From 6a656a843a629ceef6979976a353d177c97b9527 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaime.marquinez.ferrandiz@gmail.com>
Date: Wed, 4 Dec 2013 20:35:00 +0100
Subject: [PATCH 1/5] Update description value for the write_info_json test
(required after 27dcce19045670fc348ff1119c0d2283aaed3ae2)
---
test/test_write_info_json.py | 1 +
1 file changed, 1 insertion(+)
diff --git a/test/test_write_info_json.py b/test/test_write_info_json.py
index d7177611b..90426a559 100644
--- a/test/test_write_info_json.py
+++ b/test/test_write_info_json.py
@@ -33,6 +33,7 @@ TEST_ID = 'BaW_jenozKc'
INFO_JSON_FILE = TEST_ID + '.info.json'
DESCRIPTION_FILE = TEST_ID + '.mp4.description'
EXPECTED_DESCRIPTION = u'''test chars: "'/\ä↭𝕐
+test URL: https://github.com/rg3/youtube-dl/issues/1892
This is a test video for youtube-dl.
From bfb9f7bc4c5c6fd9b2d3d46be133988f70534d26 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaime.marquinez.ferrandiz@gmail.com>
Date: Wed, 4 Dec 2013 20:36:26 +0100
Subject: [PATCH 2/5] [hotnewhiphop] Update test's title
---
youtube_dl/extractor/hotnewhiphop.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/youtube_dl/extractor/hotnewhiphop.py b/youtube_dl/extractor/hotnewhiphop.py
index 3798118a7..0ee74fb38 100644
--- a/youtube_dl/extractor/hotnewhiphop.py
+++ b/youtube_dl/extractor/hotnewhiphop.py
@@ -11,7 +11,7 @@ class HotNewHipHopIE(InfoExtractor):
u'file': u'1435540.mp3',
u'md5': u'2c2cd2f76ef11a9b3b581e8b232f3d96',
u'info_dict': {
- u"title": u"Freddie Gibbs - Lay It Down"
+ u"title": u'Freddie Gibbs "Lay It Down"'
}
}
From e9bf7479d209c2623753628201ca0daffa19f3cf Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaime.marquinez.ferrandiz@gmail.com>
Date: Wed, 4 Dec 2013 23:28:40 +0100
Subject: [PATCH 3/5] Add an extractor for theplatform.com
---
youtube_dl/extractor/__init__.py | 1 +
youtube_dl/extractor/theplatform.py | 69 +++++++++++++++++++++++++++++
2 files changed, 70 insertions(+)
create mode 100644 youtube_dl/extractor/theplatform.py
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index bd996483b..900a6f02f 100644
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -144,6 +144,7 @@ from .teamcoco import TeamcocoIE
from .techtalks import TechTalksIE
from .ted import TEDIE
from .tf1 import TF1IE
+from .theplatform import ThePlatformIE
from .thisav import ThisAVIE
from .toutv import TouTvIE
from .traileraddict import TrailerAddictIE
diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py
new file mode 100644
index 000000000..d1d6a4c2c
--- /dev/null
+++ b/youtube_dl/extractor/theplatform.py
@@ -0,0 +1,69 @@
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+ xpath_with_ns,
+ find_xpath_attr,
+)
+
+_x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})
+
+
+class ThePlatformIE(InfoExtractor):
+ _VALID_URL = r'https?://link\.theplatform\.com/s/[^/]+/(?P<id>[^/\?]+)'
+
+ _TEST = {
+ # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
+ u'url': u'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
+ u'info_dict': {
+ u'id': u'e9I_cZgTgIPd',
+ u'ext': u'flv',
+ u'title': u'Blackberry\'s big, bold Z30',
+ u'description': u'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
+ u'duration': 247,
+ },
+ u'params': {
+ # rtmp download
+ u'skip_download': True,
+ },
+ }
+
+ def _get_info(self, video_id):
+ smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?'
