ascii')).digest()) shaHash = bytes_to_intlist(sha1(prefix + str(num4).encode('ascii')).digest())
# Extend 160 Bit hash to 256 Bit # Extend 160 Bit hash to 256 Bit
return shaHash + [0] * 12 return shaHash + [0] * 12
@ -79,93 +81,98 @@ class CrunchyrollIE(InfoExtractor):
def _convert_subtitles_to_srt(self, subtitles): def _convert_subtitles_to_srt(self, subtitles):
i=1 i=1
output = u'' output = ''
for start, end, text in re.findall(r'<event [^>]*?start="([^"]+)" [^>]*?end="([^"]+)" [^>]*?text="([^"]+)"[^>]*?>', subtitles): for start, end, text in re.findall(r'<event [^>]*?start="([^"]+)" [^>]*?end="([^"]+)" [^>]*?text="([^"]+)"[^>]*?>', subtitles):
start = start.replace(u'.', u',') start = start.replace('.', ',')
end = end.replace(u'.', u',') end = end.replace('.', ',')
text = clean_html(text) text = clean_html(text)
text = text.replace(u'\\N', u'\n') text = text.replace('\\N', '\n')
if not text: if not text:
continue continue
output += u'%d\n%s --> %s\n%s\n\n' % (i, start, end, text) output += '%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
i+=1 i+=1
return output return output
def _real_extract(self,url): def _real_extract(self,url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('video_id')
webpage_url = u'http://www.' + mobj.group('url') if mobj.group('prefix') == 'm':
video_id = mobj.group(u'video_id') mobile_webpage = self._download_webpage(url, video_id, 'Downloading mobile webpage')
webpage = self._download_webpage(webpage_url, video_id) webpage_url = self._search_regex(r'<link rel="canonical" href="([^"]+)" />', mobile_webpage, 'webpage_url')
note_m = self._html_search_regex(r'<div class="showmedia-trailer-notice">(.+?)</div>', webpage, u'trailer-notice', default=u'') 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'<div class="showmedia-trailer-notice">(.+?)</div>', webpage, 'trailer-notice', default='')
if note_m: if note_m:
raise ExtractorError(note_m) raise ExtractorError(note_m)
video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, u'video_title', flags=re.DOTALL) video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL)
video_title = re.sub(r' {2,}', u' ', video_title) video_title = re.sub(r' {2,}', ' ', video_title)
video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, u'video_description', default=u'') video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='')
if not video_description: if not video_description:
video_description = None video_description = None
video_upload_date = self._html_search_regex(r'<div>Availability for free users:(.+?)</div>', webpage, u'video_upload_date', fatal=False, flags=re.DOTALL) video_upload_date = self._html_search_regex(r'<div>Availability for free users:(.+?)</div>', webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)
if video_upload_date: if video_upload_date:
video_upload_date = unified_strdate(video_upload_date) video_upload_date = unified_strdate(video_upload_date)
video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, u'video_uploader', fatal=False, flags=re.DOTALL) video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, 'video_uploader', fatal=False, flags=re.DOTALL)
playerdata_url = compat_urllib_parse.unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, u'playerdata_url')) 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 = compat_urllib_request.Request(playerdata_url)
playerdata_req.data = compat_urllib_parse.urlencode({u'current_page': webpage_url}) playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url})
playerdata_req.add_header(u'Content-Type', u'application/x-www-form-urlencoded') playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
playerdata = self._download_webpage(playerdata_req, video_id, note=u'Downloading media info') playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info')
stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, u'stream_id') stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, 'stream_id')
video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, u'thumbnail', fatal=False) video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False)
formats = [] formats = []
for fmt in re.findall(r'\?p([0-9]{3,4})=1', webpage): for fmt in re.findall(r'\?p([0-9]{3,4})=1', webpage):
stream_quality, stream_format = self._FORMAT_IDS[fmt] stream_quality, stream_format = self._FORMAT_IDS[fmt]
video_format = fmt+u'p' video_format = fmt+'p'
streamdata_req = compat_urllib_request.Request(u'http://www.crunchyroll.com/xml/') streamdata_req = compat_urllib_request.Request('http://www.crunchyroll.com/xml/')
