]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/zdf.py
Update changelog.
[youtubedl] / youtube_dl / extractor / zdf.py
index 418509cb96df7822b6120909fa440e2186c677c4..98f15177bd6665bd1c6b96a071d59d4b67e5d918 100644 (file)
+# coding: utf-8
+from __future__ import unicode_literals
+
+import functools
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    ExtractorError,
-    unescapeHTML,
+    int_or_none,
+    unified_strdate,
+    OnDemandPagedList,
 )
 
+
+def extract_from_xml_url(ie, video_id, xml_url):
+    doc = ie._download_xml(
+        xml_url, video_id,
+        note='Downloading video info',
+        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)
+
+    def xml_to_format(fnode):
+        video_url = fnode.find('url').text
+        is_available = 'http://www.metafilegenerator' not in video_url
+
+        format_id = fnode.attrib['basetype']
+        format_m = re.match(r'''(?x)
+            (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
+            (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
+        ''', format_id)
+
+        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
+
+        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)
+
+        format_note = ''
+        if not format_note:
+            format_note = None
+
+        return {
+            'format_id': format_id + '-' + quality,
+            'url': video_url,
+            'ext': ext,
+            'acodec': format_m.group('acodec'),
+            'vcodec': format_m.group('vcodec'),
+            'abr': abr,
+            'vbr': vbr,
+            'width': width,
+            'height': height,
+            'filesize': int_or_none(fnode.find('./filesize').text),
+            'format_note': format_note,
+            'protocol': proto,
+            '_available': is_available,
+        }
+
+    format_nodes = doc.findall('.//formitaeten/formitaet')
+    formats = list(filter(
+        lambda f: f['_available'],
+        map(xml_to_format, format_nodes)))
+    ie._sort_formats(formats)
+
+    return {
+        'id': video_id,
+        'title': title,
+        'description': description,
+        'duration': duration,
+        'uploader': uploader,
+        'uploader_id': uploader_id,
+        'upload_date': upload_date,
+        'formats': formats,
+    }
+
+
 class ZDFIE(InfoExtractor):
-    _VALID_URL = r'^http://www\.zdf\.de\/ZDFmediathek\/(.*beitrag\/video\/)(?P<video_id>[^/\?]+)(?:\?.*)?'
-    _TITLE = r'<h1(?: class="beitragHeadline")?>(?P<title>.*)</h1>'
-    _MEDIA_STREAM = r'<a href="(?P<video_url>.+(?P<media_type>.streaming).+/zdf/(?P<quality>[^\/]+)/[^"]*)".+class="play".+>'
-    _MMS_STREAM = r'href="(?P<video_url>mms://[^"]*)"'
-    _RTSP_STREAM = r'(?P<video_url>rtsp://[^"]*.mp4)'
+    _VALID_URL = r'(?:zdf:|zdf:video:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/(.*beitrag/(?:video/)?))(?P<id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'
+
+    _TEST = {
+        '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 _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
-        video_id = mobj.group('video_id')
-
-        html = self._download_webpage(url, video_id)
-        streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)]
-        if streams is None:
-            raise ExtractorError(u'No media url found.')
-
-        # s['media_type'] == 'wstreaming' -> use 'Windows Media Player' and mms url
-        # s['media_type'] == 'hstreaming' -> use 'Quicktime' and rtsp url
-        # choose first/default media type and highest quality for now
-        for s in streams:        #find 300 - dsl1000mbit
-            if s['quality'] == '300' and s['media_type'] == 'wstreaming':
-                stream_=s
-                break
-        for s in streams:        #find veryhigh - dsl2000mbit
-            if s['quality'] == 'veryhigh' and s['media_type'] == 'wstreaming': # 'hstreaming' - rtsp is not working
-                stream_=s
-                break
-        if stream_ is None:
-            raise ExtractorError(u'No stream found.')
-
-        media_link = self._download_webpage(stream_['video_url'], video_id,'Get stream URL')
-
-        self.report_extraction(video_id)
-        mobj = re.search(self._TITLE, html)
-        if mobj is None:
-            raise ExtractorError(u'Cannot extract title')
-        title = unescapeHTML(mobj.group('title'))
-
-        mobj = re.search(self._MMS_STREAM, media_link)
-        if mobj is None:
-            mobj = re.search(self._RTSP_STREAM, media_link)
-            if mobj is None:
-                raise ExtractorError(u'Cannot extract mms:// or rtsp:// URL')
-        mms_url = mobj.group('video_url')
-
-        mobj = re.search('(.*)[.](?P<ext>[^.]+)', mms_url)
-        if mobj is None:
-            raise ExtractorError(u'Cannot extract extention')
-        ext = mobj.group('ext')
-
-        return [{'id': video_id,
-                 'url': mms_url,
-                 'title': title,
-                 'ext': ext
-                 }]
+        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)
+
+
+class ZDFChannelIE(InfoExtractor):
+    _VALID_URL = r'(?:zdf:topic:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/.*kanaluebersicht/)(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://www.zdf.de/ZDFmediathek#/kanaluebersicht/1586442/sendung/Titanic',
+        'info_dict': {
+            'id': '1586442',
+        },
+        '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)
+
+        return {
+            '_type': 'playlist',
+            'id': channel_id,
+            'entries': entries,
+        }