Re: exchange LyricHyphen with a "proper" hyphen

2016-12-09 Thread Noeck 
Dear Alexander,

a very nice score!

I don't like to spoil it but in the last line (measure 30) I see an
extra hyphen. I guess it is no problem to remove it though. It is in the
lyrics of the upper and lower staff inside the syllable "ten" of
"deuten". So probably not an issue of harms code but just "te -- n"
instead of "ten"?

Best,
Joram

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: exchange LyricHyphen with a "proper" hyphen

2016-12-09 Thread Alexander Kobel 

On 2016-12-07 23:59, Thomas Morley wrote:

2016-12-07 22:03 GMT+01:00 Simon Albrecht :

On 07.12.2016 11:45, Alexander Kobel wrote:


I'm not sure whether Harm's proof of concept also involved
pre-compile-time modifications.



It certainly didn’t; Harm is a Schemer, not a C++ guy :-)


Indeed.

So here a very first shot.
For now I let display the new _and_ the colored default LyricHyphen.
You'll see a drawback already: In tight situations the default gets
shortened until it vanishs. This is not possible or at least not
desirable for characters taken from a font, imho.
Iirc this was already mentioned in the discussion the tracker links to.
Maybe minimum-distance may help here, not tested though.


AFAICS for now: Works like a charm. Thanks a ton again.

And with this, all of this week's issues of mine are worked around, and 
I can proudly present a beautiful little setting of a beautiful little 
piece by Reger... :-)



Cheers,
Alexander


nacht.pdf
Description: Adobe PDF document
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Fixing LSR 888: center-on-words ignoring punctuation

2016-12-09 Thread Alexander Kobel 

Dear Harm,

if we ever meet, remind me that I owe you more than just one drink...
That's amazing.

AFAICS, everything works beautifully as expected, with one strange 
exception: in a chord with two equal notes, the default algorithm 
chooses the right one as reference, your method chooses the reference 
point depending on the stem direction. That's something I can perfectly 
live with.
In particular because I took the liberty to add a second variant that 
centers on chords; starting from your code, I found out how to do that. 
Please find the file attached.


On a side note, I think that at some point during the past import to the 
LSR, the font encoding of some mail must have been messed up, and the 
list of special characters passed away - that's why I chose an 
attachment. I restored what I think might be the original one, and added 
some further quotation marks (in the languages supported in my 
Frescobaldi installation) plus dagger and double dagger that might be 
used as footnote marks.



Cheers,
Alexander


On 2016-12-09 00:50, Thomas Morley wrote:

2016-12-07 1:07 GMT+01:00 Alexander Kobel :

Hi all,

yet another semi-feature request where I'm running out of knowledge...
I happily used LSR 888 (center lyrics without taking punctuation into
account) for years and only now found out that the alignment is off under
certain circumstances: The default algorithm seems to align lyrics on the
center of the note*head* of the associated voice. Which seems the right
thing to do. However, center-on-words aligns on the center of the *entire
note* - in particular, this yields different results on quavers with stems
up and "outward flags". At least, neither dots nor accidentals are taken
into account...

Even more ugly, the alignment is totally off on chords with two adjacent
notes (shifting one note). I know that this should be a forbidden situation
for vocal music, but sometimes there is a reason for writing such chords
instead of two separate voices (e.g., soprano and alto notated on a common
staff in a homorhythmic piece, with one of the voices splitting for a
cluster on just a few notes).
In such a rare case, I do not like the default algorithm's choices either
(and I'd prefer alignment on the center of the noteheads of the chord); but
center-on-words does even /way/ worse.

I attach a "test case" (including a full copy of the core of
http://lsr.di.unimi.it/LSR/Snippet?id=888), where one would expect
center-on-words to do exactly nothing, but you can notice the slight offsets
on the "nonius scale" of lyrics.

Anyone got an idea where to start to fix that? I think one would have to
figure how to get the note-column-extent without flags; but I have no clue
how to approach that.


Thanks in advance,
Alexander




Hi,

how about below?
You need a devel-version, though. Please test thoroughly. I'm too
tired to do more than the included examples.

\version "2.19.52"  %% sic !!

%% http://lsr.di.unimi.it/LSR/Item?id=888

#(define space-set
  (list->char-set
(string->list "—.?-;,:“†‘’–— */()[]{}|<>!`~&…")))

#(define (width grob text)
  (let* ((X-extent
   (ly:stencil-extent (grob-interpret-markup grob text) X)))
   (if (interval-empty? X-extent)
   0
   (cdr X-extent

#(define (remove-suspended-note-heads stem note-heads)
  (let* ((nc (ly:grob-common-refpoint stem (car note-heads) X))
 (stem-coord
   (ly:grob-relative-coordinate stem stem X))
 (half-stem-thick
   (/ (ly:grob-property stem 'thickness) 2))
 (stem-dir (ly:grob-property stem 'direction)))
(remove
  (lambda (nh)
(if (positive? stem-dir)
(> (ly:grob-relative-coordinate nh nc X)
stem-coord)
(< (ly:grob-relative-coordinate nh nc X)
   (- stem-coord half-stem-thick
  note-heads)))

#(define (center-on-word grob)
  (let* ((text (ly:grob-property-data grob 'text))
 (syllable (markup->string text))
 (word-position
   (if (string-skip syllable space-set)
   (string-skip syllable space-set)
   0))
 (word-end
   (if (string-skip-right syllable space-set)
   (+ (string-skip-right syllable space-set) 1)
   (string-length syllable)))
 (preword (substring syllable 0 word-position))
 (word (substring syllable word-position word-end))
 (preword-width (width grob preword))
 (word-width (width grob (if (string-null? syllable) text word)))
 (note-column (ly:grob-parent grob X))
 (stem (ly:grob-object note-column 'stem))
 (stem-dir (ly:grob-property stem 'direction))
 (sys (ly:grob-system grob))
 (nh-ls
   (if (ly:grob-array? (ly:grob-object note-column 'note-heads))
   (ly:grob-array->list (ly:grob-object note-column 'note-heads))
   '()))
 (note-column-width
   (interval-length

Re: Emmentaler font issues with the infinality bundle [was: on Arch Linux builds]

2016-12-09 Thread Alexander Kobel 

On 2016-12-09 23:36, Michael Gerdau wrote:

Hi Alexander,


Confirmed by a two-minute test without recompiling anything (I use
bohoomils repo for infinality, not the aur package). Could you do me a
favor and double-check whether the skylines change if you replace
{cairo,fontconfig,freetype2} by
{cairo,fontconfig,freetype2}-infinality-ultimate (possibly along with
the lib32- variants)?
No hurry, though, I'll have to find time to investigate more, and
whether the configuration can be changed to leave Lily's music fonts
alone...


I used to use only the fonts repo.


Ah, I see.


I now added
[infinality-bundle]

Server = http://bohoomil.com/repo/x86_64

[infinality-bundle-multilib]

Server = http://bohoomil.com/repo/multilib/x86_64

to my pacman.conf and installed
{cairo,fontconfig,freetype2}-infinality-ultimateall (plus lib32 variants)


Yes, that used to be my default configuration.


Result:
The skyline seems to be like in your 2.19.53 example.
Going back to ArchLinux stock repos again gives your 2.19.50 version
(which I seem to like better)


Indeed. The stock version is okay, the other one is plainly wrong IMHO. 
(The position of the octavation of treble_8 is a matter of taste, but it 
is as intended in the official Lily builds.)


I reported the issue on bohoomil's tracker at
  https://github.com/bohoomil/fontconfig-ultimate/issues/186


Thanks a lot for the confirmation. I'll let the list know when I have 
further news on the subject.



Cheers,
Alexander

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Emmentaler font issues with the infinality bundle [was: on Arch Linux builds]

2016-12-09 Thread Michael Gerdau 
Hi Alexander,

> Confirmed by a two-minute test without recompiling anything (I use
> bohoomils repo for infinality, not the aur package). Could you do me a
> favor and double-check whether the skylines change if you replace
> {cairo,fontconfig,freetype2} by
> {cairo,fontconfig,freetype2}-infinality-ultimate (possibly along with
> the lib32- variants)?
> No hurry, though, I'll have to find time to investigate more, and
> whether the configuration can be changed to leave Lily's music fonts
> alone...

I used to use only the fonts repo.

I now added
[infinality-bundle]

Server = http://bohoomil.com/repo/x86_64

[infinality-bundle-multilib]

Server = http://bohoomil.com/repo/multilib/x86_64

to my pacman.conf and installed
{cairo,fontconfig,freetype2}-infinality-ultimateall (plus lib32 variants)

Result:
The skyline seems to be like in your 2.19.53 example.
Going back to ArchLinux stock repos again gives your 2.19.50 version
(which I seem to like better)

Kind regards,
Michael
-- 
 Michael Gerdau   email: m...@qata.de
 GPG-keys available on request or at public keyserver

signature.asc
Description: This is a digitally signed message part.
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Is it possible to put alternatives inside repeats?

2016-12-09 Thread David Wright 
On Fri 09 Dec 2016 at 20:40:01 (+0100), Simon Albrecht wrote:
> On 09.12.2016 10:22, list_lilypond wrote:
> >inside the repeat there are alternatives.
> >
> >As normally the \alternative{} syntax is put outside the \repeat{}, I
> >really wonder whether this is possible at all.
> >If possible in some other way, could someone please be so kind to point
> >out how to repeat with embedded alternatives without rewriting all?
> 
> The available implementation of \repeat doesn’t allow any other use
> of \alternative than immediately _after_ the repeated expression. So
> it seems you’ll have to either
> 
> – split the repeat up into several repeats (and relocate stuff so
> that the alternative at the beginning becomes an alternative at the
> end
> – write it out entirely – that might really reduce confusion for the
> reader, except in Reich-esque pattern music,
> – or fake the appearance by setting the context property
> repeatCommands. There is an example in 
> .
> So you can do something like the attached. I’d advise to close the
> volta brackets for the 2. alternatives _inside_ the repeat, but I
> don’t know how. Anyone?

I used \allowVoltaHook in
http://lists.gnu.org/archive/html/lilypond-user/2015-10/msg00404.html
but it has its downside as it seems unresettable.

Cheers,
David.

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Is it possible to put alternatives inside repeats?

2016-12-09 Thread Simon Albrecht 

Hello anonymous guy (or gal…),


On 09.12.2016 10:22, list_lilypond wrote:

inside the repeat there are alternatives.

As normally the \alternative{} syntax is put outside the \repeat{}, I
really wonder whether this is possible at all.
If possible in some other way, could someone please be so kind to point
out how to repeat with embedded alternatives without rewriting all?


The available implementation of \repeat doesn’t allow any other use of 
\alternative than immediately _after_ the repeated expression. So it 
seems you’ll have to either


– split the repeat up into several repeats (and relocate stuff so that 
the alternative at the beginning becomes an alternative at the end
– write it out entirely – that might really reduce confusion for the 
reader, except in Reich-esque pattern music,
– or fake the appearance by setting the context property repeatCommands. 
There is an example in 
. 
So you can do something like the attached. I’d advise to close the volta 
brackets for the 2. alternatives _inside_ the repeat, but I don’t know 
how. Anyone?


Best, Simon


\version "2.18.2"
{
  \repeat volta 2 {
\set Score.repeatCommands = #'((volta "1.") start-repeat)
a1 |
\set Score.repeatCommands = #'((volta #f)(volta "2."))
b |
\set Score.repeatCommands = #'((volta #f))
c | d | e | f | g |
\set Score.repeatCommands = #'((volta "1."))
b |
\set Score.repeatCommands = #'((volta #f)(volta "2."))
c |
\set Score.repeatCommands = #'((volta #f))
d | e | f | g |
  }
  \alternative {
{ b }
{ c }
  }
}___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Emmentaler font issues with the infinality bundle [was: on Arch Linux builds]

2016-12-09 Thread Gerdau, Michael 
> So the mgd in the mail address is the same mgd who mentioned problems 
> with the infinality-bundle on aur/lilypond-devel? Well, guess what 
> resolved my problems... (Turned up as first hit after I had the correct 
> idea, and searched for "Lilypond onfinality".)

Yes, that's me :)

> Confirmed by a two-minute test without recompiling anything (I use 
> bohoomils repo for infinality, not the aur package). Could you do me a 
> favor and double-check whether the skylines change if you replace 
> {cairo,fontconfig,freetype2} by 
> {cairo,fontconfig,freetype2}-infinality-ultimate (possibly along with 
> the lib32- variants)?
> No hurry, though, I'll have to find time to investigate more, and 
> whether the configuration can be changed to leave Lily's music fonts 
> alone...

Will do that when I'll return home tonight.

Kind regards,
Michael
--
Michael Gerdau   email: m...@qata.de
GPG-keys available on request or at public keyserver

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Makefile and shell loop

2016-12-09 Thread Federico Bruni 
Il giorno ven 9 dic 2016 alle 10:53, Johan Vromans 
 ha scritto:
 where $(CPU_CORES) is treated as a string, instead of being 
executed in

 the shell:


It seems you are using GNU make, so you can write

  CPU_CORES = $(shell cat /proc/cpuinfo | grep -m1 "cpu cores" | sed 
s/".*: "//)


Yep, this evaluates to 4 (correct).
Many thanks Johan!


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Makefile and shell loop

2016-12-09 Thread Johan Vromans 
On Fri, 09 Dec 2016 10:30:32 +0100, Federico Bruni 
wrote:

> where $(CPU_CORES) is treated as a string, instead of being executed in 
> the shell:

It seems you are using GNU make, so you can write

  CPU_CORES = $(shell cat /proc/cpuinfo | grep -m1 "cpu cores" | sed s/".*: "//)

-- Johan

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: LSR 357: "Dynamics context is now included by default"

2016-12-09 Thread Thomas Morley 
2016-12-09 4:56 GMT+01:00 Joel C. Salomon :
> I’ve been using a variant of the LSR 357 “centered dynamics” template
> from , but I notice that the
> documentation says:
>
>> This snippet demonstrates how this was achieved with older versions
>> of LilyPond; the Dynamics context is now included by default (starting
>> with LilyPond 2.13.29), and does not require to be defined manually.
>
> What does this mean?

Hi Joel,

the description is misleading.
In older lily-version (2.12.3 and before) there was the need to create
the Dynamics-context manually. See
http://lilypond.org/doc/v2.12/input/lsr/lilypond-snippets/Keyboards#Piano-template-_0028simple_0029
Scroll down to
Piano template with centered dynamics
to see how it was made.
Nowadays the Dynamics-context is implemented per default in the source.

I'll fix the lsr-doc as soon as I can find some spare time.

> Is there another method recommended for centered
> dynamics marks suitable for newer versions of LilyPond?

Actually, the snippet shows the the recommended method for up-to-date versions.

> (The linked example in the LilyPond documentation does not show dynamics at 
> all.)

Scroll down to
Piano template with centered dynamics

Cheers,
  Harm

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Makefile and shell loop

2016-12-09 Thread Federico Bruni 

Hi all

Once again I have a problem with a loop within a Makefile.
I followed the example in the Usage manual, so I have these variables 
defined:


CPU_CORES=`cat /proc/cpuinfo | grep -m1 "cpu cores" | sed s/".*: "//`

LILY_CMD = lilypond -ddelete-intermediate-files \
--include=/home/fede/Documenti/spartiti/ly \
-dno-point-and-click -djob-count=$(CPU_CORES)

Then I have a for loop:

.PHONY: libri
libri:
for f in *.lybook; do $(LILY_CMD) $$f; done


where $(CPU_CORES) is treated as a string, instead of being executed in 
the shell:


$ make
for f in *.lybook; do lilypond -ddelete-intermediate-files 
--include=/home/fede/Documenti/spartiti/ly -dno-point-and-click 
-djob-count=`cat /proc/cpuinfo | grep -m1 "cpu cores" | sed s/".*: "//` 
$f; done

GNU LilyPond 2.19.51

The files are created correctly, but I guess that it's not using the 
desired number of cores.


Thanks in advance
Federico




___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Fonts http://fonts.openlilylib.org/

2016-12-09 Thread Urs Liska 


Am 09.12.2016 um 10:26 schrieb Herbert Liechti:
> Hello
>
> I read this blog
> post http://lilypondblog.org/2014/09/lilyponds-look-and-feel/ where
> the http://fonts.openlilylib.org/ link is mentioned for downloading
> patches and fonts. Unfortunately this link is not working. Anybody
> knows more about that?
>

This has been pending for quite some time, and once it was settled I
missed to update the post, sorry.

Basically Abraham had to decide moving things to a completely new
"business model", and his fonts are now available from
https://www.musictypefoundry.com/.

I will change the post ASAP.

Best
Urs

> Thanks & best regards
> Herbert
>
>
>
> ___
> lilypond-user mailing list
> lilypond-user@gnu.org
> https://lists.gnu.org/mailman/listinfo/lilypond-user

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Fonts http://fonts.openlilylib.org/

2016-12-09 Thread Herbert Liechti 
Hello

I read this blog post
http://lilypondblog.org/2014/09/lilyponds-look-and-feel/ where the
http://fonts.openlilylib.org/ link is mentioned for downloading patches and
fonts. Unfortunately this link is not working. Anybody knows more about
that?

Thanks & best regards
Herbert
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Is it possible to put alternatives inside repeats?

2016-12-09 Thread list_lilypond 
On Fri, 9 Dec 2016 09:22:50 +
list_lilypond  wrote:

> Hi, 
> 
> what I want is something like:


Sorry, typo, this had to be:

\repeat volta 2 {
  \alternative{
{ a | }
{ b | }
  }
  c | d | e | f | g |
  \alternative{
{ b | }
{ c | }
  }
  d | e | f | g |  
}
\alternative{
  { b }
  { c }
}

> So inside the repeat there are alternatives.
> 
> As normally the \alternative{} syntax is put outside the \repeat{}, I
> really wonder whether this is possible at all.
> If possible in some other way, could someone please be so kind to
> point out how to repeat with embedded alternatives without rewriting
> all?
> 
> Thanks
> 
> ___
> lilypond-user mailing list
> lilypond-user@gnu.org
> https://lists.gnu.org/mailman/listinfo/lilypond-user


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Is it possible to put alternatives inside repeats?

2016-12-09 Thread list_lilypond 
Hi, 

what I want is something like:

\repeat volta 2 {
  \alternative{
{ a | }
{ b | }
  }
  c | d | e | f | g |
  \alternative{
{ b | }
{ c | }
  d | e | f | g |  
}
\alternative{
  { b }
  { c }
}

So inside the repeat there are alternatives.

As normally the \alternative{} syntax is put outside the \repeat{}, I
really wonder whether this is possible at all.
If possible in some other way, could someone please be so kind to point
out how to repeat with embedded alternatives without rewriting all?

Thanks

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: Emmentaler font issues with the infinality bundle [was: on Arch Linux builds]

2016-12-09 Thread Alexander Kobel 

Hi Michael,

On 2016-12-09 07:17, Michael Gerdau wrote:

Hi Alexander,

I'm on ArchLinux 64bit myself and use a self compiled devel (curretly
2.19.52) myself. My skyline seems identical to your 2.19.50 version.


thanks a lot, that helped tremendeously to get me on the right track.


My fontforge and fontconfig are the binaries from the offiial repo.
I know for sure since I recently have tracked down a font scaling issue
on my system, where I made sure I had all of these packages w/o me
"fixing" anything :)


So the mgd in the mail address is the same mgd who mentioned problems 
with the infinality-bundle on aur/lilypond-devel? Well, guess what 
resolved my problems... (Turned up as first hit after I had the correct 
idea, and searched for "Lilypond onfinality".)



Therefor I would claim, nothing wrong with ArchLinux as out-of-the-box.


Confirmed by a two-minute test without recompiling anything (I use 
bohoomils repo for infinality, not the aur package). Could you do me a 
favor and double-check whether the skylines change if you replace 
{cairo,fontconfig,freetype2} by 
{cairo,fontconfig,freetype2}-infinality-ultimate (possibly along with 
the lib32- variants)?
No hurry, though, I'll have to find time to investigate more, and 
whether the configuration can be changed to leave Lily's music fonts 
alone...



Cheers,
Alexander

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user