paradizelost/youtube-dl
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[crunchyroll] Convert to new subtitles system

Browse Source
This commit is contained in:
Jaime Marquínez Ferrándiz 2015-02-15 18:21:42 +01:00
parent a504ced097
commit b5857f62e2
1 changed files with 35 additions and 31 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

66
youtube_dl/extractor/crunchyroll.py
View File

@ -9,7 +9,7 @@ import xml.etree.ElementTree
from hashlib import sha1 from hashlib import sha1
from math import pow, sqrt, floor from math import pow, sqrt, floor
from .subtitles import SubtitlesInfoExtractor from .common import InfoExtractor
from ..compat import ( from ..compat import (
compat_urllib_parse, compat_urllib_parse,
compat_urllib_request, compat_urllib_request,
@ -25,10 +25,9 @@ from ..aes import (
aes_cbc_decrypt, aes_cbc_decrypt,
inc, inc,
) )
from .common import InfoExtractor
class CrunchyrollIE(SubtitlesInfoExtractor): class CrunchyrollIE(InfoExtractor):
_VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:[^/]*/[^/?&]*?|media/\?id=)(?P<video_id>[0-9]+))(?:[/?&]|$)' _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:[^/]*/[^/?&]*?|media/\?id=)(?P<video_id>[0-9]+))(?:[/?&]|$)'
_TESTS = [{ _TESTS = [{
'url': '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',
@ -187,6 +186,38 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
return output return output
def _get_subtitles(self, video_id, webpage):
subtitles = {}
for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', 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
id = int(id)
iv = base64.b64decode(iv)
data = base64.b64decode(data)
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
sub_root = xml.etree.ElementTree.fromstring(subtitle)
subtitles[lang_code] = [
{
'ext': 'srt',
'data': self._convert_subtitles_to_srt(sub_root),
},
{
'ext': 'ass',
'data': self._convert_subtitles_to_ass(sub_root),
},
]
return subtitles
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') video_id = mobj.group('video_id')
@ -249,34 +280,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
'format_id': video_format, 'format_id': video_format,
}) })
subtitles = {} subtitles = self.extract_subtitles(video_id, webpage)
sub_format = self._downloader.params.get('subtitlesformat', 'srt')
for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', 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
id = int(id)
iv = base64.b64decode(iv)
data = base64.b64decode(data)
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
sub_root = xml.etree.ElementTree.fromstring(subtitle)
if sub_format == 'ass':
subtitles[lang_code] = self._convert_subtitles_to_ass(sub_root)
else:
subtitles[lang_code] = self._convert_subtitles_to_srt(sub_root)
if self._downloader.params.get('listsubtitles', False):
self._list_available_subtitles(video_id, subtitles)
return
return { return {
'id': video_id, 'id': video_id,