Re: Suggestion: SCM_DEBUG_REST_ARGUMENTS (fwd)

[Keisuke Nishida <kxn30 at po dot cwru dot edu>]

Dirk Herrmann <dirk@ida.ing.tu-bs.de> writes:

> However, for the long term I think we could really use macros like the
> following:
> 
> SCM_DEBUG_CELL_ACCESSES -- Check cell accesses exhaustively, i. e. verify
> that the cell pointers actually point to a cell on the heap.  scm_cellp is
> a predicate which already performs that kind of test.  We could go even
> further here, by also checking that cell elements written as objects are
> always read as objects, and that cell elements written as raw data are
> always read as raw data.  (see file doc/api.txt)
> 
> SCM_DEBUG_PAIR_ACCESSES -- This is what you suggested for pairs:  make
> sure that SCM_CAR and friends are only applied to real cons cells.

Those macros will be used like this, right?

SCM foo (SCM obj)
{
  SCM_DEBUG_PAIR_ACCESSES (obj);
  return SCM_CAR (obj);
}

In this case, I have to always think of where the object may be wrong
type.  Instead, I'd like to just write

SCM foo (SCM obj)
{
  return SCM_CAR (obj);
}

and if something goes wrong, I want to turn on the Guile's debug option.
Do you think it is possible?

> #ifndef SCM_DEBUG_DEPRECATED
> #define SCM_DEBUG_DEPRECATED SCM_DEBUG
> #endif
> 
> and make sure that SCM_DEBUG is defined as the very first of these
> macros.  This simple hack would allow to achieve what you have suggested,
> and would still allow to override those macros from the command line.
> 
> Cool, isn't it :-)

Yes, that's great :)

-- Kei