[rfc] cp-symtab.c?

[rfc] cp-symtab.c?

[David Carlton]

Elena and I were recently talking about moving
make_symbol_overload_list and friends from symtab.c to some C++ file.
It's not obvious to me where to put it, though; should I create a new
file "cp-symtab.c" for that?  If so, should I move the parts of
cp-namespace.c that deal specifically with symbol lookup (all the
functions in there with "lookup" in their name, I guess) to that file
as well?  My opinion is that the answer to both questions is "yes",
but I wanted to check with other people first.

David Carlton
carlton@math.stanford.edu

Re: [rfc] cp-symtab.c?

[Elena Zannoni]

David Carlton writes:
 > Elena and I were recently talking about moving
 > make_symbol_overload_list and friends from symtab.c to some C++ file.
 > It's not obvious to me where to put it, though; should I create a new
 > file "cp-symtab.c" for that?  If so, should I move the parts of
 > cp-namespace.c that deal specifically with symbol lookup (all the
 > functions in there with "lookup" in their name, I guess) to that file
 > as well?  My opinion is that the answer to both questions is "yes",
 > but I wanted to check with other people first.
 > 
 > David Carlton
 > carlton@math.stanford.edu

I would hold back on the new file, *for now*. I think cp-support.c
would work.  Doing a grep for ^lookup_ in *.c throws up too many
things that are not clearly symbol table related, but for instance
gdbtypes and charset related. My guess is that the term 'lookup' has
been abused, and we should introduce something more qualifying in the
function names.

elena

Re: [rfc] cp-symtab.c?

[Daniel Jacobowitz]

On Mon, Jun 02, 2003 at 03:16:15PM -0400, Elena Zannoni wrote:
> David Carlton writes:
>  > Elena and I were recently talking about moving
>  > make_symbol_overload_list and friends from symtab.c to some C++ file.
>  > It's not obvious to me where to put it, though; should I create a new
>  > file "cp-symtab.c" for that?  If so, should I move the parts of
>  > cp-namespace.c that deal specifically with symbol lookup (all the
>  > functions in there with "lookup" in their name, I guess) to that file
>  > as well?  My opinion is that the answer to both questions is "yes",
>  > but I wanted to check with other people first.
>  > 
>  > David Carlton
>  > carlton@math.stanford.edu
> 
> I would hold back on the new file, *for now*. I think cp-support.c
> would work.  Doing a grep for ^lookup_ in *.c throws up too many
> things that are not clearly symbol table related, but for instance
> gdbtypes and charset related. My guess is that the term 'lookup' has
> been abused, and we should introduce something more qualifying in the
> function names.

I agree.

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer

Re: [rfc] cp-symtab.c?

[David Carlton]

On Mon, 2 Jun 2003 15:12:27 -0400, Daniel Jacobowitz <drow@mvista.com> said:
> On Mon, Jun 02, 2003 at 03:16:15PM -0400, Elena Zannoni wrote:

>> I would hold back on the new file, *for now*. I think cp-support.c
>> would work.  Doing a grep for ^lookup_ in *.c throws up too many
>> things that are not clearly symbol table related, but for instance
>> gdbtypes and charset related. My guess is that the term 'lookup'
>> has been abused, and we should introduce something more qualifying
>> in the function names.

> I agree.

Thanks, that's what I'll do then.  I'll post an RFA in a bit (maybe
today, maybe later this week.)

David Carlton
carlton@math.stanford.edu