Adds a glyph for tied lyrics.

Adds a glyph for tied lyrics.

[Bertrand Bordage, 2011-08-06 22:59:38]

Better handling of lyric ties:
* The space between two tied words is no more fixed. Its value is word-space.
The default space is therefore decreased.
* New glyph, so that we won't need an external font.
* Increased vertical space between words and tie. It is therefore now possible
to have commas in tied lyrics without any collision (example:  verte,~et).

How to build it:
rm mf/out/feta*
make

Regards,
Bertrand

[lemzwerg, 2011-08-07 05:49:20]

Please use tabs in MF files.  Besides that, everythings looks fine.

[pkx166h, 2011-08-07 09:11:06]

a 'while you're there' nitpick.

http://codereview.appspot.com/4808074/diff/18/Documentation/notation/vocal.itely
File Documentation/notation/vocal.itely (right):

http://codereview.appspot.com/4808074/diff/18/Documentation/notation/vocal.it...
Documentation/notation/vocal.itely:612: code the tilde symbol (@code{~}) to get
a lyric tie.
Hello, I know this was not your mistake but could we (while we are touching this
file) delete the first instance of the word 'code' so that it reads

"Alternatively, you can use the tilde ..."

[Janek Warchol, 2011-08-07 12:48:56]

I've created tracker issue:
http://code.google.com/p/lilypond/issues/detail?id=1807

Could you upload some pngs showing before/after output?  It would be helpful for
those who cannot build the patch themselves.

[Bertrand Bordage, 2011-08-08 14:30:36]

Changes done.
Passes make, regtest comparison and make doc.

Bertrand

[pkx166h, 2011-08-08 21:05:45]

Passes make and reg tests

[Janek Warchol, 2011-08-08 23:23:29]

I'd make the tie slightly shorter.  I find it awkward that two adjacent ties
collide so easily, for example here  
{
  \time 3/4
  \relative c' { c2 e4 g2 e4 }
  \addlyrics { gran- de_a- mi- go }
  \addlyrics { pu- "ro y ho-" nes- to }
  \addlyrics { pu- ro~y~ho- nes- to }
}
(example from documentation), or in the example you attached on the tracker
issue page.

cheers,
Janek

[Bertrand Bordage, 2011-08-09 08:59:03]

So do I, but a shorter tie would collide with commas. The best solution is
probably to shorten the tie and lower it according to the Y-extent of the
covered letters. But this requires much more work.
Maybe having two sizes of lyric ties would also be a good idea: a short one and
the default one.

Bertrand

[pacovila, 2011-08-10 15:35:27]

2011/8/9  <lemniskata.bernoullego@gmail.com>:
> I'd make the tie slightly shorter.  I find it awkward that two adjacent
> ties collide so easily, for example here
> {
>  \time 3/4
>  \relative c' { c2 e4 g2 e4 }
>  \addlyrics { gran- de_a- mi- go }
>  \addlyrics { pu- "ro y ho-" nes- to }
>  \addlyrics { pu- ro~y~ho- nes- to }
> }
> (example from documentation), or in the example you attached on the
> tracker issue page.

I have never liked this example in the documentation.

Firstly, it seems to mix Portuguese and Spanish.

Secondly, it does not follow the policy for syllable separator, which
is '[space]--[space]', not '-[space]'.

Finally, 2nd and 3rd stanzas look _very_ improbable to me in that it
has three syllables on a single note, which requires two lyric ties.
o~y is a synalepha and it could be a in-word diphthong, y~ho is
another synalepha, but o~y~ho is not a triphthong and can not be a
three-vowel synalepha.

If nobody finds a real example in literature, I suggest to remove the
problematic case, it is too artificial.
-- 
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com

[Bertrand Bordage, 2011-08-10 17:54:52]

> Secondly, it does not follow the policy for syllable separator, which
> is '[space]--[space]', not '-[space]'.

I already fixed it in this patch (for every language).

