Ludovic Brenta <[email protected]> writes:

> I received a complaint about the error handling inside ada-goto-
> declaration.  Suppose point is on the Q in the line
>
>   type Q is new Pak1.Pak2 (parameters...);
>
> and the user types C-c C-d (ada-goto-declaration).  This runs
> gnatfind in a new buffer; gnatfind only the "spec" line for
> package Q, since there is no body.  This triggers this part
> of ada-gnat-xref-other:
>
>       (when (eobp)
>         (pop-to-buffer (current-buffer))
>         (error "gnat find did not return other item"))
>
> The pop-to-buffer part annoys people who are not interested in
> debugging ada-mode; they prefer nothing to change on their
> screen rather than seeing the buffer where gnatfind was running
> :)

Yes, that makes sense.

Maybe we need an "ada-mode-debug" flag for this; 0 for normal use, 1 for
some debug, 2 for intense debug, as for wisi-debug.

> It could be argued that there is no error to report since not
> all entities in Ada have an "other item" to find.

It's actually a user error; what were they expecting C-c C-d to do,
since there is no full type for Q?

But there should be a gentler response to the user error. Just a
message: "no corresponding item found".

-- 
-- Stephe

_______________________________________________
Emacs-ada-mode mailing list
[email protected]
http://host114.hostmonster.com/mailman/listinfo/emacs-ada-mode_stephe-leake.org

Reply via email to