From 0987f2ddb27a27506c697ad9dae2ccbf24fc786d Mon Sep 17 00:00:00 2001 From: Timendum Date: Tue, 14 Nov 2017 16:34:45 +0100 Subject: [PATCH] [vshare] Fix extraction (closes #14473) --- youtube_dl/extractor/generic.py | 16 +++++++++++++++ youtube_dl/extractor/vshare.py | 36 +++++++++++++++++++++++++-------- 2 files changed, 44 insertions(+), 8 deletions(-) diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 2a9c3e2de..31564e550 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -102,6 +102,7 @@ from .joj import JojIE from .megaphone import MegaphoneIE from .vzaar import VzaarIE from .channel9 import Channel9IE +from .vshare import VShareIE class GenericIE(InfoExtractor): @@ -1921,6 +1922,16 @@ class GenericIE(InfoExtractor): 'title': 'Rescue Kit 14 Free Edition - Getting started', }, 'playlist_count': 4, + }, + { + # vshare embed + 'url': 'https://youtube-dl-demo.neocities.org/vshare.html', + 'md5': '17b39f55b5497ae8b59f5fbce8e35886', + 'info_dict': { + 'id': '0f64ce6', + 'title': 'vl14062007715967', + 'ext': 'mp4', + } } # { # # TODO: find another test @@ -2879,6 +2890,11 @@ class GenericIE(InfoExtractor): return self.playlist_from_matches( channel9_urls, video_id, video_title, ie=Channel9IE.ie_key()) + vshare_urls = VShareIE._extract_urls(webpage) + if vshare_urls: + return self.playlist_from_matches( + vshare_urls, video_id, video_title, ie=VShareIE.ie_key()) + def merge_dicts(dict1, dict2): merged = {} for k, v in dict1.items(): diff --git a/youtube_dl/extractor/vshare.py b/youtube_dl/extractor/vshare.py index 5addbc280..ea39a9051 100644 --- a/youtube_dl/extractor/vshare.py +++ b/youtube_dl/extractor/vshare.py @@ -1,14 +1,18 @@ # coding: utf-8 from __future__ import unicode_literals +import re + from .common import InfoExtractor +from ..compat import compat_chr +from ..utils import decode_packed_codes class VShareIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?vshare\.io/[dv]/(?P[^/?#&]+)' _TESTS = [{ 'url': 'https://vshare.io/d/0f64ce6', - 'md5': '16d7b8fef58846db47419199ff1ab3e7', + 'md5': '17b39f55b5497ae8b59f5fbce8e35886', 'info_dict': { 'id': '0f64ce6', 'title': 'vl14062007715967', @@ -19,20 +23,36 @@ class VShareIE(InfoExtractor): 'only_matching': True, }] + def _extract_packed(self, webpage): + packed = self._search_regex(r'(eval\(function.+)', webpage, 'packed code') + unpacked = decode_packed_codes(packed) + digits = self._search_regex(r'\[((?:\d+,?)+)\]', unpacked, 'digits') + digits = digits.split(',') + digits = [int(digit) for digit in digits] + key_digit = self._search_regex(r'fromCharCode\(.+?(\d+)\)}', unpacked, 'key digit') + chars = [compat_chr(d - int(key_digit)) for d in digits] + return ''.join(chars) + def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage( - 'https://vshare.io/d/%s' % video_id, video_id) + 'https://vshare.io/v/%s/width-650/height-430/1' % video_id, video_id) - title = self._html_search_regex( - r'(?s)
(.+?)
', webpage, 'title') - video_url = self._search_regex( - r']+href=(["\'])(?P(?:https?:)?//.+?)\1[^>]*>[Cc]lick\s+here', - webpage, 'video url', group='url') + title = self._html_search_regex(r'([^<]+)', webpage, 'title') + title = title.split(' - ')[0] + unpacked = self._extract_packed(webpage) + video_urls = re.findall(r']+?src=["\'](?P(?:https?:)?//(?:www\.)?vshare\.io/v/[^/?#&]+)', + webpage)