> Finally, 2nd and 3rd stanzas look _very_ improbable to me in that it
> has three syllables on a single note, which requires two lyric ties.
> o~y is a synalepha and it could be a in-word diphthong, y~ho is
> another synalepha, but o~y~ho is not a triphthong and can not be a
> three-vowel synalepha.
> 
> If nobody finds a real example in literature, I suggest to remove the
> problematic case, it is too artificial.

I agree, I never saw such a case.

[reinhold_kainhofer.com, 2011-08-10 21:00:33]

Am Mittwoch, 10. August 2011, 19:54:52 schrieb bordage.bertrand@gmail.com:
> > Finally, 2nd and 3rd stanzas look _very_ improbable to me in that it
> > has three syllables on a single note, which requires two lyric ties.
> > o~y is a synalepha and it could be a in-word diphthong, y~ho is
> > another synalepha, but o~y~ho is not a triphthong and can not be a
> > three-vowel synalepha.
> > 
> > If nobody finds a real example in literature, I suggest to remove the
> > problematic case, it is too artificial.
> 
> I agree, I never saw such a case.

I can't find it now, but I definitely remember having seen three syllables in 
a soprano aria (I think it was Italian). It was something like "-- to e in".

Unfortunately, I really can't find it any more. I thought it was in Rossini's 
Stabat Mater, but that was wrong.

Cheers,
Reinhold

-- 
------------------------------------------------------------------
Reinhold Kainhofer, reinhold@kainhofer.com, http://reinhold.kainhofer.com/
 * Financial & Actuarial Math., Vienna Univ. of Technology, Austria
 * http://www.fam.tuwien.ac.at/, DVR: 0005886
 * LilyPond, Music typesetting, http://www.lilypond.org

[wl_gnu.org, 2011-08-10 21:22:10]

> I can't find it now, but I definitely remember having seen three
> syllables in a soprano aria (I think it was Italian). It was
> something like "-- to e in".

An example is the second aria of Susanna in Mozart's `Le Nozze di
Figaro', bar 16:

   f2     f8  e8       g8        c8

  fin -- chè l'a -- "ria è an" -- cor

Almost all singers I've met during my work as a coach have problems if
they sing it the first time :-)

However, I've *never* seen lyric ties in any edition of Mozart, be it
Urtext or something else.


    Werner

[wl_gnu.org, 2011-08-10 21:27:04]

> An example is the second aria of Susanna in Mozart's `Le Nozze di
> Figaro', bar 16:
> 
>    f2     f8  e8       g8        c8
> 
>   fin -- chè l'a -- "ria è an" -- cor

BTW, the next bar is

  a8[ d8]     f8     g,8[ b8] d8

  bru --  "na, e il" mon --   do

:-)


     Werner

[pacovila, 2011-08-10 22:11:07]

2011/8/10 Werner LEMBERG <wl@gnu.org>:
>> I can't find it now, but I definitely remember having seen three
>> syllables in a soprano aria (I think it was Italian). It was
>> something like "-- to e in".
>
> An example is the second aria of Susanna in Mozart's `Le Nozze di
> Figaro', bar 16:
>
>   f2     f8  e8       g8        c8
>
>  fin -- chè l'a -- "ria è an" -- cor
>
> Almost all singers I've met during my work as a coach have problems if
> they sing it the first time :-)

That makes four vowels! i+a+e+a

Here http://cosinasdeleon.blogspot.com/2009/07/hinmo-leon.html are two
different instances of 'ioa' but they come from two words, not three.
First "-- gio~a" , then "Dio~a".

-- 
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com

[pacovila, 2011-08-10 22:12:47]

2011/8/10 Werner LEMBERG <wl@gnu.org>:
>> I can't find it now, but I definitely remember having seen three
>> syllables in a soprano aria (I think it was Italian). It was
>> something like "-- to e in".
>
> An example is the second aria of Susanna in Mozart's `Le Nozze di
> Figaro', bar 16:
>
>   f2     f8  e8       g8        c8
>
>  fin -- chè l'a -- "ria è an" -- cor
>
> Almost all singers I've met during my work as a coach have problems if
> they sing it the first time :-)

That makes four vowels! i+a+e+a

Here http://cosinasdeleon.blogspot.com/2009/07/hinmo-leon.html are two
different instances of 'ioa' but they come from two words, not three.
First "-- gio~a" , then "Dio~a".

-- 
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com

[pacovila, 2011-08-10 22:24:45]

2011/8/10 Werner LEMBERG <wl@gnu.org>:
>> An example is the second aria of Susanna in Mozart's `Le Nozze di
>> Figaro', bar 16:
>>
>>    f2     f8  e8       g8        c8
>>
>>   fin -- chè l'a -- "ria è an" -- cor
>
> BTW, the next bar is
>
>  a8[ d8]     f8     g,8[ b8] d8
>
>  bru --  "na, e il" mon --   do
>
> :-)

Thanks. It is available on page 346 of

 
http://216.129.110.22/files/imglnks/usimg/6/66/IMSLP25315-PMLP03845-Mozart_Fi...

redirected from

  http://imslp.org/wiki/Special:ImagefromIndex/25315

linked in

  http://imslp.org/wiki/Le_nozze_di_Figaro,_K.492_(Mozart,_Wolfgang_Amadeus)

-- 
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com

[wl_gnu.org, 2011-08-11 05:35:14]

>> An example is the second aria of Susanna in Mozart's `Le Nozze di
>> Figaro', bar 16:
>>
>>   f2     f8  e8       g8        c8
>>
>>  fin -- chè l'a -- "ria è an" -- cor
>>
>> Almost all singers I've met during my work as a coach have problems
>> if they sing it the first time :-)
> 
> That makes four vowels! i+a+e+a

Yes :-) Since this is completely unexpected (and I don't know any
other work of Mozart with a similar situation), people are stumbling
there.

> Here http://cosinasdeleon.blogspot.com/2009/07/hinmo-leon.html are
> two different instances of 'ioa' but they come from two words, not
> three.  First "-- gio~a" , then "Dio~a".

No lyric ties either...


    Werner

[pacovila, 2011-08-11 11:16:54]

2011/8/11 Werner LEMBERG <wl@gnu.org>:
>
>>> An example is the second aria of Susanna in Mozart's `Le Nozze di
>>> Figaro', bar 16:
>>>
>>>   f2     f8  e8       g8        c8
>>>
>>>  fin -- chè l'a -- "ria è an" -- cor
>>>
>>> Almost all singers I've met during my work as a coach have problems
>>> if they sing it the first time :-)
>>
>> That makes four vowels! i+a+e+a
>
> Yes :-) Since this is completely unexpected (and I don't know any
> other work of Mozart with a similar situation), people are stumbling
> there.
>
>> Here http://cosinasdeleon.blogspot.com/2009/07/hinmo-leon.html are
>> two different instances of 'ioa' but they come from two words, not
>> three.  First "-- gio~a" , then "Dio~a".
>
> No lyric ties either...

So your view is that lyric ties are not used in the real world?  I
still feel them as a pedagogy resource for young musicians or
something.  Or maybe old scores did not use them and they are now more
often used. I don't know.
-- 
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com

[Jean-Charles, 2011-08-11 11:37:46]

Le 11/08/2011 13:16, Francisco Vila disait :
> 2011/8/11 Werner LEMBERG<wl@gnu.org>:
>>
>>>> An example is the second aria of Susanna in Mozart's `Le Nozze
>>>> di Figaro', bar 16:
>>>>
>>>> f2     f8  e8       g8        c8
>>>>
>>>> fin -- chè l'a -- "ria è an" -- cor
>>>>
>>>> Almost all singers I've met during my work as a coach have
>>>> problems if they sing it the first time :-)
>>>
>>> That makes four vowels! i+a+e+a
>>
>> Yes :-) Since this is completely unexpected (and I don't know any
>> other work of Mozart with a similar situation), people are
>> stumbling there.
>>
>>> Here http://cosinasdeleon.blogspot.com/2009/07/hinmo-leon.html
>>> are two different instances of 'ioa' but they come from two
>>> words, not three.  First "-- gio~a" , then "Dio~a".
>>
>> No lyric ties either...
>
> So your view is that lyric ties are not used in the real world?  I
> still feel them as a pedagogy resource for young musicians or
> something.  Or maybe old scores did not use them and they are now
> more often used. I don't know.

