From: bpfoley <bpfoley>
Date: Wed, 30 Dec 2015 21:50:26 +0000 (+0000)
Subject: [extractor/rte.py] Add support for RTE radio player
X-Git-Url: https://git.cielonegro.org/gitweb.cgi?a=commitdiff_plain;h=896c7a23cd9fbfc589f26e9a7bbada38a325e2ad;p=youtube-dl.git

[extractor/rte.py] Add support for RTE radio player

While here, stop RteIE changing filename extensions to .mp4. The files
saved are .flv containers with h264 video.
---

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index fb7151443..fe3c32874 100644
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -556,7 +556,7 @@ from .ro220 import Ro220IE
 from .rottentomatoes import RottenTomatoesIE
 from .roxwel import RoxwelIE
 from .rtbf import RTBFIE
-from .rte import RteIE
+from .rte import RteIE, RteRadioIE
 from .rtlnl import RtlNlIE
 from .rtl2 import RTL2IE
 from .rtp import RTPIE
diff --git a/youtube_dl/extractor/rte.py b/youtube_dl/extractor/rte.py
index d9cfbf180..25abcee92 100644
--- a/youtube_dl/extractor/rte.py
+++ b/youtube_dl/extractor/rte.py
@@ -5,16 +5,19 @@ from .common import InfoExtractor
 
 from ..utils import (
     float_or_none,
+    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': 'mp4',
+            'ext': 'flv',
             'title': 'Watch iWitness  online',
             'thumbnail': 're:^https?://.*\.jpg$',
             'description': 'iWitness : The spirit of Ireland, one voice and one minute at a time.',
@@ -44,13 +47,6 @@ class RteIE(InfoExtractor):
         # 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)
-        f4m_formats = [{
-            'format_id': f['format_id'],
-            'url': f['url'],
-            'ext': 'mp4',
-            'width': f['width'],
-            'height': f['height'],
-        } for f in f4m_formats]
 
         return {
             'id': video_id,
@@ -60,3 +56,68 @@ class RteIE(InfoExtractor):
             'thumbnail': thumbnail,
             'duration': duration,
         }
+
+
+
+class RteRadioIE(InfoExtractor):
+    IE_NAME = 'rte:radio'
+    IE_DESC = 'Raidió Teilifís Éireann radio'
+    # Radioplayer URLs have the specifier #!rii=<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=(?:[0-9]*)(?:%3A|:)(?P<id>[0-9]+)'
+
+    _TEST = {
+        'url': 'http://www.rte.ie/radio/utils/radioplayer/rteradioweb.html#!rii=16:10507902:2414:27-12-2015:',
+        'info_dict': {
+            'id': '10507902',
+            'ext': 'flv',
+            'title': 'Gloria',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'description': 'Tim Thurston guides you through a millennium of sacred music featuring Gregorian chant, pure solo voices and choral masterpieces, framed around the glorious music of J.S. Bach.',
+            'duration': 7230.0,
+        },
+        'params': {
+            'skip_download': 'f4m fails with --test atm'
+        }
+    }
+
+    def _real_extract(self, url):
+        item_id = self._match_id(url)
+        feeds_url = 'http://www.rte.ie/rteavgen/getplaylist/?type=web&format=json&id=' + item_id
+        json_string = self._download_json(feeds_url, 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)
+
+        mg = show['media:group'][0]
+
+        formats = []
+
+        if mg.get('url') and not mg['url'].startswith('rtmpe:'):
+            formats.append({'url': mg.get('url')})
+
+        if mg.get('hls_server') and mg.get('hls_url'):
+            hls_url = mg['hls_server'] +  mg['hls_url']
+            hls_formats = self._extract_m3u8_formats(
+                    hls_url, item_id, 'mp4', m3u8_id='hls', fatal=False)
+            formats.extend(hls_formats)
+
+        if mg.get('hds_server') and mg.get('hds_url'):
+            f4m_url = mg['hds_server'] + mg['hds_url']
+            f4m_formats = self._extract_f4m_formats(
+                    f4m_url, item_id, f4m_id='hds', fatal=False)
+            formats.extend(f4m_formats)
+
+        return {
+            'id': item_id,
+            'title': title,
+            'formats': formats,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+        }