>> we shouldn’t abandon the standard because it’s difficult to keep it
>>up.
>
> We should abandon the approach when its cost is higher than its
> benefit though. If this is just an occasional nudge, it's fine.
> But if I remember right, I think Werner was ta
o
explicitly add a reminder to the 2.26.0 announcement note to use
`convert-ly`.
Werner
If so, how could this be
improved?
Werner
is an inconvenience, yes, but I think that taking care of normal
users is more important.
Werner
e never heard
of before. Is it sufficient to install the two packages via
`package-install` in Emacs?
Werner
>>>> We’d definitely need more opinions, especially from folks like
>>>> Werner who have worked on the docs/appendices a lot, as well as
>>>> from users: how do they feel about it? How much do they use which
>>>> index?
>>
>> I use Appe
seems like a
> "bug" that has existed at least since version 2.14.
AFAIK, this is limitation of `texi2any`; see my report at
https://lists.gnu.org/archive/html/bug-texinfo/2025-01/msg00019.html
Werner
> The document attached here suggests edits to 8 examples in NR4 and
> NR5.
Thanks, see
https://gitlab.com/lilypond/lilypond/-/merge_requests/2559
Changes to snippets are visible in the documentation after the next
LSR import.
Werner
> I have a few copy edits for the notes in the changelog. [...]
Thanks, this is now
https://gitlab.com/lilypond/lilypond/-/merge_requests/2556
Werner
h
>> intertwined and LilyPond receives more regular updates, I think Urs
>> deserves to have a LilyPond stable release named after him - given
>> that he was also one of the most prolific advocates of real-world
>> use of LilyPond. What do you all think?
>
> I'd say it's a good idea.
+1
Werner
recipes for building the development version on both
Homebrew and MacPorts should be updated, too. At least the one from
MacPorts still uses guile2.x, which is no longer supported in the 2.25
series.
Werner
> The document attached here suggests edits to 36 examples in NR1 and
> NR2.
Most of your suggestions are now incorporated in
https://gitlab.com/lilypond/lilypond/-/merge_requests/2549
Please have a look – and thanks again for your report!
Werner
LilyPond meeting, where almost all major developers took
part of...
Sigh.
Werner
ightmare.
The Python3 port (which is dblatex version 0.3.12, AFAIK) builds fine
with MacPorts:
https://ports.macports.org/port/dblatex/details/
https://sourceforge.net/p/dblatex/dblatex-py3/ci/default/tree/
Werner
nspositions
> Appendix B: Lead-Sheet Symbols
> Appendix C: Set Class List
> Appendix D: Answers to Self-Tests
> Index of Music Examples
> Subject Index
It's not clear to me what these examples should demonstrate. Please
elaborate.
Werner
x for authors not overlapping with the general
index (and the 'LaTeX companion' has such a separate index).
Werner
`
command.
https://opensource.com/article/21/11/linux-diff-patch
Werner
>> We’d definitely need more opinions, especially from folks like
>> Werner who have worked on the docs/appendices a lot, as well as
>> from users: how do they feel about it? How much do they use which
>> index?
I use Appendix E only.
> While people are thinking,
just include all of the musical
> input inline with the example, just like all the other examples in
> the NR.
+1 This will add less than a page to the NR, which is fine with me.
Werner
> The document attached here suggests edits to 36 examples in NR1 and
> NR2.
Thanks a lot. Do you have time (and energy) to create a Merge Request
by yourself? Otherwise I can take care of it.
Werner
> Can the spelling of to-barline be changed to to-bar-line to better
> match the rest of the code?
Sounds sensible to me. Please open an issue.
Werner
p anyway, so with a concise description the shorter names should
be fine IMHO.
And what about `proportionalNotationBaseLength`?
Werner
pets/new breathing-signs.ly and
> unfretted-headword.ly
>
> Both fail, because scripts.upbow was changed to scripts.u/dupbow
> (same for downbow). Please find fixed files attached, too tired
> to patch myself
Will take care of them; I will also do an LSR import soon.
Werner
ge.
https://github.com/nybbs2003/lilypond-Jianpu/blob/master/jianpu10a.ly
To summarize: It makes sense to file a feature request for the
Nashville Numbering System, but to do that you have to find better –
and correct – samples.
Werner
\version "2.19.40"
\include "jianpu1
ld have to do this
by your own. Otherwise please present *typeset*, high-quality scores
that we can use to illustrate a new issue.
Werner
> However, after following your suggestions, I tried looking for the
> latest 2.25.22 version on GitLab but was unable to find a
> downloadable release.
2.25.22 has not been released yet.
Werner
Thanks!
> I don't think so.
>
>> Just curious: Is the 'top-posting' restriction on the LilyPond
>> mailing lists still active? If yes, where and how is this
>> enforced?
Just curious: Is the 'top-posting' restriction on the LilyPond mailing
lists still active? If yes, where and how is this enforced?
Werner
ds interesting. However, I have to see this in action and try it
by myself to give a qualified comment.
Werner
ht as well see if
> other ways to represent this same information work better for
> people.
What might be helpful is to provide a 'halo' around changed pixels so
that the attention is immediately drawn to the right spot.
Werner
l/lilypond-user/2020-07/msg00081.html
says:
;; at 101.178, a staff space is exactly 7 pixels.
Werner
ed on my system (which is thus 'IMv7', too), and I don't get
this message. Apparently this warning got introduced in version
7.1.1.
> Do I have to bother about it?
Not yet :-) In the long run, we have to adjust the `configure` script
to get the right ImageMagick tool name.
Werner
> To follow up the replacement of make-moment by \musicLength [1], I
> think it would make sense to rename baseMoment to beatBaseLength.
> Are there any concerns or other suggestions?
Sounds good to me.
Werner
Dear LilyPonders,
>> in case you are an owner or a user of a recent Sibelius or Dorico
>> version, please send me privately screenshots of an import of the
>> two attached MusicXML files [...]
four helpful guys sent me their results for both programs – thanks a
lot!
Werner
ng XML, making it difficult to handle them. For this reason I'm
resending the files, this time in a zip folder.
Please retry!
Werner
<>
x27;t need renderings for these two programs.]
I'm working on `musicxml2ly` improvements, and I need to check how
certain MusicXML elements are handled in various music notation
programs.
Werner
http://www.musicxml.org/dtds/partwise.dtd";>
XXX
/configure --prefix=/usr/local/opt \
--disable-shared \
--without-threads \
--disable-networking
```
Werner
prefix=/usr/local/opt
exec_prefix=/usr/local/opt
bindir=/usr/local/opt/bin
libdir=/usr/local/opt/lib64
includedir=/usr/local/opt/include
datarootdir=
> In a number of years monitoring patches/MRs, today is the first time
> there is nothing to report: all outstanding MRs need_work.
Vacation time, I guess :-)
Anyway, thans *a lot* for your service of taking care of the
countdowns!
Werner
> We are happy to announce the release of LilyPond 2.25.19. [...]
Thanks as usual for doing this job!
Werner
> I'm building Frescobaldi flatpak using a more recent Sdk (for Qt6)
> and I noticed that building guile3 from sources requires hours (!).
> [...]
Irrespective of the patch mentioned by Michael: You do a multi-job
build, right? [*]
Werner
[*] I.e., using `make`'s `-jX
> I'm going to make a new release within a week. Is there anything
> urgent that must be added/fixed/whatever?
Ooops, wrong mailing list :-)
Sorry for the noise.
Werner
Folks,
I'm going to make a new release within a week. Is there anything
urgent that must be added/fixed/whatever?
Werner
e a blinking comparison
mechanism that helps find such differences quickly by visual
inspection.
Werner
27;d checkout HEAD and get my diffs? Or would this be
> unadvisable?
This is described in the Contributor's Guide:
https://lilypond.org/development.html
Please read that and ask questions afterwards if necessary :-)
Werner
diameter of only 1 pixel, but
> missing notehead is much larger.
Sounds good, too. However, my knowledge of such algorithms is far too
limited to even have an opinion.
Werner
is one test below the threshold, and this is
the one in question!
Werner
[*] After a MR has been merged you have to create a branch locally and
apply the changes manually.
y shifted images and/or objects and much larger demerits
for appearing or disappearing objects.
It doesn't make any sense right now to muse about completely changing
the process.
Werner
, also maintaining them for
the next 20 years :-).
Right now, the only thing that needs improvement is the threshold
algorithm.
Werner
uca
and Han-Wen) want to comment on it.
https://computergraphics.stackexchange.com/questions/14143/search-for-special-image-difference-metric/14145
Werner
> I do like Michael's idea, used for selected tests (saybe maybe 50 to
> 80% of the tests). Agreed with Werner that there is the issue we
> don't have such a backend at the moment, although there is a
> question as to whether a normalizing postprocess on the SVG ou
nly be very beneficial. The usual question: Who is
going to write such a thing, together with all the necessary
infrastructure stuff and, as you correctly state, the many additional
tests? And most importantly, is this person willing to maintain it
for at least some time until the backend is stable, mature, and well
documented?
