From 7b4137c351222a94f46f854bf490a299e4124acc Mon Sep 17 00:00:00 2001 From: remitamine Date: Wed, 9 Sep 2015 10:42:47 +0100 Subject: [PATCH] [fktv] fix info extraction --- youtube_dl/extractor/__init__.py | 5 +- youtube_dl/extractor/fktv.py | 78 ++++++++------------------------ 2 files changed, 20 insertions(+), 63 deletions(-) diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 57f55b479..f8d4c8462 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -170,10 +170,7 @@ from .firsttv import FirstTVIE from .fivemin import FiveMinIE from .fivetv import FiveTVIE -from .fktv import ( - FKTVIE, - FKTVPosteckeIE, -) +from .fktv import FKTVIE from .flickr import FlickrIE from .folketinget import FolketingetIE from .footyroom import FootyRoomIE diff --git a/youtube_dl/extractor/fktv.py b/youtube_dl/extractor/fktv.py index 190d9f9ad..c2aa23aa2 100644 --- a/youtube_dl/extractor/fktv.py +++ b/youtube_dl/extractor/fktv.py @@ -1,13 +1,11 @@ from __future__ import unicode_literals import re -import random -import json from .common import InfoExtractor from ..utils import ( - get_element_by_id, clean_html, + determine_ext, ) @@ -17,66 +15,28 @@ class FKTVIE(InfoExtractor): _TEST = { 'url': 'http://fernsehkritik.tv/folge-1', + 'md5': '21f0b0c99bce7d5b524eb1b17b1c6d79', 'info_dict': { - 'id': '00011', - 'ext': 'flv', + 'id': '1', + 'ext': 'mp4', 'title': 'Folge 1 vom 10. April 2007', - 'description': 'md5:fb4818139c7cfe6907d4b83412a6864f', }, } def _real_extract(self, url): - episode = int(self._match_id(url)) + episode = self._match_id(url) - video_thumbnail = 'http://fernsehkritik.tv/images/magazin/folge%s.jpg' % episode - start_webpage = self._download_webpage('http://fernsehkritik.tv/folge-%s/Start' % episode, - episode) - playlist = self._search_regex(r'playlist = (\[.*?\]);', start_webpage, - 'playlist', flags=re.DOTALL) - files = json.loads(re.sub('{[^{}]*?}', '{}', playlist)) - - videos = [] - for i, _ in enumerate(files, 1): - video_id = '%04d%d' % (episode, i) - video_url = 'http://fernsehkritik.tv/js/directme.php?file=%s%s.flv' % (episode, '' if i == 1 else '-%d' % i) - videos.append({ - 'ext': 'flv', - 'id': video_id, - 'url': video_url, - 'title': clean_html(get_element_by_id('eptitle', start_webpage)), - 'description': clean_html(get_element_by_id('contentlist', start_webpage)), - 'thumbnail': video_thumbnail - }) - return { - '_type': 'multi_video', - 'entries': videos, - 'id': 'folge-%s' % episode, - } - - -class FKTVPosteckeIE(InfoExtractor): - IE_NAME = 'fernsehkritik.tv:postecke' - _VALID_URL = r'http://(?:www\.)?fernsehkritik\.tv/inline-video/postecke\.php\?(.*&)?ep=(?P[0-9]+)(&|$)' - _TEST = { - 'url': 'http://fernsehkritik.tv/inline-video/postecke.php?iframe=true&width=625&height=440&ep=120', - 'md5': '262f0adbac80317412f7e57b4808e5c4', - 'info_dict': { - 'id': '0120', - 'ext': 'flv', - 'title': 'Postecke 120', - } - } - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - episode = int(mobj.group('ep')) - - server = random.randint(2, 4) - video_id = '%04d' % episode - video_url = 'http://dl%d.fernsehkritik.tv/postecke/postecke%d.flv' % (server, episode) - video_title = 'Postecke %d' % episode - return { - 'id': video_id, - 'url': video_url, - 'title': video_title, - } + webpage = self._download_webpage('http://fernsehkritik.tv/folge-%s/play' % episode, episode) + title = clean_html(self._html_search_regex('

([^<]+?)

', webpage, 'title')) + matchs = re.search(r'(?s)]*poster="([^"]+)"[^>]*>(.*?)', webpage) + if matchs: + poster, sources = matchs.groups() + urls = re.findall(r'(?s)]*src="([^"]+)"[^>]*>', sources) + if sources: + formats = [{'url': url, 'format_id': determine_ext(url)} for url in urls] + return { + 'id': episode, + 'title': title, + 'formats': formats, + 'thumbnail': poster, + }