[pbs] Add support for article pages (Fixes #870)

test/test_signatures.py

@@ -0,0 +1,75 @@
+#!/usr/bin/env python
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+
+import io
+import re
+import string
+
+from youtube_dl.extractor import YoutubeIE
+from youtube_dl.utils import compat_str, compat_urlretrieve
+
+_TESTS = [
+    (
+        u'https://s.ytimg.com/yts/jsbin/html5player-vflHOr_nV.js',
+        u'js',
+        86,
+        u'>=<;:/.-[+*)(\'&%$#"!ZYX0VUTSRQPONMLKJIHGFEDCBA\\yxwvutsrqponmlkjihgfedcba987654321',
+    ),
+    (
+        u'https://s.ytimg.com/yts/jsbin/html5player-vfldJ8xgI.js',
+        u'js',
+        85,
+        u'3456789a0cdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRS[UVWXYZ!"#$%&\'()*+,-./:;<=>?@',
+    ),
+]
+
+
+class TestSignature(unittest.TestCase):
+    def setUp(self):
+        TEST_DIR = os.path.dirname(os.path.abspath(__file__))
+        self.TESTDATA_DIR = os.path.join(TEST_DIR, 'testdata')
+        if not os.path.exists(self.TESTDATA_DIR):
+            os.mkdir(self.TESTDATA_DIR)
+
+
+def make_tfunc(url, stype, sig_length, expected_sig):
+    basename = url.rpartition('/')[2]
+    m = re.match(r'.*-([a-zA-Z0-9_-]+)\.[a-z]+$', basename)
+    assert m, '%r should follow URL format' % basename
+    test_id = m.group(1)
+
+    def test_func(self):
+        fn = os.path.join(self.TESTDATA_DIR, basename)
+
+        if not os.path.exists(fn):
+            compat_urlretrieve(url, fn)
+
+        ie = YoutubeIE()
+        if stype == 'js':
+            with io.open(fn, encoding='utf-8') as testf:
+                jscode = testf.read()
+            func = ie._parse_sig_js(jscode)
+        else:
+            assert stype == 'swf'
+            with open(fn, 'rb') as testf:
+                swfcode = testf.read()
+            func = ie._parse_sig_swf(swfcode)
+        src_sig = compat_str(string.printable[:sig_length])
+        got_sig = func(src_sig)
+        self.assertEqual(got_sig, expected_sig)
+
+    test_func.__name__ = str('test_signature_' + stype + '_' + test_id)
+    setattr(TestSignature, test_func.__name__, test_func)
+
+for test_spec in _TESTS:
+    make_tfunc(*test_spec)
+
+
+if __name__ == '__main__':
+    unittest.main()

youtube_dl/extractor/pbs.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -5,30 +7,63 @@ from .common import InfoExtractor
 
 
 class PBSIE(InfoExtractor):
-    _VALID_URL = r'https?://video\.pbs\.org/video/(?P<id>\d+)/?'
+    _VALID_URL = r'''(?x)https?://
+        (?:
+            # Direct video URL
+            video\.pbs\.org/video/(?P<id>[0-9]+)/? |
+            # Article with embedded player
+           (?:www\.)?pbs\.org/(?:[^/]+/){2,5}(?P<presumptive_id>[^/]+)/?(?:$|[?\#]) |
+           # Player
+           video\.pbs\.org/partnerplayer/(?P<player_id>[^/]+)/
+        )
+    '''
 
     _TEST = {
-        u'url': u'http://video.pbs.org/video/2365006249/',
+        'url': 'http://www.pbs.org/tpt/constitution-usa-peter-sagal/watch/a-more-perfect-union/',
-        u'file': u'2365006249.mp4',
+        'md5': 'ce1888486f0908d555a8093cac9a7362',
-        u'md5': 'ce1888486f0908d555a8093cac9a7362',
+        'info_dict': {
-        u'info_dict': {
+            'id': '2365006249',
-            u'title': u'A More Perfect Union',
+            'ext': 'mp4',
-            u'description': u'md5:ba0c207295339c8d6eced00b7c363c6a',
+            'title': 'A More Perfect Union',
-            u'duration': 3190,
+            'description': 'md5:ba0c207295339c8d6eced00b7c363c6a',
+            'duration': 3190,
         },
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+
+        presumptive_id = mobj.group('presumptive_id')
+        display_id = presumptive_id
+        if presumptive_id:
+            webpage = self._download_webpage(url, display_id)
+            url = self._search_regex(
+                r'<iframe\s+id=["\']partnerPlayer["\'].*?\s+src=["\'](.*?)["\']>',
+                webpage, 'player URL')
+            mobj = re.match(self._VALID_URL, url)
+
+        player_id = mobj.group('player_id')
+        if not display_id:
+            display_id = player_id
+        if player_id:
+            player_page = self._download_webpage(
+                url, display_id, note='Downloading player page',
+                errnote='Could not download player page')
+            video_id = self._search_regex(
+                r'<div\s+id="video_([0-9]+)"', player_page, 'video ID')
+        else:
+            video_id = mobj.group('id')
+            display_id = video_id
+
         info_url = 'http://video.pbs.org/videoInfo/%s?format=json' % video_id
-        info_page = self._download_webpage(info_url, video_id)
+        info = self._download_json(info_url, display_id)
-        info =json.loads(info_page)
+
-        return {'id': video_id,
+        return {
-                'title': info['title'],
+            'id': video_id,
-                'url': info['alternate_encoding']['url'],
+            'title': info['title'],
-                'ext': 'mp4',
+            'url': info['alternate_encoding']['url'],
-                'description': info['program'].get('description'),
+            'ext': 'mp4',
-                'thumbnail': info.get('image_url'),
+            'description': info['program'].get('description'),
-                'duration': info.get('duration'),
+            'thumbnail': info.get('image_url'),
-                }
+            'duration': info.get('duration'),
+        }