aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl/extractor
diff options
context:
space:
mode:
Diffstat (limited to 'youtube_dl/extractor')
-rw-r--r--youtube_dl/extractor/__init__.py1
-rw-r--r--youtube_dl/extractor/generic.py19
-rw-r--r--youtube_dl/extractor/livestream.py5
-rw-r--r--youtube_dl/extractor/theplatform.py21
-rw-r--r--youtube_dl/extractor/udn.py64
-rw-r--r--youtube_dl/extractor/youtube.py62
6 files changed, 142 insertions, 30 deletions
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index e6fdf1297..8df1db83e 100644
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -557,6 +557,7 @@ from .udemy import (
UdemyIE,
UdemyCourseIE
)
+from .udn import UDNEmbedIE
from .ultimedia import UltimediaIE
from .unistra import UnistraIE
from .urort import UrortIE
diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 2ff002643..6c212efac 100644
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -34,6 +34,7 @@ from .ooyala import OoyalaIE
from .rutv import RUTVIE
from .smotri import SmotriIE
from .condenast import CondeNastIE
+from .udn import UDNEmbedIE
class GenericIE(InfoExtractor):
@@ -650,6 +651,17 @@ class GenericIE(InfoExtractor):
'title': "PFT Live: New leader in the 'new-look' defense",
'description': 'md5:65a19b4bbfb3b0c0c5768bed1dfad74e',
},
+ },
+ # UDN embed
+ {
+ 'url': 'http://www.udn.com/news/story/7314/822787',
+ 'md5': 'de06b4c90b042c128395a88f0384817e',
+ 'info_dict': {
+ 'id': '300040',
+ 'ext': 'mp4',
+ 'title': '生物老師男變女 全校挺"做自己"',
+ 'thumbnail': 're:^https?://.*\.jpg$',
+ }
}
]
@@ -1268,6 +1280,13 @@ class GenericIE(InfoExtractor):
if nbc_sports_url:
return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
+ # Look for UDN embeds
+ mobj = re.search(
+ r'<iframe[^>]+src="(?P<url>%s)"' % UDNEmbedIE._VALID_URL, webpage)
+ if mobj is not None:
+ return self.url_result(
+ compat_urlparse.urljoin(url, mobj.group('url')), 'UDNEmbed')
+
def check_video(vurl):
if YoutubeIE.suitable(vurl):
return True
diff --git a/youtube_dl/extractor/livestream.py b/youtube_dl/extractor/livestream.py
index 2467f8bdd..ec309dadd 100644
--- a/youtube_dl/extractor/livestream.py
+++ b/youtube_dl/extractor/livestream.py
@@ -21,7 +21,7 @@ from ..utils import (
class LivestreamIE(InfoExtractor):
IE_NAME = 'livestream'
- _VALID_URL = r'https?://new\.livestream\.com/.*?/(?P<event_name>.*?)(/videos/(?P<id>[0-9]+)(?:/player)?)?/?(?:$|[?#])'
+ _VALID_URL = r'https?://(?:new\.)?livestream\.com/.*?/(?P<event_name>.*?)(/videos/(?P<id>[0-9]+)(?:/player)?)?/?(?:$|[?#])'
_TESTS = [{
'url': 'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370',
'md5': '53274c76ba7754fb0e8d072716f2292b',
@@ -51,6 +51,9 @@ class LivestreamIE(InfoExtractor):
}, {
'url': 'https://new.livestream.com/accounts/362/events/3557232/videos/67864563/player?autoPlay=false&height=360&mute=false&width=640',
'only_matching': True,
+ }, {
+ 'url': 'http://livestream.com/bsww/concacafbeachsoccercampeonato2015',
+ 'only_matching': True,
}]
def _parse_smil(self, video_id, smil_url):
diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py
index 0e3e627f4..2d2178331 100644
--- a/youtube_dl/extractor/theplatform.py
+++ b/youtube_dl/extractor/theplatform.py
@@ -28,7 +28,7 @@ class ThePlatformIE(InfoExtractor):
(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/)?
