Rewrites last-grob-cause function

Rewrites last-grob-cause function

[dak, 2013-01-16 10:11:49]

https://codereview.appspot.com/7092058/diff/1/scm/output-lib.scm
File scm/output-lib.scm (right):

https://codereview.appspot.com/7092058/diff/1/scm/output-lib.scm#newcode57
scm/output-lib.scm:57: (define-public (last-grob-cause grob)
The name is quite clearer, and the semantics make sense.  It is conceivable that
it would be more useful to return #f instead of the grob itself when there is no
preceding grob cause.  But I am not really sure about that, and this variant
_is_ consistent as well.  However, I should be surprised if previous callers
won't need to be notified of the renaming as well.  This has been committed so
recently that, in spite of this being a public function, a convert-ly rule seems
like overkill to me (it would basically cover one version or less).  But the
callers introduced with the same commit will need renaming, anyway.

[mike7, 2013-01-16 12:12:40]

On 16 janv. 2013, at 11:11, dak@gnu.org wrote:

> 
> https://codereview.appspot.com/7092058/diff/1/scm/output-lib.scm
> File scm/output-lib.scm (right):
> 
> https://codereview.appspot.com/7092058/diff/1/scm/output-lib.scm#newcode57
> scm/output-lib.scm:57: (define-public (last-grob-cause grob)
> The name is quite clearer, and the semantics make sense.  It is
> conceivable that it would be more useful to return #f instead of the
> grob itself when there is no preceding grob cause.  But I am not really
> sure about that, and this variant _is_ consistent as well.  However, I
> should be surprised if previous callers won't need to be notified of the
> renaming as well.  This has been committed so recently that, in spite of
> this being a public function, a convert-ly rule seems like overkill to
> me (it would basically cover one version or less).  But the callers
> introduced with the same commit will need renaming, anyway.
> 
> https://codereview.appspot.com/7092058/

Ah, this is cruft.  Just did a git grep and didn't find it used anywhere.  This
must have been left-over from an old patch set.
I'll just post a patch deleting it.

Cheers,
MS

[dak, 2013-01-16 12:37:23]

On 2013/01/16 12:12:40, mike7 wrote:

> Ah, this is cruft.  Just did a git grep and didn't find it used anywhere. 
This
> must have been left-over from an old patch set.
> I'll just post a patch deleting it.

Well, for the record, I found this function when looking for something to use in
the context of a discussion thread for removing an event-cause from a grob, and
of course you'd exactly need last-grob-cause for doing that.

On the other hand, a random and incomplete set of define-public utility
functions easily written on-demand and without an actual documentation string
and mention in the manual is not really going to help significantly.  Deletion
is probably the sanest course.  Likely for a whole lot of other "utility"
functions almost duplicating Guile library functionality as well.

[dak, 2013-01-21 09:28:25]

Well, this just removes an unused buggy function previously undocumented.  In my
book, this does not need a countdown but can be pushed straight to staging.  It
would make sense, however, to rename this issue to "Removes last-grob-cause
function" in case people look for it any time after.