Re: scm_must_malloc() vs. malloc()

[Martin Grabmueller]

> From: Dirk Herrmann <address@hidden>
> Date: Tue, 13 Mar 2001 17:48:39 +0100 (MET)
> 
> On Tue, 13 Mar 2001, Martin Grabmueller wrote:
> 
> > Additionally, gh_data.c contains a lot of code like this:
> > 
> >       if (m == 0)
> >     m = (short *) malloc (n * sizeof (short));
> >       for (i = 0; i < n; ++i)
> >     m[i] = SCM_INUM (SCM_VELTS (obj)[i]);
> > 
> > I think these should get replaced by calls to scm_must_malloc too.
> > We don't want segfaults, do we?
> 
> These should, IMO, _not_ be replaced by scm_must_malloc.  Let me try
> to give an explanation about when scm_must_malloc should be used: 

[snip explanation]

So the rule of thumb is: If we return something which the caller must
free() herself, we use malloc, otherwise scm_must_malloc()?

> Now, looking at the code in gh_data.c: Some functions convert scheme
> vectors into a corresponding C level array.  The C level array is
> created with malloc.  But, these arrays are not under control of the
> garbage collector.  It completely depends on the calling function
> what it will do with them.  Thus, in these cases you should not use
> scm_must_malloc.

Then I suppose it's correct to change `gh_scm2newstr', `symbol2newstr'
to use malloc(), because it returns malloc()ed memory which the caller
must free.  I'll fix that.

Regards,
  'martin