Re: Issue 881 Arpeggios may collide with laissezVibrer ties (was Re: Critical issues)

[Carl Sorensen]

On 5/15/10 1:12 AM, "Karl Hammar" <address@hidden> wrote:

> Carl Sorensen:
> ...
>> I've posted a patch on Rietveld.  Can you do the
>> regression test?
>> http://codereview.appspot.com/1195044
> 
> After a make test-redo I get:
> 
> . the "mandatory" output-distance.
> . a diff of tree.gittext, showing Carls patch
> . 314 below threshold
> . 2062 unchanged
> 
> From this I assume Carls patch does not affect any present regression
> test.

Yes, this is correct.  Thanks for doing that!

> 
> ***
> 
> But I also notice that there are no test for the present problem.
> Perhaps one should be added?

Yes, it should.  I forgot to do that.  The one present in the report would
be the proper one to use.  Thanks for the catch!

> ***
> 
> I also note that input/regression/laissez-vibrer-ties.ly has a bar
> check failure:
> 
> \relative c' {
>   <c e g>\laissezVibrer r4
>   <c f g>\laissezVibrer r
>   <c d f g>\laissezVibrer r
>   <c d f g>4.\laissezVibrer r % here we get a bar check failure
> 
> In Documentation/snippets/laissez-vibrer-ties.ly we have a r8 at that
> point, otherwise thoose two are basically the same.
> 
> What was the point of having a bar check failure in the regression
> test?

I suspect nothing -- it's just an error.

Thanks,

Carl