From 17b598d30cae2c287f3556f874ddf0fc5d028aec Mon Sep 17 00:00:00 2001
From: Yen Chi Hsuan <yan12125@gmail.com>
Date: Sat, 30 Jan 2016 21:05:55 +0800
Subject: [common] _parse_dash_manifest() from youtube.py
---
youtube_dl/extractor/common.py | 52 ++++++++++++++++++++++++++++++++++++++++++
1 file changed, 52 insertions(+)
(limited to 'youtube_dl/extractor/common.py')
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index b3d57dfce..7ad255672 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -1330,6 +1330,58 @@ class InfoExtractor(object):
})
return entries
+ def _parse_dash_manifest(self, video_id, dash_doc, fatal=True):
+ formats = []
+ for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
+ mime_type = a.attrib.get('mimeType')
+ for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
+ url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
+ if url_el is None:
+ continue
+ if mime_type == 'text/vtt':
+ # TODO implement WebVTT downloading
+ pass
+ elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
+ segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
+ format_id = r.attrib['id']
+ video_url = url_el.text
+ filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
+ f = {
+ 'format_id': format_id,
+ 'url': video_url,
+ 'width': int_or_none(r.attrib.get('width')),
+ 'height': int_or_none(r.attrib.get('height')),
+ 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
+ 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
+ 'filesize': filesize,
+ 'fps': int_or_none(r.attrib.get('frameRate')),
+ }
+ if segment_list is not None:
+ f.update({
+ 'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'],
+ 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
+ 'protocol': 'http_dash_segments',
+ })
+ try:
+ existing_format = next(
+ fo for fo in formats
+ if fo['format_id'] == format_id)
+ except StopIteration:
+ full_info = self._formats.get(format_id, {}).copy()
+ full_info.update(f)
+ codecs = r.attrib.get('codecs')
+ if codecs:
+ if full_info.get('acodec') == 'none':
+ full_info['vcodec'] = codecs
+ elif full_info.get('vcodec') == 'none':
+ full_info['acodec'] = codecs
+ formats.append(full_info)
+ else:
+ existing_format.update(f)
+ else:
+ self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
+ return formats
+
def _live_title(self, name):
""" Generate the title for a live video """
now = datetime.datetime.now()
--
cgit v1.2.3
From b323e1707d8f058b88a5f15f3418b31cf969399d Mon Sep 17 00:00:00 2001
From: Yen Chi Hsuan <yan12125@gmail.com>
Date: Sat, 30 Jan 2016 21:27:43 +0800
Subject: [common] Modify _parse_dash_manifest for use in Facebook
---
youtube_dl/extractor/common.py | 29 +++++++++++++++++------------
1 file changed, 17 insertions(+), 12 deletions(-)
(limited to 'youtube_dl/extractor/common.py')
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 7ad255672..83628a68f 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -1330,22 +1330,24 @@ class InfoExtractor(object):
})
return entries
- def _parse_dash_manifest(self, video_id, dash_doc, fatal=True):
+ def _parse_dash_manifest(self, video_id, dash_doc, default_ns='urn:mpeg:DASH:schema:MPD:2011', formats_dict={}, fatal=True):
+ def _add_ns(tag):
+ return '{%s}%s' % (default_ns, tag)
+
formats = []
- for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
+ for a in dash_doc.findall('.//' + _add_ns('AdaptationSet')):
mime_type = a.attrib.get('mimeType')
- for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
- url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
- if url_el is None:
- continue
+ for r in a.findall(_add_ns('Representation')):
+ mime_type = r.attrib.get('mimeType') or mime_type
+ url_el = r.find(_add_ns('BaseURL'))
if mime_type == 'text/vtt':
# TODO implement WebVTT downloading
pass
elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
- segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
+ segment_list = r.find(_add_ns('SegmentList'))
format_id = r.attrib['id']
- video_url = url_el.text
- filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
+ video_url = url_el.text if url_el else None
+ filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el else None)
f = {
'format_id': format_id,
'url': video_url,
@@ -1357,17 +1359,20 @@ class InfoExtractor(object):
'fps': int_or_none(r.attrib.get('frameRate')),
}
if segment_list is not None:
+ initialization_url = segment_list.find(_add_ns('Initialization')).attrib['sourceURL']
f.update({
- 'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'],
- 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
+ 'initialization_url': initialization_url,
+ 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall(_add_ns('SegmentURL'))],
'protocol': 'http_dash_segments',
})
+ if not f.get('url'):
+ f['url'] = initialization_url
try:
existing_format = next(
fo for fo in formats
if fo['format_id'] == format_id)
except StopIteration:
- full_info = self._formats.get(format_id, {}).copy()
+ full_info = formats_dict.get(format_id, {}).copy()
full_info.update(f)
codecs = r.attrib.get('codecs')