The first time I've seen tied syllables was in English engravings, but
I'm unable to mention where.

Cheers,
Jean-Charles

[Bertrand Bordage, 2011-08-11 11:50:46]

I forgot these cases (o~è~in; a~è~in; o~è~an...).  They are often elided
("quest'in" instead of "questa~è~in") by editors and composers.

> So your view is that lyric ties are not used in the real world?  I
> still feel them as a pedagogy resource for young musicians or
> something.  Or maybe old scores did not use them and they are now more
> often used. I don't know.

Several famous editors use them, with no particular rule.
Let's take the example of Ricordi Milano.

In this score, published around 1982, there's no ties (we can see a "na è in" in
the beginning):
http://petrucci.mus.auth.gr/imglnks/usimg/2/20/IMSLP81325-PMLP165637-la_danza...

In this one, published around 1925, we can find ties (and it's not a beginner
score!):
http://216.129.110.22/files/imglnks/usimg/2/29/IMSLP80146-PMLP162554-Vivaldi_...

In every recent score I have from them, they always use ties. Those which were
done using traditional engraving as those using Finale.

I will have to do a better patch...

[wl_gnu.org, 2011-08-12 05:22:52]

>> No lyric ties either...
> 
> So your view is that lyric ties are not used in the real world?

No.  I simply can't discuss how lyric ties should look like since I
haven't seen real-world examples in the scores I've used up to now.
It seems that German engraving for classical music can live without
it.


    Werner

[mail_philholmes.net, 2011-08-12 09:44:22]

----- Original Message ----- 
From: "Werner LEMBERG" <wl@gnu.org>
To: <paconet.org@gmail.com>
Cc: <reply@codereview.appspotmail.com>; <lemzwerg@googlemail.com>; 
<lemniskata.bernoullego@gmail.com>; <lilypond-devel@gnu.org>; 
<pkx166h@gmail.com>
Sent: Friday, August 12, 2011 6:23 AM
Subject: Re: Adds a glyph for tied lyrics. (issue4808074)


>
>>> No lyric ties either...
>>
>> So your view is that lyric ties are not used in the real world?
>
> No.  I simply can't discuss how lyric ties should look like since I
> haven't seen real-world examples in the scores I've used up to now.
> It seems that German engraving for classical music can live without
> it.

They're definitely used in English 16C music, although I can't find an 
example right now.

--
Phil Holmes

[Bertrand Bordage, 2011-08-13 10:12:50]

Here's how a better solution should work:
* create a smaller tie in addition to this one
* use a scm regexp to find this special case; something like:

#(use-modules (ice-9 regex))
#(display (regexp-substitute #f (string-match "([^:])~([^:])~([^:])"
"questa~è~in")
'pre 1 " " 2 " " 3 'post))

But I don't have time to do it right now.
I'll push this patch when the countdown will be over and make another patch
later for this special case.

[Neil Puttock, 2011-08-13 15:51:33]

LGTM.

http://codereview.appspot.com/4808074/diff/7001/scm/define-markup-commands.scm
File scm/define-markup-commands.scm (right):

http://codereview.appspot.com/4808074/diff/7001/scm/define-markup-commands.sc...
scm/define-markup-commands.scm:943: (tie-str (markup #:hspace (/ word-space 2)
use separate binding for (/ word-space 2)

[Bertrand Bordage, 2011-08-14 08:08:48]

> use separate binding for (/ word-space 2)

Done.

[Bertrand Bordage, 2011-08-14 10:30:42]

Pushed as 8d148ea05fa4b34f8cc3407e112363d715b27ad8

Bertrand