Add an extractor for orf.at (closes #1346)

Make find_xpath_attr also accept numbers in the value

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index 6b5037c8c18092c8fdafd3675b0baa9d1327cadd..90f1a4418b1267cf630a60dca3cd23266133f59c 100644 (file)
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -59,6 +59,7 @@ from .myvideo import MyVideoIE
 from .nba import NBAIE
 from .nbc import NBCNewsIE
 from .ooyala import OoyalaIE
+from .orf import ORFIE
 from .pbs import PBSIE
 from .photobucket import PhotobucketIE
 from .pornotube import PornotubeIE

diff --git a/youtube_dl/extractor/orf.py b/youtube_dl/extractor/orf.py
new file mode 100644 (file)
index 0000000..8da0a2c
--- /dev/null
+++ b/youtube_dl/extractor/orf.py
@@ -0,0 +1,65 @@
+import re
+import xml.etree.ElementTree
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    ExtractorError,
+    find_xpath_attr,
+)
+
+class ORFIE(InfoExtractor):
+    _VALID_URL = r'https?://tvthek.orf.at/(programs/.+?/episodes|topics/.+?)/(?P<id>\d+)'
+
+    _TEST = {
+        u'url': u'http://tvthek.orf.at/programs/1171769-Wetter-ZIB/episodes/6557323-Wetter',
+        u'file': u'6566957.flv',
+        u'info_dict': {
+            u'title': u'Wetter',
+            u'description': u'Christa Kummer, Marcus Wadsak und Kollegen  präsentieren abwechselnd ihre täglichen Wetterprognosen für Österreich.\r \r Mehr Wetter unter wetter.ORF.at',
+        },
+        u'params': {
+            # It uses rtmp
+            u'skip_download': True,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('id')
+        webpage = self._download_webpage(url, playlist_id)
+
+        flash_xml = self._search_regex('ORF.flashXML = \'(.+?)\'', webpage, u'flash xml')
+        flash_xml = compat_urlparse.parse_qs('xml='+flash_xml)['xml'][0]
+        flash_config = xml.etree.ElementTree.fromstring(flash_xml.encode('utf-8'))
+        playlist_json = self._search_regex(r'playlist\': \'(\[.*?\])\'', webpage, u'playlist').replace(r'\"','"')
+        playlist = json.loads(playlist_json)
+
+        videos = []
+        ns = '{http://tempuri.org/XMLSchema.xsd}'
+        xpath = '%(ns)sPlaylist/%(ns)sItems/%(ns)sItem' % {'ns': ns}
+        webpage_description = self._og_search_description(webpage)
+        for (i, (item, info)) in enumerate(zip(flash_config.findall(xpath), playlist), 1):
+            # Get best quality url
+            rtmp_url = None
+            for q in ['Q6A', 'Q4A', 'Q1A']:
+                video_url = find_xpath_attr(item, '%sVideoUrl' % ns, 'quality', q)
+                if video_url is not None:
+                    rtmp_url = video_url.text
+                    break
+            if rtmp_url is None:
+                raise ExtractorError(u'Couldn\'t get video url: %s' % info['id'])
+            description = self._html_search_regex(
+                r'id="playlist_entry_%s".*?<p>(.*?)</p>' % i, webpage,
+                u'description', default=webpage_description, flags=re.DOTALL)
+            videos.append({
+                '_type': 'video',
+                'id': info['id'],
+                'title': info['title'],
+                'url': rtmp_url,
+                'ext': 'flv',
+                'description': description,
+                })
+
+        return videos

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index b3d0f64ea9162876f5cd3d517ef77a9b631c781e..201802cee6e56cbfffeed573c7fc42592a33fdab 100644 (file)
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -213,7 +213,7 @@ if sys.version_info >= (2,7):
     def find_xpath_attr(node, xpath, key, val):
         """ Find the xpath xpath[@key=val] """
         assert re.match(r'^[a-zA-Z]+$', key)
-        assert re.match(r'^[a-zA-Z@\s]*$', val)
+        assert re.match(r'^[a-zA-Z0-9@\s]*$', val)
         expr = xpath + u"[@%s='%s']" % (key, val)
         return node.find(expr)
 else: