nhyatt/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp synced 2025-07-03 20:42:53 -05:00
Code Issues Projects Releases Wiki Activity

[downloader/mhtml] Add new downloader (#343)

Browse Source 
This downloader is intended to be used for streams that consist of a
timed sequence of stand-alone images, such as slideshows or thumbnail
streams

This can be used for implementing:

https://github.com/ytdl-org/youtube-dl/issues/4974#issue-58006762
https://github.com/ytdl-org/youtube-dl/issues/4540#issuecomment-69574231
https://github.com/ytdl-org/youtube-dl/pull/11185#issuecomment-335554239

https://github.com/ytdl-org/youtube-dl/issues/9868
https://github.com/ytdl-org/youtube-dl/pull/14951


Authored by: fstirlitz
This commit is contained in:
felix
2021-05-23 18:34:49 +02:00
committed by pukkandan
parent 4d85fbbdbb
commit cdb19aa4c2
Notes: pukkandan 2021-06-13 22:41:29 +05:30 
This also adds extracting storyboards from DASH manifest as mhtml
6 changed files with 248 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

29
yt_dlp/extractor/common.py
View File

@ -2126,6 +2126,7 @@ class InfoExtractor(object):
format_id.append(str(format_index))
f = {
'format_id': '-'.join(format_id),
'format_note': name,
'format_index': format_index,
'url': manifest_url,
'manifest_url': m3u8_url,
@ -2637,7 +2638,7 @@ class InfoExtractor(object):
mime_type = representation_attrib['mimeType']
content_type = representation_attrib.get('contentType', mime_type.split('/')[0])
if content_type in ('video', 'audio', 'text'):
if content_type in ('video', 'audio', 'text') or mime_type == 'image/jpeg':
base_url = ''
for element in (representation, adaptation_set, period, mpd_doc):
base_url_e = element.find(_add_ns('BaseURL'))
@ -2654,9 +2655,15 @@ class InfoExtractor(object):
url_el = representation.find(_add_ns('BaseURL'))
filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el is not None else None)
bandwidth = int_or_none(representation_attrib.get('bandwidth'))
if representation_id is not None:
format_id = representation_id
else:
format_id = content_type
if mpd_id:
format_id = mpd_id + '-' + format_id
if content_type in ('video', 'audio'):
f = {
'format_id': '%s-%s' % (mpd_id, representation_id) if mpd_id else representation_id,
'format_id': format_id,
'manifest_url': mpd_url,
'ext': mimetype2ext(mime_type),
'width': int_or_none(representation_attrib.get('width')),
@ -2676,6 +2683,17 @@ class InfoExtractor(object):
'manifest_url': mpd_url,
'filesize': filesize,
}
elif mime_type == 'image/jpeg':
# See test case in VikiIE
# https://www.viki.com/videos/1175236v-choosing-spouse-by-lottery-episode-1
f = {
'format_id': format_id,
'ext': 'mhtml',
'manifest_url': mpd_url,
'format_note': 'DASH storyboards (jpeg)',
'acodec': 'none',
'vcodec': 'none',
}
representation_ms_info = extract_multisegment_info(representation, adaption_set_ms_info)
def prepare_template(template_name, identifiers):
@ -2694,7 +2712,8 @@ class InfoExtractor(object):
t += c
# Next, $...$ templates are translated to their
# %(...) counterparts to be used with % operator
t = t.replace('$RepresentationID$', representation_id)
if representation_id is not None:
t = t.replace('$RepresentationID$', representation_id)
t = re.sub(r'\$(%s)\$' % '|'.join(identifiers), r'%(\1)d', t)
t = re.sub(r'\$(%s)%%([^$]+)\$' % '|'.join(identifiers), r'%(\1)\2', t)
t.replace('$$', '$')
@ -2811,7 +2830,7 @@ class InfoExtractor(object):
'url': mpd_url or base_url,
'fragment_base_url': base_url,
'fragments': [],
'protocol': 'http_dash_segments',
'protocol': 'http_dash_segments' if mime_type != 'image/jpeg' else 'mhtml',
})
if 'initialization_url' in representation_ms_info:
initialization_url = representation_ms_info['initialization_url']
@ -2822,7 +2841,7 @@ class InfoExtractor(object):
else:
# Assuming direct URL to unfragmented media.
f['url'] = base_url
if content_type in ('video', 'audio'):
if content_type in ('video', 'audio') or mime_type == 'image/jpeg':
formats.append(f)
elif content_type == 'text':
subtitles.setdefault(lang or 'und', []).append(f)