Mirroring
/
yt-dlc
mirror of https://github.com/blackjack4494/yt-dlc
1
0
Fork 0
Code Issues Releases Wiki Activity

[nrk] Rework extractor (Closes #9470)

Browse Source
pull/2/head
Sergey M․ 9 years ago
parent 0df79d552a
commit d8d540cf0d
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D
1 changed files with 197 additions and 240 deletions
Show Stats Download Patch File Download Diff File

395
youtube_dl/extractor/nrk.py
Unescape Escape View File

@ -4,169 +4,151 @@ from __future__ import unicode_literals
import re import re
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import ( from ..compat import compat_urllib_parse_unquote
compat_urlparse,
compat_urllib_parse_unquote,
)
from ..utils import ( from ..utils import (
determine_ext,
ExtractorError, ExtractorError,
float_or_none, int_or_none,
parse_age_limit,
parse_duration, parse_duration,
unified_strdate,
) )
class NRKIE(InfoExtractor): class NRKBaseIE(InfoExtractor):
_VALID_URL = r'(?:nrk:|https?://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)' def _extract_formats(self, manifest_url, video_id, fatal=True):
return self._extract_f4m_formats(
_TESTS = [ manifest_url + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81',
{ video_id, f4m_id='hds', fatal=fatal)
'url': 'http://www.nrk.no/video/PS*150533',
# MD5 is unstable
'info_dict': {
'id': '150533',
'ext': 'flv',
'title': 'Dompap og andre fugler i Piip-Show',
'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
'duration': 263,
}
},
{
'url': 'http://www.nrk.no/video/PS*154915',
# MD5 is unstable
'info_dict': {
'id': '154915',
'ext': 'flv',
'title': 'Slik høres internett ut når du er blind',
'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
'duration': 20,
}
},
]
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
data = self._download_json( data = self._download_json(
'http://v8.psapi.nrk.no/mediaelement/%s' % video_id, 'http://%s/mediaelement/%s' % (self._API_HOST, video_id),
video_id, 'Downloading media JSON') video_id, 'Downloading mediaelement JSON')
media_url = data.get('mediaUrl') title = data.get('fullTitle') or data.get('mainTitle') or data['title']
video_id = data.get('id') or video_id
if not media_url: entries = []
if data['usageRights']['isGeoBlocked']:
raise ExtractorError(
'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
expected=True)
if determine_ext(media_url) == 'f4m': media_assets = data.get('mediaAssets')
formats = self._extract_f4m_formats( if media_assets and isinstance(media_assets, list):
media_url + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81', video_id, f4m_id='hds') def video_id_and_title(idx):
return ((video_id, title) if len(media_assets) == 1
else ('%s-%d' % (video_id, idx), '%s (Part %d)' % (title, idx)))
for num, asset in enumerate(media_assets, 1):
asset_url = asset.get('url')
if not asset_url:
continue
formats = self._extract_formats(asset_url, video_id, fatal=False)
if not formats:
continue
self._sort_formats(formats) self._sort_formats(formats)
else: entry_id, entry_title = video_id_and_title(num)
formats = [{ duration = parse_duration(asset.get('duration'))
'url': media_url, subtitles = {}
'ext': 'flv', for subtitle in ('webVtt', 'timedText'):
}] subtitle_url = asset.get('%sSubtitlesUrl' % subtitle)
if subtitle_url:
subtitles.setdefault('no', []).append({'url': subtitle_url})
entries.append({
'id': asset.get('carrierId') or entry_id,
'title': entry_title,
'duration': duration,
'subtitles': subtitles,
'formats': formats,
})
if not entries:
media_url = data.get('mediaUrl')
if media_url:
formats = self._extract_formats(media_url, video_id)
self._sort_formats(formats)
duration = parse_duration(data.get('duration')) duration = parse_duration(data.get('duration'))
entries = [{
images = data.get('images')
if images:
thumbnails = images['webImages']
