commit 66398056f17d0a9c04861a98b6e95d70ab08a353
parent b9c76aa1a9c80013ad4f2f3cd20b98ab1be430ac
Author: Philipp Hagemeister <phihag@phihag.de>
Date: Sat, 12 Apr 2014 17:15:16 +0200
Merge branch 'master' of github.com:rg3/youtube-dl
Diffstat:
8 files changed, 127 insertions(+), 27 deletions(-)
diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py
@@ -92,6 +92,8 @@ from .extractor import gen_extractors
from .version import __version__
from .YoutubeDL import YoutubeDL
from .postprocessor import (
+ AtomicParsleyPP,
+ FFmpegAudioFixPP,
FFmpegMetadataPP,
FFmpegVideoConvertor,
FFmpegExtractAudioPP,
@@ -503,6 +505,8 @@ def parseOpts(overrideArguments=None):
help='do not overwrite post-processed files; the post-processed files are overwritten by default')
postproc.add_option('--embed-subs', action='store_true', dest='embedsubtitles', default=False,
help='embed subtitles in the video (only for mp4 videos)')
+ postproc.add_option('--embed-thumbnail', action='store_true', dest='embedthumbnail', default=False,
+ help='embed thumbnail in the audio as cover art')
postproc.add_option('--add-metadata', action='store_true', dest='addmetadata', default=False,
help='write metadata to the video file')
postproc.add_option('--xattrs', action='store_true', dest='xattrs', default=False,
@@ -808,6 +812,10 @@ def _real_main(argv=None):
ydl.add_post_processor(FFmpegEmbedSubtitlePP(subtitlesformat=opts.subtitlesformat))
if opts.xattrs:
ydl.add_post_processor(XAttrMetadataPP())
+ if opts.embedthumbnail:
+ if not opts.addmetadata:
+ ydl.add_post_processor(FFmpegAudioFixPP())
+ ydl.add_post_processor(AtomicParsleyPP())
# Update version
if opts.update_self:
diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py
@@ -140,7 +140,11 @@ class BrightcoveIE(InfoExtractor):
url_m = re.search(r'<meta\s+property="og:video"\s+content="(http://c.brightcove.com/[^"]+)"', webpage)
if url_m:
- return [unescapeHTML(url_m.group(1))]
+ url = unescapeHTML(url_m.group(1))
+ # Some sites don't add it, we can't download with this url, for example:
+ # http://www.ktvu.com/videos/news/raw-video-caltrain-releases-video-of-man-almost/vCTZdY/
+ if 'playerKey' in url:
+ return [url]
matches = re.findall(
r'''(?sx)<object
diff --git a/youtube_dl/extractor/firstpost.py b/youtube_dl/extractor/firstpost.py
@@ -6,7 +6,6 @@ from .common import InfoExtractor
class FirstpostIE(InfoExtractor):
- IE_NAME = 'Firstpost.com'
_VALID_URL = r'http://(?:www\.)?firstpost\.com/[^/]+/.*-(?P<id>[0-9]+)\.html'
_TEST = {
@@ -16,7 +15,6 @@ class FirstpostIE(InfoExtractor):
'id': '1025403',
'ext': 'mp4',
'title': 'India to launch indigenous aircraft carrier INS Vikrant today',
- 'description': 'Its flight deck is over twice the size of a football field, its power unit can light up the entire Kochi city and the cabling is enough to cover the distance between here to Delhi.',
}
}
@@ -24,15 +22,26 @@ class FirstpostIE(InfoExtractor):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
- webpage = self._download_webpage(url, video_id)
- video_url = self._html_search_regex(
- r'<div.*?name="div_video".*?flashvars="([^"]+)">',
- webpage, 'video URL')
+ data = self._download_xml(
+ 'http://www.firstpost.com/getvideoxml-%s.xml' % video_id, video_id,
+ 'Downloading video XML')
+
+ item = data.find('./playlist/item')
+ thumbnail = item.find('./image').text
+ title = item.find('./title').text
+
+ formats = [
+ {
+ 'url': details.find('./file').text,
+ 'format_id': details.find('./label').text.strip(),
+ 'width': int(details.find('./width').text.strip()),
+ 'height': int(details.find('./height').text.strip()),
+ } for details in item.findall('./source/file_details') if details.find('./file').text
+ ]
return {
'id': video_id,
- 'url': video_url,
- 'title': self._og_search_title(webpage),
- 'description': self._og_search_description(webpage),
- 'thumbnail': self._og_search_thumbnail(webpage),
+ 'title': title,
+ 'thumbnail': thumbnail,
+ 'formats': formats,
}
diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py
@@ -160,6 +160,7 @@ class ProSiebenSat1IE(InfoExtractor):
_CLIPID_REGEXES = [
r'"clip_id"\s*:\s+"(\d+)"',
r'clipid: "(\d+)"',
+ r'clipId=(\d+)',
]
_TITLE_REGEXES = [
r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
diff --git a/youtube_dl/extractor/weibo.py b/youtube_dl/extractor/weibo.py
@@ -1,10 +1,11 @@
# coding: utf-8
+from __future__ import unicode_literals
import re
-import json
from .common import InfoExtractor
+
class WeiboIE(InfoExtractor):
"""
The videos in Weibo come from different sites, this IE just finds the link
@@ -13,16 +14,16 @@ class WeiboIE(InfoExtractor):
_VALID_URL = r'https?://video\.weibo\.com/v/weishipin/t_(?P<id>.+?)\.htm'
_TEST = {
- u'add_ie': ['Sina'],