if codecs:
--
cgit v1.2.3
From 5ea1eb78f5e8d9b2ede35504dc3b999f5f89bbaa Mon Sep 17 00:00:00 2001
From: Yen Chi Hsuan <yan12125@gmail.com>
Date: Sat, 30 Jan 2016 21:36:01 +0800
Subject: [common] Fix for youtube
---
youtube_dl/extractor/common.py | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
(limited to 'youtube_dl/extractor/common.py')
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 83628a68f..243db71dc 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -1346,8 +1346,8 @@ class InfoExtractor(object):
elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
segment_list = r.find(_add_ns('SegmentList'))
format_id = r.attrib['id']
- video_url = url_el.text if url_el else None
- filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el else None)
+ video_url = url_el.text if url_el is not None else None
+ filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el is not None else None)
f = {
'format_id': format_id,
'url': video_url,
--
cgit v1.2.3
From df374b52228e8a083d045f9bc56847e418ef452e Mon Sep 17 00:00:00 2001
From: Yen Chi Hsuan <yan12125@gmail.com>
Date: Sat, 30 Jan 2016 21:42:27 +0800
Subject: [common] Prefer the manifest than formats_dict in determining codecs
---
youtube_dl/extractor/common.py | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
(limited to 'youtube_dl/extractor/common.py')
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 243db71dc..f1313ef04 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -1376,10 +1376,15 @@ class InfoExtractor(object):
full_info.update(f)
codecs = r.attrib.get('codecs')
if codecs:
- if full_info.get('acodec') == 'none':
- full_info['vcodec'] = codecs
- elif full_info.get('vcodec') == 'none':
- full_info['acodec'] = codecs
+ if mime_type.startswith('video/'):
+ vcodec, acodec = codecs, 'none'
+ else: # mime_type.startswith('audio/')
+ vcodec, acodec = 'none', codecs
+
+ full_info.update({
+ 'vcodec': vcodec,
+ 'acodec': acodec,
+ })
formats.append(full_info)
else:
existing_format.update(f)
--
cgit v1.2.3
From 16f38a699f7c6d2820720d704a2373ba28c568b1 Mon Sep 17 00:00:00 2001
From: Yen Chi Hsuan <yan12125@gmail.com>
Date: Sat, 30 Jan 2016 22:40:13 +0800
Subject: [common] Rename to namespace
For consistency with _parse_smil_*
---
youtube_dl/extractor/common.py | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
(limited to 'youtube_dl/extractor/common.py')
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index f1313ef04..a05efec9e 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -1330,9 +1330,9 @@ class InfoExtractor(object):
})
return entries
- def _parse_dash_manifest(self, video_id, dash_doc, default_ns='urn:mpeg:DASH:schema:MPD:2011', formats_dict={}, fatal=True):
- def _add_ns(tag):
- return '{%s}%s' % (default_ns, tag)
+ def _parse_dash_manifest(self, video_id, dash_doc, namespace=None, formats_dict={}, fatal=True):
+ def _add_ns(path):
+ return self._xpath_ns(path, namespace)
formats = []
for a in dash_doc.findall('.//' + _add_ns('AdaptationSet')):
--
cgit v1.2.3
From c94678957fbe4483b2b7c8b3e6824cb7a215d42d Mon Sep 17 00:00:00 2001
From: Yen Chi Hsuan <yan12125@gmail.com>
Date: Sat, 30 Jan 2016 22:45:16 +0800
Subject: [common] Remove unused arguments
---
youtube_dl/extractor/common.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
(limited to 'youtube_dl/extractor/common.py')
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index a05efec9e..5a2b7a721 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -1330,7 +1330,7 @@ class InfoExtractor(object):
})
return entries
- def _parse_dash_manifest(self, video_id, dash_doc, namespace=None, formats_dict={}, fatal=True):
+ def _parse_dash_manifest(self, dash_doc, namespace=None, formats_dict={}):
def _add_ns(path):
return self._xpath_ns(path, namespace)
--
cgit v1.2.3
From 2d2fa82d172a10a49fb5449fa35bc409de778f05 Mon Sep 17 00:00:00 2001
From: Yen Chi Hsuan <yan12125@gmail.com>
Date: Sat, 30 Jan 2016 22:52:23 +0800
Subject: [common] Add _extract_dash_manifest_formats
---
youtube_dl/extractor/common.py | 15 +++++++++++++++
1 file changed, 15 insertions(+)
(limited to 'youtube_dl/extractor/common.py')
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 5a2b7a721..199a04d1c 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -1330,6 +1330,21 @@ class InfoExtractor(object):
})
return entries
+ def _download_dash_manifest(self, dash_manifest_url, video_id, fatal=True):
+ return self._download_xml(
+ dash_manifest_url, video_id,
+ note='Downloading DASH manifest',
+ errnote='Could not download DASH manifest',
+ fatal=fatal)
+
+ def _extract_dash_manifest_formats(self, dash_manifest_url, video_id, fatal=True, namespace=None, formats_dict={}):
+ dash_doc = self._download_dash_manifest(dash_manifest_url, video_id, fatal)
+ if dash_doc is False:
+ return []
+
+ return self._parse_dash_manifest(
+ dash_doc, namespace=namespace, formats_dict=formats_dict)
+
def _parse_dash_manifest(self, dash_doc, namespace=None, formats_dict={}):
def _add_ns(path):
return self._xpath_ns(path, namespace)
--
cgit v1.2.3