X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Futils.py;h=8ccf254895133f38d1c8e095718a0d53d53bd061;hb=43ebf77df3bbd93dbbd0336b0243d8d50895ab72;hp=0d30075aa1dd982a6b0a95e717575ce217453533;hpb=9d30c2132acf2d12bfa8e559987c341c76d9cd24;p=youtube-dl

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 0d30075aa..8ccf25489 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -46,6 +46,7 @@ from .compat import (
     compat_html_entities,
     compat_html_entities_html5,
     compat_http_client,
+    compat_integer_types,
     compat_kwargs,
     compat_os_name,
     compat_parse_qs,
@@ -2794,6 +2795,15 @@ class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor):
     https_response = http_response
 
 
+class YoutubeDLRedirectHandler(compat_urllib_request.HTTPRedirectHandler):
+    if sys.version_info[0] < 3:
+        def redirect_request(self, req, fp, code, msg, headers, newurl):
+            # On python 2 urlh.geturl() may sometimes return redirect URL
+            # as byte string instead of unicode. This workaround allows
+            # to force it always return unicode.
+            return compat_urllib_request.HTTPRedirectHandler.redirect_request(self, req, fp, code, msg, headers, compat_str(newurl))
+
+
 def extract_timezone(date_str):
     m = re.search(
         r'^.{8,}?(?P<tz>Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)',
@@ -3519,10 +3529,11 @@ def str_or_none(v, default=None):
 
 def str_to_int(int_str):
     """ A more relaxed version of int_or_none """
-    if int_str is None:
-        return None
-    int_str = re.sub(r'[,\.\+]', '', int_str)
-    return int(int_str)
+    if isinstance(int_str, compat_integer_types):
+        return int_str
+    elif isinstance(int_str, compat_str):
+        int_str = re.sub(r'[,\.\+]', '', int_str)
+        return int_or_none(int_str)
 
 
 def float_or_none(v, scale=1, invscale=1, default=None):
@@ -5383,6 +5394,19 @@ def decode_packed_codes(code):
         obfucasted_code)
 
 
+def caesar(s, alphabet, shift):
+    if shift == 0:
+        return s
+    l = len(alphabet)
+    return ''.join(
+        alphabet[(alphabet.index(c) + shift) % l] if c in alphabet else c
+        for c in s)
+
+
+def rot47(s):
+    return caesar(s, r'''!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~''', 47)
+
+
 def parse_m3u8_attributes(attrib):
     info = {}
     for (key, val) in re.findall(r'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib):