[MINC-development] patches for tmpnam

[Vicka Corey]

On Wed, Mar 05, 2003 at 10:07:06PM -0500, Steve ROBBINS wrote:
> On Mon, Mar 03, 2003 at 05:06:51PM -0500, Vicka Corey wrote:
> > Hi -- here are the changes I've put in to get rid of gcc warnings
> > from calls to tmpnam().
> 
> The changes you propose will cause file descriptor leakage which
> is not acceptable, especially not in a library.
> 
> It's true that you can get around that by closing the file
> descriptor returned by mkstemp(), but that really misses the
> point of the warnings.  The linker is warning you of a race
> condition.  If the tmpnam() calls are going to be replaced,
> the race ought to be fixed.  There's no point in doing it just
> to get rid of a link warning.
 
Since the warnings were causing various of our developers to modify
private copies of your code and use different versions of it and of 
the compilers, actually getting rid of the warnings is *very* 
important to us.  It's hard to debug your code when you have to
wade through a lot of irrelevant output, which is why I'm trying
to request changes in the MNI code in the first place.

Fixing race condition and leaks is worth doing too, of course, but 
please don't minimize the need for code to compile cleanly.  

Did the tmpnam() and tempnam() calls in the code clean up after themselves?
If so, the same closes should still work; if not it is at least no worse.

> P.S.  GCC 3 no longer emits this particular warning, anyway.

See above about developers using different versions of compilers.

> P.P.S. The man page for mkstemp() on my system says "Don't use this function,
> use tmpfile(3) instead.  It is better defined and more portable."

Under gcc 3.1, the tmpfile() man page says,
BUGS
     These interfaces are provided for System V and ANSI compatibility only.
     The mkstemp(3) interface is strongly preferred.
[...]

--vicka