aboutsummaryrefslogtreecommitdiffstats
path: root/youtube_dl/extractor
diff options
context:
space:
mode:
authordst <dstftw@gmail.com>2013-12-12 15:19:23 +0700
committerdst <dstftw@gmail.com>2013-12-12 15:19:23 +0700
commit4d2ebb6bd771edaa26ebb5e76f9ee8e880a4d152 (patch)
tree9de7463e9ea65068648df91012bcaa2570ef5355 /youtube_dl/extractor
parentdf5374743615fe54178b3942b115e7168ce0cd97 (diff)
downloadyoutube-dl-4d2ebb6bd771edaa26ebb5e76f9ee8e880a4d152.zip
youtube-dl-4d2ebb6bd771edaa26ebb5e76f9ee8e880a4d152.tar.gz
youtube-dl-4d2ebb6bd771edaa26ebb5e76f9ee8e880a4d152.tar.bz2
[channel9] Cleanup
Diffstat (limited to 'youtube_dl/extractor')
-rw-r--r--youtube_dl/extractor/channel9.py94
1 files changed, 5 insertions, 89 deletions
diff --git a/youtube_dl/extractor/channel9.py b/youtube_dl/extractor/channel9.py
index 016c449..b4b586c 100644
--- a/youtube_dl/extractor/channel9.py
+++ b/youtube_dl/extractor/channel9.py
@@ -3,10 +3,7 @@
import re
from .common import InfoExtractor
-from ..utils import (
- format_bytes,
- ExtractorError,
-)
+from ..utils import ExtractorError
class Channel9IE(InfoExtractor):
'''
@@ -51,7 +48,6 @@ class Channel9IE(InfoExtractor):
]
_RSS_URL = 'http://channel9.msdn.com/%s/RSS'
- _EXTRACT_ENTRY_ITEMS_FROM_RSS = False
# Sorted by quality
_known_formats = ['MP3', 'MP4', 'Mid Quality WMV', 'Mid Quality MP4', 'High Quality WMV', 'High Quality MP4']
@@ -90,37 +86,6 @@ class Channel9IE(InfoExtractor):
formats.sort(key=lambda fmt: self._known_formats.index(fmt['format_id']))
return formats
- def _formats_from_rss_item(self, item):
-
- def process_formats(elem):
- formats = []
- for media_content in elem.findall('./{http://search.yahoo.com/mrss/}content'):
- url = media_content.attrib['url']
- # Ignore unrelated media
- if url.endswith('.ism/manifest'):
- continue
- format_note = media_content.attrib['type']
- filesize = int(media_content.attrib['fileSize'])
- formats.append({'url': url,
- 'format_note': format_note,
- 'format': '%s %s' % (format_note, format_bytes(filesize)),
- 'filesize': filesize,
- })
- return formats
-
- formats = []
-
- for media_group in item.findall('./{http://search.yahoo.com/mrss/}group'):
- formats.extend(process_formats(media_group))
-
- # Sometimes there are no media:groups in item, but there is media:content
- # right in item (usually when there is the only media source)
- formats.extend(process_formats(item))
-
- # Sort by file size
- formats.sort(key=lambda fmt: fmt['filesize'])
- return formats
-
def _extract_title(self, html):
title = self._html_search_meta(u'title', html, u'title')
if title is None:
@@ -274,61 +239,12 @@ class Channel9IE(InfoExtractor):
return contents
- def _extract_content_rss(self, rss):
- '''
- Extracts links to entry items right out of RSS feed.
- This approach is faster than extracting from web pages
- one by one, but suffers from some problems.
- Pros:
- - no need to download additional pages
- - provides more media links
- - accurate file size
- Cons:
- - fewer meta data provided
- - links to media files have no appropriate data that may be used as format_id
- - RSS does not contain links to presentation materials (slides, zip)
- '''
- entries = []
- for item in rss.findall('./channel/item'):
- url = item.find('./link').text
- video_id = url.split('/')[-1]
- formats = self._formats_from_rss_item(item)
-
- if len(formats) == 0:
- self._downloader.report_warning(u'The recording for session %s is not yet available' % video_id)
- continue
-
- title = item.find('./title').text
- description = item.find('./description').text
-
- thumbnail = item.find('./{http://search.yahoo.com/mrss/}thumbnail').text
-
- duration_e = item.find('./{http://www.itunes.com/dtds/podcast-1.0.dtd}duration')
- duration = duration_e.text if duration_e is not None else 0
-
- speakers_e = item.find('./{http://purl.org/dc/elements/1.1/}creator')
- speakers = speakers_e.text.split(', ') if speakers_e is not None and speakers_e.text else []
-
- entries.append({'_type': 'video',
- 'id': video_id,
- 'formats': formats,
- 'title': title,
- 'description': description,
- 'thumbnail': thumbnail,
- 'duration': duration,
- 'session_speakers': speakers,
- })
- return entries
-
def _extract_list(self, content_path):
rss = self._download_xml(self._RSS_URL % content_path, content_path, u'Downloading RSS')
- if self._EXTRACT_ENTRY_ITEMS_FROM_RSS:
- return self._extract_content_rss(rss)
- else:
- entries = [self.url_result(session_url.text, 'Channel9')
- for session_url in rss.findall('./channel/item/link')]
- title_text = rss.find('./channel/title').text
- return self.playlist_result(entries, content_path, title_text)
+ entries = [self.url_result(session_url.text, 'Channel9')
+ for session_url in rss.findall('./channel/item/link')]
+ title_text = rss.find('./channel/title').text
+ return self.playlist_result(entries, content_path, title_text)
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)