X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Flibraryofcongress.py;h=03f2051444d63bb046b6b5943cc5400085c6d8d3;hb=HEAD;hp=49351759ed318bda771a6648f7f33744fe62c4b7;hpb=c917106be4d6d98ce7504d71a32b58ddca2bc03d;p=youtube-dl

diff --git a/youtube_dl/extractor/libraryofcongress.py b/youtube_dl/extractor/libraryofcongress.py
index 49351759e..03f205144 100644
--- a/youtube_dl/extractor/libraryofcongress.py
+++ b/youtube_dl/extractor/libraryofcongress.py
@@ -1,28 +1,30 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 
 from ..utils import (
     determine_ext,
     float_or_none,
     int_or_none,
+    parse_filesize,
 )
 
 
 class LibraryOfCongressIE(InfoExtractor):
     IE_NAME = 'loc'
     IE_DESC = 'Library of Congress'
-    _VALID_URL = r'https?://(?:www\.)?loc\.gov/(?:item/|today/cyberlc/feature_wdesc\.php\?.*\brec=)(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?loc\.gov/(?:item/|today/cyberlc/feature_wdesc\.php\?.*\brec=)(?P<id>[0-9a-z_.]+)'
     _TESTS = [{
         # embedded via <div class="media-player"
         'url': 'http://loc.gov/item/90716351/',
-        'md5': '353917ff7f0255aa6d4b80a034833de8',
+        'md5': '6ec0ae8f07f86731b1b2ff70f046210a',
         'info_dict': {
             'id': '90716351',
             'ext': 'mp4',
             'title': "Pa's trip to Mars",
-            'thumbnail': 're:^https?://.*\.jpg$',
             'duration': 0,
             'view_count': int,
         },
@@ -40,6 +42,26 @@ class LibraryOfCongressIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+    }, {
+        # with direct download links
+        'url': 'https://www.loc.gov/item/78710669/',
+        'info_dict': {
+            'id': '78710669',
+            'ext': 'mp4',
+            'title': 'La vie et la passion de Jesus-Christ',
+            'duration': 0,
+            'view_count': int,
+            'formats': 'mincount:4',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.loc.gov/item/ihas.200197114/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.loc.gov/item/afc1981005_afs20503/',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -50,16 +72,20 @@ class LibraryOfCongressIE(InfoExtractor):
             (r'id=(["\'])media-player-(?P<id>.+?)\1',
              r'<video[^>]+id=(["\'])uuid-(?P<id>.+?)\1',
              r'<video[^>]+data-uuid=(["\'])(?P<id>.+?)\1',
-             r'mediaObjectId\s*:\s*(["\'])(?P<id>.+?)\1'),
+             r'mediaObjectId\s*:\s*(["\'])(?P<id>.+?)\1',
+             r'data-tab="share-media-(?P<id>[0-9A-F]{32})"'),
             webpage, 'media id', group='id')
 
         data = self._download_json(
             'https://media.loc.gov/services/v1/media?id=%s&context=json' % media_id,
-            video_id)['mediaObject']
+            media_id)['mediaObject']
 
         derivative = data['derivatives'][0]
         media_url = derivative['derivativeUrl']
 
+        title = derivative.get('shortName') or data.get('shortName') or self._og_search_title(
+            webpage)
+
         # Following algorithm was extracted from setAVSource js function
         # found in webpage
         media_url = media_url.replace('rtmp', 'https')
@@ -69,22 +95,42 @@ class LibraryOfCongressIE(InfoExtractor):
         if ext not in ('mp4', 'mp3'):
             media_url += '.mp4' if is_video else '.mp3'
 
-        if 'vod/mp4:' in media_url:
-            formats = [{
-                'url': media_url.replace('vod/mp4:', 'hls-vod/media/') + '.m3u8',
+        formats = []
+        if '/vod/mp4:' in media_url:
+            formats.append({
+                'url': media_url.replace('/vod/mp4:', '/hls-vod/media/') + '.m3u8',
                 'format_id': 'hls',
                 'ext': 'mp4',
                 'protocol': 'm3u8_native',
-            }]
-        elif 'vod/mp3:' in media_url:
-            formats = [{
-                'url': media_url.replace('vod/mp3:', ''),
-                'vcodec': 'none',
-            }]
+                'quality': 1,
+            })
+        http_format = {
+            'url': re.sub(r'(://[^/]+/)(?:[^/]+/)*(?:mp4|mp3):', r'\1', media_url),
+            'format_id': 'http',
+            'quality': 1,
+        }
+        if not is_video:
+            http_format['vcodec'] = 'none'
+        formats.append(http_format)
+
+        download_urls = set()
+        for m in re.finditer(
+                r'<option[^>]+value=(["\'])(?P<url>.+?)\1[^>]+data-file-download=[^>]+>\s*(?P<id>.+?)(?:(?:&nbsp;|\s+)\((?P<size>.+?)\))?\s*<', webpage):
+            format_id = m.group('id').lower()
+            if format_id in ('gif', 'jpeg'):
+                continue
+            download_url = m.group('url')
+            if download_url in download_urls:
+                continue
+            download_urls.add(download_url)
+            formats.append({
+                'url': download_url,
+                'format_id': format_id,
+                'filesize_approx': parse_filesize(m.group('size')),
+            })
 
         self._sort_formats(formats)
 
-        title = derivative.get('shortName') or data.get('shortName') or self._og_search_title(webpage)
         duration = float_or_none(data.get('duration'))
         view_count = int_or_none(data.get('viewCount'))