X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fzdf.py;h=c619a75e2a2d6617fd083c44bf05114d9e4c83bc;hb=78be2eca7cb2806c3a51547da14968336febb57c;hp=07f830e80793e5ac08432831e1179708461e22fc;hpb=31cb6d8fefaa42aff293b03e37471bb05d5b8391;p=youtube-dl diff --git a/youtube_dl/extractor/zdf.py b/youtube_dl/extractor/zdf.py index 07f830e80..c619a75e2 100644 --- a/youtube_dl/extractor/zdf.py +++ b/youtube_dl/extractor/zdf.py @@ -1,111 +1,250 @@ -import operator +# coding: utf-8 +from __future__ import unicode_literals + +import functools import re from .common import InfoExtractor from ..utils import ( - parse_xml_doc, + int_or_none, unified_strdate, + OnDemandPagedList, + xpath_text, + determine_ext, + qualities, + float_or_none, + ExtractorError, ) class ZDFIE(InfoExtractor): - _VALID_URL = r'^http://www\.zdf\.de\/ZDFmediathek(?P#)?\/(.*beitrag\/video\/)(?P[^/\?]+)(?:\?.*)?' - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('video_id') - - xml_url = u'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id - info_xml = self._download_webpage( - xml_url, video_id, note=u'Downloading video info') - doc = parse_xml_doc(info_xml) + _VALID_URL = r'(?:zdf:|zdf:video:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/(.*beitrag/(?:video/)?))(?P[0-9]+)(?:/[^/?]+)?(?:\?.*)?' + + _TESTS = [{ + 'url': 'http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt', + 'info_dict': { + 'id': '2037704', + 'ext': 'webm', + 'title': 'ZDFspezial - Ende des Machtpokers', + 'description': 'Union und SPD haben sich auf einen Koalitionsvertrag geeinigt. Aber was bedeutet das für die Bürger? Sehen Sie hierzu das ZDFspezial "Ende des Machtpokers - Große Koalition für Deutschland".', + 'duration': 1022, + 'uploader': 'spezial', + 'uploader_id': '225948', + 'upload_date': '20131127', + }, + 'skip': 'Videos on ZDF.de are depublicised in short order', + }] + + def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None): + param_groups = {} + for param_group in smil.findall(self._xpath_ns('./head/paramGroup', namespace)): + group_id = param_group.attrib.get(self._xpath_ns('id', 'http://www.w3.org/XML/1998/namespace')) + params = {} + for param in param_group: + params[param.get('name')] = param.get('value') + param_groups[group_id] = params + + formats = [] + for video in smil.findall(self._xpath_ns('.//video', namespace)): + src = video.get('src') + if not src: + continue + bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) + group_id = video.get('paramGroup') + param_group = param_groups[group_id] + for proto in param_group['protocols'].split(','): + formats.append({ + 'url': '%s://%s' % (proto, param_group['host']), + 'app': param_group['app'], + 'play_path': src, + 'ext': 'flv', + 'format_id': '%s-%d' % (proto, bitrate), + 'tbr': bitrate, + }) + self._sort_formats(formats) + return formats + + def extract_from_xml_url(self, video_id, xml_url): + doc = self._download_xml( + xml_url, video_id, + note='Downloading video info', + errnote='Failed to download video info') + + status_code = doc.find('./status/statuscode') + if status_code is not None and status_code.text != 'ok': + code = status_code.text + if code == 'notVisibleAnymore': + message = 'Video %s is not available' % video_id + else: + message = '%s returned error: %s' % (self.IE_NAME, code) + raise ExtractorError(message, expected=True) title = doc.find('.//information/title').text - description = doc.find('.//information/detail').text - uploader_node = doc.find('.//details/originChannelTitle') - uploader = None if uploader_node is None else uploader_node.text - duration_str = doc.find('.//details/length').text - duration_m = re.match(r'''(?x)^ - (?P[0-9]{2}) - :(?P[0-9]{2}) - :(?P[0-9]{2}) - (?:\.(?P[0-9]+)?) - ''', duration_str) - duration = ( - ( - (int(duration_m.group('hours')) * 60 * 60) + - (int(duration_m.group('minutes')) * 60) + - int(duration_m.group('seconds')) - ) - if duration_m - else None - ) - upload_date = unified_strdate(doc.find('.