[go90] Improve extraction

- add metadata for 'series', 'episode', 'season', 'season_id',
  'season_number', and 'episode_number'
- integrate series title into the title
- extract subtitles  (fallback to `vtt` if the subtitle file type
  detection fails as that is the most likely extension)
This commit is contained in:
Jeremie J. Jarosh 2017-04-15 10:12:08 -05:00
parent 3dfceb286c
commit 334f41e0d8

View file

@ -5,6 +5,7 @@
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
determine_ext,
int_or_none, int_or_none,
parse_iso8601, parse_iso8601,
) )
@ -18,7 +19,7 @@ class Go90IE(InfoExtractor):
'info_dict': { 'info_dict': {
'id': '84BUqjLpf9D', 'id': '84BUqjLpf9D',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Inside The Utah Coalition Against Pornography Convention', 'title': 'Daily VICE - Inside The Utah Coalition Against Pornography Convention',
'description': 'VICE\'s Karley Sciortino meets with activists who discuss the state\'s strong anti-porn stance. Then, VICE Sports explains NFL contracts.', 'description': 'VICE\'s Karley Sciortino meets with activists who discuss the state\'s strong anti-porn stance. Then, VICE Sports explains NFL contracts.',
'timestamp': 1491868800, 'timestamp': 1491868800,
'upload_date': '20170411', 'upload_date': '20170411',
@ -32,11 +33,28 @@ def _real_extract(self, url):
video_id, headers={ video_id, headers={
'Content-Type': 'application/json; charset=utf-8', 'Content-Type': 'application/json; charset=utf-8',
}, data=b'{"client":"web","device_type":"pc"}') }, data=b'{"client":"web","device_type":"pc"}')
title = video_data['title']
main_video_asset = video_data['main_video_asset'] main_video_asset = video_data['main_video_asset']
episode_number = int_or_none(video_data.get('episode_number'))
series = None
season = None
season_id = None
season_number = None
for metadata in video_data.get('__children', {}).get('Item', {}).values():
if metadata.get('type') == 'show':
series = metadata.get('title')
elif metadata.get('type') == 'season':
season = metadata.get('title')
season_id = metadata.get('id')
season_number = int_or_none(metadata.get('season_number'))
title = episode = video_data.get('title') or series
if series and series != title:
title = '%s - %s' % (series, title)
thumbnails = [] thumbnails = []
formats = [] formats = []
subtitles = {}
for asset in video_data.get('assets'): for asset in video_data.get('assets'):
if asset.get('id') == main_video_asset: if asset.get('id') == main_video_asset:
for source in asset.get('sources', []): for source in asset.get('sources', []):
@ -70,6 +88,15 @@ def _real_extract(self, url):
'height': int_or_none(source.get('height')), 'height': int_or_none(source.get('height')),
'tbr': int_or_none(source.get('bitrate')), 'tbr': int_or_none(source.get('bitrate')),
}) })
for caption in asset.get('caption_metadata', []):
caption_url = caption.get('source_url')
if not caption_url:
continue
subtitles.setdefault(caption.get('language', 'en'), []).append({
'url': caption_url,
'ext': determine_ext(caption_url, 'vtt'),
})
elif asset.get('type') == 'image': elif asset.get('type') == 'image':
asset_location = asset.get('location') asset_location = asset.get('location')
if not asset_location: if not asset_location:
@ -89,4 +116,11 @@ def _real_extract(self, url):
'description': video_data.get('short_description'), 'description': video_data.get('short_description'),
'like_count': int_or_none(video_data.get('like_count')), 'like_count': int_or_none(video_data.get('like_count')),
'timestamp': parse_iso8601(video_data.get('released_at')), 'timestamp': parse_iso8601(video_data.get('released_at')),
'series': series,
'episode': episode,
'season': season,
'season_id': season_id,
'season_number': season_number,
'episode_number': episode_number,
'subtitles': subtitles,
} }