Ludovic Brenta <[email protected]> writes:

> Stephen Leake <[email protected]> writes:
>> Stephen Leake <[email protected]> writes:
>>
>>> To investigate further:
>>>
>>> <snip>
>>
>> Or we can do a GotoMeeting session, so I can watch you reproduce it.
>
> That won't be necessary; I think the problem was in one of the bugs you
> solved.  I reverted my change to gnat-core.el and verified that the
> variable 'default-directory' indeed has the value "/tmp", designating
> the directory where the project file lives; gnat find uses this project
> file all right.

Ok, good.

> Later, I found out that generating a .gpr project file that extends the
> top-level project file was the wrong approach.  As soon as a developer
> checks a file out of a lower-level project, gnat find and gnat ls
> complain that a source file cannot belong to two projects (one copy of
> the file is in the developer's view and belongs to the generated,
> extending project, and another copy is in the read-only baseline and
> belongs to the baseline project which is not top-level), so navigation
> is again broken.  So, I've reverted to generating an Ada mode project
> path instead and it seems to work like a charm, so far.

Something you can't do in GPS; Emacs to the rescue :).

> Thanks for your help!

You're welcome.

-- 
-- Stephe

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

Reply via email to