Browse Source

[pbs] Add support for flp frontlines (Closes #7369)

Sergey M․ 9 năm trước cách đây
mục cha
commit
686f98816e
1 tập tin đã thay đổi với 18 bổ sung0 xóa
  1. 18 0
      youtube_dl/extractor/pbs.py

+ 18 - 0
youtube_dl/extractor/pbs.py

@@ -8,6 +8,7 @@ from ..utils import (
     ExtractorError,
     ExtractorError,
     determine_ext,
     determine_ext,
     int_or_none,
     int_or_none,
+    strip_jsonp,
     unified_strdate,
     unified_strdate,
     US_RATINGS,
     US_RATINGS,
 )
 )
@@ -191,6 +192,23 @@ class PBSIE(InfoExtractor):
             if media_id:
             if media_id:
                 return media_id, presumptive_id, upload_date
                 return media_id, presumptive_id, upload_date
 
 
+            # Fronline video embedded via flp
+            video_id = self._search_regex(
+                r'videoid\s*:\s*"([\d+a-z]{7,})"', webpage, 'videoid')
+            if video_id:
+                # pkg_id calculation is reverse engineered from
+                # http://www.pbs.org/wgbh/pages/frontline/js/flp2012.js
+                prg_id = self._search_regex(
+                    r'videoid\s*:\s*"([\d+a-z]{7,})"', webpage, 'videoid')[7:]
+                if 'q' in prg_id:
+                    prg_id = prg_id.split('q')[1]
+                prg_id = int(prg_id, 16)
+                getdir = self._download_json(
+                    'http://www.pbs.org/wgbh/pages/frontline/.json/getdir/getdir%d.json' % prg_id,
+                    presumptive_id, 'Downloading getdir JSON',
+                    transform_source=strip_jsonp)
+                return getdir['mid'], presumptive_id, upload_date
+
             for iframe in re.findall(r'(?s)<iframe(.+?)></iframe>', webpage):
             for iframe in re.findall(r'(?s)<iframe(.+?)></iframe>', webpage):
                 url = self._search_regex(
                 url = self._search_regex(
                     r'src=(["\'])(?P<url>.+?partnerplayer.+?)\1', iframe,
                     r'src=(["\'])(?P<url>.+?partnerplayer.+?)\1', iframe,