[filmon] improve extraction

This commit is contained in:
Remita Amine 2017-02-03 10:15:03 +01:00
parent a0758dfa1a
commit 4ce3407d08
2 changed files with 139 additions and 102 deletions

View File

@ -287,7 +287,10 @@
FC2EmbedIE, FC2EmbedIE,
) )
from .fczenit import FczenitIE from .fczenit import FczenitIE
from .filmon import FilmOnIE, FilmOnVODIE from .filmon import (
FilmOnIE,
FilmOnChannelIE,
)
from .firstpost import FirstpostIE from .firstpost import FirstpostIE
from .firsttv import FirstTVIE from .firsttv import FirstTVIE
from .fivemin import FiveMinIE from .fivemin import FiveMinIE

View File

@ -2,74 +2,21 @@
from __future__ import unicode_literals from __future__ import unicode_literals
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import qualities from ..compat import (
from ..compat import compat_urllib_request compat_str,
compat_HTTPError,
)
_QUALITY = qualities(('low', 'high')) from ..utils import (
qualities,
strip_or_none,
int_or_none,
ExtractorError,
)
class FilmOnIE(InfoExtractor): class FilmOnIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?filmon\.com/(?:tv|channel)/(?P<id>[a-z0-9-]+)' IE_NAME = 'filmon'
_TESTS = [{ _VALID_URL = r'(?:https?://(?:www\.)?filmon\.com/vod/view/|filmon:)(?P<id>\d+)'
'url': 'https://www.filmon.com/channel/filmon-sports',
'only_matching': True,
}, {
'url': 'https://www.filmon.com/tv/2894',
'only_matching': True,
}]
def _real_extract(self, url):
channel_id = self._match_id(url)
request = compat_urllib_request.Request('https://www.filmon.com/channel/%s' % (channel_id))
request.add_header('X-Requested-With', 'XMLHttpRequest')
channel_info = self._download_json(request, channel_id)
now_playing = channel_info['now_playing']
thumbnails = []
for thumb in now_playing.get('images', ()):
if thumb['type'] != '2':
continue
thumbnails.append({
'url': thumb['url'],
'width': int(thumb['width']),
'height': int(thumb['height']),
})
formats = []
for stream in channel_info['streams']:
formats.append({
'format_id': str(stream['id']),
# this is an m3u8 stream, but we are deliberately not using _extract_m3u8_formats
# because 0) it doesn't have bitrate variants anyway, and 1) the ids generated
# by that method are highly unstable (because the bitrate is variable)
'url': stream['url'],
'resolution': stream['name'],
'format_note': 'expires after %u seconds' % int(stream['watch-timeout']),
'ext': 'mp4',
'quality': _QUALITY(stream['quality']),
'preference': int(stream['watch-timeout']),
})
self._sort_formats(formats)
return {
'id': str(channel_info['id']),
'display_id': channel_info['alias'],
'formats': formats,
# XXX: use the channel description (channel_info['description'])?
'uploader_id': channel_info['alias'],
'uploader': channel_info['title'], # XXX: kinda stretching it...
'title': now_playing.get('programme_name') or channel_info['title'],
'description': now_playing.get('programme_description'),
'thumbnails': thumbnails,
'is_live': True,
}
class FilmOnVODIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?filmon\.com/vod/view/(?P<id>\d+)'
_TESTS = [{ _TESTS = [{
'url': 'https://www.filmon.com/vod/view/24869-0-plan-9-from-outer-space', 'url': 'https://www.filmon.com/vod/view/24869-0-plan-9-from-outer-space',
'info_dict': { 'info_dict': {
@ -83,62 +30,149 @@ class FilmOnVODIE(InfoExtractor):
'info_dict': { 'info_dict': {
'id': '2825', 'id': '2825',
'title': 'Popeye Series 1', 'title': 'Popeye Series 1',
'description': 'The original series of Popeye.',
}, },
'playlist_count': 8, 'playlist_mincount': 8,
}] }]
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
result = self._download_json('https://www.filmon.com/api/vod/movie?id=%s' % (video_id), video_id) try:
if result['code'] != 200: response = self._download_json(
raise ExtractorError('FilmOn said: %s' % (result['reason']), expected=True) 'https://www.filmon.com/api/vod/movie?id=%s' % video_id,
video_id)['response']
except ExtractorError as e:
if isinstance(e.cause, compat_HTTPError):
errmsg = self._parse_json(e.cause.read().decode(), video_id)['reason']
raise ExtractorError('%s said: %s' % (self.IE_NAME, errmsg), expected=True)
raise
response = result['response'] title = response['title']
description = strip_or_none(response.get('description'))
