- # coding: utf-8
- from __future__ import unicode_literals
-
- import re
-
- from .common import InfoExtractor
- from ..utils import (
- float_or_none,
- parse_iso8601,
- unescapeHTML,
- )
-
-
- class RteIE(InfoExtractor):
- IE_NAME = 'rte'
- IE_DESC = 'Raidió Teilifís Éireann TV'
- _VALID_URL = r'https?://(?:www\.)?rte\.ie/player/[^/]{2,3}/show/[^/]+/(?P<id>[0-9]+)'
- _TEST = {
- 'url': 'http://www.rte.ie/player/ie/show/iwitness-862/10478715/',
- 'info_dict': {
- 'id': '10478715',
- 'ext': 'flv',
- 'title': 'Watch iWitness online',
- 'thumbnail': 're:^https?://.*\.jpg$',
- 'description': 'iWitness : The spirit of Ireland, one voice and one minute at a time.',
- 'duration': 60.046,
- },
- 'params': {
- 'skip_download': 'f4m fails with --test atm'
- }
- }
-
- def _real_extract(self, url):
- video_id = self._match_id(url)
- webpage = self._download_webpage(url, video_id)
-
- title = self._og_search_title(webpage)
- description = self._html_search_meta('description', webpage, 'description')
- duration = float_or_none(self._html_search_meta(
- 'duration', webpage, 'duration', fatal=False), 1000)
-
- thumbnail = None
- thumbnail_meta = self._html_search_meta('thumbnail', webpage)
- if thumbnail_meta:
- thumbnail_id = self._search_regex(
- r'uri:irus:(.+)', thumbnail_meta,
- 'thumbnail id', fatal=False)
- if thumbnail_id:
- thumbnail = 'http://img.rasset.ie/%s.jpg' % thumbnail_id
-
- feeds_url = self._html_search_meta('feeds-prefix', webpage, 'feeds url') + video_id
- json_string = self._download_json(feeds_url, video_id)
-
- # f4m_url = server + relative_url
- f4m_url = json_string['shows'][0]['media:group'][0]['rte:server'] + json_string['shows'][0]['media:group'][0]['url']
- f4m_formats = self._extract_f4m_formats(f4m_url, video_id)
- self._sort_formats(f4m_formats)
-
- return {
- 'id': video_id,
- 'title': title,
- 'formats': f4m_formats,
- 'description': description,
- 'thumbnail': thumbnail,
- 'duration': duration,
- }
-
-
- class RteRadioIE(InfoExtractor):
- IE_NAME = 'rte:radio'
- IE_DESC = 'Raidió Teilifís Éireann radio'
- # Radioplayer URLs have two distinct specifier formats,
- # the old format #!rii=<channel_id>:<id>:<playable_item_id>:<date>:
- # the new format #!rii=b<channel_id>_<id>_<playable_item_id>_<date>_
- # where the IDs are int/empty, the date is DD-MM-YYYY, and the specifier may be truncated.
- # An <id> uniquely defines an individual recording, and is the only part we require.
- _VALID_URL = r'https?://(?:www\.)?rte\.ie/radio/utils/radioplayer/rteradioweb\.html#!rii=(?:b?[0-9]*)(?:%3A|:|%5F|_)(?P<id>[0-9]+)'
-
- _TESTS = [{
- # Old-style player URL; HLS and RTMPE formats
- 'url': 'http://www.rte.ie/radio/utils/radioplayer/rteradioweb.html#!rii=16:10507902:2414:27-12-2015:',
- 'info_dict': {
- 'id': '10507902',
- 'ext': 'mp4',
- 'title': 'Gloria',
- 'thumbnail': 're:^https?://.*\.jpg$',
- 'description': 'md5:9ce124a7fb41559ec68f06387cabddf0',
- 'timestamp': 1451203200,
- 'upload_date': '20151227',
- 'duration': 7230.0,
- },
- 'params': {
- 'skip_download': 'f4m fails with --test atm'
- }
- }, {
- # New-style player URL; RTMPE formats only
- 'url': 'http://rte.ie/radio/utils/radioplayer/rteradioweb.html#!rii=b16_3250678_8861_06-04-2012_',
- 'info_dict': {
- 'id': '3250678',
- 'ext': 'flv',
- 'title': 'The Lyric Concert with Paul Herriott',
- 'thumbnail': 're:^https?://.*\.jpg$',
- 'description': '',
- 'timestamp': 1333742400,
- 'upload_date': '20120406',
- 'duration': 7199.016,
- },
- 'params': {
- 'skip_download': 'f4m fails with --test atm'
- }
- }]
-
- def _real_extract(self, url):
- item_id = self._match_id(url)
-
- json_string = self._download_json(
- 'http://www.rte.ie/rteavgen/getplaylist/?type=web&format=json&id=' + item_id,
- item_id)
-
- # NB the string values in the JSON are stored using XML escaping(!)
- show = json_string['shows'][0]
- title = unescapeHTML(show['title'])
- description = unescapeHTML(show.get('description'))
- thumbnail = show.get('thumbnail')
- duration = float_or_none(show.get('duration'), 1000)
- timestamp = parse_iso8601(show.get('published'))
-
- mg = show['media:group'][0]
-
- formats = []
-
- if mg.get('url'):
- m = re.match(r'(?P<url>rtmpe?://[^/]+)/(?P<app>.+)/(?P<playpath>mp4:.*)', mg['url'])
- if m:
- m = m.groupdict()
- formats.append({
- 'url': m['url'] + '/' + m['app'],
- 'app': m['app'],
- 'play_path': m['playpath'],
- 'player_url': url,
- 'ext': 'flv',
- 'format_id': 'rtmp',
- })
-
- if mg.get('hls_server') and mg.get('hls_url'):
- formats.extend(self._extract_m3u8_formats(
- mg['hls_server'] + mg['hls_url'], item_id, 'mp4',
- entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
-
- if mg.get('hds_server') and mg.get('hds_url'):
- formats.extend(self._extract_f4m_formats(
- mg['hds_server'] + mg['hds_url'], item_id,
- f4m_id='hds', fatal=False))
-
- self._sort_formats(formats)
-
- return {
- 'id': item_id,
- 'title': title,
- 'description': description,
- 'thumbnail': thumbnail,
- 'timestamp': timestamp,
- 'duration': duration,
- 'formats': formats,
- }
|