[tv4] improve extraction(closes #11698)

- remove check for requires_subscription
- extract more formats
- extract subtitles
This commit is contained in:
Remita Amine 2017-01-13 10:19:53 +01:00
parent c4251b9aaa
commit 1f393a3241

View file

@ -4,11 +4,10 @@
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import compat_str from ..compat import compat_str
from ..utils import ( from ..utils import (
ExtractorError,
int_or_none, int_or_none,
parse_iso8601, parse_iso8601,
try_get, try_get,
update_url_query, determine_ext,
) )
@ -28,7 +27,7 @@ class TV4IE(InfoExtractor):
_TESTS = [ _TESTS = [
{ {
'url': 'http://www.tv4.se/kalla-fakta/klipp/kalla-fakta-5-english-subtitles-2491650', 'url': 'http://www.tv4.se/kalla-fakta/klipp/kalla-fakta-5-english-subtitles-2491650',
'md5': '909d6454b87b10a25aa04c4bdd416a9b', 'md5': 'cb837212f342d77cec06e6dad190e96d',
'info_dict': { 'info_dict': {
'id': '2491650', 'id': '2491650',
'ext': 'mp4', 'ext': 'mp4',
@ -40,7 +39,7 @@ class TV4IE(InfoExtractor):
}, },
{ {
'url': 'http://www.tv4play.se/iframe/video/3054113', 'url': 'http://www.tv4play.se/iframe/video/3054113',
'md5': '77f851c55139ffe0ebd41b6a5552489b', 'md5': 'cb837212f342d77cec06e6dad190e96d',
'info_dict': { 'info_dict': {
'id': '3054113', 'id': '3054113',
'ext': 'mp4', 'ext': 'mp4',
@ -75,11 +74,10 @@ def _real_extract(self, url):
# If is_geo_restricted is true, it doesn't necessarily mean we can't download it # If is_geo_restricted is true, it doesn't necessarily mean we can't download it
if info.get('is_geo_restricted'): if info.get('is_geo_restricted'):
self.report_warning('This content might not be available in your country due to licensing restrictions.') self.report_warning('This content might not be available in your country due to licensing restrictions.')
if info.get('requires_subscription'):
raise ExtractorError('This content requires subscription.', expected=True)
title = info['title'] title = info['title']
subtitles = {}
formats = [] formats = []
# http formats are linked with unresolvable host # http formats are linked with unresolvable host
for kind in ('hls', ''): for kind in ('hls', ''):
@ -87,26 +85,41 @@ def _real_extract(self, url):
'https://prima.tv4play.se/api/web/asset/%s/play.json' % video_id, 'https://prima.tv4play.se/api/web/asset/%s/play.json' % video_id,
video_id, 'Downloading sources JSON', query={ video_id, 'Downloading sources JSON', query={
'protocol': kind, 'protocol': kind,
'videoFormat': 'MP4+WEBVTTS+WEBVTT', 'videoFormat': 'MP4+WEBVTT',
}) })
item = try_get(data, lambda x: x['playback']['items']['item'], dict) items = try_get(data, lambda x: x['playback']['items']['item'])
manifest_url = item.get('url') if not items:
if not isinstance(manifest_url, compat_str):
continue continue
if kind == 'hls': if isinstance(items, dict):
formats.extend(self._extract_m3u8_formats( items = [items]
manifest_url, video_id, 'mp4', entry_protocol='m3u8_native', for item in items:
m3u8_id=kind, fatal=False)) manifest_url = item.get('url')
else: if not isinstance(manifest_url, compat_str):
formats.extend(self._extract_f4m_formats( continue
update_url_query(manifest_url, {'hdcore': '3.8.0'}), ext = determine_ext(manifest_url)
video_id, f4m_id='hds', fatal=False)) if ext == 'm3u8':
formats.extend(self._extract_m3u8_formats(
manifest_url, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id=kind, fatal=False))
elif ext == 'f4m':
formats.extend(self._extract_akamai_formats(
manifest_url, video_id, {
'hls': 'tv4play-i.akamaihd.net',
}))
elif ext == 'webvtt':
subtitles = self._merge_subtitles(
subtitles, {
'sv': [{
'url': manifest_url,
'ext': 'vtt',
}]})
self._sort_formats(formats) self._sort_formats(formats)
return { return {
'id': video_id, 'id': video_id,
'title': title, 'title': title,
'formats': formats, 'formats': formats,
'subtitles': subtitles,
'description': info.get('description'), 'description': info.get('description'),
'timestamp': parse_iso8601(info.get('broadcast_date_time')), 'timestamp': parse_iso8601(info.get('broadcast_date_time')),
'duration': int_or_none(info.get('duration')), 'duration': int_or_none(info.get('duration')),