Issue 800 melisma extenders

[Nicolas Sceaux]

Hi,

As was reported on <http://code.google.com/p/lilypond/issues/detail?id=800 
 >
a regression was introduced by patch  
7531ea6b3bb254f4d82aabde347a91671533b45a:

     Fix lyric extenders to end properly when there are notes but no  
more lyrics.

     The previous extender code only ends extenders when the next
     lyric is encountered. This can lead to issues when there are  
additional note
     but no more lyrics. This patch adds a check to see if a
     melisma is ending, completetizing the extender if appropriate.

As this patch aims at fixing a robustness, somewhat unusual, case, and  
breaks
the general behavior of melisma extender lines, may I propose to  
revert this patch?
It's really annoying.

Nicolas



_______________________________________________
bug-lilypond mailing list
bug-lilypond@...
http://lists.gnu.org/mailman/listinfo/bug-lilypond

[Chris Snyder-5]

Nicolas Sceaux wrote:
 > Hi,
 >
 > As was reported on
<http://code.google.com/p/lilypond/issues/detail?id=800>
 > a regression was introduced by patch
 > 7531ea6b3bb254f4d82aabde347a91671533b45a:
 >
 >     Fix lyric extenders to end properly when there are notes but no more
 > lyrics.
 >
 > As this patch aims at fixing a robustness, somewhat unusual, case, and
 > breaks
 > the general behavior of melisma extender lines, may I propose to revert
 > this patch?
 > It's really annoying.

In my opinion, the issue it addressed is much more annoying than
regression - I had multiple scores that would have extender lines going
for pages at a time, well after the lyrics were finished, to the end of
the piece.

I offered to fix this regression a few months ago, and sent a couple of
messages to lilypond-devel asking for help in determining the correct
approach to take. Unfortunately, my requests went unanswered (and my
second request, sent a month after the first, was rebuked as being too
demanding, even though I wasn't asking for someone to fix my pet issue
for me but was asking for advice from those more experienced with the
code so that my patch wouldn't be DOA).

Chris Snyder
Adoro Music Publishing
1-616-828-4436 x800
http://www.adoromusicpub.com





_______________________________________________
bug-lilypond mailing list
bug-lilypond@...
http://lists.gnu.org/mailman/listinfo/bug-lilypond

[Nicolas Sceaux]

Le 9 nov. 2009 à 19:43, Chris Snyder a écrit :

> Nicolas Sceaux wrote:
> > Hi,
> >
> > As was reported on <http://code.google.com/p/lilypond/issues/detail?id=800 
> >
> > a regression was introduced by patch
> > 7531ea6b3bb254f4d82aabde347a91671533b45a:
> >
> >     Fix lyric extenders to end properly when there are notes but  
> no more
> > lyrics.
> >
> > As this patch aims at fixing a robustness, somewhat unusual, case,  
> and
> > breaks
> > the general behavior of melisma extender lines, may I propose to  
> revert
> > this patch?
> > It's really annoying.
>
> In my opinion, the issue it addressed is much more annoying than  
> regression

I beg to disagree: it fixes a special case, but breaks extender lines in
general.

> I offered to fix this regression a few months ago, and sent a couple  
> of messages to lilypond-devel asking for help in determining the  
> correct approach to take. Unfortunately, my requests went unanswered  
> (and my second request, sent a month after the first, was rebuked as  
> being too demanding, even though I wasn't asking for someone to fix  
> my pet issue for me but was asking for advice from those more  
> experienced with the code so that my patch wouldn't be DOA).

I did not meant to offense you, but have your opinion: I have no idea  
of the
background of this modification. But melisma extender lines are just  
broken
now, this is something to consider. I can live without seeing your patch
reverted in master: I work with a custom LilyPond anyway, I just  
reverted it
on my working copy. But I don't think LilyPond is better now,  
regarding vocal
music.

Nicolas



_______________________________________________
bug-lilypond mailing list
bug-lilypond@...
http://lists.gnu.org/mailman/listinfo/bug-lilypond

[Chris Snyder-5]

Nicolas Sceaux wrote:
 >> In my opinion, the issue it addressed is much more annoying than
 >> regression
 >
 > I beg to disagree: it fixes a special case, but breaks extender lines in
 > general.

I disagree here - the cases where I encountered the bug were not
anything out of the ordinary (it's quite common for lyrics to end before
the end of a piece of music).

 > I did not meant to offense you, but have your opinion: I have no idea of
 > the
 > background of this modification. But melisma extender lines are just
broken
 > now, this is something to consider.

I contend that they were also broken before - just for a different use case.

What we can agree on is that the best solution would be for both bugs to
be fixed. I'll take a look and see what I can come up with.

(Sorry for the confrontational tone in my last message - I was
frustrated with the lack of response to my earlier requests for help in
addressing the issue, not anything you said)

Chris Snyder
Adoro Music Publishing
1-616-828-4436 x800
http://www.adoromusicpub.com


>
>> I offered to fix this regression a few months ago, and sent a couple
>> of messages to lilypond-devel asking for help in determining the
>> correct approach to take. Unfortunately, my requests went unanswered
>> (and my second request, sent a month after the first, was rebuked as
>> being too demanding, even though I wasn't asking for someone to fix my
>> pet issue for me but was asking for advice from those more experienced
>> with the code so that my patch wouldn't be DOA).
>
  I can live without seeing your patch
> reverted in master: I work with a custom LilyPond anyway, I just
> reverted it
> on my working copy. But I don't think LilyPond is better now, regarding
> vocal
> music.
>
> Nicolas
>


_______________________________________________
bug-lilypond mailing list
bug-lilypond@...
http://lists.gnu.org/mailman/listinfo/bug-lilypond