[crackle] Add new extractor

3 files changed, 102 insertions, 7 deletions

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index 18951c287..f08f27480 100644
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -126,6 +126,7 @@ from .comcarcoff import ComCarCoffIE
 from .commonmistakes import CommonMistakesIE, UnicodeBOMIE
 from .condenast import CondeNastIE
 from .cracked import CrackedIE
+from .crackle import CrackleIE
 from .criterion import CriterionIE
 from .crooksandliars import CrooksAndLiarsIE
 from .crunchyroll import (
diff --git a/youtube_dl/extractor/comcarcoff.py b/youtube_dl/extractor/comcarcoff.py
index 2efa200b5..7dff68492 100644
--- a/youtube_dl/extractor/comcarcoff.py
+++ b/youtube_dl/extractor/comcarcoff.py
@@ -2,6 +2,7 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     int_or_none,
     parse_duration,
@@ -14,14 +15,13 @@ class ComCarCoffIE(InfoExtractor):
     _TESTS = [{
         'url': 'http://comediansincarsgettingcoffee.com/miranda-sings-happy-thanksgiving-miranda/',
         'info_dict': {
-            'id': 'miranda-sings-happy-thanksgiving-miranda',
+            'id': '2494164',
             'ext': 'mp4',
             'upload_date': '20141127',
             'timestamp': 1417107600,
             'duration': 1232,
             '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',
         },
         'params': {
             'skip_download': 'requires ffmpeg',
@@ -39,15 +39,14 @@ class ComCarCoffIE(InfoExtractor):
                 r'window\.app\s*=\s*({.+?});\n', webpage, 'full data json'),
             display_id)['videoData']
 
-        video_id = full_data['activeVideo']['video']
-        video_data = full_data.get('videos', {}).get(video_id) or full_data['singleshots'][video_id]
+        display_id = full_data['activeVideo']['video']
+        video_data = full_data.get('videos', {}).get(display_id) or full_data['singleshots'][display_id]
+        video_id = compat_str(video_data['mediaId'])
         thumbnails = [{
             'url': video_data['images']['thumb'],
         }, {
             'url': video_data['images']['poster'],
         }]
-        formats = self._extract_m3u8_formats(
-            video_data['mediaUrl'], video_id, ext='mp4')
 
         timestamp = int_or_none(video_data.get('pubDateTime')) or parse_iso8601(
             video_data.get('pubDate'))
@@ -55,6 +54,8 @@ class ComCarCoffIE(InfoExtractor):
             video_data.get('duration'))
 
         return {
+            '_type': 'url_transparent',
+            'url': 'crackle:%s' % video_id,
             'id': video_id,
             'display_id': display_id,
             'title': video_data['title'],
@@ -62,6 +63,7 @@ class ComCarCoffIE(InfoExtractor):
             'timestamp': timestamp,
             'duration': duration,
             'thumbnails': thumbnails,
-            'formats': formats,
+            'season_number': int_or_none(video_data.get('season')),
+            'episode_number': int_or_none(video_data.get('episode')),
             'webpage_url': 'http://comediansincarsgettingcoffee.com/%s' % (video_data.get('urlSlug', video_data.get('slug'))),
         }
diff --git a/youtube_dl/extractor/crackle.py b/youtube_dl/extractor/crackle.py
new file mode 100644
index 000000000..a478333a2
--- /dev/null
+++ b/youtube_dl/extractor/crackle.py
@@ -0,0 +1,92 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import int_or_none
+
+
+class CrackleIE(InfoExtractor):
+    _VALID_URL = r'(?:crackle:|https?://(?:www\.)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://www.crackle.com/the-art-of-more/2496419',
+        'info_dict': {
+            'id': '2496419',
+            'ext': 'mp4',
+            'title': 'Heavy Lies the Head',
+            'description': 'md5:bb56aa0708fe7b9a4861535f15c3abca',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        }
+    }
+
+    # extracted from http://legacyweb-us.crackle.com/flash/QueryReferrer.ashx
+    _SUBTITLE_SERVER = 'http://web-us-az.crackle.com'
+    _UPLYNK_OWNER_ID = 'e8773f7770a44dbd886eee4fca16a66b'
+    _THUMBNAIL_TEMPLATE = 'http://images-us-am.crackle.com/%stnl_1920x1080.jpg?ts=20140107233116?c=635333335057637614'
+
+    # extracted from http://legacyweb-us.crackle.com/flash/ReferrerRedirect.ashx
+    _MEDIA_FILE_SLOTS = {
+        'c544.flv': {
+            'width': 544,
+            'height': 306,
+        },
+        '360p.mp4': {
+            'width': 640,
+            'height': 360,
+        },
+        '480p.mp4': {
+            'width': 852,
+            'height': 478,
+        },
+        '480p_1mbps.mp4': {
+            'width': 852,
+            'height': 478,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        item = self._download_xml(
+            'http://legacyweb-us.crackle.com/app/revamp/vidwallcache.aspx?flags=-1&fm=%s' % video_id, video_id).find('i')
+        title = item.attrib['t']
+
+        thumbnail = None
+        subtitles = {}
+        formats = self._extract_m3u8_formats('http://content.uplynk.com/ext/%s/%s.m3u8' % (self._UPLYNK_OWNER_ID, video_id), video_id, 'mp4', fatal=None)
+        path = item.attrib.get('p')
+        if path:
+            thumbnail = self._THUMBNAIL_TEMPLATE % path
+            http_base_url = 'http://ahttp.crackle.com/' + path
+            for mfs_path, mfs_info in self._MEDIA_FILE_SLOTS.items():
+                formats.append({
+                    'url': http_base_url + mfs_path,
+                    'format_id': mfs_path.split('.')[0],
+                    'width': mfs_info['width'],
+                    'height': mfs_info['height'],
+                })
+            for cc in item.findall('cc'):
+                locale = cc.attrib.get('l')
+                v = cc.attrib.get('v')
+                if locale and v:
+                    if locale not in subtitles:
+                        subtitles[locale] = []
+                    subtitles[locale] = [{
+                        'url': '%s/%s%s_%s.xml' % (self._SUBTITLE_SERVER, path, locale, v),
+                        'ext': 'ttml',
+                    }]
+        self._sort_formats(formats, ('width', 'height', 'tbr'))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': item.attrib.get('d'),
+            'duration': int(item.attrib.get('r'), 16) if item.attrib.get('r') else None,
+            'series': item.attrib.get('sn'),
+            'season_number': int_or_none(item.attrib.get('se')),
+            'episode_number': int_or_none(item.attrib.get('ep')),
+            'thumbnail': thumbnail,
+            'subtitles': subtitles,
+            'formats': formats,
+        }