midgard/youtube-dl
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

[crunchyroll] Extract subtitles extraction routine

Browse source
This commit is contained in:
Sergey M․ 2015-05-30 14:12:58 +06:00
parent e26be70bca
commit 0385d64223
1 changed files with 13 additions and 17 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

30
youtube_dl/extractor/crunchyroll.py
View file

@ -76,8 +76,8 @@ class CrunchyrollIE(InfoExtractor):
self._login() self._login()
def _decrypt_subtitles(self, data, iv, id): def _decrypt_subtitles(self, data, iv, id):
data = bytes_to_intlist(data) data = bytes_to_intlist(base64.b64decode(data))
iv = bytes_to_intlist(iv) iv = bytes_to_intlist(base64.b64decode(iv))
id = int(id) id = int(id)
def obfuscate_key_aux(count, modulo, start): def obfuscate_key_aux(count, modulo, start):
@ -179,6 +179,16 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
return output return output
def _extract_subtitles(self, subtitle):
sub_root = xml.etree.ElementTree.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): def _get_subtitles(self, video_id, webpage):
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):
@ -190,25 +200,11 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
data = self._search_regex(r'<data>([^<]+)', sub_page, '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)
iv = base64.b64decode(iv)
data = base64.b64decode(data)
subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8') subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, '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
sub_root = xml.etree.ElementTree.fromstring(subtitle) subtitles[lang_code] = self._extract_subtitles(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 return subtitles
def _real_extract(self, url): def _real_extract(self, url):