Werner
ifferences, or make another
> option for capturing additions and removals.
Honestly, I have no idea how to achieve that – sich algorithms is not
my forte. Luca has offered to come up with something (he is an expert
in this area), so let's wait for that.
Werner
; so perhaps what you have can be modified to add that.
Yes, this might be the solution to go: On top of the MAE algorithm we
currently use, another algorithm might increase the penalty for small
feature differences so that it is above our threshold.
Werner
t all. We 'just'
have to find one that fits our needs.
Werner
n't provide what we need for LilyPond.
Werner
ant an algorithm that *reduces* the metric for such
cases but not eliminate it.
> In other scenarios, you may also want to normalize brightness /
> color / white balanve and other stuff,
I think we can safely ignore this for LilyPond.
Werner
nce.py`) is written in Python,
too. As you say, a dependency on NumPy shouldn't be too much of a
problem, but Jonas can evaluate this better than me.
Werner
> Werner the case you have on SE seems to indicate you need a
> translation invariant test, I think.
Whatever you say :-) Great that there are people on this list who can
actually contribute to the topic.
> In your case you could compute min diff over all possible
> translation
I've posted a question on StackExchange, searching for a better
regtest comparison algorithm
https://computergraphics.stackexchange.com/questions/14143/search-for-special-image-difference-metric
Werner
> I think I tend towards skipping all of them for stable/2.24 at the
> current state.
OK.
Werner
> The tentative plan would be to release 2.25.18 next weekend and then
> 2.24.4 the week after, maybe again with binaries produced on
> Wednesday or Thursday (to be seen).
+1
Werner
>> I also suggest !2359.
>
> So far I didn't backport any documentation improvements to
> stable/2.24. Is there anything special about this one?
I think these index entries complete the list of essential syntax
elements in the NR.
Werner
> [...] we could think about having LilyPond 2.24.4 in early / mid
> July. What do you think?
+1
I also suggest !2359.
Werner
> For added fun, also try
>
>>>> import antigravity
:-)
More easter eggs:
https://stackoverflow.com/questions/34472872/purpose-of-python-antigravity-module
Werner
> [...]
>
> Why would the Slur be positioned differently in these two cases?
Welcome to the mystery world of slur penalties, and sorry for not
having a better answer or an explanation.
Werner
Something completely off-topic: In case you don't know this, try
```
$ python
>>> import this
```
Enjoy :-)
Werner
improving the metrics, maybe we could improve the
> review workflow by adding dynamic control to the diff web page to
> expand or collapse the entries based on a variable threshold.
Yes, this would be a nice feature – please submit an issue for that.
Werner
in
image analysis, and a quick search in the internet didn't help me in
finding a good metrics. I think we would need something that checks
whether a shape completely disappears and/or reappears in the images
(contrary to shapes that are only slightly shifted and/or distorted),
almost regardless of its size.
Werner
> Anyway—shouldn’t it be on
> <http://lilypond.org/doc/v2.25/Documentation/web/manuals.html> as
> well?
It eventually will, as soon as version 2.26 gets released.
Werner
rg → Manuals 2.25.15 → Visual Index
Werner
MNX so that a future conversion from and to LilyPond source code
can be done as easily as possible, addressing the needs of LilyPond
where necessary.
Werner
ing at a particular place. I don't see any increased
computation time for this.
Do you see *any* advantage of having `\autoBeamOff` being delayed?
Werner
end a auto-beam in progress.
Thanks for digging! I've now done some `git blame` calls and found
out that this behaviour was there from the very beginning of the
auto-beamer's existence (in 1999). Honestly, I don't see any reason
for it – who likes delayed actions like this?
> Using \noBeam should let you terminate the beam early: [...]
Very nice! Maybe we should add your fix to the git repository?
Werner
I consider this behaviour rather
counter-intuitive.
Could this be fixed?
Werner
> While revising the documentation for proportional spacing [...]
This is now available as
https://gitlab.com/lilypond/lilypond/-/merge_requests/2279
Werner
perty (musical_column, "shortest-playing-duration",
> proportional);
> set_property (musical_column, "shortest-starter-duration",
> proportional);
> set_property (musical_column, "used", SCM_BOOL_T);
> return;
> }
>
> so it looks like you can remove it.
Thanks for confirmation.