- u'url': u'http://video.weibo.com/v/weishipin/t_zjUw2kZ.htm',
- u'file': u'98322879.flv',
- u'info_dict': {
- u'title': u'魔声耳机最新广告“All Eyes On Us”',
+ 'url': 'http://video.weibo.com/v/weishipin/t_zjUw2kZ.htm',
+ 'info_dict': {
+ 'id': '98322879',
+ 'ext': 'flv',
+ 'title': '魔声耳机最新广告“All Eyes On Us”',
},
- u'note': u'Sina video',
- u'params': {
- u'skip_download': True,
+ 'params': {
+ 'skip_download': True,
},
+ 'add_ie': ['Sina'],
}
# Additional example videos from different sites
@@ -33,17 +34,16 @@ class WeiboIE(InfoExtractor):
mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
video_id = mobj.group('id')
info_url = 'http://video.weibo.com/?s=v&a=play_list&format=json&mix_video_id=t_%s' % video_id
- info_page = self._download_webpage(info_url, video_id)
- info = json.loads(info_page)
+ info = self._download_json(info_url, video_id)
videos_urls = map(lambda v: v['play_page_url'], info['result']['data'])
- #Prefer sina video since they have thumbnails
- videos_urls = sorted(videos_urls, key=lambda u: u'video.sina.com' in u)
+ # Prefer sina video since they have thumbnails
+ videos_urls = sorted(videos_urls, key=lambda u: 'video.sina.com' in u)
player_url = videos_urls[-1]
- m_sina = re.match(r'https?://video.sina.com.cn/v/b/(\d+)-\d+.html', player_url)
+ m_sina = re.match(r'https?://video\.sina\.com\.cn/v/b/(\d+)-\d+\.html',
+ player_url)
if m_sina is not None:
self.to_screen('Sina video detected')
sina_id = m_sina.group(1)
player_url = 'http://you.video.sina.com.cn/swf/quotePlayer.swf?vid=%s' % sina_id
return self.url_result(player_url)
-
diff --git a/youtube_dl/postprocessor/__init__.py b/youtube_dl/postprocessor/__init__.py
@@ -1,5 +1,7 @@
+from .atomicparsley import AtomicParsleyPP
from .ffmpeg import (
+ FFmpegAudioFixPP,
FFmpegMergerPP,
FFmpegMetadataPP,
FFmpegVideoConvertor,
@@ -9,6 +11,8 @@ from .ffmpeg import (
from .xattrpp import XAttrMetadataPP
__all__ = [
+ 'AtomicParsleyPP',
+ 'FFmpegAudioFixPP',
'FFmpegMergerPP',
'FFmpegMetadataPP',
'FFmpegVideoConvertor',
diff --git a/youtube_dl/postprocessor/atomicparsley.py b/youtube_dl/postprocessor/atomicparsley.py
@@ -0,0 +1,56 @@
+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals
+
+
+import os
+import subprocess
+
+from .common import PostProcessor
+
+from ..utils import (
+ check_executable,
+ compat_urlretrieve,
+ encodeFilename,
+ PostProcessingError,
+ prepend_extension,
+ shell_quote
+)
+
+
+class AtomicParsleyPPError(PostProcessingError):
+ pass
+
+
+class AtomicParsleyPP(PostProcessor):
+ def run(self, info):
+ if not check_executable('AtomicParsley', ['-v']):
+ raise AtomicParsleyPPError('AtomicParsley was not found. Please install.')
+
+ filename = info['filepath']
+ temp_filename = prepend_extension(filename, 'temp')
+ temp_thumbnail = prepend_extension(filename, 'thumb')
+
+ if not info.get('thumbnail'):
+ raise AtomicParsleyPPError('Thumbnail was not found. Nothing to do.')
+
+ compat_urlretrieve(info['thumbnail'], temp_thumbnail)
+
+ cmd = ['AtomicParsley', filename, '--artwork', temp_thumbnail, '-o', temp_filename]
+
+ self._downloader.to_screen('[atomicparsley] Adding thumbnail to "%s"' % filename)
+
+ if self._downloader.params.get('verbose', False):
+ self._downloader.to_screen('[debug] AtomicParsley command line: %s' % shell_quote(cmd))
+
+ p = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+ stdout, stderr = p.communicate()
+
+ if p.returncode != 0:
+ msg = stderr.decode('utf-8', 'replace').strip()
+ raise AtomicParsleyPPError(msg)
+
+ os.remove(encodeFilename(filename))
+ os.remove(encodeFilename(temp_thumbnail))
+ os.rename(encodeFilename(temp_filename), encodeFilename(filename))
+
+ return True, info
diff --git a/youtube_dl/postprocessor/ffmpeg.py b/youtube_dl/postprocessor/ffmpeg.py
@@ -464,7 +464,11 @@ class FFmpegMetadataPP(FFmpegPostProcessor):
filename = info['filepath']
temp_filename = prepend_extension(filename, 'temp')
- options = ['-c', 'copy']
+ if info['ext'] == u'm4a':
+ options = ['-vn', '-acodec', 'copy']
+ else:
+ options = ['-c', 'copy']
+
for (name, value) in metadata.items():
options.extend(['-metadata', '%s=%s' % (name, value)])
@@ -483,3 +487,17 @@ class FFmpegMergerPP(FFmpegPostProcessor):
self.run_ffmpeg_multiple_files(info['__files_to_merge'], filename, args)
return True, info
+
+class FFmpegAudioFixPP(FFmpegPostProcessor):
+ def run(self, info):
+ filename = info['filepath']
+ temp_filename = prepend_extension(filename, 'temp')
+
+ options = ['-vn', '-acodec', 'copy']
+ self._downloader.to_screen(u'[ffmpeg] Fixing audio file "%s"' % filename)
+ self.run_ffmpeg(filename, temp_filename, options)
+
+ os.remove(encodeFilename(filename))
+ os.rename(encodeFilename(temp_filename), encodeFilename(filename))
+
+ return True, info
