midgard/youtube-dl
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

[br] add support for br-klassik.de and improve extraction

Browse source 
- extend _VALID_URL to match both br.de and br-klassik.de
- extract all formats(hls,hds and rtmp)
- use xpath_element and xpath_text for xml info extraction
This commit is contained in:
remitamine 2015-12-21 21:06:10 +01:00
parent 5ef5d25b15
commit 5625bd0617
1 changed files with 78 additions and 49 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

127
youtube_dl/extractor/br.py
View file

@ -1,18 +1,21 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
ExtractorError, ExtractorError,
int_or_none, int_or_none,
parse_duration, parse_duration,
xpath_element,
xpath_text,
) )
class BRIE(InfoExtractor): class BRIE(InfoExtractor):
IE_DESC = 'Bayerischer Rundfunk Mediathek' IE_DESC = 'Bayerischer Rundfunk Mediathek'
_VALID_URL = r'https?://(?:www\.)?br\.de/(?:[a-z0-9\-_]+/)+(?P<id>[a-z0-9\-_]+)\.html' _VALID_URL = r'(?P<base_url>https?://(?:www\.)?br(?:-klassik)?\.de)/(?:[a-z0-9\-_]+/)+(?P<id>[a-z0-9\-_]+)\.html'
_BASE_URL = 'http://www.br.de'
_TESTS = [ _TESTS = [
{ {
@ -22,7 +25,7 @@ class BRIE(InfoExtractor):
'id': '48f656ef-287e-486f-be86-459122db22cc', 'id': '48f656ef-287e-486f-be86-459122db22cc',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Die böse Überraschung', 'title': 'Die böse Überraschung',
'description': 'Betriebliche Altersvorsorge: Die böse Überraschung', 'description': 'md5:ce9ac81b466ce775b8018f6801b48ac9',
'duration': 180, 'duration': 180,
'uploader': 'Reinhard Weber', 'uploader': 'Reinhard Weber',
'upload_date': '20150422', 'upload_date': '20150422',
@ -30,23 +33,23 @@ class BRIE(InfoExtractor):
}, },
{ {
'url': 'http://www.br.de/nachrichten/oberbayern/inhalt/muenchner-polizeipraesident-schreiber-gestorben-100.html', 'url': 'http://www.br.de/nachrichten/oberbayern/inhalt/muenchner-polizeipraesident-schreiber-gestorben-100.html',
'md5': 'a44396d73ab6a68a69a568fae10705bb', 'md5': 'af3a3a4aa43ff0ce6a89504c67f427ef',
'info_dict': { 'info_dict': {
'id': 'a4b83e34-123d-4b81-9f4e-c0d3121a4e05', 'id': 'a4b83e34-123d-4b81-9f4e-c0d3121a4e05',
'ext': 'mp4', 'ext': 'flv',
'title': 'Manfred Schreiber ist tot', 'title': 'Manfred Schreiber ist tot',
'description': 'Abendschau kompakt: Manfred Schreiber ist tot', 'description': 'md5:b454d867f2a9fc524ebe88c3f5092d97',
'duration': 26, 'duration': 26,
} }
}, },
{ {
'url': 'http://www.br.de/radio/br-klassik/sendungen/allegro/premiere-urauffuehrung-the-land-2015-dance-festival-muenchen-100.html', 'url': 'https://www.br-klassik.de/audio/peeping-tom-premierenkritik-dance-festival-muenchen-100.html',
'md5': '8b5b27c0b090f3b35eac4ab3f7a73d3d', 'md5': '8b5b27c0b090f3b35eac4ab3f7a73d3d',
'info_dict': { 'info_dict': {
'id': '74c603c9-26d3-48bb-b85b-079aeed66e0b', 'id': '74c603c9-26d3-48bb-b85b-079aeed66e0b',
'ext': 'aac', 'ext': 'aac',
'title': 'Kurzweilig und sehr bewegend', 'title': 'Kurzweilig und sehr bewegend',
'description': '"The Land" von Peeping Tom: Kurzweilig und sehr bewegend', 'description': 'md5:0351996e3283d64adeb38ede91fac54e',
'duration': 296, 'duration': 296,
} }
}, },
@ -57,7 +60,7 @@ class BRIE(InfoExtractor):
'id': '6ba73750-d405-45d3-861d-1ce8c524e059', 'id': '6ba73750-d405-45d3-861d-1ce8c524e059',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Umweltbewusster Häuslebauer', 'title': 'Umweltbewusster Häuslebauer',
'description': 'Uwe Erdelt: Umweltbewusster Häuslebauer', 'description': 'md5:d52dae9792d00226348c1dbb13c9bae2',
'duration': 116, 'duration': 116,
} }
}, },
@ -68,7 +71,7 @@ class BRIE(InfoExtractor):
'id': 'd982c9ce-8648-4753-b358-98abb8aec43d', 'id': 'd982c9ce-8648-4753-b358-98abb8aec43d',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Folge 1 - Metaphysik', 'title': 'Folge 1 - Metaphysik',
'description': 'Kant für Anfänger: Folge 1 - Metaphysik', 'description': 'md5:bb659990e9e59905c3d41e369db1fbe3',
'duration': 893, 'duration': 893,
'uploader': 'Eva Maria Steimle', 'uploader': 'Eva Maria Steimle',
'upload_date': '20140117', 'upload_date': '20140117',
@ -77,28 +80,31 @@ class BRIE(InfoExtractor):
] ]
def _real_extract(self, url): def _real_extract(self, url):
display_id = self._match_id(url) base_url, display_id = re.search(self._VALID_URL, url).groups()
page = self._download_webpage(url, display_id) page = self._download_webpage(url, display_id)
xml_url = self._search_regex( xml_url = self._search_regex(
r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/(?:[a-z0-9\-]+/)+[a-z0-9/~_.-]+)'}\)\);", page, 'XMLURL') r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/(?:[a-z0-9\-]+/)+[a-z0-9/~_.-]+)'}\)\);", page, 'XMLURL')
xml = self._download_xml(self._BASE_URL + xml_url, None) xml = self._download_xml(base_url + xml_url, display_id)
