[youtube] Fix extraction.

[youtube-dl] / youtube_dl / extractor / archiveorg.py

diff --git a/youtube_dl/extractor/archiveorg.py b/youtube_dl/extractor/archiveorg.py
index 497ce97acbef0c20f3050bed69bba410d7f8767c..c79c58e828150b9608bc2ac74c34a3428deb821a 100644 (file)
--- a/youtube_dl/extractor/archiveorg.py
+++ b/youtube_dl/extractor/archiveorg.py
@@ -1,63 +1,65 @@
 from __future__ import unicode_literals
 
-import json
-import re
-
 from .common import InfoExtractor
 from ..utils import (
-    determine_ext,
     unified_strdate,
+    clean_html,
 )
 
 
 class ArchiveOrgIE(InfoExtractor):
     IE_NAME = 'archive.org'
     IE_DESC = 'archive.org videos'
-    _VALID_URL = r'(?:https?://)?(?:www\.)?archive\.org/details/(?P<id>[^?/]+)(?:[?].*)?$'
-    _TEST = {
-        "url": "http://archive.org/details/XD300-23_68HighlightsAResearchCntAugHumanIntellect",
-        'file': 'XD300-23_68HighlightsAResearchCntAugHumanIntellect.ogv',
+    _VALID_URL = r'https?://(?:www\.)?archive\.org/(?:details|embed)/(?P<id>[^/?#]+)(?:[?].*)?$'
+    _TESTS = [{
+        'url': 'http://archive.org/details/XD300-23_68HighlightsAResearchCntAugHumanIntellect',
         'md5': '8af1d4cf447933ed3c7f4871162602db',
         'info_dict': {
-            "title": "1968 Demo - FJCC Conference Presentation Reel #1",
-            "description": "Reel 1 of 3: Also known as the \"Mother of All Demos\", Doug Engelbart's presentation at the Fall Joint Computer Conference in San Francisco, December 9, 1968 titled \"A Research Center for Augmenting Human Intellect.\" For this presentation, Doug and his team astonished the audience by not only relating their research, but demonstrating it live. This was the debut of the mouse, interactive computing, hypermedia, computer supported software engineering, video teleconferencing, etc. See also <a href=\"http://dougengelbart.org/firsts/dougs-1968-demo.html\" rel=\"nofollow\">Doug's 1968 Demo page</a> for more background, highlights, links, and the detailed paper published in this conference proceedings. Filmed on 3 reels: Reel 1 | <a href=\"http://www.archive.org/details/XD300-24_68HighlightsAResearchCntAugHumanIntellect\" rel=\"nofollow\">Reel 2</a> | <a href=\"http://www.archive.org/details/XD300-25_68HighlightsAResearchCntAugHumanIntellect\" rel=\"nofollow\">Reel 3</a>",
-            "upload_date": "19681210",
-            "uploader": "SRI International"
+            'id': 'XD300-23_68HighlightsAResearchCntAugHumanIntellect',
+            'ext': 'ogg',
+            'title': '1968 Demo - FJCC Conference Presentation Reel #1',
+            'description': 'md5:da45c349df039f1cc8075268eb1b5c25',
+            'upload_date': '19681210',
+            'uploader': 'SRI International'
         }
-    }
-
+    }, {
+        'url': 'https://archive.org/details/Cops1922',
+        'md5': '0869000b4ce265e8ca62738b336b268a',
+        'info_dict': {
+            'id': 'Cops1922',
+            'ext': 'mp4',
+            'title': 'Buster Keaton\'s "Cops" (1922)',
+            'description': 'md5:89e7c77bf5d965dd5c0372cfb49470f6',
+        }
+    }, {
+        'url': 'http://archive.org/embed/XD300-23_68HighlightsAResearchCntAugHumanIntellect',
+        '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)
+        webpage = self._download_webpage(
+            'http://archive.org/embed/' + video_id, video_id)
+        jwplayer_playlist = self._parse_json(self._search_regex(
+            r"(?s)Play\('[^']+'\s*,\s*(\[.+\])\s*,\s*{.*?}\)",
+            webpage, 'jwplayer playlist'), video_id)
+        info = self._parse_jwplayer_data(
+            {'playlist': jwplayer_playlist}, video_id, base_url=url)
 
-        json_url = url + ('?' if '?' in url else '&') + 'output=json'
-        json_data = self._download_webpage(json_url, video_id)
-        data = json.loads(json_data)
+        def get_optional(metadata, field):
+            return metadata.get(field, [None])[0]
 
-        title = data['metadata']['title'][0]
-        description = data['metadata']['description'][0]
-        uploader = data['metadata']['creator'][0]
-        upload_date = unified_strdate(data['metadata']['date'][0])
-
-        formats = [{
-                'format': fdata['format'],
-                'url': 'http://' + data['server'] + data['dir'] + fn,
-                'file_size': int(fdata['size']),
-            }
-            for fn,fdata in data['files'].items()
-            if 'Video' in fdata['format']]
-        formats.sort(key=lambda fdata: fdata['file_size'])
-        for f in formats:
-            f['ext'] = determine_ext(f['url'])
-
-        return {
-            '_type': 'video',
-            'id': video_id,
-            'title': title,
-            'formats': formats,
-            'description': description,
-            'uploader': uploader,
-            'upload_date': upload_date,
-            'thumbnail': data.get('misc', {}).get('image'),
-        }
+        metadata = self._download_json(
+            'http://archive.org/details/' + video_id, video_id, query={
+                'output': 'json',
+            })['metadata']
+        info.update({
+            'title': get_optional(metadata, 'title') or info.get('title'),
+            'description': clean_html(get_optional(metadata, 'description')),
+        })
+        if info.get('_type') != 'playlist':
+            info.update({
+                'uploader': get_optional(metadata, 'creator'),
+                'upload_date': unified_strdate(get_optional(metadata, 'date')),
+            })
+        return info