Re: we now have lilypond organization on GitHub
Urs Liska: Am 17.09.2013 18:21, schrieb David Kastrup: Janek Warchołjanek.lilyp...@gmail.com writes: 2013/9/16 David Kastrupd...@gnu.org: So the question is what we should be telling the Savannah operators to make working on GNU projects using Git more feasible. Here you go: A web interface with online editor, allowing to create commits from a web browser, on any branch (as well as creating branches), integrated with pull requests and automatic forking (so that when someone without write access tries to modify a file, Savannah automatically forks the repository for him and when he's done making the change, a pull request is sent to the original repository). Also, being able to post comments on commits, and integration of pull requests with code review. And the ability to update a pull request with new commits. As far as i know Savannah doesn't have these features. When it will have, i'll be happy to switch, but i don't expect it will happen today or tomorrow, so i'm going to use Github for now. Now basically we have to split these into two different sets of requirements: Savannah does not provide accounts or services to the general public; its services will be restricted to actual developers. But what you list above mostly is _not_ related to participating with the project but rather with managing one's own repository using a graphical interface that just happens to be provided as a web service, in a similar vein to Gmail providing a web interface to handling a mail account. I think that's not quite true. IIUC Janek lists options that allow people to contribute to LilyPond without officially applying to be developers. You are doing code reviews through a web interface already, isn't it? And this is because that's a quite natural way to communicate, comment on code etc. What's natural is different for different people. Web interfaces are not natural for me, to the contrary, for me they appear constrained. You can't do _that_ with plain Git. Github isn't plain git either. I'm perfectly fine with git and email, your requirments might differ from mine. /// I don't know about github, but I wouldn't be surprised if their rules would be something that I wouldn't accept. You might be more liberal in what you tolerate from 3rd parties. Regards, /Karl Hammar --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: website suggestion: live Lilypond preview
Janek: when i was talking with MuseScore developers some time ago, they mentioned that it would be great to have a try lilypond in your browser thing on our website, similar to these: http://tryhaskell.org/ http://try.github.io/levels/1/challenges/1 I have no idea how difficult that would be, but i agree that it would be great! At the very least we could perhaps have a prominent link to tunefl/weblily/lilybin/mediawiki score extension live preview. (should we have an issue for this?) tunefl is the top item on http://www.lilypond.org/easier-editing.html perhaps you could simply add the others to that page. Regards, /Karl Hammar --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Suggestions for participating institutions?
Trevor Daniels: Han-Wen Nienhuys wrote Tuesday, March 26, 2013 10:52 AM On Tue, Mar 26, 2013 at 9:52 AM, David Kastrup d...@gnu.org wrote: I met a former colleague in the bus to Chemnitz, and he is at least knowledgeable about EU research programmes. Do people here have ideas about possible institutions who could be made to participate? I think Take in mind that EU research programmes come with an incredible amount of burocracy and require both academic and industry partners, the more the merrier. The projects that get funded are buzzword compliant, but often nobody knows what they set out to do, except divert EU money into the partnering institutions. Have a look at Indeed. When I was involved in bidding for EU research funding in what now seems to be a previous life the process involved many meetings with potential and later actual partners, who have to come from, I believe at least 3 separate countries, plus presentations in Brussels, all involving lots of international travel. And at the end of that expense of time, effort and money we were unsuccessful. This is not like bidding for an Arts grant. I think we can meet the 3 seperate countries requirement. I would like to try to set up something here with possible the University in Uppsala and/or the Church. I don't know if I qualify as an industry (note. eu defines an industry as something that manuf. phycical goods, no I don't have the ref. handy) partner, tought I have a company that have been involved in typesetting for pay. And I don't know if I can get any interest from the Univ. (they used Sibelius a few years ago) or the Church. Regards, /Karl Hammar --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
ice-9/psyntax.scm:1417:32: Syntax error:
Do anyone have a clue what to do about this error: $ git describe release/2.15.33-1-4-gb04051d $ ./autogen.sh http://turkos.aspodata.se/tmp/log.autogen.sh $ make all http://turkos.aspodata.se/tmp/log.make 21 $ tail log.make ;;; compiling /var/home/karl/Net/git/lilypond/out/share/lilypond/current/scm/part-combiner.scm ;;; WARNING: compilation of /var/home/karl/Net/git/lilypond/out/share/lilypond/current/scm/part-combiner.scm failed: ;;; ERROR: Syntax error: ;;; unknown location: source expression failed to match any pattern in form when ice-9/psyntax.scm:1417:32: In procedure expand-macro: ice-9/psyntax.scm:1417:32: Syntax error: unknown location: source expression failed to match any pattern in form when make[1]: *** [out/internals.texi] Error 1 make[1]: Leaving directory `/var/home/karl/Net/git/lilypond/Documentation' make: *** [all] Error 2 $ Regards, /Karl Hammar --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: ice-9/psyntax.scm:1417:32: Syntax error:
David: k...@aspodata.se (Karl Hammar) writes: Do anyone have a clue what to do about this error: ... $ ./autogen.sh http://turkos.aspodata.se/tmp/log.autogen.sh ... Guilev2 is not supported yet. What does autoconf tell? $ grep -iC2 guile log.autogen.sh checking for working metafont mode... ljfour checking for kpsewhich... kpsewhich checking for guile-config... guile-config checking guile-config version... 2.0.5 checking guile compile flags... -pthread -I/usr/include/guile/2.0 checking guile link flags... -lguile-2.0 -lgc checking libguile.h usability... yes checking libguile.h presence... yes checking for libguile.h... yes checking for scm_boot_guile in -lguile... no checking for scm_boot_guile... yes checking for scm_t_hash_fold_fn... yes checking for scm_t_hash_handle_fn... yes checking for scm_t_subr... yes checking for usable C++ demangler... yes checking GUILE rational bugfix... ok checking for python-config... python-config checking Python.h usability... yes -- checking for -windres... no checking for windres... no checking for guile... guile checking for guile... /usr/bin/guile checking for perl... perl checking for perl... /usr/local/bin/perl $ Ok, I'll try to downgrade guile then, thanks. Regards, /Karl Hammar --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Moving away from make
David Kastrup: ... The main problem I see is that dependencies don't work out, and that presumably is mostly because the temporary/work files of lilypond-book are not in the rules and get stomped over by parallel make. ... I have given up on lilypond-book and make. Instead I've made two scripts depend_ly and depen_tex [1] which finds out what depends on what (think gcc -M), and make [2] takes care of the rest. Would it be good to make lilypond print out a files dependancies like gcc? Regards, /Karl Hammar [1] http://turkos.aspodata.se/git/musik/bin/ [2] http://turkos.aspodata.se/git/musik/include/Makefile --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: What's the deal with String_convert::reverse?
