X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fzdf.py;h=a795f56b37bbd710c895b0e255342057a5aa354f;hb=c240ab6ecfc06fe98c03900c59861b84dce338b9;hp=221f16686ae4e92a41588ebc8c7e2c5f2bbc08d2;hpb=cf68bcaeff77f5a5784c7c5473ce4fea0b8c73ae;p=youtube-dl diff --git a/youtube_dl/extractor/zdf.py b/youtube_dl/extractor/zdf.py index 221f16686..a795f56b3 100644 --- a/youtube_dl/extractor/zdf.py +++ b/youtube_dl/extractor/zdf.py @@ -1,12 +1,15 @@ # 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, + xpath_text, ) @@ -17,13 +20,11 @@ def extract_from_xml_url(ie, video_id, xml_url): errnote='Failed to download video info') title = doc.find('.//information/title').text - description = doc.find('.//information/detail').text - duration = int(doc.find('.//details/lengthSec').text) - uploader_node = doc.find('.//details/originChannelTitle') - uploader = None if uploader_node is None else uploader_node.text - uploader_id_node = doc.find('.//details/originChannelId') - uploader_id = None if uploader_id_node is None else uploader_id_node.text - upload_date = unified_strdate(doc.find('.//details/airtime').text) + description = xpath_text(doc, './/information/detail', 'description') + duration = int_or_none(xpath_text(doc, './/details/lengthSec', 'duration')) + uploader = xpath_text(doc, './/details/originChannelTitle', 'uploader') + uploader_id = xpath_text(doc, './/details/originChannelId', 'uploader id') + upload_date = unified_strdate(xpath_text(doc, './/details/airtime', 'upload date')) def xml_to_format(fnode): video_url = fnode.find('url').text @@ -38,15 +39,14 @@ def extract_from_xml_url(ie, video_id, xml_url): ext = format_m.group('container') proto = format_m.group('proto').lower() - quality = fnode.find('./quality').text - abr = int(fnode.find('./audioBitrate').text) // 1000 - vbr_node = fnode.find('./videoBitrate') - vbr = None if vbr_node is None else int(vbr_node.text) // 1000 + quality = xpath_text(fnode, './quality', 'quality') + abr = int_or_none(xpath_text(fnode, './audioBitrate', 'abr'), 1000) + vbr = int_or_none(xpath_text(fnode, './videoBitrate', 'vbr'), 1000) - width_node = fnode.find('./width') - width = None if width_node is None else int_or_none(width_node.text) - height_node = fnode.find('./height') - height = None if height_node is None else int_or_none(height_node.text) + width = int_or_none(xpath_text(fnode, './width', 'width')) + height = int_or_none(xpath_text(fnode, './height', 'height')) + + filesize = int_or_none(xpath_text(fnode, './filesize', 'filesize')) format_note = '' if not format_note: @@ -62,12 +62,31 @@ def extract_from_xml_url(ie, video_id, xml_url): 'vbr': vbr, 'width': width, 'height': height, - 'filesize': int_or_none(fnode.find('./filesize').text), + 'filesize': filesize, 'format_note': format_note, 'protocol': proto, '_available': is_available, } + def xml_to_thumbnails(fnode): + thumbnails = [] + for node in fnode: + thumbnail_url = node.text + if not thumbnail_url: + continue + thumbnail = { + 'url': thumbnail_url, + } + if 'key' in node.attrib: + m = re.match('^([0-9]+)x([0-9]+)$', node.attrib['key']) + if m: + thumbnail['width'] = int(m.group(1)) + thumbnail['height'] = int(m.group(2)) + thumbnails.append(thumbnail) + return thumbnails + + thumbnails = xml_to_thumbnails(doc.findall('.//teaserimages/teaserimage')) + format_nodes = doc.findall('.//formitaeten/formitaet') formats = list(filter( lambda f: f['_available'], @@ -79,6 +98,7 @@ def extract_from_xml_url(ie, video_id, xml_url): 'title': title, 'description': description, 'duration': duration, + 'thumbnails': thumbnails, 'uploader': uploader, 'uploader_id': uploader_id, 'upload_date': upload_date, @@ -86,28 +106,8 @@ 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 +124,54 @@ 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, } + _PAGE_SIZE = 50 + + 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') + + 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): + channel_id = self._match_id(url) + entries = OnDemandPagedList( + functools.partial(self._fetch_page, channel_id), self._PAGE_SIZE) - 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 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']], + 'id': channel_id, + 'entries': entries, } - - def _real_extract(self, url): - return self._extract_channel(self._match_id(url))