Werner
/msg00107.html
it seems that it was necessary then. However, the examples given in
these e-mails seem to work just fine with 2.25.13 if I omit setting
the `used` property.
Does anybody know more? Can someone confirm this observation?
Werner
is the default as specified in `define-grobs.scm`.
Werner
==
\version "2.25.14"
music = {
\compressEmptyMeasures
R1*20
\tempo "Tempo" R1*20
\tempo "Tempo" \key b\minor R1*20
\
In December I asked the question below on the 'lilypond-user' list
(https://lists.gnu.org/archive/html/lilypond-user/2023-12/msg00014.html),
alas, without an answer – I try again, this time on 'lilypond-devel'.
I'm also attaching the image for completeness.
> I had proven at the time to James that hel-arabic.ly could be used
> without modifying anything, he had wanted the integration of this
> hel-arabic file and I was forced to modify certain scheme files to
> that the integration goes well
You were 'forced'? I doubt that. You could always have said 'no, I
don't want that'.
Werner
`hel-arabic.ly` file somewhere that might interfere.
If you can demonstrate that an input file doesn't work correctly with
the code in the merge request, we are going to fix this. Up to now,
however, I only see faulty tests on your side, insults, and unfounded
accusations.
Werner
es – especially for Arabic, since
this was the very first language supported by this service –, this is
no longer an excuse.
Maybe it's really better if you don't submit new code if such
communication is the result.
Werner
ode compiles fine – this means
that it works for previous versions, too.
The changes to `hel-arabic.ly` happened in version 2.25.2; earlier
versions are *not* affected.
Werner
> I did some tests on version 2.24.3, 2.21.0-1 , 2.23.6-1
>
> it doesn't work either.
What *exactly* did you do?
Werner
e current development version without
my fix. Again: please have a look at
https://gitlab.com/lilypond/lilypond/-/merge_requests/2268
and try the code from this MR.
Werner
tput compiles fine with the patch from
https://gitlab.com/lilypond/lilypond/-/merge_requests/2268
applied. Please look at this MR and comment.
Werner
> \include "hel_arabic.ly"
It's `hel-arabic.ly`, not `hel_arabic.ly`.
Werner
ould ensure full backward compatibility if you process your old
files with `convert-ly`.
Werner
Indeed edb and sdb are no longer recognized
It's not screwed up at all, however, current LilyPond is not backward
compatible. I'm already working on a patch.
Werner
arabic.ly was truncated.
> for example the makam rast, an important maqam of Arabic music, was
> removed hel-arabic.ly does not depend in any way on arabic.ly.
Well, `hel-arabic.ly` *does* now depend on `arabic.ly`. Why is that a
problem? Please elaborate.
Werner
> inline constexpr auto VPOS = vsize (-1);
This is now part of
https://gitlab.com/lilypond/lilypond/-/merge_requests/2267
Werner
ghostpdl 'master' (commit b809058c5ea0cc5689d8b7c15f191a440c9bb426),
everything works fine; your example PS file compiles correctly, and
'qdf --check' (built from current git) passes, too. I'm on a
GNU/Linux box in case this matters.
Werner
true, the
embedding status is known (PDF interpreter *only*) and the font was
NOT embedded in the input, then we don't embed it in the output. This is
a change in behaviour, previously we would embed a substitute.
Other interpreters should not see a change in behaviour.
```
Sigh. Masamichi-san, please have a look!
Werner
ide's type is exactly the one on the left-hand side, and I can stop
thinking about it immediately.
Werner
ne?
This sounds sensible to me, since `size_t` is always unsigned...
Werner
[-Wsign-conversion]
inline constexpr vsize VPOS (-1);
^~
```
for almost all C++ fies, making the log file quite noisy. Dan, any
idea how to fix this gracefully?
Werner
https://gitlab.com/lilypond/lilypond/-/issues/6699
Werner
-fonts -O TeX-GS
```
However,...
> If I do `lilypond --eps input.ly` I get a much bigger EPS file, that
> seemingly has the fonts attached as binary data.
... it doesn't matter for the problem at hand: you also get the
differences with normal `lilypond --eps` calls.
Werner
t that we have a problem with thread support on
macOS (Carl runs on macOS Sonoma which identifies itself as Darwin
23.2 on arm64e).
Any idea how to tackle this? BTW, `make doc` without `CPU_COUNT=10`
succeeds on his computer.
Werner
ect) I will prepare something more specific
for further tests.
Werner
1 - 100 of 1457 matches
Mail list logo
