# New Ticket Created by Zefram
# Please include the string: [perl #128511]
# in the subject line of all future correspondence about this issue.
# <URL: https://rt.perl.org/Ticket/Display.html?id=128511 >
A decode-then-encode cycle through the utf8-c8 encoding is meant to
round-trip an octet string. But sometimes it adds a spurious NUL to
the end of the string:
> Blob[uint8].new(233).decode("utf8-c8").encode("utf8-c8").perl
Blob[uint8].new(233,0)
> Blob[uint8].new(233, 128).decode("utf8-c8").encode("utf8-c8").perl
Blob[uint8].new(233,128,0)
This seems to happen whenever the end of the input string is a truncation
of the UTF-8 representation of a character in the Unicode range, and not
in any other case. In particular, for single-octet strings it happens
iff the octet value is between 194 (0xc2) and 244 (0xf4) inclusive.
-zefram
