d/control: Build-depend on flake8 and python3-nose.

[youtubedl] / youtube_dl / extractor / iprima.py

diff --git a/youtube_dl/extractor/iprima.py b/youtube_dl/extractor/iprima.py
index 61a0de47283f426879d2fdad1c3620ac979ac0d0..11bbeb5922a9d85e05977196c822a076c8b45ec3 100644 (file)
--- a/youtube_dl/extractor/iprima.py
+++ b/youtube_dl/extractor/iprima.py
@@ -1,4 +1,4 @@
-# -*- coding: utf-8 -*-
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -6,12 +6,14 @@ import time
 
 from .common import InfoExtractor
 from ..utils import (
-    sanitized_Request,
+    determine_ext,
+    js_to_json,
 )
 
 
 class IPrimaIE(InfoExtractor):
-    _VALID_URL = r'https?://play\.iprima\.cz/(?:.+/)?(?P<id>[^?#]+)'
+    _VALID_URL = r'https?://(?:[^/]+)\.iprima\.cz/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+    _GEO_BYPASS = False
 
     _TESTS = [{
         'url': 'http://play.iprima.cz/gondici-s-r-o-33',
@@ -27,25 +29,97 @@ class IPrimaIE(InfoExtractor):
     }, {
         'url': 'http://play.iprima.cz/particka/particka-92',
         'only_matching': True,
+    }, {
+        # geo restricted
+        'url': 'http://play.iprima.cz/closer-nove-pripady/closer-nove-pripady-iv-1',
+        'only_matching': True,
+    }, {
+        # iframe api.play-backend.iprima.cz
+        'url': 'https://prima.iprima.cz/my-little-pony/mapa-znameni-2-2',
+        'only_matching': True,
+    }, {
+        # iframe prima.iprima.cz
+        'url': 'https://prima.iprima.cz/porady/jak-se-stavi-sen/rodina-rathousova-praha',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.iprima.cz/filmy/desne-rande',
+        'only_matching': True,
+    }, {
+        'url': 'https://zoom.iprima.cz/10-nejvetsich-tajemstvi-zahad/posvatna-mista-a-stavby',
+        'only_matching': True,
+    }, {
+        'url': 'https://krimi.iprima.cz/mraz-0/sebevrazdy',
+        'only_matching': True,
+    }, {
+        'url': 'https://cool.iprima.cz/derava-silnice-nevadi',
+        'only_matching': True,
+    }, {
+        'url': 'https://love.iprima.cz/laska-az-za-hrob/slib-dany-bratrovi',
+        'only_matching': True,
+    }, {
+        'url': 'https://autosalon.iprima.cz/motorsport/7-epizoda-1',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
+
+        self._set_cookie('play.iprima.cz', 'ott_adult_confirmed', '1')
 
         webpage = self._download_webpage(url, video_id)
 
-        video_id = self._search_regex(r'data-product="([^"]+)">', webpage, 'real id')
+        video_id = self._search_regex(
+            (r'<iframe[^>]+\bsrc=["\'](?:https?:)?//(?:api\.play-backend\.iprima\.cz/prehravac/embedded|prima\.iprima\.cz/[^/]+/[^/]+)\?.*?\bid=(p\d+)',
+             r'data-product="([^"]+)">'),
+            webpage, 'real id')
+
+        playerpage = self._download_webpage(
+            'http://play.iprima.cz/prehravac/init',
+            video_id, note='Downloading player', query={
+                '_infuse': 1,
+                '_ts': round(time.time()),
+                'productId': video_id,
+            }, headers={'Referer': url})
+
+        formats = []
+
+        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)
 
-        req = sanitized_Request(
-            'http://play.iprima.cz/prehravac/init?_infuse=1'
-            '&_ts=%s&productId=%s' % (round(time.time()), video_id))
-        req.add_header('Referer', url)
-        playerpage = self._download_webpage(req, video_id, note='Downloading player')
+        options = self._parse_json(
+            self._search_regex(
+                r'(?s)(?:TDIPlayerOptions|playerOptions)\s*=\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'))
 
-        m3u8_url = self._search_regex(r"'src': '([^']+\.m3u8)'", playerpage, 'm3u8 url')
+        if not formats:
+            for _, src in re.findall(r'src["\']\s*:\s*(["\'])(.+?)\1', playerpage):
+                extract_formats(src)
 
-        formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4')
+        if not formats and '>GEO_IP_NOT_ALLOWED<' in playerpage:
+            self.raise_geo_restricted(countries=['CZ'])
 
         self._sort_formats(formats)