Re: Eliminates pure-print-callbacks list (issue 7300082)

[David Kastrup]

"address@hidden" <address@hidden> writes:

> On 20 févr. 2013, at 14:06, David Kastrup <address@hidden> wrote:
>
>> 
>> (define-public (constant-grob-callback callback)
>>  "Wraps a grob @var{callback} into an unpure-pure-container
>> delivering the same results for pure and unpure calculations.  This
>> means that the result of the callback must not depend on the line
>> spacing."
>>  (ly:make-unpure-pure-container
>>   callback
>>   (lambda (grob start end) (callback grob))))
>> 
>> Then look-and-behold, we can define
>> 
>> (define-public grob::constant-height-from-stencil
>>    (constant-grob-callback grob::stencil-height))
>> 
>> Now look-and-behold, we can get along without a separate comment for
>> each such definition since constant-grob-callback already tells us all
>> that we need to know (by virtue of its comment).
>> 
>
> This is an excellent idea - hadn't crossed my mind.

It is _exactly_ what I proposed in
<URL:https://codereview.appspot.com/7300082#msg3> in review.

> Please propose a patch if you have time.

There is no point in doing so.  As mentioned further down, the whole
thing can be obliterated by just extending unpure-pure-containers in a
natural way.  And anyway, you state

> I'm proposing another patch set in a day or so that eliminates the
> rest of the lists in define-grobs.scm and does everything via
> unpure-pure-containers.  After that I'll have a better idea if this is
> feasible.

So that would just be conflicting commits.  And I really fail to see the
point in me proposing patches to clean up after you when I already
invested the work of proposing fixes in review.

Where is the point in reviewing the stuff, proposing fixes, and then
having to clean up anything by oneself anyway?  If you are not
interested in maintaining your own code, how can you expect anybody else
to be?

-- 
David Kastrup