]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/rte.py
Rapsys Git / youtubedl / blobdiff
? search:
summary | shortlog | log | commit | commitdiff | tree
raw | inline | side by side
Initiate new release.
[youtubedl] / youtube_dl / extractor / rte.py
diff --git a/youtube_dl/extractor/rte.py b/youtube_dl/extractor/rte.py
index ebe563ebb89e86e28a6bf55669cd066aca44d851..1fbc72915ea9b45cfcfe4f17e418a5958d44093c 100644 (file)
--- a/youtube_dl/extractor/rte.py
+++ b/youtube_dl/extractor/rte.py
@@ -4,69 +4,126 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_HTTPError
 from ..utils import (
     float_or_none,
     parse_iso8601,
 from ..utils import (
     float_or_none,
     parse_iso8601,
+    str_or_none,
+    try_get,
     unescapeHTML,
     unescapeHTML,
+    url_or_none,
+    ExtractorError,
 )
 
 
 )
 
 
-class RteIE(InfoExtractor):
+class RteBaseIE(InfoExtractor):
+    def _real_extract(self, url):
+        item_id = self._match_id(url)
+
+        info_dict = {}
+        formats = []
+
+        ENDPOINTS = (
+            'https://feeds.rasset.ie/rteavgen/player/playlist?type=iptv&format=json&showId=',
+            'http://www.rte.ie/rteavgen/getplaylist/?type=web&format=json&id=',
+        )
+
+        for num, ep_url in enumerate(ENDPOINTS, start=1):
+            try:
+                data = self._download_json(ep_url + item_id, item_id)
+            except ExtractorError as ee:
+                if num < len(ENDPOINTS) or formats:
+                    continue
+                if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404:
+                    error_info = self._parse_json(ee.cause.read().decode(), item_id, fatal=False)
+                    if error_info:
+                        raise ExtractorError(
+                            '%s said: %s' % (self.IE_NAME, error_info['message']),
+                            expected=True)
+                raise
+
+            # NB the string values in the JSON are stored using XML escaping(!)
+            show = try_get(data, lambda x: x['shows'][0], dict)
+            if not show:
+                continue
+
+            if not info_dict:
+                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'))
+                info_dict = {
+                    'id': item_id,
+                    'title': title,
+                    'description': description,
+                    'thumbnail': thumbnail,
+                    'timestamp': timestamp,
+                    'duration': duration,
+                }
+
+            mg = try_get(show, lambda x: x['media:group'][0], dict)
+            if not mg:
+                continue
+
+            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))
+
+            mg_rte_server = str_or_none(mg.get('rte:server'))
+            mg_url = str_or_none(mg.get('url'))
+            if mg_rte_server and mg_url:
+                hds_url = url_or_none(mg_rte_server + mg_url)
+                if hds_url:
+                    formats.extend(self._extract_f4m_formats(
+                        hds_url, item_id, f4m_id='hds', fatal=False))
+
+        self._sort_formats(formats)
+
+        info_dict['formats'] = formats
+        return info_dict
+
+
+class RteIE(RteBaseIE):
     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/',
     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/',
+        'md5': '4a76eb3396d98f697e6e8110563d2604',
         'info_dict': {
             'id': '10478715',
         'info_dict': {
             'id': '10478715',
-            'ext': 'flv',
-            'title': 'Watch iWitness  online',
-            'thumbnail': 're:^https?://.*\.jpg$',
-            'description': 'iWitness : The spirit of Ireland, one voice and one minute at a time.',
+            'ext': 'mp4',
+            'title': 'iWitness',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'description': 'The spirit of Ireland, one voice and one minute at a time.',
             'duration': 60.046,
             'duration': 60.046,
+            'upload_date': '20151012',
+            'timestamp': 1444694160,
         },
         },
-        '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):
+
+class RteRadioIE(RteBaseIE):
     IE_NAME = 'rte:radio'
     IE_DESC = 'Raidió Teilifís Éireann radio'
     # Radioplayer URLs have two distinct specifier formats,
     IE_NAME = 'rte:radio'
     IE_DESC = 'Raidió Teilifís Éireann radio'
     # Radioplayer URLs have two distinct specifier formats,
@@ -79,19 +136,17 @@ class RteRadioIE(InfoExtractor):
     _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:',
     _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:',
+        'md5': 'c79ccb2c195998440065456b69760411',
         'info_dict': {
             'id': '10507902',
             'ext': 'mp4',
             'title': 'Gloria',
         'info_dict': {
             'id': '10507902',
             'ext': 'mp4',
             'title': 'Gloria',
-            'thumbnail': 're:^https?://.*\.jpg$',
+            'thumbnail': r're:^https?://.*\.jpg$',
             'description': 'md5:9ce124a7fb41559ec68f06387cabddf0',
             'timestamp': 1451203200,
             'upload_date': '20151227',
             'duration': 7230.0,
         },
             '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_',
     }, {
         # New-style player URL; RTMPE formats only
         'url': 'http://rte.ie/radio/utils/radioplayer/rteradioweb.html#!rii=b16_3250678_8861_06-04-2012_',
@@ -99,67 +154,14 @@ class RteRadioIE(InfoExtractor):
             'id': '3250678',
             'ext': 'flv',
             'title': 'The Lyric Concert with Paul Herriott',
             'id': '3250678',
             'ext': 'flv',
             'title': 'The Lyric Concert with Paul Herriott',
-            'thumbnail': 're:^https?://.*\.jpg$',
+            'thumbnail': r're:^https?://.*\.jpg$',
             'description': '',
             'timestamp': 1333742400,
             'upload_date': '20120406',
             'duration': 7199.016,
         },
         'params': {
             'description': '',
             'timestamp': 1333742400,
             'upload_date': '20120406',
             'duration': 7199.016,
         },
         'params': {
-            'skip_download': 'f4m fails with --test atm'
-        }
+            # rtmp download
+            'skip_download': True,
+        },
     }]
     }]
-
-    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,
-        }
Backup clone of https://github.com/ytdl-org unavailable since RIAA fucking raid...