Re: oops? read/write vs type of length parameter

[Eli Zaretskii]

> Date: Tue, 12 Apr 2011 01:19:10 -0700
> From: Paul Eggert <address@hidden>
> CC: address@hidden, Jim Meyering <address@hidden>
> 
> I added a runtime check for this, which I don't think
> will ever fail, but I've been surprised in the past.

If it ever fails, aborting is too harsh, I think.  The original code
was well defended against that possibility, see write-region.  It
would signal an IO error.

> With that check in place we might as well use size_t for the size,

Which will cause annoying compiler warnings, at least with some
optional switches.

> with the goal of removing the runtime checks once we have
> carefully checked that they aren't needed.

Which will never happen, so these aborts will stay in the code
forever.

> Here's the patch I installed for that.

I don't understand why you went ahead and installed such a change,
although it was clear that your opinion on this is being disputed, and
at least I explicitly expressed my disagreement with changing that
argument to an unsigned type.  As long as you are not the head
maintainer, I think such unilateral actions are inappropriate.

But if I'm the only one who objects to that (both the change and
disregarding the disagreement), then so be it.