#2834: Change re module semantics, so that str and bytes mixing is forbidden,
and str (unicode) patterns get full unicode matching by default. The re.ASCII
flag is also introduced to ask for ASCII matching instead.
diff --git a/Lib/base64.py b/Lib/base64.py
index fc80835..4308fb4 100755
--- a/Lib/base64.py
+++ b/Lib/base64.py
@@ -39,7 +39,7 @@
return s.translate(translation)
-
+
# Base64 encoding/decoding uses binascii
def b64encode(s, altchars=None):
@@ -126,7 +126,7 @@
return b64decode(s, b'-_')
-
+
# Base32 encoding/decoding must be done in Python
_b32alphabet = {
0: b'A', 9: b'J', 18: b'S', 27: b'3',
@@ -225,7 +225,7 @@
# characters because this will tell us how many null bytes to remove from
# the end of the decoded string.
padchars = 0
- mo = re.search('(?P<pad>[=]*)$', s)
+ mo = re.search(b'(?P<pad>[=]*)$', s)
if mo:
padchars = len(mo.group('pad'))
if padchars > 0:
@@ -262,7 +262,7 @@
return b''.join(parts)
-
+
# RFC 3548, Base 16 Alphabet specifies uppercase, but hexlify() returns
# lowercase. The RFC also recommends against accepting input case
# insensitively.
@@ -291,12 +291,12 @@
raise TypeError("expected bytes, not %s" % s.__class__.__name__)
if casefold:
s = s.upper()
- if re.search('[^0-9A-F]', s):
+ if re.search(b'[^0-9A-F]', s):
raise binascii.Error('Non-base16 digit found')
return binascii.unhexlify(s)
-
+
# Legacy interface. This code could be cleaned up since I don't believe
# binascii has any line length limitations. It just doesn't seem worth it
# though. The files should be opened in binary mode.
@@ -353,7 +353,7 @@
return binascii.a2b_base64(s)
-
+
# Usable as a script...
def main():
"""Small main program"""