thumbnails.sort(key=lambda image: image['pixelWidth'])
thumbnail = thumbnails[-1]['imageUrl']
else:
thumbnail = None
return {
'id': video_id, 'id': video_id,
'title': data['title'], 'title': title,
'description': data['description'],
'duration': duration, 'duration': duration,
'thumbnail': thumbnail,
'formats': formats, 'formats': formats,
}
class NRKPlaylistIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P<id>[^/]+)'
_TESTS = [{
'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
'info_dict': {
'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
'title': 'Gjenopplev den historiske solformørkelsen',
'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
},
'playlist_count': 2,
}, {
'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
'info_dict': {
'id': 'rivertonprisen-til-karin-fossum-1.12266449',
'title': 'Rivertonprisen til Karin Fossum',
'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
},
'playlist_count': 5,
}] }]
def _real_extract(self, url): if not entries:
playlist_id = self._match_id(url) if data.get('usageRights', {}).get('isGeoBlocked'):
raise ExtractorError(
'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
expected=True)
webpage = self._download_webpage(url, playlist_id) conviva = data.get('convivaStatistics') or {}
series = conviva.get('seriesName') or data.get('seriesTitle')
episode = conviva.get('episodeName') or data.get('episodeNumberOrDate')
entries = [ thumbnails = None
self.url_result('nrk:%s' % video_id, 'NRK') images = data.get('images')
for video_id in re.findall( if images and isinstance(images, dict):
r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"', web_images = images.get('webImages')
webpage) if isinstance(web_images, list):
] thumbnails = [{
'url': image['imageUrl'],
'width': int_or_none(image.get('width')),
'height': int_or_none(image.get('height')),
} for image in web_images if image.get('imageUrl')]
description = data.get('description')
common_info = {
'description': description,
'series': series,
'episode': episode,
'age_limit': parse_age_limit(data.get('legalAge')),
'thumbnails': thumbnails,
}
playlist_title = self._og_search_title(webpage) vcodec = 'none' if data.get('mediaType') == 'Audio' else None
playlist_description = self._og_search_description(webpage)
return self.playlist_result( # TODO: extract chapters when https://github.com/rg3/youtube-dl/pull/9409 is merged
entries, playlist_id, playlist_title, playlist_description)
for entry in entries:
entry.update(common_info)
for f in entry['formats']:
f['vcodec'] = vcodec
return self.playlist_result(entries, video_id, title, description)
class NRKSkoleIE(InfoExtractor):
IE_DESC = 'NRK Skole'
_VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/klippdetalj?.*\btopic=(?P<id>[^/?#&]+)'
class NRKIE(NRKBaseIE):
_VALID_URL = r'(?:nrk:|https?://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)'
_API_HOST = 'v8.psapi.nrk.no'
_TESTS = [{ _TESTS = [{
'url': 'http://nrk.no/skole/klippdetalj?topic=nrk:klipp/616532', # video
'md5': '04cd85877cc1913bce73c5d28a47e00f', 'url': 'http://www.nrk.no/video/PS*150533',
# MD5 is unstable
'info_dict': { 'info_dict': {
'id': '6021', 'id': '150533',
'ext': 'flv', 'ext': 'flv',
'title': 'Genetikk og eneggede tvillinger', 'title': 'Dompap og andre fugler i Piip-Show',
'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d', 'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
'duration': 399, 'duration': 263,
}, }
}, {
'url': 'http://www.nrk.no/skole/klippdetalj?topic=nrk%3Aklipp%2F616532#embed',
'only_matching': True,
}, { }, {
'url': 'http://www.nrk.no/skole/klippdetalj?topic=urn:x-mediadb:21379', # audio
'only_matching': True, 'url': 'http://www.nrk.no/video/PS*154915',
# MD5 is unstable
'info_dict': {
'id': '154915',
'ext': 'flv',
'title': 'Slik høres internett ut når du er blind',
