Merge pull request #8348 from remitamine/dfxp2srt-text

[youtube-dl] / youtube_dl / utils.py

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 0ed6c45c8090e7d328ec4e09d12875adc768c0f9..18dbe28bb5ecbcdb922f292490a43ccf6662ac9b 100644 (file)
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -70,6 +70,21 @@ ENGLISH_MONTH_NAMES = [
     'January', 'February', 'March', 'April', 'May', 'June',
     'July', 'August', 'September', 'October', 'November', 'December']
 
+KNOWN_EXTENSIONS = (
+    'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac',
+    'flv', 'f4v', 'f4a', 'f4b',
+    'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus',
+    'mkv', 'mka', 'mk3d',
+    'avi', 'divx',
+    'mov',
+    'asf', 'wmv', 'wma',
+    '3gp', '3g2',
+    'mp3',
+    'flac',
+    'ape',
+    'wav',
+    'f4f', 'f4m', 'm3u8', 'smil')
+
 
 def preferredencoding():
     """Get preferred encoding.
@@ -942,20 +957,8 @@ def determine_ext(url, default_ext='unknown_video'):
     guess = url.partition('?')[0].rpartition('.')[2]
     if re.match(r'^[A-Za-z0-9]+$', guess):
         return guess
-    elif guess.rstrip('/') in (
-            'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac',
-            'flv', 'f4v', 'f4a', 'f4b',
-            'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus',
-            'mkv', 'mka', 'mk3d',
-            'avi', 'divx',
-            'mov',
-            'asf', 'wmv', 'wma',
-            '3gp', '3g2',
-            'mp3',
-            'flac',
-            'ape',
-            'wav',
-            'f4f', 'f4m', 'm3u8', 'smil'):
+    # Try extract ext from URLs like http://example.com/foo/bar.mp4/?download
+    elif guess.rstrip('/') in KNOWN_EXTENSIONS:
         return guess.rstrip('/')
     else:
         return default_ext
@@ -981,7 +984,7 @@ def date_from_str(date_str):
         if sign == '-':
             time = -time
         unit = match.group('unit')
-        # A bad aproximation?
+        # A bad approximation?
         if unit == 'month':
             unit = 'day'
             time *= 30
@@ -1304,7 +1307,7 @@ def parse_filesize(s):
     if s is None:
         return None
 
-    # The lower-case forms are of course incorrect and inofficial,
+    # The lower-case forms are of course incorrect and unofficial,
     # but we support those too
     _UNIT_TABLE = {
         'B': 1,
@@ -1825,9 +1828,11 @@ def mimetype2ext(mt):
     _, _, res = mt.rpartition('/')
 
     return {
-        'x-ms-wmv': 'wmv',
-        'x-mp4-fragmented': 'mp4',
+        '3gpp': '3gp',
         'ttml+xml': 'ttml',
+        'x-flv': 'flv',
+        'x-mp4-fragmented': 'mp4',
+        'x-ms-wmv': 'wmv',
     }.get(res, res)
 
 
@@ -2012,20 +2017,27 @@ def dfxp2srt(dfxp_data):
         'ttaf1': 'http://www.w3.org/2006/10/ttaf1',
     })
 
-    def parse_node(node):
-        str_or_empty = functools.partial(str_or_none, default='')
+    class TTMLPElementParser:
+        out = ''
 
-        out = str_or_empty(node.text)
+        def start(self, tag, attrib):
+            if tag in (_x('ttml:br'), _x('ttaf1:br'), 'br'):
+                self.out += '\n'
 
-        for child in node:
-            if child.tag in (_x('ttml:br'), _x('ttaf1:br'), 'br'):
-                out += '\n' + str_or_empty(child.tail)
-            elif child.tag in (_x('ttml:span'), _x('ttaf1:span'), 'span'):
-                out += str_or_empty(parse_node(child))
-            else:
-                out += str_or_empty(xml.etree.ElementTree.tostring(child))
+        def end(self, tag):
+            pass
+
+        def data(self, data):
+            self.out += data
 
-        return out
+        def close(self):
+            return self.out.strip()
+
+    def parse_node(node):
+        target = TTMLPElementParser()
+        parser = xml.etree.ElementTree.XMLParser(target=target)
+        parser.feed(xml.etree.ElementTree.tostring(node))
+        return parser.close()
 
     dfxp = compat_etree_fromstring(dfxp_data.encode('utf-8'))
     out = []