[liveleak] extract multiple videos
This commit is contained in:
aeph6Ee0
parent
31208a07c2
commit
0b0fb6fd42
@ -53,9 +53,26 @@ class LiveLeakIE(InfoExtractor):
|
||||
}
|
||||
}]
|
||||
|
||||
video_count = 0
|
||||
def _video_count(self):
|
||||
self.video_count += 1
|
||||
if self.video_count == 1:
|
||||
return ''
|
||||
else:
|
||||
return '-' + str(self.video_count-1)
|
||||
|
||||
# Removing '.h264_*.mp4' gives the raw video, which is essentially
|
||||
# the same video without the LiveLeak logo at the top (see
|
||||
# https://github.com/rg3/youtube-dl/pull/4768)
|
||||
def _get_orig_video_url(self, url):
|
||||
return re.sub(r'\.h264_.+?\.mp4', '', url)
|
||||
|
||||
def _real_extract(self, url):
|
||||
video_id = self._match_id(url)
|
||||
webpage = self._download_webpage(url, video_id)
|
||||
|
||||
entries = list() # collect all found videos
|
||||
|
||||
page_id = self._match_id(url)
|
||||
webpage = self._download_webpage(url, page_id)
|
||||
|
||||
video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
|
||||
video_description = self._og_search_description(webpage)
|
||||
@ -65,55 +82,94 @@ class LiveLeakIE(InfoExtractor):
|
||||
r'you confirm that you are ([0-9]+) years and over.',
|
||||
webpage, 'age limit', default=None))
|
||||
|
||||
# extracts native video #1 (single video, maybe multiple formats)
|
||||
sources_raw = self._search_regex(
|
||||
r'(?s)sources:\s*(\[.*?\]),', webpage, 'video URLs', default=None)
|
||||
if sources_raw is None:
|
||||
alt_source = self._search_regex(
|
||||
r'(file: ".*?"),', webpage, 'video URL', default=None)
|
||||
if alt_source:
|
||||
sources_raw = '[{ %s}]' % alt_source
|
||||
else:
|
||||
# Maybe an embed?
|
||||
embed_url = self._search_regex(
|
||||
r'<iframe[^>]+src="(http://www.prochan.com/embed\?[^"]+)"',
|
||||
webpage, 'embed URL')
|
||||
return {
|
||||
'_type': 'url_transparent',
|
||||
'url': embed_url,
|
||||
'id': video_id,
|
||||
'title': video_title,
|
||||
'description': video_description,
|
||||
'uploader': video_uploader,
|
||||
'age_limit': age_limit,
|
||||
}
|
||||
if sources_raw:
|
||||
sources_json = re.sub(r'\s([a-z]+):\s', r'"\1": ', sources_raw)
|
||||
sources = json.loads(sources_json)
|
||||
|
||||
sources_json = re.sub(r'\s([a-z]+):\s', r'"\1": ', sources_raw)
|
||||
sources = json.loads(sources_json)
|
||||
formats = [{
|
||||
'format_id': '%s' % i,
|
||||
'format_note': s.get('label'),
|
||||
'url': s['file'],
|
||||
} for i, s in enumerate(sources)]
|
||||
for i, s in enumerate(sources):
|
||||
orig_url = self._get_orig_video_url(s['file'])
|
||||
if s['file'] != orig_url:
|
||||
formats.append({
|
||||
'format_id': 'original-%s' % i,
|
||||
'format_note': s.get('label'),
|
||||
'url': orig_url,
|
||||
'preference': 1,
|
||||
})
|
||||
self._sort_formats(formats)
|
||||
|
||||
formats = [{
|
||||
'format_id': '%s' % i,
|
||||
'format_note': s.get('label'),
|
||||
'url': s['file'],
|
||||
} for i, s in enumerate(sources)]
|
||||
for i, s in enumerate(sources):
|
||||
# Removing '.h264_*.mp4' gives the raw video, which is essentially
|
||||
# the same video without the LiveLeak logo at the top (see
|
||||
# https://github.com/rg3/youtube-dl/pull/4768)
|
||||
orig_url = re.sub(r'\.h264_.+?\.mp4', '', s['file'])
|
||||
if s['file'] != orig_url:
|
||||
entries.append({
|
||||
'id': page_id,
|
||||
'title': video_title,
|
||||
'description': video_description,
|
||||
'uploader': video_uploader,
|
||||
'formats': formats,
|
||||
'age_limit': age_limit,
|
||||
})
|
||||
|
||||
# extracts native videos #2 (maybe multiple videos, single format)
|
||||
sources = re.findall(r'(?s)jwplayer\("file_[0-9a-f]+"\).+?file: "(.*?)"', webpage)
|
||||
for url in sources:
|
||||
formats = [{
|
||||
'format_id': '0',
|
||||
'format_note': 'standard quality (with logo)',
|
||||
'url': url,
|
||||
}]
|
||||
orig_url = self._get_orig_video_url(url)
|
||||
if orig_url != url:
|
||||
formats.append({
|
||||
'format_id': 'original-%s' % i,
|
||||
'format_note': s.get('label'),
|
||||
'format_id': '1',
|
||||
'format_note': 'high quality (no logo)',
|
||||
'url': orig_url,
|
||||
'preference': 1,
|
||||
})
|
||||
self._sort_formats(formats)
|
||||
entries.append({
|
||||
'id': page_id + self._video_count(),
|
||||
'title': video_title,
|
||||
'description': video_description,
|
||||
'uploader': video_uploader,
|
||||
'formats': formats,
|
||||
'age_limit': age_limit,
|
||||
})
|
||||
|
||||
return {
|
||||
'id': video_id,
|
||||
'title': video_title,
|
||||
'description': video_description,
|
||||
'uploader': video_uploader,
|
||||
'formats': formats,
|
||||
'age_limit': age_limit,
|
||||
}
|
||||
|
||||
# collect embedded videos:
|
||||
embed_urls = list()
|
||||
|
||||
# prochan.com:
|
||||
embed_prochan = (re.findall(
|
||||
r'<iframe[^>]+src="(http://www.prochan.com/embed\?[^"]+)"',
|
||||
webpage))
|
||||
if len(embed_prochan) > 0:
|
||||
for embed in embed_prochan:
|
||||
embed_urls.append(embed)
|
||||
|
||||
# add all collected embed urls
|
||||
for embed_url in embed_urls:
|
||||
entries.append({
|
||||
'_type': 'url_transparent',
|
||||
'id': page_id + self._video_count(),
|
||||
'url': embed_url,
|
||||
'title': video_title,
|
||||
'description': video_description,
|
||||
'uploader': video_uploader,
|
||||
'age_limit': age_limit,
|
||||
})
|
||||
|
||||
if len(entries) == 0:
|
||||
raise ExtractorError('No videos found')
|
||||
if len(entries) == 1:
|
||||
return entries[0]
|
||||
else:
|
||||
return {
|
||||
'_type': 'multi_video',
|
||||
'id': page_id,
|
||||
'entries': entries,
|
||||
}
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user