+ 'format=smil&mbr=true'.format(video_id))
+ meta = self._download_xml(smil_url, video_id)
+ info_url = 'http://link.theplatform.com/s/dJ5BDC/{0}?format=preview'.format(video_id)
+ info_json = self._download_webpage(info_url, video_id)
+ info = json.loads(info_json)
+
+ head = meta.find(_x('smil:head'))
+ body = meta.find(_x('smil:body'))
+ base_url = head.find(_x('smil:meta')).attrib['base']
+ switch = body.find(_x('smil:switch'))
+ formats = []
+ for f in switch.findall(_x('smil:video')):
+ attr = f.attrib
+ formats.append({
+ 'url': base_url,
+ 'play_path': 'mp4:' + attr['src'],
+ 'ext': 'flv',
+ 'width': int(attr['width']),
+ 'height': int(attr['height']),
+ 'vbr': int(attr['system-bitrate']),
+ })
+ formats.sort(key=lambda f: (f['height'], f['width'], f['vbr']))
+
+ return {
+ 'id': video_id,
+ 'title': info['title'],
+ 'formats': formats,
+ 'description': info['description'],
+ 'thumbnail': info['defaultThumbnailUrl'],
+ 'duration': info['duration']//1000,
+ }
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+ video_id = mobj.group('id')
+ return self._get_info(video_id)
From b9a2c53833a3cebc32df908aad74f7c5a3537aa1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaime.marquinez.ferrandiz@gmail.com>
Date: Wed, 4 Dec 2013 23:43:50 +0100
Subject: [PATCH 4/5] [metacafe] Add support for cbs videos (fixes #1838)
They use theplatform.com
---
youtube_dl/extractor/metacafe.py | 29 +++++++++++++++++++++++++----
youtube_dl/extractor/theplatform.py | 2 +-
2 files changed, 26 insertions(+), 5 deletions(-)
diff --git a/youtube_dl/extractor/metacafe.py b/youtube_dl/extractor/metacafe.py
index 91480ba87..e59bdd604 100644
--- a/youtube_dl/extractor/metacafe.py
+++ b/youtube_dl/extractor/metacafe.py
@@ -69,6 +69,21 @@ class MetacafeIE(InfoExtractor):
u'age_limit': 18,
},
},
+ # cbs video
+ {
+ u'url': u'http://www.metacafe.com/watch/cb-0rOxMBabDXN6/samsung_galaxy_note_2_samsungs_next_generation_phablet/',
+ u'info_dict': {
+ u'id': u'0rOxMBabDXN6',
+ u'ext': u'flv',
+ u'title': u'Samsung Galaxy Note 2: Samsung\'s next-generation phablet',
+ u'description': u'md5:54d49fac53d26d5a0aaeccd061ada09d',
+ u'duration': 129,
+ },
+ u'params': {
+ # rtmp download
+ u'skip_download': True,
+ },
+ },
]
@@ -106,10 +121,16 @@ class MetacafeIE(InfoExtractor):
video_id = mobj.group(1)
- # Check if video comes from YouTube
- mobj2 = re.match(r'^yt-(.*)$', video_id)
- if mobj2 is not None:
- return [self.url_result('http://www.youtube.com/watch?v=%s' % mobj2.group(1), 'Youtube')]
+ # the video may come from an external site
+ m_external = re.match('^(\w{2})-(.*)$', video_id)
+ if m_external is not None:
+ prefix, ext_id = m_external.groups()
+ # Check if video comes from YouTube
+ if prefix == 'yt':
+ return self.url_result('http://www.youtube.com/watch?v=%s' % ext_id, 'Youtube')
+ # CBS videos use theplatform.com
+ if prefix == 'cb':
+ return self.url_result('theplatform:%s' % ext_id, 'ThePlatform')
# Retrieve video webpage to extract further information
req = compat_urllib_request.Request('http://www.metacafe.com/watch/%s/' % video_id)
diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py
index d1d6a4c2c..920689511 100644
--- a/youtube_dl/extractor/theplatform.py
+++ b/youtube_dl/extractor/theplatform.py
@@ -11,7 +11,7 @@ _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language
class ThePlatformIE(InfoExtractor):
- _VALID_URL = r'https?://link\.theplatform\.com/s/[^/]+/(?P<id>[^/\?]+)'
+ _VALID_URL = r'(?:https?://link\.theplatform\.com/s/[^/]+/|theplatform:)(?P<id>[^/\?]+)'
_TEST = {
# from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
From 673d1273ff6f6d3267728fbe6f79c9c801598fd2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?=
<jaime.marquinez.ferrandiz@gmail.com>
Date: Thu, 5 Dec 2013 12:41:58 +0100
Subject: [PATCH 5/5] [vevo] Support '/watch/{id}' urls
---
youtube_dl/extractor/vevo.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/youtube_dl/extractor/vevo.py b/youtube_dl/extractor/vevo.py
index d8bfcd155..3eedcf7dd 100644
--- a/youtube_dl/extractor/vevo.py
+++ b/youtube_dl/extractor/vevo.py
@@ -15,7 +15,7 @@ class VevoIE(InfoExtractor):
Accepts urls from vevo.com or in the format 'vevo:{id}'
(currently used by MTVIE)
"""
- _VALID_URL = r'((http://www\.vevo\.com/watch/.*?/.*?/)|(vevo:))(?P<id>.*?)(\?|$)'
+ _VALID_URL = r'((http://www\.vevo\.com/watch/(?:[^/]+/[^/]+/)?)|(vevo:))(?P<id>.*?)(\?|$)'
_TESTS = [{
u'url': u'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
u'file': u'GB1101300280.mp4',