Re: Issue 2026 (markup utility) should keep empty-markup positionless

[David Kastrup]

James <address@hidden> writes:

> On 3 February 2012 21:00, Marek Klein <address@hidden> wrote:
>> Hello,
>>
>> 2012/2/1 Robin Bannister <address@hidden>
>>
>>> In
>>> http://code.google.com/p/**lilypond/issues/detail?id=**2026#c32<http://code.google.com/p/lilypond/issues/detail?id=2026#c32>Ian
>>> Hulin wrote:
>>>
>>>> This is fixed if we move the definition of the \null markup to the before
>>>> the empty-markup declaration and change empty-markup to
>>>> (define-public (empty-markup)
>>>>   make-null-markup))
>>>> ...
>>>> It's built the docs successfully and run reg-tests OK.
>>>>
>>>
>>>
>>> OK, the docs say \null markup is an "empty markup", merely a single point,
>>> but it has _position_, which empty-markup does not.
>>> The following snippet demonstrates this difference:
>>> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>>> offset = \markup \translate #'(9 . 9) "offset"
>>> \markup \line {  "concat with empty-markup:" \box \concat { \empty-markup
>>> \offset }  " concat with null markup: "\box \concat { \null \offset }
>>> }
>>> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>>>
>>
>> Is this a bug report? I am not sure; 2026 is still open, is it not better
>> to add a coment to the issue?
>
> I think so, but to be honest only David or Ian can probably answer that.
>
> I'd hold off from creating a new tracker for now.

Well, David has no clue about markups.  If you are uncertain about
whether those issues are connected or interdependent in their
resolution, you can add an explicit issue for this one and enter the
other in the "Blocked-on" field of it.  If the person working on 2026
thinks the relation is the other way round, he can change that around
and/or add comments if the situation is not going to clear up
significantly soon enough to be able to mark everything as "Fixed".

-- 
David Kastrup