[theplatform] Correctly extract videos that don't use f4m or rtmp (reported in #3176)

This commit is contained in:
Jaime Marquínez Ferrándiz 2014-09-21 16:08:38 +02:00
parent 224ce0d872
commit e35cb78c40
2 changed files with 31 additions and 19 deletions

View File

@ -21,7 +21,7 @@ class SBSIE(InfoExtractor):
'md5': '3150cf278965eeabb5b4cea1c963fe0a', 'md5': '3150cf278965eeabb5b4cea1c963fe0a',
'info_dict': { 'info_dict': {
'id': '320403011771', 'id': '320403011771',
'ext': 'flv', 'ext': 'mp4',
'title': 'Dingo Conservation', 'title': 'Dingo Conservation',
'description': 'Dingoes are on the brink of extinction; most of the animals we think are dingoes are in fact crossbred with wild dogs. This family run a dingo conservation park to prevent their extinction', 'description': 'Dingoes are on the brink of extinction; most of the animals we think are dingoes are in fact crossbred with wild dogs. This family run a dingo conservation park to prevent their extinction',
'thumbnail': 're:http://.*\.jpg', 'thumbnail': 're:http://.*\.jpg',

View File

@ -5,6 +5,7 @@ import json
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
compat_str,
ExtractorError, ExtractorError,
xpath_with_ns, xpath_with_ns,
) )
@ -55,7 +56,7 @@ class ThePlatformIE(InfoExtractor):
body = meta.find(_x('smil:body')) body = meta.find(_x('smil:body'))
f4m_node = body.find(_x('smil:seq//smil:video')) f4m_node = body.find(_x('smil:seq//smil:video'))
if f4m_node is not None: if f4m_node is not None and '.f4m' in f4m_node.attrib['src']:
f4m_url = f4m_node.attrib['src'] f4m_url = f4m_node.attrib['src']
if 'manifest.f4m?' not in f4m_url: if 'manifest.f4m?' not in f4m_url:
f4m_url += '?' f4m_url += '?'
@ -64,9 +65,10 @@ class ThePlatformIE(InfoExtractor):
f4m_url += '&g=UXWGVKRWHFSP&hdcore=3.0.3' f4m_url += '&g=UXWGVKRWHFSP&hdcore=3.0.3'
formats = self._extract_f4m_formats(f4m_url, video_id) formats = self._extract_f4m_formats(f4m_url, video_id)
else: else:
base_url = head.find(_x('smil:meta')).attrib['base']
switch = body.find(_x('smil:switch'))
formats = [] formats = []
switch = body.find(_x('smil:switch'))
if switch is not None:
base_url = head.find(_x('smil:meta')).attrib['base']
for f in switch.findall(_x('smil:video')): for f in switch.findall(_x('smil:video')):
attr = f.attrib attr = f.attrib
width = int(attr['width']) width = int(attr['width'])
@ -82,6 +84,16 @@ class ThePlatformIE(InfoExtractor):
'height': height, 'height': height,
'vbr': vbr, 'vbr': vbr,
}) })
else:
switch = body.find(_x('smil:seq//smil:switch'))
for f in switch.findall(_x('smil:video')):
attr = f.attrib
vbr = int(attr['system-bitrate']) // 1000
formats.append({
'format_id': compat_str(vbr),
'url': attr['src'],
'vbr': vbr,
})
self._sort_formats(formats) self._sort_formats(formats)
return { return {