[vshare] Improve extraction, fix formats sorting and carry long lines

This commit is contained in:
Sergey M․ 2017-11-14 22:49:25 +07:00
parent ff31f2d5c3
commit a2b6aba8de
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D

View file

@ -26,13 +26,20 @@ class VShareIE(InfoExtractor):
'only_matching': True, 'only_matching': True,
}] }]
@staticmethod
def _extract_urls(webpage):
return re.findall(
r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?vshare\.io/v/[^/?#&]+)',
webpage)
def _extract_packed(self, webpage): def _extract_packed(self, webpage):
packed = self._search_regex(r'(eval\(function.+)', webpage, 'packed code') packed = self._search_regex(
r'(eval\(function.+)', webpage, 'packed code')
unpacked = decode_packed_codes(packed) unpacked = decode_packed_codes(packed)
digits = self._search_regex(r'\[((?:\d+,?)+)\]', unpacked, 'digits') digits = self._search_regex(r'\[((?:\d+,?)+)\]', unpacked, 'digits')
digits = digits.split(',') digits = [int(digit) for digit in digits.split(',')]
digits = [int(digit) for digit in digits] key_digit = self._search_regex(
key_digit = self._search_regex(r'fromCharCode\(.+?(\d+)\)}', unpacked, 'key digit') r'fromCharCode\(.+?(\d+)\)}', unpacked, 'key digit')
chars = [compat_chr(d - int(key_digit)) for d in digits] chars = [compat_chr(d - int(key_digit)) for d in digits]
return ''.join(chars) return ''.join(chars)
@ -40,9 +47,11 @@ def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
webpage = self._download_webpage( webpage = self._download_webpage(
'https://vshare.io/v/%s/width-650/height-430/1' % video_id, video_id) 'https://vshare.io/v/%s/width-650/height-430/1' % video_id,
video_id)
title = self._html_search_regex(r'<title>([^<]+)</title>', webpage, 'title') title = self._html_search_regex(
r'<title>([^<]+)</title>', webpage, 'title')
title = title.split(' - ')[0] title = title.split(' - ')[0]
error = self._html_search_regex( error = self._html_search_regex(
@ -51,17 +60,15 @@ def _real_extract(self, url):
if error: if error:
raise ExtractorError(error, expected=True) raise ExtractorError(error, expected=True)
unpacked = self._extract_packed(webpage) info = self._parse_html5_media_entries(
video_urls = re.findall(r'<source src="([^"]+)', unpacked) url, '<video>%s</video>' % self._extract_packed(webpage),
formats = [{'url': video_url} for video_url in video_urls] video_id)[0]
return {
self._sort_formats(info['formats'])
info.update({
'id': video_id, 'id': video_id,
'title': title, 'title': title,
'formats': formats, })
}
@staticmethod return info
def _extract_urls(webpage):
return re.findall(
r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?vshare\.io/v/[^/?#&]+)',
webpage)