|theplatform:)(?P<id>[^/\?&]+)'''
- _TEST = {
+ _TESTS = [{
# from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
'url': 'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
'info_dict': {
@@ -42,7 +42,20 @@ class ThePlatformIE(InfoExtractor):
# rtmp download
'skip_download': True,
},
- }
+ # from http://www.cnet.com/videos/tesla-model-s-a-second-step-towards-a-cleaner-motoring-future/
+ }, {
+ 'url': 'http://link.theplatform.com/s/kYEXFC/22d_qsQ6MIRT',
+ 'info_dict': {
+ 'id': '22d_qsQ6MIRT',
+ 'ext': 'flv',
+ 'description': 'md5:ac330c9258c04f9d7512cf26b9595409',
+ 'title': 'Tesla Model S: A second step towards a cleaner motoring future',
+ },
+ 'params': {
+ # rtmp download
+ 'skip_download': True,
+ }
+ }]
@staticmethod
def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False):
@@ -127,6 +140,10 @@ class ThePlatformIE(InfoExtractor):
else:
formats = []
switch = body.find(_x('smil:switch'))
+ if switch is None:
+ switch = body.find(_x('smil:par//smil:switch'))
+ if switch is None:
+ switch = body.find(_x('smil:par'))
if switch is not None:
base_url = head.find(_x('smil:meta')).attrib['base']
for f in switch.findall(_x('smil:video')):
diff --git a/youtube_dl/extractor/udn.py b/youtube_dl/extractor/udn.py
new file mode 100644
index 000000000..bba25bb58
--- /dev/null
+++ b/youtube_dl/extractor/udn.py
@@ -0,0 +1,64 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import json
+from .common import InfoExtractor
+from ..utils import js_to_json
+from ..compat import compat_urlparse
+
+
+class UDNEmbedIE(InfoExtractor):
+ _VALID_URL = r'(?:https?:)?//video\.udn\.com/embed/news/(?P<id>\d+)'
+ _TESTS = [{
+ 'url': 'http://video.udn.com/embed/news/300040',
+ 'md5': 'de06b4c90b042c128395a88f0384817e',
+ 'info_dict': {
+ 'id': '300040',
+ 'ext': 'mp4',
+ 'title': '生物老師男變女 全校挺"做自己"',
+ 'thumbnail': 're:^https?://.*\.jpg$',
+ }
+ }, {
+ 'url': '//video.udn.com/embed/news/300040',
+ 'only_matching': True,
+ }]
+
+ def _real_extract(self, url):
+ video_id = self._match_id(url)
+
+ page = self._download_webpage(url, video_id)
+
+ options = json.loads(js_to_json(self._html_search_regex(
+ r'var options\s*=\s*([^;]+);', page, 'video urls dictionary')))
+
+ video_urls = options['video']
+
+ if video_urls.get('youtube'):
+ return self.url_result(video_urls.get('youtube'), 'Youtube')
+
+ try:
+ del video_urls['youtube']
+ except KeyError:
+ pass
+
+ formats = [{
+ 'url': self._download_webpage(
+ compat_urlparse.urljoin(url, api_url), video_id,
+ 'retrieve url for %s video' % video_type),
+ 'format_id': video_type,
+ 'preference': 0 if video_type == 'mp4' else -1,
+ } for video_type, api_url in video_urls.items()]
+
+ self._sort_formats(formats)
+
+ thumbnail = None
+
+ if options.get('gallery') and len(options['gallery']):
+ thumbnail = options['gallery'][0].get('original')
+
+ return {
+ 'id': video_id,
+ 'formats': formats,
+ 'title': options['title'],
+ 'thumbnail': thumbnail
+ }
diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index 79ce39aa4..2774ec30b 100644
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -788,33 +788,41 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
errnote='Could not download DASH manifest')
formats = []
- for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
- url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
- if url_el is None:
- continue
- format_id = r.attrib['id']
- video_url = url_el.text
- filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
- f = {
- 'format_id': format_id,
- 'url': video_url,
- 'width': int_or_none(r.attrib.get('width')),
- 'height': int_or_none(r.attrib.get('height')),
- 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
- 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
- 'filesize': filesize,
- 'fps': int_or_none(r.attrib.get('frameRate')),
- }
- try:
- existing_format = next(
- fo for fo in formats
- if fo['format_id'] == format_id)
- except StopIteration:
- full_info = self._formats.get(format_id, {}).copy()
- full_info.update(f)
- formats.append(full_info)
- else:
- existing_format.update(f)
+ for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
+ mime_type = a.attrib.get('mimeType')
+ for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
+ url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
+ if url_el is None:
+ continue
+ if mime_type == 'text/vtt':
+ # TODO implement WebVTT downloading
+ pass
+ elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
+ format_id = r.attrib['id']
+ video_url = url_el.text
+ filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
+ f = {
+ 'format_id': format_id,
+ 'url': video_url,
+ 'width': int_or_none(r.attrib.get('width')),
+ 'height': int_or_none(r.attrib.get('height')),
+ 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
+ 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
+ 'filesize': filesize,
+ 'fps': int_or_none(r.attrib.get('frameRate')),
+ }
+ try:
+ existing_format = next(
+ fo for fo in formats
+ if fo['format_id'] == format_id)
+ except StopIteration:
+ full_info = self._formats.get(format_id, {}).copy()
+ full_info.update(f)
+ formats.append(full_info)
+ else:
+ existing_format.update(f)
+ else:
+ self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
return formats
def _real_extract(self, url):