On Sat, Sep 24, 2016 at 04:13:19PM -0700, Kevin J. McCarthy wrote: > The first is general. Mutt uses fixed size buffers for a lot of things, > but I think this might be one of the cases where we don't want to use a > LONG_STRING fixed length buffer. I realize this makes the > implementation even more complex, so I don't insist on it, but just > wanted to mention it. (I would much rather have your patch and put that > on the TODO list to fix than wait for your free time situation to > improve!)
Actually, after looking at the later patches using mutt_apply_replace(), this isn't as big a deal as I thought. They are using it for a display-only copy of the value, not changing the original value. If we were to add an option in the future to have ~s search disp_subj, then this could be an issue. But for now, again, I'd rather just have the buffer overruns fixed and submit as-is. -- Kevin J. McCarthy GPG Fingerprint: 8975 A9B3 3AA3 7910 385C 5308 ADEF 7684 8031 6BDA
signature.asc
Description: PGP signature
