Re: Fwd: report 3/3: latent unterminated buffer in imap_wordcasecmp()

On Fri, Jun 19, 2026 at 13:02 +0800, Kevin J. McCarthy wrote:

>However, this "utility" function is pretty lousy, and as Ian mentions 
>I can't think of any reason to copy b over to a tmp[] array.

Agreed.

>If we're going to rewrite it, why don't we just compute length of a, 
>grab the strcpn() retval as length of b, and if they match then just 
>use ascii_strncmp()?

That ought to work.

First thing I thought of was to combine everything in a single loop: 
caseless compare of a pair of characters, with a terminated by NUL, and 
b terminated by either NUL or whitespace. (Also, this function doesn't 
have to return strcmp()-style < == or >; can just return boolean, match 
or not.)

Is there some reason not to code a caseless compare of two characters 
in this function? Some reason we have to use ascii_strncmp() or whatever?