Merge remote-tracking branch 'Boris-de/wdrmaus_fix#8562'

[youtube-dl] / youtube_dl / utils.py

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 25a9f33c02ff1aae6e13688c97109bc128fabb56..89234b39df9cdfe3edbce0420fee5ec5b67c4575 100644 (file)
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -105,9 +105,9 @@ KNOWN_EXTENSIONS = (
     'f4f', 'f4m', 'm3u8', 'smil')
 
 # needed for sanitizing filenames in restricted mode
-ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØŒÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõöøœùúûüýþÿ',
-                        itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOO', ['OE'], 'UUUUYP', ['ss'],
-                                        'aaaaaa', ['ae'], 'ceeeeiiiionoooooo', ['oe'], 'uuuuypy')))
+ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ',
+                        itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUYP', ['ss'],
+                                        'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuypy')))
 
 
 def preferredencoding():
@@ -861,9 +861,13 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
                 # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3
                 if sys.version_info >= (3, 0):
                     location = location.encode('iso-8859-1').decode('utf-8')
+                else:
+                    location = location.decode('utf-8')
                 location_escaped = escape_url(location)
                 if location != location_escaped:
                     del resp.headers['Location']
+                    if sys.version_info < (3, 0):
+                        location_escaped = location_escaped.encode('utf-8')
                     resp.headers['Location'] = location_escaped
         return resp
 
@@ -883,12 +887,17 @@ def make_socks_conn_class(base_class, socks_proxy):
     elif url_components.scheme.lower() == 'socks4a':
         socks_type = ProxyType.SOCKS4A
 
+    def unquote_if_non_empty(s):
+        if not s:
+            return s
+        return compat_urllib_parse_unquote_plus(s)
+
     proxy_args = (
         socks_type,
         url_components.hostname, url_components.port or 1080,
         True,  # Remote DNS
-        compat_urllib_parse_unquote_plus(url_components.username),
-        compat_urllib_parse_unquote_plus(url_components.password),
+        unquote_if_non_empty(url_components.username),
+        unquote_if_non_empty(url_components.password),
     )
 
     class SocksConnection(base_class):
@@ -1030,6 +1039,7 @@ def unified_strdate(date_str, day_first=True):
         format_expressions.extend([
             '%d-%m-%Y',
             '%d.%m.%Y',
+            '%d.%m.%y',
             '%d/%m/%Y',
             '%d/%m/%y',
             '%d/%m/%Y %H:%M:%S',
@@ -1050,7 +1060,10 @@ def unified_strdate(date_str, day_first=True):
     if upload_date is None:
         timetuple = email.utils.parsedate_tz(date_str)
         if timetuple:
-            upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d')
+            try:
+                upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d')
+            except ValueError:
+                pass
     if upload_date is not None:
         return compat_str(upload_date)
 
@@ -1544,15 +1557,11 @@ def setproctitle(title):
 
 
 def remove_start(s, start):
-    if s.startswith(start):
-        return s[len(start):]
-    return s
+    return s[len(start):] if s is not None and s.startswith(start) else s
 
 
 def remove_end(s, end):
-    if s.endswith(end):
-        return s[:-len(end)]
-    return s
+    return s[:-len(end)] if s is not None and s.endswith(end) else s
 
 
 def remove_quotes(s):
@@ -1906,7 +1915,7 @@ def parse_age_limit(s):
 
 def strip_jsonp(code):
     return re.sub(
-        r'(?s)^[a-zA-Z0-9_.]+\s*\(\s*(.*)\);?\s*?(?://[^\n]*)*$', r'\1', code)
+        r'(?s)^[a-zA-Z0-9_.$]+\s*\(\s*(.*)\);?\s*?(?://[^\n]*)*$', r'\1', code)
 
 
 def js_to_json(code):
@@ -1926,14 +1935,14 @@ def js_to_json(code):
             }.get(m.group(0), m.group(0)), v[1:-1])
 
         INTEGER_TABLE = (
-            (r'^(0[xX][0-9a-fA-F]+)', 16),
-            (r'^(0+[0-7]+)', 8),
+            (r'^0[xX][0-9a-fA-F]+', 16),
+            (r'^0+[0-7]+', 8),
         )
 
         for regex, base in INTEGER_TABLE:
             im = re.match(regex, v)
             if im:
-                i = int(im.group(1), base)
+                i = int(im.group(0), base)
                 return '"%d":' % i if v.endswith(':') else '%d' % i
 
         return '"%s"' % v
@@ -2031,11 +2040,7 @@ def mimetype2ext(mt):
 
 
 def urlhandle_detect_ext(url_handle):
-    try:
-        url_handle.headers
-        getheader = lambda h: url_handle.headers[h]
-    except AttributeError:  # Python < 3
-        getheader = url_handle.info().getheader
+    getheader = url_handle.headers.get
 
     cd = getheader('Content-Disposition')
     if cd: