Re: [PATCH] Fix -save-temps from clobbering input file@apple.com

[Daniel Jacobowitz <drow at mvista dot com>]

On Wed, Oct 31, 2001 at 12:58:10AM +0100, degger@fhm.edu wrote:
> On 30 Oct, Ira Ruben wrote:
> 
> > But that isn't the issue.  IMO, the compiler never has a "right" to 
> > implicitly clobber user files.
> 
> I tend to disagree here. The user should be grown up enough to decide
> what the computer should do and what not. You could argue the same about
> object files, they're also implicitely clobbered by gcc when using the
> -c switch but that's exactly what one expects. When I use -save-temps
> (which in fact I do very often to check the produced assembly) then
> I know that it'll wipe away all .i, .s, and .ii files and if I need them
> (for example to produce diffs) I have to back them up. I would find it
> quite disgusting to notice that gcc -save-temps doesn't overwrite .s
> files because they are already existant from the last run....

That's not what Ira's patch addresses, if I understand correctly.  He
was preprocessing a .s file and -save-temps was writing out a .s file
with the same name as the specified input file.  The string comparison
he adds is case sensitive, and against the specified -input- file.

-- 
Daniel Jacobowitz                           Carnegie Mellon University
MontaVista Software                         Debian GNU/Linux Developer