X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/46113edab215c2211a604c06245c16d5d4e57dcf..abc36f0474141930338eef35e2444365fab932b6:/youtube_dl/extractor/vidzi.py diff --git a/youtube_dl/extractor/vidzi.py b/youtube_dl/extractor/vidzi.py index 669979e..d055629 100644 --- a/youtube_dl/extractor/vidzi.py +++ b/youtube_dl/extractor/vidzi.py @@ -1,12 +1,20 @@ -#coding: utf-8 +# coding: utf-8 from __future__ import unicode_literals +import re + from .common import InfoExtractor +from ..utils import ( + decode_packed_codes, + js_to_json, + NO_DEFAULT, + PACKED_CODES_RE, +) class VidziIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?vidzi\.tv/(?P\w+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?vidzi\.(?:tv|cc)/(?:embed-)?(?P[0-9a-zA-Z]+)' + _TESTS = [{ 'url': 'http://vidzi.tv/cghql9yq6emu.html', 'md5': '4f16c71ca0c8c8635ab6932b5f3f1660', 'info_dict': { @@ -14,20 +22,39 @@ class VidziIE(InfoExtractor): 'ext': 'mp4', 'title': 'youtube-dl test video 1\\\\2\'3/4<5\\\\6ä7↭', }, - } + 'params': { + # m3u8 download + 'skip_download': True, + }, + }, { + 'url': 'http://vidzi.tv/embed-4z2yb0rzphe9-600x338.html', + 'skip_download': True, + }, { + 'url': 'http://vidzi.cc/cghql9yq6emu.html', + 'skip_download': True, + }] def _real_extract(self, url): video_id = self._match_id(url) - - webpage = self._download_webpage(url, video_id) - video_url = self._html_search_regex( - r'{\s*file\s*:\s*"([^"]+)"\s*}', webpage, 'video url') + + webpage = self._download_webpage( + 'http://vidzi.tv/%s' % video_id, video_id) title = self._html_search_regex( r'(?s)

(.*?)

', webpage, 'title') - - return { - 'id': video_id, - 'title': title, - 'url': video_url, - } - \ No newline at end of file + + packed_codes = [mobj.group(0) for mobj in re.finditer( + PACKED_CODES_RE, webpage)] + for num, pc in enumerate(packed_codes, 1): + code = decode_packed_codes(pc).replace('\\\'', '\'') + jwplayer_data = self._parse_json( + self._search_regex( + r'setup\(([^)]+)\)', code, 'jwplayer data', + default=NO_DEFAULT if num == len(packed_codes) else '{}'), + video_id, transform_source=js_to_json) + if jwplayer_data: + break + + info_dict = self._parse_jwplayer_data(jwplayer_data, video_id, require_title=False) + info_dict['title'] = title + + return info_dict