On Tuesday 19 February 2002 4:53 pm, Angus Leeming wrote:
> On Tuesday 19 February 2002 4:42 pm, Jules Bean wrote:
> > On Tue, Feb 19, 2002 at 04:28:39PM +0000, Angus Leeming wrote:
> > > On Tuesday 19 February 2002 4:18 pm, John Levon wrote:
> > > > On Tue, Feb 19, 2002 at 03:59:09PM +0000, Angus Leeming wrote:
> > > >
> > > > > I'd like to commit the attached patch.
> > > >
> > > > are you trying to confuse :)
> > > >
> > > > + if (pid<=0) { // Fork failed.
> > > > + retval = 1;
> > >
> > > Sorry, I don't follow you. We tried to fork and failed, so return an
> > > indication of such to the function that invoked startscript. Why is
that
> > > confusing.
> >
> > The comment is wrong. <=0 is not failure. <0 is failure, but ==0 is
> > 'this is the child process'.
> >
> > Jules
>
> Ok, thank you. This confusion is cleaned up by redefining the member
function
> int fork();
> as
> bool fork_child();
Or alternatively, changing the comment to:
// Fork and/or execvp failed.
Angus
