Merge pull request #5523 from jaimeMF/remove-format-limit

[youtube-dl] / youtube_dl / utils.py

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index c69d3e165fc1fa9b0d19c029ee74ae41a1c49d3f..7d15eab64e8bf7ef779bd3b6bd8a64fbfcd8ece7 100644 (file)
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -371,6 +371,18 @@ def unescapeHTML(s):
         r'&([^;]+);', lambda m: _htmlentity_transform(m.group(1)), s)
 
 
+def get_subprocess_encoding():
+    if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5:
+        # For subprocess calls, encode with locale encoding
+        # Refer to http://stackoverflow.com/a/9951851/35070
+        encoding = preferredencoding()
+    else:
+        encoding = sys.getfilesystemencoding()
+    if encoding is None:
+        encoding = 'utf-8'
+    return encoding
+
+
 def encodeFilename(s, for_subprocess=False):
     """
     @param s The name of the file
@@ -382,21 +394,24 @@ def encodeFilename(s, for_subprocess=False):
     if sys.version_info >= (3, 0):
         return s
 
-    if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5:
-        # Pass '' directly to use Unicode APIs on Windows 2000 and up
-        # (Detecting Windows NT 4 is tricky because 'major >= 4' would
-        # match Windows 9x series as well. Besides, NT 4 is obsolete.)
-        if not for_subprocess:
-            return s
-        else:
-            # For subprocess calls, encode with locale encoding
-            # Refer to http://stackoverflow.com/a/9951851/35070
-            encoding = preferredencoding()
-    else:
-        encoding = sys.getfilesystemencoding()
-    if encoding is None:
-        encoding = 'utf-8'
-    return s.encode(encoding, 'ignore')
+    # Pass '' directly to use Unicode APIs on Windows 2000 and up
+    # (Detecting Windows NT 4 is tricky because 'major >= 4' would
+    # match Windows 9x series as well. Besides, NT 4 is obsolete.)
+    if not for_subprocess and sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5:
+        return s
+
+    return s.encode(get_subprocess_encoding(), 'ignore')
+
+
+def decodeFilename(b, for_subprocess=False):
+
+    if sys.version_info >= (3, 0):
+        return b
+
+    if not isinstance(b, bytes):
+        return b
+
+    return b.decode(get_subprocess_encoding(), 'ignore')
 
 
 def encodeArgument(s):
@@ -408,6 +423,10 @@ def encodeArgument(s):
     return encodeFilename(s, True)
 
 
+def decodeArgument(b):
+    return decodeFilename(b, True)
+
+
 def decodeOption(optval):
     if optval is None:
         return optval
@@ -1791,6 +1810,59 @@ def match_filter_func(filter_str):
     return _match_func
 
 
+def parse_dfxp_time_expr(time_expr):
+    if not time_expr:
+        return 0.0
+
+    mobj = re.match(r'^(?P<time_offset>\d+(?:\.\d+)?)s?$', time_expr)
+    if mobj:
+        return float(mobj.group('time_offset'))
+
+    mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:\.\d+)?)$', time_expr)
+    if mobj:
+        return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3))
+
+
+def format_srt_time(seconds):
+    (mins, secs) = divmod(seconds, 60)
+    (hours, mins) = divmod(mins, 60)
+    millisecs = (secs - int(secs)) * 1000
+    secs = int(secs)
+    return '%02d:%02d:%02d,%03d' % (hours, mins, secs, millisecs)
+
+
+def dfxp2srt(dfxp_data):
+    _x = functools.partial(xpath_with_ns, ns_map={'ttml': 'http://www.w3.org/ns/ttml'})
+
+    def parse_node(node):
+        str_or_empty = functools.partial(str_or_none, default='')
+
+        out = str_or_empty(node.text)
+
+        for child in node:
+            if child.tag == _x('ttml:br'):
+                out += '\n' + str_or_empty(child.tail)
+            elif child.tag == _x('ttml:span'):
+                out += str_or_empty(parse_node(child))
+            else:
+                out += str_or_empty(xml.etree.ElementTree.tostring(child))
+
+        return out
+
+    dfxp = xml.etree.ElementTree.fromstring(dfxp_data.encode('utf-8'))
+    out = []
+    paras = dfxp.findall(_x('.//ttml:p'))
+
+    for para, index in zip(paras, itertools.count(1)):
+        out.append('%d\n%s --> %s\n%s\n\n' % (
+            index,
+            format_srt_time(parse_dfxp_time_expr(para.attrib.get('begin'))),
+            format_srt_time(parse_dfxp_time_expr(para.attrib.get('end'))),
+            parse_node(para)))
+
+    return ''.join(out)
+
+
 class PerRequestProxyHandler(compat_urllib_request.ProxyHandler):
     def __init__(self, proxies=None):
         # Set default handlers