'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
'duration': 20,
}
}] }]
def _real_extract(self, url):
video_id = compat_urllib_parse_unquote(self._match_id(url))
webpage = self._download_webpage(url, video_id)
nrk_id = self._search_regex(r'data-nrk-id=["\'](\d+)', webpage, 'nrk id')
return self.url_result('nrk:%s' % nrk_id)
class NRKTVIE(InfoExtractor): class NRKTVIE(NRKBaseIE):
IE_DESC = 'NRK TV and NRK Radio' IE_DESC = 'NRK TV and NRK Radio'
_VALID_URL = r'(?P<baseurl>https?://(?:tv|radio)\.nrk(?:super)?\.no/)(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?' _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
_API_HOST = 'psapi-we.nrk.no'
_TESTS = [ _TESTS = [{
{
'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014', 'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
'info_dict': { 'info_dict': {
'id': 'MUHH48000314', 'id': 'MUHH48000314',
@ -180,8 +162,7 @@ class NRKTVIE(InfoExtractor):
# m3u8 download # m3u8 download
'skip_download': True, 'skip_download': True,
}, },
}, }, {
{
'url': 'https://tv.nrk.no/program/mdfp15000514', 'url': 'https://tv.nrk.no/program/mdfp15000514',
'info_dict': { 'info_dict': {
'id': 'mdfp15000514', 'id': 'mdfp15000514',
@ -195,8 +176,7 @@ class NRKTVIE(InfoExtractor):
# m3u8 download # m3u8 download
'skip_download': True, 'skip_download': True,
}, },
}, }, {
{
# single playlist video # single playlist video
'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2', 'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
'md5': 'adbd1dbd813edaf532b0a253780719c2', 'md5': 'adbd1dbd813edaf532b0a253780719c2',
@ -208,11 +188,9 @@ class NRKTVIE(InfoExtractor):
'upload_date': '20150106', 'upload_date': '20150106',
}, },
'skip': 'Only works from Norway', 'skip': 'Only works from Norway',
}, }, {
{
'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015', 'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
'playlist': [ 'playlist': [{
{
'md5': '9480285eff92d64f06e02a5367970a7a', 'md5': '9480285eff92d64f06e02a5367970a7a',
'info_dict': { 'info_dict': {
'id': 'MSPO40010515-part1', 'id': 'MSPO40010515-part1',
@ -221,8 +199,7 @@ class NRKTVIE(InfoExtractor):
'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
'upload_date': '20150106', 'upload_date': '20150106',
}, },
}, }, {
{
'md5': 'adbd1dbd813edaf532b0a253780719c2', 'md5': 'adbd1dbd813edaf532b0a253780719c2',
'info_dict': { 'info_dict': {
'id': 'MSPO40010515-part2', 'id': 'MSPO40010515-part2',
@ -231,107 +208,87 @@ class NRKTVIE(InfoExtractor):
'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
'upload_date': '20150106', 'upload_date': '20150106',
}, },
}, }],
],
'info_dict': { 'info_dict': {
'id': 'MSPO40010515', 'id': 'MSPO40010515',
'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn', 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn',
'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26', 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
'upload_date': '20150106', 'upload_date': '20150106',
'duration': 6947.5199999999995, 'duration': 6947.52,
}, },
'skip': 'Only works from Norway', 'skip': 'Only works from Norway',
}, }, {
{
'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#', 'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
'only_matching': True, 'only_matching': True,
} }]
]
def _extract_f4m(self, manifest_url, video_id):
return self._extract_f4m_formats( class NRKPlaylistIE(InfoExtractor):
manifest_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', video_id, f4m_id='hds') _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P<id>[^/]+)'
_TESTS = [{
'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
'info_dict': {
'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
'title': 'Gjenopplev den historiske solformørkelsen',
'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
},
'playlist_count': 2,
}, {
'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
'info_dict': {
'id': 'rivertonprisen-til-karin-fossum-1.12266449',
'title': 'Rivertonprisen til Karin Fossum',
'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
},
'playlist_count': 5,
}]
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) playlist_id = self._match_id(url)
video_id = mobj.group('id')
part_id = mobj.group('part_id')
base_url = mobj.group('baseurl')
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, playlist_id)
title = self._html_search_meta( entries = [
'title', webpage, 'title') self.url_result('nrk:%s' % video_id, 'NRK')
description = self._html_search_meta( for video_id in re.findall(
'description', webpage, 'description') r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"',
webpage)
thumbnail = self._html_search_regex( ]
r'data-posterimage="([^"]+)"',
webpage, 'thumbnail', fatal=False)
upload_date = unified_strdate(self._html_search_meta(
'rightsfrom', webpage, 'upload date', fatal=False))
duration = float_or_none(self._html_search_regex(
r'data-duration="([^"]+)"',
webpage, 'duration', fatal=False))
# playlist
parts = re.findall(
r'<a href="#del=(\d+)"[^>]+data-argument="([^"]+)">([^<]+)</a>', webpage)
if parts:
entries = []
for current_part_id, stream_url, part_title in parts:
if part_id and current_part_id != part_id:
continue
video_part_id = '%s-part%s' % (video_id, current_part_id)
formats = self._extract_f4m(stream_url, video_part_id)
entries.append({
'id': video_part_id,
'title': part_title,
'description': description,
'thumbnail': thumbnail,
'upload_date': upload_date,
'formats': formats,
})
if part_id:
if entries:
return entries[0]
else:
playlist = self.playlist_result(entries, video_id, title, description)
playlist.update({
'thumbnail': thumbnail,
'upload_date': upload_date,
'duration': duration,
})
return playlist
formats = [] playlist_title = self._og_search_title(webpage)
playlist_description = self._og_search_description(webpage)
f4m_url = re.search(r'data-media="([^"]+)"', webpage) return self.playlist_result(
if f4m_url: entries, playlist_id, playlist_title, playlist_description)
formats.extend(self._extract_f4m(f4m_url.group(1), video_id))
m3u8_url = re.search(r'data-hls-media="([^"]+)"', webpage)
if m3u8_url:
formats.extend(self._extract_m3u8_formats(m3u8_url.group(1), video_id, 'mp4', m3u8_id='hls'))
self._sort_formats(formats)
subtitles_url = self._html_search_regex( class NRKSkoleIE(InfoExtractor):
r'data-subtitlesurl\s*=\s*(["\'])(?P<url>.+?)\1', IE_DESC = 'NRK Skole'
webpage, 'subtitle URL', default=None, group='url') _VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/klippdetalj?.*\btopic=(?P<id>[^/?#&]+)'
subtitles = {}
if subtitles_url: _TESTS = [{
subtitles['no'] = [{ 'url': 'http://nrk.no/skole/klippdetalj?topic=nrk:klipp/616532',
'ext': 'ttml', 'md5': '04cd85877cc1913bce73c5d28a47e00f',
'url': compat_urlparse.urljoin(base_url, subtitles_url), 'info_dict': {
'id': '6021',
'ext': 'flv',
'title': 'Genetikk og eneggede tvillinger',
'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d',
'duration': 399,
},
}, {
'url': 'http://www.nrk.no/skole/klippdetalj?topic=nrk%3Aklipp%2F616532#embed',
'only_matching': True,
}, {
'url': 'http://www.nrk.no/skole/klippdetalj?topic=urn:x-mediadb:21379',
'only_matching': True,
}] }]
return { def _real_extract(self, url):
'id': video_id, video_id = compat_urllib_parse_unquote(self._match_id(url))
'title': title,
'description': description, webpage = self._download_webpage(url, video_id)
'thumbnail': thumbnail,
'upload_date': upload_date, nrk_id = self._search_regex(r'data-nrk-id=["\'](\d+)', webpage, 'nrk id')
'duration': duration, return self.url_result('nrk:%s' % nrk_id)
'formats': formats,
'subtitles': subtitles,
}

Write Preview
Loading…
Cancel
Save