[elpais] Add extractor

youtube_dl/extractor/__init__.py

@@ -54,6 +54,7 @@
 from .ehow import EHowIE
 from .eighttracks import EightTracksIE
 from .eitb import EitbIE
+from .elpais import ElPaisIE
 from .escapist import EscapistIE
 from .everyonesmixtape import EveryonesMixtapeIE
 from .exfm import ExfmIE

youtube_dl/extractor/elpais.py

@@ -0,0 +1,58 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import unified_strdate
+
+
+class ElPaisIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:[^.]+\.)?elpais\.com/.*/(?P<id>[^/#?]+)\.html(?:$|[?#])'
+    IE_DESCR = 'El País'
+
+    _TEST = {
+        'url': 'http://blogs.elpais.com/la-voz-de-inaki/2014/02/tiempo-nuevo-recetas-viejas.html',
+        'md5': '98406f301f19562170ec071b83433d55',
+        'info_dict': {
+            'id': 'tiempo-nuevo-recetas-viejas',
+            'ext': 'mp4',
+            'title': 'Tiempo nuevo, recetas viejas',
+            'description': 'De lunes a viernes, a partir de las ocho de la mañana, Iñaki Gabilondo nos cuenta su visión de la actualidad nacional e internacional.',
+            'upload_date': '20140206',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        prefix = self._html_search_regex(
+            r'var url_cache = "([^"]+)";', webpage, 'URL prefix')
+        video_suffix = self._search_regex(
+            r"URLMediaFile = url_cache \+ '([^']+)'", webpage, 'video URL')
+        video_url = prefix + video_suffix
+        thumbnail_suffix = self._search_regex(
+            r"URLMediaStill = url_cache \+ '([^']+)'", webpage, 'thumbnail URL',
+            fatal=False)
+        thumbnail = (
+            None if thumbnail_suffix is None
+            else prefix + thumbnail_suffix)
+        title = self._html_search_regex(
+            '<h2 class="entry-header entry-title.*?>(.*?)</h2>',
+            webpage, 'title')
+        date_str = self._search_regex(
+            r'<p class="date-header date-int updated"\s+title="([^"]+)">',
+            webpage, 'upload date', fatal=False)
+        upload_date = (None if date_str is None else unified_strdate(date_str))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'description': self._og_search_description(webpage),
+            'thumbnail': thumbnail,
+            'upload_date': upload_date,
+        }

youtube_dl/utils.py

@@ -751,13 +751,14 @@ def http_response(self, req, resp):
     https_request = http_request
     https_response = http_response
 
+
 def unified_strdate(date_str):
     """Return a string with the date in the format YYYYMMDD"""
     upload_date = None
     #Replace commas
     date_str = date_str.replace(',',' ')
     # %z (UTC offset) is only supported in python>=3.2
-    date_str = re.sub(r' (\+|-)[\d]*$', '', date_str)
+    date_str = re.sub(r' ?(\+|-)[0-9:]*$', '', date_str)
     format_expressions = [
         '%d %B %Y',
         '%B %d %Y',
@@ -771,11 +772,12 @@ def unified_strdate(date_str):
         '%Y-%m-%dT%H:%M:%S.%fZ',
         '%Y-%m-%dT%H:%M:%S.%f0Z',
         '%Y-%m-%dT%H:%M:%S',
+        '%Y-%m-%dT%H:%M',
     ]
     for expression in format_expressions:
         try:
             upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d')
-        except:
+        except ValueError:
             pass
     if upload_date is None:
         timetuple = email.utils.parsedate_tz(date_str)