Re: [PATCH] Allow 'ptype/o' for assembly

[Philippe Waroquiers <philippe.waroquiers@skynet.be>]

On Wed, 2022-11-09 at 08:28 -0700, Tom Tromey via Gdb-patches wrote:
> > > PR exp/28359 points out that 'ptype/o' does not work when the current
> > > language is "asm".
> 
> Keith> I hesitate to bring this up, but...
> 
> Keith> Why would the current language matter in this case?
> Keith> [I understand that the print options are set before
> Keith> the expression is evaluated...]
> 
> Keith> Since we have the type of the symbol, couldn't we
> Keith> simply use that? Maybe it's not worth being so
> Keith> pedantic?
> 
> Ordinarily this would be fine, but for non-C languages it's sometimes
> useful to "set lang c" and see what the C type-printer thinks.  For
> example, in Ada there are "fat pointers" that are really just a kind of
> struct.  Ada will always print these as ordinary access types, but it
> can be handy to look under the hood and see that it's really a struct
> type.
> 
> Maybe there's some better way to do this kind of thing.  Like we could
> add "ptype/c" for "C style printing".
At work, I relatively often use ptype/o for Ada types.
As it only works in C mode, as part of our zillions of site aliases,
we have added:
  alias Wc = with language c -- with annotate 0 --
(where annotate 0 avoids spurious language switch messages in the output).

So, waiting for Adacore to implement ptype/o for Ada (hint hint :)),
we use e.g.  
  (gdb) Wc ptype/o sometype

Of course, as the type will be searched in C mode, we have to use the
"underlying" type name e.g.   This_Is_A_Package__This_Is_The_Ada_Type
instead of This_Is_A_Package.This_Is_The_Ada_Type

Of course, Wc will work with whatever you want to do with C language temporarily set.
(thanks again to Pedro for the "with" implementation).

Thanks
Philippe