aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl/extractor
diff options
context:
space:
mode:
authorPhilipp Hagemeister <phihag@phihag.de>2013-12-26 13:49:44 +0100
committerPhilipp Hagemeister <phihag@phihag.de>2013-12-26 13:49:44 +0100
commit608d11f515c9668fa069b422b5850a41e17f52be (patch)
treec658405e840ef3cec33c5a2bf542d0eaf5e234d4 /youtube_dl/extractor
parent723f83991198c09cf3519abd3306704f64b09af0 (diff)
[cnn] Add multiple formats, duration, and upload_date
Diffstat (limited to 'youtube_dl/extractor')
-rw-r--r--youtube_dl/extractor/cnn.py71
1 files changed, 56 insertions, 15 deletions
diff --git a/youtube_dl/extractor/cnn.py b/youtube_dl/extractor/cnn.py
index a034bb2fb..ecac5e0e9 100644
--- a/youtube_dl/extractor/cnn.py
+++ b/youtube_dl/extractor/cnn.py
@@ -1,7 +1,10 @@
import re
from .common import InfoExtractor
-from ..utils import determine_ext
+from ..utils import (
+ int_or_none,
+ parse_duration,
+)
class CNNIE(InfoExtractor):
@@ -15,6 +18,8 @@ class CNNIE(InfoExtractor):
u'info_dict': {
u'title': u'Nadal wins 8th French Open title',
u'description': u'World Sport\'s Amanda Davies chats with 2013 French Open champion Rafael Nadal.',
+ u'duration': 135,
+ u'upload_date': u'20130609',
},
},
{
@@ -35,22 +40,58 @@ class CNNIE(InfoExtractor):
info = self._download_xml(info_url, page_title)
formats = []
+ rex = re.compile(r'''(?x)
+ (?P<width>[0-9]+)x(?P<height>[0-9]+)
+ (?:_(?P<bitrate>[0-9]+)k)?
+ ''')
for f in info.findall('files/file'):
- mf = re.match(r'(\d+)x(\d+)(?:_(.*)k)?',f.attrib['bitrate'])
- if mf is not None:
- formats.append((int(mf.group(1)), int(mf.group(2)), int(mf.group(3) or 0), f.text))
- formats = sorted(formats)
- (_,_,_, video_path) = formats[-1]
- video_url = 'http://ht.cdn.turner.com/cnn/big%s' % video_path
+ video_url = 'http://ht.cdn.turner.com/cnn/big%s' % (f.text.strip())
+ fdct = {
+ 'format_id': f.attrib['bitrate'],
+ 'url': video_url,
+ }
+
+ mf = rex.match(f.attrib['bitrate'])
+ if mf:
+ fdct['width'] = int(mf.group('width'))
+ fdct['height'] = int(mf.group('height'))
+ fdct['tbr'] = int_or_none(mf.group('bitrate'))
+ else:
+ mf = rex.search(f.text)
+ if mf:
+ fdct['width'] = int(mf.group('width'))
+ fdct['height'] = int(mf.group('height'))
+ fdct['tbr'] = int_or_none(mf.group('bitrate'))
+ else:
+ mi = re.match(r'ios_(audio|[0-9]+)$', f.attrib['bitrate'])
+ if mi:
+ if mi.group(1) == 'audio':
+ fdct['vcodec'] = 'none'
+ fdct['ext'] = 'm4a'
+ else:
+ fdct['tbr'] = int(mi.group(1))
+
+ formats.append(fdct)
+
+ self._sort_formats(formats)
thumbnails = sorted([((int(t.attrib['height']),int(t.attrib['width'])), t.text) for t in info.findall('images/image')])
thumbs_dict = [{'resolution': res, 'url': t_url} for (res, t_url) in thumbnails]
- return {'id': info.attrib['id'],
- 'title': info.find('headline').text,
- 'url': video_url,
- 'ext': determine_ext(video_url),
- 'thumbnail': thumbnails[-1][1],
- 'thumbnails': thumbs_dict,
- 'description': info.find('description').text,
- }
+ metas_el = info.find('metas')
+ upload_date = (
+ metas_el.attrib.get('version') if metas_el is not None else None)
+
+ duration_el = info.find('length')
+ duration = parse_duration(duration_el.text)
+
+ return {
+ 'id': info.attrib['id'],
+ 'title': info.find('headline').text,
+ 'formats': formats,
+ 'thumbnail': thumbnails[-1][1],
+ 'thumbnails': thumbs_dict,
+ 'description': info.find('description').text,
+ 'duration': duration,
+ 'upload_date': upload_date,
+ }