From 9abd500a7402fe1f66e4177b61a7327b416b6b9d Mon Sep 17 00:00:00 2001 From: Philipp Hagemeister Date: Thu, 11 Dec 2014 17:07:59 +0100 Subject: [PATCH] [zdf:channel] Simplify (#4427) --- youtube_dl/extractor/zdf.py | 111 +++++++++++++----------------------- 1 file changed, 40 insertions(+), 71 deletions(-) diff --git a/youtube_dl/extractor/zdf.py b/youtube_dl/extractor/zdf.py index 221f16686a..74c76a9a04 100644 --- a/youtube_dl/extractor/zdf.py +++ b/youtube_dl/extractor/zdf.py @@ -1,12 +1,14 @@ # coding: utf-8 from __future__ import unicode_literals +import functools import re from .common import InfoExtractor from ..utils import ( int_or_none, unified_strdate, + OnDemandPagedList, ) @@ -86,28 +88,8 @@ def extract_from_xml_url(ie, video_id, xml_url): } -def extract_channel_from_xml_url(ie, channel_id, xml_url): - doc = ie._download_xml( - xml_url, channel_id, - note='Downloading channel info', - errnote='Failed to download channel info') - - title = doc.find('.//information/title').text - description = doc.find('.//information/detail').text - assets = [{'id': asset.find('./details/assetId').text, - 'type': asset.find('./type').text, - } for asset in doc.findall('.//teasers/teaser')] - - return { - 'id': channel_id, - 'title': title, - 'description': description, - 'assets': assets, - } - - class ZDFIE(InfoExtractor): - _VALID_URL = r'^https?://www\.zdf\.de/ZDFmediathek(?P#)?/(.*beitrag/(?:video/)?)(?P[0-9]+)(?:/[^/?]+)?(?:\?.*)?' + _VALID_URL = r'(?:zdf:|zdf:video:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/(.*beitrag/(?:video/)?))(?P[0-9]+)(?:/[^/?]+)?(?:\?.*)?' _TEST = { 'url': 'http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt', @@ -124,67 +106,54 @@ class ZDFIE(InfoExtractor): 'skip': 'Videos on ZDF.de are depublicised in short order', } - def _extract_video(self, video_id): + def _real_extract(self, url): + video_id = self._match_id(url) xml_url = 'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id return extract_from_xml_url(self, video_id, xml_url) - def _real_extract(self, url): - return self._extract_video(self._match_id(url)) - - -class ZDFChannelIE(ZDFIE): - _VALID_URL = r'^https?://www\.zdf\.de/ZDFmediathek(?P#)?/(.*kanaluebersicht/)(?P[0-9]+)' +class ZDFChannelIE(InfoExtractor): + _VALID_URL = r'(?:zdf:topic:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/.*kanaluebersicht/)(?P[0-9]+)' _TEST = { 'url': 'http://www.zdf.de/ZDFmediathek#/kanaluebersicht/1586442/sendung/Titanic', 'info_dict': { 'id': '1586442', - 'title': 'Titanic', - 'description': 'md5:444c048cfe3fdc2561be7de4bcbf1d04', }, - 'playlist_count': 3, + 'playlist_count': 4, } + _PAGE_SIZE = 50 - def _extract_channel(self, channel_id): - def load_chunks(channel_id, chunk_length): - offset = 0 - while True: - url = ('http://www.zdf.de/ZDFmediathek/xmlservice/web/aktuellste?ak=web&offset=%d&maxLength=%d&id=%s' - % (offset, chunk_length, channel_id)) - result = extract_channel_from_xml_url(self, channel_id, url) - yield result - if len(result['assets']) < chunk_length: - return - offset += chunk_length + def _fetch_page(self, channel_id, page): + offset = page * self._PAGE_SIZE + xml_url = ( + 'http://www.zdf.de/ZDFmediathek/xmlservice/web/aktuellste?ak=web&offset=%d&maxLength=%d&id=%s' + % (offset, self._PAGE_SIZE, channel_id)) + doc = self._download_xml( + xml_url, channel_id, + note='Downloading channel info', + errnote='Failed to download channel info') - def load_channel(channel_id): - chunks = list(load_chunks(channel_id, 50)) # The server rejects higher values - assets = [asset for chunk in chunks for asset in chunk['assets']] - video_ids = [asset['id'] for asset in - filter(lambda asset: asset['type'] == 'video', - assets)] - topic_ids = [asset['id'] for asset in - filter(lambda asset: asset['type'] == 'thema', - assets)] - if topic_ids: - video_ids = reduce(list.__add__, - [load_channel(topic_id)['video_ids'] - for topic_id in topic_ids], - video_ids) - - result = chunks[0] - result['video_ids'] = video_ids - return result - - channel = load_channel(channel_id) - return { - '_type': 'playlist', - 'id': channel['id'], - 'title': channel['title'], - 'description': channel['description'], - 'entries': [self._extract_video(video_id) - for video_id in channel['video_ids']], - } + title = doc.find('.//information/title').text + description = doc.find('.//information/detail').text + for asset in doc.findall('.//teasers/teaser'): + a_type = asset.find('./type').text + a_id = asset.find('./details/assetId').text + if a_type not in ('video', 'topic'): + continue + yield { + '_type': 'url', + 'playlist_title': title, + 'playlist_description': description, + 'url': 'zdf:%s:%s' % (a_type, a_id), + } def _real_extract(self, url): - return self._extract_channel(self._match_id(url)) + channel_id = self._match_id(url) + entries = OnDemandPagedList( + functools.partial(self._fetch_page, channel_id), self._PAGE_SIZE) + + return { + '_type': 'playlist', + 'id': channel_id, + 'entries': entries, + }