On 10 Jun 2016, at 17:34, Stephen Leake <[email protected]> wrote:
> 
> Simon Wright <[email protected]> writes:

>>> But (setq ada-indent-comment-gnat t) fixes that:
>>> 
>>>  type A is new B with
>>>  -- a comment; TAB on this line results in error
>>>     record
>> 
>> Here it made no difference??
> 
> Hmm; it worked for me. It's buffer-local; did you set it in the correct 
> buffer?

Yes, and checked.

If I manually indent to a multiple of 3 columns then sure enough TAB will leave 
it there, but otherwise TAB uses the computed value (5 in this case) which 
gives the style warning. Is this as designed? (no complaints, just a question!)

>> But then I noticed that GNAT complains about this as well:
>> 
>>   type A3 is
>>     --  a comment.
>>      record
>>         C : Integer;
>>      end record;
> 
> I would prefer to align the comment with 'record' in these cases; that
> would require looking ahead to see if 'record' is there.

I've never come across this problem myself, because I've never felt the urge to 
put a comment in such a place!
i agree about aligning with 'record'.

But since GPS is just as inconsistent; iin the original case it chooses our 
current layout which gives GNAT style errors, in the case in my para above (A3) 
it aligns the comment with 'type'.


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

Reply via email to