if response.get('episodes'): if response.get('type_id') == 1:
return { entries = [self.url_result('filmon:' + episode_id) for episode_id in response.get('episodes', [])]
'_type': 'playlist', return self.playlist_result(entries, video_id, title, description)
'id': video_id,
'title': response['title'],
'entries': [{
'_type': 'url',
'url': 'https://www.filmon.com/vod/view/%s' % (ep),
} for ep in response['episodes']]
}
QUALITY = qualities(('low', 'high'))
formats = [] formats = []
for (id, stream) in response['streams'].items(): for format_id, stream in response.get('streams', {}).items():
stream_url = stream.get('url')
if not stream_url:
continue
formats.append({ formats.append({
'format_id': id, 'format_id': format_id,
'url': stream['url'], 'url': stream_url,
'resolution': stream['name'],
'format_note': 'expires after %u seconds' % int(stream['watch-timeout']),
'ext': 'mp4', 'ext': 'mp4',
'quality': _QUALITY(stream['quality']), 'quality': QUALITY(stream.get('quality')),
'preference': int(stream['watch-timeout']), 'protocol': 'm3u8_native',
}) })
self._sort_formats(formats) self._sort_formats(formats)
poster = response['poster'] thumbnails = []
thumbnails = [{ poster = response.get('poster', {})
'id': 'poster', thumbs = poster.get('thumbs', {})
'url': poster['url'], thumbs['poster'] = poster
'width': poster['width'], for thumb_id, thumb in thumbs.items():
'height': poster['height'], thumb_url = thumb.get('url')
}] if not thumb_url:
for (id, thumb) in poster['thumbs'].items(): continue
thumbnails.append({ thumbnails.append({
'id': id, 'id': thumb_id,
'url': thumb['url'], 'url': thumb_url,
'width': thumb['width'], 'width': int_or_none(thumb.get('width')),
'height': thumb['height'], 'height': int_or_none(thumb.get('height')),
}) })
return { return {
'id': video_id, 'id': video_id,
'title': response['title'], 'title': title,
'formats': formats, 'formats': formats,
'description': response['description'], 'description': description,
'thumbnails': thumbnails, 'thumbnails': thumbnails,
} }
class FilmOnChannelIE(InfoExtractor):
IE_NAME = 'filmon:channel'
_VALID_URL = r'https?://(?:www\.)?filmon\.com/(?:tv|channel)/(?P<id>[a-z0-9-]+)'
_TESTS = [{
# VOD
'url': 'http://www.filmon.com/tv/sports-haters',
'info_dict': {
'id': '4190',
'ext': 'mp4',
'title': 'Sports Haters',
'description': 'md5:dabcb4c1d9cfc77085612f1a85f8275d',
},
}, {
# LIVE
'url': 'https://www.filmon.com/channel/filmon-sports',
'only_matching': True,
}, {
'url': 'https://www.filmon.com/tv/2894',
'only_matching': True,
}]
_THUMBNAIL_RES = [
('logo', 56, 28),
('big_logo', 106, 106),
('extra_big_logo', 300, 300),
]
def _real_extract(self, url):
channel_id = self._match_id(url)
try:
channel_data = self._download_json(
'http://www.filmon.com/api-v2/channel/' + channel_id, channel_id)['data']
except ExtractorError as e:
if isinstance(e.cause, compat_HTTPError):
errmsg = self._parse_json(e.cause.read().decode(), channel_id)['message']
raise ExtractorError('%s said: %s' % (self.IE_NAME, errmsg), expected=True)
raise
channel_id = compat_str(channel_data['id'])
is_live = not channel_data.get('is_vod') and not channel_data.get('is_vox')
title = channel_data['title']
QUALITY = qualities(('low', 'high'))
formats = []
for stream in channel_data.get('streams', []):
stream_url = stream.get('url')
if not stream_url:
continue
if not is_live:
formats.extend(self._extract_wowza_formats(
stream_url, channel_id, skip_protocols=['dash', 'rtmp', 'rtsp']))
continue
quality = stream.get('quality')
formats.append({
'format_id': quality,
# this is an m3u8 stream, but we are deliberately not using _extract_m3u8_formats
# because it doesn't have bitrate variants anyway
'url': stream_url,
'ext': 'mp4',
'quality': QUALITY(quality),
})
self._sort_formats(formats)
thumbnails = []
for name, width, height in self._THUMBNAIL_RES:
thumbnails.append({
'id': name,
'url': 'http://static.filmon.com/assets/channels/%s/%s.png' % (channel_id, name),
'width': width,
'height': height,
})
return {
'id': channel_id,
'display_id': channel_data.get('alias'),
'title': self._live_title(title) if is_live else title,
'description': channel_data.get('description'),
'thumbnails': thumbnails,
'formats': formats,
'is_live': is_live,
}