Fix for #1444: utf_8_sig.StreamReader was (indirectly through decode())

calling codecs.utf_8_decode() with final==True, which falled with incomplete
byte sequences. Fix and test by James G. Sack.
This commit is contained in:
Walter Dörwald 2007-11-19 12:41:10 +00:00
parent fc7e72d1c6
commit 183744d6b9
2 changed files with 55 additions and 5 deletions

View file

@ -84,12 +84,18 @@ class StreamReader(codecs.StreamReader):
pass
def decode(self, input, errors='strict'):
if len(input) < 3 and codecs.BOM_UTF8.startswith(input):
# not enough data to decide if this is a BOM
# => try again on the next call
return (u"", 0)
if len(input) < 3:
if codecs.BOM_UTF8.startswith(input):
# not enough data to decide if this is a BOM
# => try again on the next call
return (u"", 0)
elif input[:3] == codecs.BOM_UTF8:
self.decode = codecs.utf_8_decode
(output, consumed) = codecs.utf_8_decode(input[3:],errors)
return (output, consumed+3)
# (else) no BOM present
self.decode = codecs.utf_8_decode
return decode(input, errors)
return codecs.utf_8_decode(input, errors)
### encodings module API