X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fccc.py;h=6924eac704cd5cf02d266bd60125dad9cf13e765;hb=40101dc311909523852a88ba69df76be9b6bc920;hp=809725f1251ddffddca399dd3c36f6d873779ba1;hpb=46d9760f5ebd86ee5b34fd669f44b124f34fffc1;p=youtube-dl

diff --git a/youtube_dl/extractor/ccc.py b/youtube_dl/extractor/ccc.py
index 809725f12..6924eac70 100644
--- a/youtube_dl/extractor/ccc.py
+++ b/youtube_dl/extractor/ccc.py
@@ -1,46 +1,99 @@
-# encoding: utf-8
+from __future__ import unicode_literals
+
 import re
-import json
 
 from .common import InfoExtractor
 from ..utils import (
+    int_or_none,
+    qualities,
     unified_strdate,
 )
 
-class ComCarCoffIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?comediansincarsgettingcoffee\.com/(?P<id>[a-z0-9\-]+)/?'
-    _TESTS = [
-        {
-            'url': 'http://comediansincarsgettingcoffee.com/miranda-sings-happy-thanksgiving-miranda/',
-            'info_dict': {
-                'id': 'miranda-sings-happy-thanksgiving-miranda',
-                'upload_date': '20141127',
-                'title': 'Happy Thanksgiving Miranda',
-                'description': 'Jerry Seinfeld and his special guest Miranda Sings cruise around town in search of coffee, complaining and apologizing along the way.',
-                'thumbnail': 'http://ccc.crackle.com/images/s5e4_thumb.jpg',
-            },
+
+class CCCIE(InfoExtractor):
+    IE_NAME = 'media.ccc.de'
+    _VALID_URL = r'https?://(?:www\.)?media\.ccc\.de/[^?#]+/[^?#/]*?_(?P<id>[0-9]{8,})._[^?#/]*\.html'
+
+    _TEST = {
+        'url': 'http://media.ccc.de/browse/congress/2013/30C3_-_5443_-_en_-_saal_g_-_201312281830_-_introduction_to_processor_design_-_byterazor.html#video',
+        'md5': '3a1eda8f3a29515d27f5adb967d7e740',
+        'info_dict': {
+            'id': '20131228183',
+            'ext': 'mp4',
+            'title': 'Introduction to Processor Design',
+            'description': 'md5:5ddbf8c734800267f2cee4eab187bc1b',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'view_count': int,
+            'upload_date': '20131229',
         }
-    ]
+    }
 
     def _real_extract(self, url):
-        display_id = self._match_id(url)
-        webpage = self._download_webpage(url, display_id)
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        if self._downloader.params.get('prefer_free_formats'):
+            preference = qualities(['mp3', 'opus', 'mp4-lq', 'webm-lq', 'h264-sd', 'mp4-sd', 'webm-sd', 'mp4', 'webm', 'mp4-hd', 'h264-hd', 'webm-hd'])
+        else:
+            preference = qualities(['opus', 'mp3', 'webm-lq', 'mp4-lq', 'webm-sd', 'h264-sd', 'mp4-sd', 'webm', 'mp4', 'webm-hd', 'mp4-hd', 'h264-hd'])
+
+        title = self._html_search_regex(
+            r'(?s)<h1>(.*?)</h1>', webpage, 'title')
+        description = self._html_search_regex(
+            r"(?s)<p class='description'>(.*?)</p>",
+            webpage, 'description', fatal=False)
+        upload_date = unified_strdate(self._html_search_regex(
+            r"(?s)<span class='[^']*fa-calendar-o'></span>(.*?)</li>",
+            webpage, 'upload date', fatal=False))
+        view_count = int_or_none(self._html_search_regex(
+            r"(?s)<span class='[^']*fa-eye'></span>(.*?)</li>",
+            webpage, 'view count', fatal=False))
+
+        matches = re.finditer(r'''(?xs)
+            <(?:span|div)\s+class='label\s+filetype'>(?P<format>.*?)</(?:span|div)>\s*
+            <a\s+download\s+href='(?P<http_url>[^']+)'>\s*
+            (?:
+                .*?
+                <a\s+href='(?P<torrent_url>[^']+\.torrent)'
+            )?''', webpage)
+        formats = []
+        for m in matches:
+            format = m.group('format')
+            format_id = self._search_regex(
+                r'.*/([a-z0-9_-]+)/[^/]*$',
+                m.group('http_url'), 'format id', default=None)
+            vcodec = 'h264' if 'h264' in format_id else (
+                'none' if format_id in ('mp3', 'opus') else None
+            )
+            formats.append({
+                'format_id': format_id,
+                'format': format,
+                'url': m.group('http_url'),
+                'vcodec': vcodec,
+                'preference': preference(format_id),
+            })
 
-        full_data = json.loads(self._search_regex(
-            r'<script type="application/json" id="videoData">(?P<json>.+?)</script>',
-            webpage, 'json'))
+            if m.group('torrent_url'):
+                formats.append({
+                    'format_id': 'torrent-%s' % (format if format_id is None else format_id),
+                    'format': '%s (torrent)' % format,
+                    'proto': 'torrent',
+                    'format_note': '(unsupported; will just download the .torrent file)',
+                    'vcodec': vcodec,
+                    'preference': -100 + preference(format_id),
+                    'url': m.group('torrent_url'),
+                })
+        self._sort_formats(formats)
 
-        video_id = full_data['activeVideo']['video']
-        video_data = full_data['videos'][video_id]
+        thumbnail = self._html_search_regex(
+            r"<video.*?poster='([^']+)'", webpage, 'thumbnail', fatal=False)
 
         return {
             'id': video_id,
-            'display_id': display_id,
-            'title': video_data['title'],
-            'description': video_data['description'],
-            # XXX: the original datum is a full ISO timestamp... why convert it to a worse format?
-            'upload_date': unified_strdate(video_data['pubDate']),
-            'thumbnail': video_data['images']['thumb'],
-            # XXX: what do we do with video_data['images']['poster']?
-            'formats': self._extract_m3u8_formats(video_data['mediaUrl'], video_id),
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'view_count': view_count,
+            'upload_date': upload_date,
+            'formats': formats,
         }