Merge pull request #8876 from remitamine/html5_media

[extractor/common] add helper method to extract html5 media entries

2 files changed, 94 insertions, 0 deletions

diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 816baa424..df546da27 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -55,6 +55,8 @@ from ..utils import (
     update_Request,
     update_url_query,
     parse_m3u8_attributes,
+    extract_attributes,
+    parse_codecs,
 )
 
 
@@ -1635,6 +1637,62 @@ class InfoExtractor(object):
                         self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
         return formats
 
+    def _parse_html5_media_entries(self, base_url, webpage):
+        def absolute_url(video_url):
+            return compat_urlparse.urljoin(base_url, video_url)
+
+        def parse_content_type(content_type):
+            if not content_type:
+                return {}
+            ctr = re.search(r'(?P<mimetype>[^/]+/[^;]+)(?:;\s*codecs="?(?P<codecs>[^"]+))?', content_type)
+            if ctr:
+                mimetype, codecs = ctr.groups()
+                f = parse_codecs(codecs)
+                f['ext'] = mimetype2ext(mimetype)
+                return f
+            return {}
+
+        entries = []
+        for media_tag, media_type, media_content in re.findall(r'(?s)(<(?P<tag>video|audio)[^>]*>)(.*?)</(?P=tag)>', webpage):
+            media_info = {
+                'formats': [],
+                'subtitles': {},
+            }
+            media_attributes = extract_attributes(media_tag)
+            src = media_attributes.get('src')
+            if src:
+                media_info['formats'].append({
+                    'url': absolute_url(src),
+                    'vcodec': 'none' if media_type == 'audio' else None,
+                })
+            media_info['thumbnail'] = media_attributes.get('poster')
+            if media_content:
+                for source_tag in re.findall(r'<source[^>]+>', media_content):
+                    source_attributes = extract_attributes(source_tag)
+                    src = source_attributes.get('src')
+                    if not src:
+                        continue
+                    f = parse_content_type(source_attributes.get('type'))
+                    f.update({
+                        'url': absolute_url(src),
+                        'vcodec': 'none' if media_type == 'audio' else None,
+                    })
+                    media_info['formats'].append(f)
+                for track_tag in re.findall(r'<track[^>]+>', media_content):
+                    track_attributes = extract_attributes(track_tag)
+                    kind = track_attributes.get('kind')
+                    if not kind or kind == 'subtitles':
+                        src = track_attributes.get('src')
+                        if not src:
+                            continue
+                        lang = track_attributes.get('srclang') or track_attributes.get('lang') or track_attributes.get('label')
+                        media_info['subtitles'].setdefault(lang, []).append({
+                            'url': absolute_url(src),
+                        })
+            if media_info['formats']:
+                entries.append(media_info)
+        return entries
+
     def _live_title(self, name):
         """ Generate the title for a live video """
         now = datetime.datetime.now()
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 3498697b6..4c1d0d526 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -2126,6 +2126,42 @@ def mimetype2ext(mt):
     }.get(res, res)
 
 
+def parse_codecs(codecs_str):
+    # http://tools.ietf.org/html/rfc6381
+    if not codecs_str:
+        return {}
+    splited_codecs = list(filter(None, map(
+        lambda str: str.strip(), codecs_str.strip().strip(',').split(','))))
+    vcodec, acodec = None, None
+    for full_codec in splited_codecs:
+        codec = full_codec.split('.')[0]
+        if codec in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v'):
+            if not vcodec:
+                vcodec = full_codec
+        elif codec in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac'):
+            if not acodec:
+                acodec = full_codec
+        else:
+            write_string('WARNING: Unknown codec %s' % full_codec, sys.stderr)
+    if not vcodec and not acodec:
+        if len(splited_codecs) == 2:
+            return {
+                'vcodec': vcodec,
+                'acodec': acodec,
+            }
+        elif len(splited_codecs) == 1:
+            return {
+                'vcodec': 'none',
+                'acodec': vcodec,
+            }
+    else:
+        return {
+            'vcodec': vcodec or 'none',
+            'acodec': acodec or 'none',
+        }
+    return {}
+
+
 def urlhandle_detect_ext(url_handle):
     getheader = url_handle.headers.get