debian/copyright: Update my copyright years.

[youtubedl] / youtube_dl / extractor / escapist.py

diff --git a/youtube_dl/extractor/escapist.py b/youtube_dl/extractor/escapist.py
index b1242f6bc457a41a9c8413eb851671acd05cc8c0..e47f3e27a57aa14e3eee526af8998230b524bb4f 100644 (file)
--- a/youtube_dl/extractor/escapist.py
+++ b/youtube_dl/extractor/escapist.py
@@ -1,84 +1,128 @@
-import json
-import re
+from __future__ import unicode_literals

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

-from ..utils import (
-    compat_str,
+from ..compat import (

     compat_urllib_parse,
     compat_urllib_parse,

-
+    compat_urllib_request,
+)
+from ..utils import (

     ExtractorError,
     ExtractorError,

+    js_to_json,
+    parse_duration,

 )
 
 
 class EscapistIE(InfoExtractor):
 )
 
 
 class EscapistIE(InfoExtractor):

-    _VALID_URL = r'^https?://?(www\.)?escapistmagazine\.com/videos/view/(?P<showname>[^/]+)/(?P<episode>[^/?]+)[/?]?.*$'
+    _VALID_URL = r'https?://?(www\.)?escapistmagazine\.com/videos/view/[^/?#]+/(?P<id>[0-9]+)-[^/?#]*(?:$|[?#])'
+    _USER_AGENT = 'Mozilla/5.0 (Windows NT 6.1; WOW64; Trident/7.0; rv:11.0) like Gecko'

     _TEST = {
     _TEST = {

-        u'url': u'http://www.escapistmagazine.com/videos/view/the-escapist-presents/6618-Breaking-Down-Baldurs-Gate',
-        u'file': u'6618-Breaking-Down-Baldurs-Gate.mp4',
-        u'md5': u'ab3a706c681efca53f0a35f1415cf0d1',
-        u'info_dict': {
-            u"description": u"Baldur's Gate: Original, Modded or Enhanced Edition? I'll break down what you can expect from the new Baldur's Gate: Enhanced Edition.", 
-            u"uploader": u"the-escapist-presents", 
-            u"title": u"Breaking Down Baldur's Gate"
+        'url': 'http://www.escapistmagazine.com/videos/view/the-escapist-presents/6618-Breaking-Down-Baldurs-Gate',
+        'md5': 'ab3a706c681efca53f0a35f1415cf0d1',
+        'info_dict': {
+            'id': '6618',
+            'ext': 'mp4',
+            'description': "Baldur's Gate: Original, Modded or Enhanced Edition? I'll break down what you can expect from the new Baldur's Gate: Enhanced Edition.",
+            'uploader_id': 'the-escapist-presents',
+            'uploader': 'The Escapist Presents',
+            'title': "Breaking Down Baldur's Gate",
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 264,

         }
     }
 
     def _real_extract(self, url):
         }
     }
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        showName = mobj.group('showname')
-        videoId = mobj.group('episode')
-
-        self.report_extraction(videoId)
-        webpage = self._download_webpage(url, videoId)
-
-        videoDesc = self._html_search_regex(
-            r'<meta name="description" content="([^"]*)"',
-            webpage, u'description', fatal=False)
-
-        playerUrl = self._og_search_video_url(webpage, name=u'player URL')
-
-        title = self._html_search_regex(
-            r'<meta name="title" content="([^"]*)"',
-            webpage, u'title').split(' : ')[-1]
-
-        configUrl = self._search_regex('config=(.*)$', playerUrl, u'config URL')
-        configUrl = compat_urllib_parse.unquote(configUrl)
+        video_id = self._match_id(url)
+        webpage_req = compat_urllib_request.Request(url)
+        webpage_req.add_header('User-Agent', self._USER_AGENT)
+        webpage = self._download_webpage(webpage_req, video_id)
+
+        uploader_id = self._html_search_regex(
+            r"<h1\s+class='headline'>\s*<a\s+href='/videos/view/(.*?)'",
+            webpage, 'uploader ID', fatal=False)
+        uploader = self._html_search_regex(
+            r"<h1\s+class='headline'>(.*?)</a>",
+            webpage, 'uploader', fatal=False)
+        description = self._html_search_meta('description', webpage)
+        duration = parse_duration(self._html_search_meta('duration', webpage))
+
+        raw_title = self._html_search_meta('title', webpage, fatal=True)
+        title = raw_title.partition(' : ')[2]
+
+        config_url = compat_urllib_parse.unquote(self._html_search_regex(
+            r'''(?x)
+            (?:
+                <param\s+name="flashvars".*?\s+value="config=|
+                flashvars=&quot;config=
+            )
+            (https?://[^"&]+)
+            ''',
+            webpage, 'config URL'))

 
         formats = []
 
         formats = []

+        ad_formats = []

 
 

-        def _add_format(name, cfgurl):
-            configJSON = self._download_webpage(
-                cfgurl, videoId,
-                u'Downloading ' + name + ' configuration',
-                u'Unable to download ' + name + ' configuration')
+        def _add_format(name, cfg_url, quality):
+            cfg_req = compat_urllib_request.Request(cfg_url)
+            cfg_req.add_header('User-Agent', self._USER_AGENT)
+            config = self._download_json(
+                cfg_req, video_id,
+                'Downloading ' + name + ' configuration',
+                'Unable to download ' + name + ' configuration',
+                transform_source=js_to_json)

 
 

-            # Technically, it's JavaScript, not JSON
-            configJSON = configJSON.replace("'", '"')
-
-            try:
-                config = json.loads(configJSON)
-            except (ValueError,) as err:
-                raise ExtractorError(u'Invalid JSON in configuration file: ' + compat_str(err))

             playlist = config['playlist']
             playlist = config['playlist']

-            formats.append({
-                'url': playlist[1]['url'],
-                'format_id': name,
-            })
-
-        _add_format(u'normal', configUrl)
-        hq_url = (configUrl +
-                  ('&hq=1' if '?' in configUrl else configUrl + '?hq=1'))
+            for p in playlist:
+                if p.get('eventCategory') == 'Video':
+                    ar = formats
+                elif p.get('eventCategory') == 'Video Postroll':
+                    ar = ad_formats
+                else:
+                    continue
+
+                ar.append({
+                    'url': p['url'],
+                    'format_id': name,
+                    'quality': quality,
+                    'http_headers': {
+                        'User-Agent': self._USER_AGENT,
+                    },
+                })
+
+        _add_format('normal', config_url, quality=0)
+        hq_url = (config_url +
+                  ('&hq=1' if '?' in config_url else config_url + '?hq=1'))

         try:
         try:

-            _add_format(u'hq', hq_url)
+            _add_format('hq', hq_url, quality=1)

         except ExtractorError:
             pass  # That's fine, we'll just use normal quality
         except ExtractorError:
             pass  # That's fine, we'll just use normal quality

+        self._sort_formats(formats)
+
+        if '/escapist/sales-marketing/' in formats[-1]['url']:
+            raise ExtractorError('This IP address has been blocked by The Escapist', expected=True)

 
 

-        return {
-            'id': videoId,
+        res = {
+            'id': video_id,

             'formats': formats,
             'formats': formats,

-            'uploader': showName,
+            'uploader': uploader,
+            'uploader_id': uploader_id,

             'title': title,
             'thumbnail': self._og_search_thumbnail(webpage),
             'title': title,
             'thumbnail': self._og_search_thumbnail(webpage),

-            'description': videoDesc,
-            'player_url': playerUrl,
+            'description': description,
+            'duration': duration,

         }
         }

+
+        if self._downloader.params.get('include_ads') and ad_formats:
+            self._sort_formats(ad_formats)
+            ad_res = {
+                'id': '%s-ad' % video_id,
+                'title': '%s (Postroll)' % title,
+                'formats': ad_formats,
+            }
+            return {
+                '_type': 'playlist',
+                'entries': [res, ad_res],
+                'title': title,
+                'id': video_id,
+            }
+
+        return res