[iprima] Fix extraction (Closes #8953)

youtube_dl/extractor/iprima.py

@@ -1,4 +1,4 @@
-# -*- coding: utf-8 -*-
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -6,6 +6,8 @@ import time
 
 from .common import InfoExtractor
 from ..utils import (
+    determine_ext,
+    js_to_json,
     sanitized_Request,
 )
 
@@ -30,8 +32,7 @@ class IPrimaIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
 
@@ -43,9 +44,42 @@ class IPrimaIE(InfoExtractor):
         req.add_header('Referer', url)
         playerpage = self._download_webpage(req, video_id, note='Downloading player')
 
-        m3u8_url = self._search_regex(r"'src': '([^']+\.m3u8)'", playerpage, 'm3u8 url')
+        formats = []
 
-        formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4')
+        def extract_formats(format_url, format_key=None, lang=None):
+            ext = determine_ext(format_url)
+            new_formats = []
+            if format_key == 'hls' or ext == 'm3u8':
+                new_formats = self._extract_m3u8_formats(
+                    format_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False)
+            elif format_key == 'dash' or ext == 'mpd':
+                return
+                new_formats = self._extract_mpd_formats(
+                    format_url, video_id, mpd_id='dash', fatal=False)
+            if lang:
+                for f in new_formats:
+                    if not f.get('language'):
+                        f['language'] = lang
+            formats.extend(new_formats)
+
+        options = self._parse_json(
+            self._search_regex(
+                r'(?s)var\s+playerOptions\s*=\s*({.+?});',
+                playerpage, 'player options', default='{}'),
+            video_id, transform_source=js_to_json, fatal=False)
+        if options:
+            for key, tracks in options.get('tracks', {}).items():
+                if not isinstance(tracks, list):
+                    continue
+                for track in tracks:
+                    src = track.get('src')
+                    if src:
+                        extract_formats(src, key.lower(), track.get('lang'))
+
+        if not formats:
+            for _, src in re.findall(r'src["\']\s*:\s*(["\'])(.+?)\1', playerpage):
+                extract_formats(src)
 
         self._sort_formats(formats)