medias = [] medias = []
for xml_media in xml.findall('video') + xml.findall('audio'): for xml_media in xml.findall('video') + xml.findall('audio'):
media_id = xml_media.get('externalId')
media = { media = {
'id': xml_media.get('externalId'), 'id': media_id,
'title': xml_media.find('title').text, 'title': xpath_text(xml_media, 'title', 'title', True),
'duration': parse_duration(xml_media.find('duration').text), 'duration': parse_duration(xpath_text(xml_media, 'duration')),
'formats': self._extract_formats(xml_media.find('assets')), 'formats': self._extract_formats(xpath_element(
'thumbnails': self._extract_thumbnails(xml_media.find('teaserImage/variants')), xml_media, 'assets'), media_id),
'description': ' '.join(xml_media.find('shareTitle').text.splitlines()), 'thumbnails': self._extract_thumbnails(xpath_element(
'webpage_url': xml_media.find('permalink').text xml_media, 'teaserImage/variants'), base_url),
'description': xpath_text(xml_media, 'desc'),
'webpage_url': xpath_text(xml_media, 'permalink'),
'uploader': xpath_text(xml_media, 'author'),
} }
if xml_media.find('author').text: broadcast_date = xpath_text(xml_media, 'broadcastDate')
media['uploader'] = xml_media.find('author').text if broadcast_date:
if xml_media.find('broadcastDate').text: media['upload_date'] = ''.join(reversed(broadcast_date.split('.')))
media['upload_date'] = ''.join(reversed(xml_media.find('broadcastDate').text.split('.')))
medias.append(media) medias.append(media)
if len(medias) > 1: if len(medias) > 1:
@ -109,35 +115,58 @@ class BRIE(InfoExtractor):
raise ExtractorError('No media entries found') raise ExtractorError('No media entries found')
return medias[0] return medias[0]
def _extract_formats(self, assets): def _extract_formats(self, assets, media_id):
formats = []
def text_or_none(asset, tag): for asset in assets.findall('asset'):
elem = asset.find(tag) format_url = xpath_text(asset, ['downloadUrl', 'url'])
return None if elem is None else elem.text asset_type = asset.get('type')
if asset_type == 'HDS':
formats = [{ f4m_formats = self._extract_f4m_formats(
'url': text_or_none(asset, 'downloadUrl'), format_url + '?hdcore=3.2.0', media_id, f4m_id='hds', fatal=False)
'ext': text_or_none(asset, 'mediaType'), if f4m_formats:
'format_id': asset.get('type'), formats.extend(f4m_formats)
'width': int_or_none(text_or_none(asset, 'frameWidth')), elif asset_type == 'HLS':
'height': int_or_none(text_or_none(asset, 'frameHeight')), m3u8_formats = self._extract_m3u8_formats(
'tbr': int_or_none(text_or_none(asset, 'bitrateVideo')), format_url, media_id, 'mp4', 'm3u8_native', m3u8_id='hds', fatal=False)
'abr': int_or_none(text_or_none(asset, 'bitrateAudio')), if m3u8_formats:
'vcodec': text_or_none(asset, 'codecVideo'), formats.extend(m3u8_formats)
'acodec': text_or_none(asset, 'codecAudio'), else:
'container': text_or_none(asset, 'mediaType'), format_info = {
'filesize': int_or_none(text_or_none(asset, 'size')), 'ext': xpath_text(asset, 'mediaType'),
} for asset in assets.findall('asset') 'width': int_or_none(xpath_text(asset, 'frameWidth')),
if asset.find('downloadUrl') is not None] 'height': int_or_none(xpath_text(asset, 'frameHeight')),
'tbr': int_or_none(xpath_text(asset, 'bitrateVideo')),
'abr': int_or_none(xpath_text(asset, 'bitrateAudio')),
'vcodec': xpath_text(asset, 'codecVideo'),
'acodec': xpath_text(asset, 'codecAudio'),
'container': xpath_text(asset, 'mediaType'),
'filesize': int_or_none(xpath_text(asset, 'size')),
}
format_url = self._proto_relative_url(format_url)
if format_url:
http_format_info = format_info.copy()
http_format_info.update({
'url': format_url,
'format_id': 'http-%s' % asset_type,
})
formats.append(http_format_info)
server_prefix = xpath_text(asset, 'serverPrefix')
if server_prefix:
rtmp_format_info = format_info.copy()
rtmp_format_info.update({
'url': server_prefix,
'play_path': xpath_text(asset, 'fileName'),
'format_id': 'rtmp-%s' % asset_type,
})
formats.append(rtmp_format_info)
self._sort_formats(formats) self._sort_formats(formats)
return formats return formats
def _extract_thumbnails(self, variants): def _extract_thumbnails(self, variants, base_url):
thumbnails = [{ thumbnails = [{
'url': self._BASE_URL + variant.find('url').text, 'url': base_url + xpath_text(variant, 'url'),
'width': int_or_none(variant.find('width').text), 'width': int_or_none(xpath_text(variant, 'width')),
'height': int_or_none(variant.find('height').text), 'height': int_or_none(xpath_text(variant, 'height')),
} for variant in variants.findall('variant')] } for variant in variants.findall('variant') if xpath_text(variant, 'url')]
thumbnails.sort(key=lambda x: x['width'] * x['height'], reverse=True) thumbnails.sort(key=lambda x: x['width'] * x['height'], reverse=True)
return thumbnails return thumbnails