# urlencode doesn't work! # urlencode doesn't work!
streamdata_req.data = u'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality='+stream_quality+u'&media%5Fid='+stream_id+u'&video%5Fformat='+stream_format streamdata_req.data = 'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality='+stream_quality+'&media%5Fid='+stream_id+'&video%5Fformat='+stream_format
streamdata_req.add_header(u'Content-Type', u'application/x-www-form-urlencoded') streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
streamdata_req.add_header(u'Content-Length', str(len(streamdata_req.data))) streamdata_req.add_header('Content-Length', str(len(streamdata_req.data)))
streamdata = self._download_webpage(streamdata_req, video_id, note=u'Downloading media info for '+video_format) streamdata = self._download_webpage(streamdata_req, video_id, note='Downloading media info for '+video_format)
video_url = self._search_regex(r'<host>([^<]+)', streamdata, u'video_url') video_url = self._search_regex(r'<host>([^<]+)', streamdata, 'video_url')
video_play_path = self._search_regex(r'<file>([^<]+)', streamdata, u'video_play_path') video_play_path = self._search_regex(r'<file>([^<]+)', streamdata, 'video_play_path')
formats.append({ formats.append({
u'url': video_url, 'url': video_url,
u'play_path': video_play_path, 'play_path': video_play_path,
u'ext': 'flv', 'ext': 'flv',
u'format': video_format, 'format': video_format,
u'format_id': video_format, 'format_id': video_format,
}) })
subtitles = {} subtitles = {}
for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage): for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage):
sub_page = self._download_webpage(u'http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id='+sub_id,\ sub_page = self._download_webpage('http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id='+sub_id,\
video_id, note=u'Downloading subtitles for '+sub_name) video_id, note='Downloading subtitles for '+sub_name)
id = self._search_regex(r'id=\'([0-9]+)', sub_page, u'subtitle_id', fatal=False) id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False)
iv = self._search_regex(r'<iv>([^<]+)', sub_page, u'subtitle_iv', fatal=False) iv = self._search_regex(r'<iv>([^<]+)', sub_page, 'subtitle_iv', fatal=False)
data = self._search_regex(r'<data>([^<]+)', sub_page, u'subtitle_data', fatal=False) data = self._search_regex(r'<data>([^<]+)', sub_page, 'subtitle_data', fatal=False)
if not id or not iv or not data: if not id or not iv or not data:
continue continue
id = int(id) id = int(id)
iv = base64.b64decode(iv) iv = base64.b64decode(iv)
data = base64.b64decode(data) data = base64.b64decode(data)
subtitle = self._decrypt_subtitles(data, iv, id).decode(u'utf-8') subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
lang_code = self._search_regex(r'lang_code=\'([^\']+)', subtitle, u'subtitle_lang_code', fatal=False) lang_code = self._search_regex(r'lang_code=\'([^\']+)', subtitle, 'subtitle_lang_code', fatal=False)
if not lang_code: if not lang_code:
continue continue
subtitles[lang_code] = self._convert_subtitles_to_srt(subtitle) subtitles[lang_code] = self._convert_subtitles_to_srt(subtitle)
return { return {
u'id': video_id, 'id': video_id,
u'title': video_title, 'title': video_title,
u'description': video_description, 'description': video_description,
u'thumbnail': video_thumbnail, 'thumbnail': video_thumbnail,
u'uploader': video_uploader, 'uploader': video_uploader,
u'upload_date': video_upload_date, 'upload_date': video_upload_date,
u'subtitles': subtitles, 'subtitles': subtitles,
u'formats': formats, 'formats': formats,
} }