//details/airtime').text) - - def xml_to_format(fnode): - video_url = fnode.find('url').text - is_available = u'http://www.metafilegenerator' not in video_url + 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_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') + quality = qualities(['veryhigh', 'high', 'med', 'low']) + + def get_quality(elem): + return quality(xpath_text(elem, 'quality')) + format_nodes.sort(key=get_quality) + format_ids = [] + formats = [] + for fnode in format_nodes: + video_url = fnode.find('url').text + is_available = 'http://www.metafilegenerator' not in video_url + if not is_available: + continue format_id = fnode.attrib['basetype'] + quality = xpath_text(fnode, './quality', 'quality') format_m = re.match(r'''(?x) (?P[^_]+)_(?P[^_]+)_(?P[^_]+)_ (?P[^_]+)_(?P[^_]+)_(?P[^_]+) ''', format_id) - ext = format_m.group('container') - is_supported = ext != 'f4f' - - PROTO_ORDER = ['http', 'rtmp', 'rtsp'] - try: - proto_pref = -PROTO_ORDER.index(format_m.group('proto')) - except ValueError: - proto_pref = 999 - - quality = fnode.find('./quality').text - QUALITY_ORDER = ['veryhigh', '300', 'high', 'med', 'low'] - try: - quality_pref = -QUALITY_ORDER.index(quality) - except ValueError: - quality_pref = 999 - - abr = int(fnode.find('./audioBitrate').text) // 1000 - vbr = int(fnode.find('./videoBitrate').text) // 1000 - pref = (is_available, is_supported, - proto_pref, quality_pref, vbr, abr) - - format_note = u'' - if not is_supported: - format_note += u'(unsupported)' - if not format_note: - format_note = None - - return { - 'format_id': format_id + u'-' + quality, - 'url': video_url, - 'ext': ext, - 'acodec': format_m.group('acodec'), - 'vcodec': format_m.group('vcodec'), - 'abr': abr, - 'vbr': vbr, - 'width': int(fnode.find('./width').text), - 'height': int(fnode.find('./height').text), - 'filesize': int(fnode.find('./filesize').text), - 'format_note': format_note, - '_pref': pref, - '_available': is_available, - } - - format_nodes = doc.findall('.//formitaeten/formitaet') - formats = sorted(filter(lambda f: f['_available'], - map(xml_to_format, format_nodes)), - key=operator.itemgetter('_pref')) + ext = determine_ext(video_url, None) or format_m.group('container') + if ext not in ('smil', 'f4m', 'm3u8'): + format_id = format_id + '-' + quality + if format_id in format_ids: + continue + + if ext == 'meta': + continue + elif ext == 'smil': + formats.extend(self._extract_smil_formats( + video_url, video_id, fatal=False)) + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + video_url, video_id, 'mp4', m3u8_id=format_id, fatal=False)) + elif ext == 'f4m': + formats.extend(self._extract_f4m_formats( + video_url, video_id, f4m_id=format_id, fatal=False)) + else: + proto = format_m.group('proto').lower() + + abr = int_or_none(xpath_text(fnode, './audioBitrate', 'abr'), 1000) + vbr = int_or_none(xpath_text(fnode, './videoBitrate', 'vbr'), 1000) + + 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: + format_note = None + + formats.append({ + 'format_id': format_id, + 'url': video_url, + 'ext': ext, + 'acodec': format_m.group('acodec'), + 'vcodec': format_m.group('vcodec'), + 'abr': abr, + 'vbr': vbr, + 'width': width, + 'height': height, + 'filesize': filesize, + 'format_note': format_note, + 'protocol': proto, + '_available': is_available, + }) + format_ids.append(format_id) + + self._sort_formats(formats) return { 'id': video_id, 'title': title, - 'formats': formats, 'description': description, - 'uploader': uploader, 'duration': duration, + 'thumbnails': thumbnails, + 'uploader': uploader, + 'uploader_id': uploader_id, 'upload_date': upload_date, + 'formats': formats, + } + + 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 self.extract_from_xml_url(video_id, xml_url) + + +class ZDFChannelIE(InfoExtractor): + _VALID_URL = r'(?:zdf:topic:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/.*kanaluebersicht/(?:[^/]+/)?)(?P[0-9]+)' + _TESTS = [{ + 'url': 'http://www.zdf.de/ZDFmediathek#/kanaluebersicht/1586442/sendung/Titanic', + 'info_dict': { + 'id': '1586442', + }, + 'playlist_count': 3, + }, { + 'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/aktuellste/332', + 'only_matching': True, + }, { + 'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/meist-gesehen/332', + 'only_matching': True, + }, { + 'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/_/1798716?bc=nrt;nrm?flash=off', + 'only_matching': True, + }] + _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) + + return { + '_type': 'playlist', + 'id': channel_id, + 'entries': entries, }