[m-dev.] for review: rename base_type_* to type_ctor_* (part 1/2)

[Tyson Dowd]

On 18-Mar-1999, Fergus Henderson <fjh at cs.mu.OZ.AU> wrote:
> On 18-Mar-1999, Tyson Dowd <trd at cs.mu.OZ.AU> wrote:
> > On 18-Mar-1999, Fergus Henderson <fjh at cs.mu.OZ.AU> wrote:
> > > On 17-Mar-1999, Tyson Dowd <trd at cs.mu.OZ.AU> wrote:
> > > > During the writing of the RTTI paper, we decided that type_ctor_info
> > > > was a much better name than base_type_info.
> > > > 
> > > > Rename base_type* as type_ctor*, except we don't rename the modules
> > > > base_type_info and base_type_layout just yet.
> > > 
> > > The sed script you posted has
> > > 
> > > 	s/type_ctor_info__/base_type_info__/g
> > > 	s/type_ctor_layout__/base_type_layout__/g
> > > 
> > > so it looks there are some additional exceptions.
> > > Could you explain why?
> > 
> > These aren't extra, that's just to turn the module qualifier
> > `type_ctor_info__' back into `base_type_info__' (similarly for
> > `type_ctor_layout__'.
> 
> Oh, I see.
> 
> > > Did you modify util/mdemangle.c and profiler/demangle.m?
> > 
> > No, not yet.  I will do as a separate change if that's OK.
> 
> I suppose that would be OK at a pinch.  But ideally I think it would be
> better to do that at the same time, because if committed alone, this
> change would effectively break the demangler.

Ok, I'll do that now then.

> 
> > The only hand-written changes were:
> > 	- changes to mercury_bootstrap.h
> > 	- change type_ctor_info in :- import_module back to
> > 	  base_type_info.
> 
> Fine.  Incidentally, for the second one, I think the sed command 
> 
> 	/import_module/s/base_type_info/type_ctor_info/g
> 
> would do the trick.

Actually.
	/import_module/s/type_ctor_info/base_type_info/g
since we want to change type_ctor_info *back* into base_type_info.

I'll put this into the sed command list in the logfile.

-- 
The quantum sort: 
	while (!sorted) { do_nothing(); }
Tyson Dowd   <tyson at tyse.net>   http://tyse.net/