diff options
Diffstat (limited to 'hypervideo_dl/extractor/mediasite.py')
| -rw-r--r-- | hypervideo_dl/extractor/mediasite.py | 69 |
1 files changed, 60 insertions, 9 deletions
diff --git a/hypervideo_dl/extractor/mediasite.py b/hypervideo_dl/extractor/mediasite.py index d6eb157..ace86c2 100644 --- a/hypervideo_dl/extractor/mediasite.py +++ b/hypervideo_dl/extractor/mediasite.py @@ -26,7 +26,7 @@ _ID_RE = r'(?:[0-9a-f]{32,34}|[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0 class MediasiteIE(InfoExtractor): - _VALID_URL = r'(?xi)https?://[^/]+/Mediasite/(?:Play|Showcase/(?:default|livebroadcast)/Presentation)/(?P<id>%s)(?P<query>\?[^#]+|)' % _ID_RE + _VALID_URL = r'(?xi)https?://[^/]+/Mediasite/(?:Play|Showcase/[^/#?]+/Presentation)/(?P<id>%s)(?P<query>\?[^#]+|)' % _ID_RE _TESTS = [ { 'url': 'https://hitsmediaweb.h-its.org/mediasite/Play/2db6c271681e4f199af3c60d1f82869b1d', @@ -122,9 +122,55 @@ class MediasiteIE(InfoExtractor): r'(?xi)<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:(?:https?:)?//[^/]+)?/Mediasite/Play/%s(?:\?.*?)?)\1' % _ID_RE, webpage)] + def __extract_slides(self, *, stream_id, snum, Stream, duration, images): + slide_base_url = Stream['SlideBaseUrl'] + + fname_template = Stream['SlideImageFileNameTemplate'] + if fname_template != 'slide_{0:D4}.jpg': + self.report_warning('Unusual slide file name template; report a bug if slide downloading fails') + fname_template = re.sub(r'\{0:D([0-9]+)\}', r'{0:0\1}', fname_template) + + fragments = [] + for i, slide in enumerate(Stream['Slides']): + if i == 0: + if slide['Time'] > 0: + default_slide = images.get('DefaultSlide') + if default_slide is None: + default_slide = images.get('DefaultStreamImage') + if default_slide is not None: + default_slide = default_slide['ImageFilename'] + if default_slide is not None: + fragments.append({ + 'path': default_slide, + 'duration': slide['Time'] / 1000, + }) + + next_time = try_get(None, [ + lambda _: Stream['Slides'][i + 1]['Time'], + lambda _: duration, + lambda _: slide['Time'], + ], expected_type=(int, float)) + + fragments.append({ + 'path': fname_template.format(slide.get('Number', i + 1)), + 'duration': (next_time - slide['Time']) / 1000 + }) + + return { + 'format_id': '%s-%u.slides' % (stream_id, snum), + 'ext': 'mhtml', + 'url': slide_base_url, + 'protocol': 'mhtml', + 'acodec': 'none', + 'vcodec': 'none', + 'format_note': 'Slides', + 'fragments': fragments, + 'fragment_base_url': slide_base_url, + } + def _real_extract(self, url): url, data = unsmuggle_url(url, {}) - mobj = re.match(self._VALID_URL, url) + mobj = self._match_valid_url(url) resource_id = mobj.group('id') query = mobj.group('query') @@ -198,15 +244,20 @@ class MediasiteIE(InfoExtractor): 'ext': mimetype2ext(VideoUrl.get('MimeType')), }) - # TODO: if Stream['HasSlideContent']: - # synthesise an MJPEG video stream '%s-%u.slides' % (stream_type, snum) - # from Stream['Slides'] - # this will require writing a custom downloader... + if Stream.get('HasSlideContent', False): + images = player_options['PlayerLayoutOptions']['Images'] + stream_formats.append(self.__extract_slides( + stream_id=stream_id, + snum=snum, + Stream=Stream, + duration=presentation.get('Duration'), + images=images, + )) # disprefer 'secondary' streams if stream_type != 0: for fmt in stream_formats: - fmt['preference'] = -1 + fmt['quality'] = -10 thumbnail_url = Stream.get('ThumbnailUrl') if thumbnail_url: @@ -276,7 +327,7 @@ class MediasiteCatalogIE(InfoExtractor): }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) + mobj = self._match_valid_url(url) mediasite_url = mobj.group('url') catalog_id = mobj.group('catalog_id') current_folder_id = mobj.group('current_folder_id') or catalog_id @@ -352,7 +403,7 @@ class MediasiteNamedCatalogIE(InfoExtractor): }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) + mobj = self._match_valid_url(url) mediasite_url = mobj.group('url') catalog_name = mobj.group('catalog_name') |