Reinhold:
Am Tuesday, 23. August 2011, 22:30:24 schrieb David Kastrup:
...
Is this used anywhere?
A git grep didn't show any usage, but maybe I'm missing something.
Make does not complain about this patch (on my box).
$ git diff |cat
diff --git a/flower/include/string-convert.hh b/flower/include/string-convert.hh
index 55c1486..0675384 100644
--- a/flower/include/string-convert.hh
+++ b/flower/include/string-convert.hh
@@ -49,7 +49,7 @@ public:
static string i64_string (I64, char const *fmt = 0);
static string to_lower (string s);
static string to_upper (string s);
- static string reverse (string s);
+ static string reverseX (string s);
};
#endif // __STRING_CONVERT_HH //
diff --git a/flower/string-convert.cc b/flower/string-convert.cc
index d46426e..01cf3e5 100644
--- a/flower/string-convert.cc
+++ b/flower/string-convert.cc
@@ -362,7 +362,7 @@ String_convert::to_lower (string s)
}
string
-String_convert::reverse (string s)
+String_convert::reverseX (string s)
{
return (char *) memrev ((unsigned char *)s.data (), s.length ());
}
Regards,
/Karl Hammar
---
Aspö Data
Lilla Aspö 148
S-742 94 Östhammar
Sweden
+46 173 140 57
___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Comand token too long?
Reinhold Kainhofer: Am Sunday, 14. August 2011, 03:36:05 schrieb Karl Hammar: Works here: Doesn't work here on a German Kubuntu system: ... $ echo $LANG C $ echo $LANG de_AT.UTF-8 $ LANG=de_AT.UTF-8 $ echo 'z:/' | egrep '^(/|[A-z]:/)' egrep: Das Ende des angegebenen Intervalls ist nicht gültig $ echo 'z:/' | egrep '^(/|[A-Za-z]:/)' z:/ $ export LC_ALL=C $ echo 'z:/' | egrep '^(/|[A-z]:/)' z:/ $ So that means that this bug is due to collation order, and as you wrote yesterday the interval should be changed to A-Za-z. The egrep thing is there to detect if a filename is absolute, i.e. starting with / as in /usr/bin/true (unix style), or C:/ (or another device letter) as in C:/WINDOWS/SYSTEM32/FILE.DLL (Dos/Windows style). As written in man grep: To obtain the traditional interpretation of bracket expressions, you can use the C locale by setting the LC_ALL environment variable to the value C. we can go around this bug by setting LC_ALL to C. This bug is fixed in [1]: - echo $command_line_filename | $EGREP '^(/|[A-z]:/)' 6 \ + echo $command_line_filename | LC_ALL=C $EGREP '^(/|[A-Za-z]:/)' 6 \ We might go around the bug with the attaced patch (texi2dvi is called from texi2pdf), but I have not had time to test it. Regards, /Karl Hammar [1] http://ftp.gnu.org/gnu/texinfo/texi2dvi --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 diff --git a/make/doc-i18n-root-rules.make b/make/doc-i18n-root-rules.make index f510ded..83f2ee6 100644 --- a/make/doc-i18n-root-rules.make +++ b/make/doc-i18n-root-rules.make @@ -18,7 +18,7 @@ $(top-build-dir)/Documentation/$(outdir)/%.$(ISOLANG).html: $(outdir)/%.texi $(X $(top-build-dir)/Documentation/$(outdir)/%.$(ISOLANG).pdf: $(outdir)/%.texi cd $(outdir) \ - texi2pdf $(TEXI2PDF_FLAGS) $(TEXINFO_PAPERSIZE_OPTION) $*.texi \ + LC_ALL=C texi2pdf $(TEXI2PDF_FLAGS) $(TEXINFO_PAPERSIZE_OPTION) $*.texi \ mkdir -p $(dir $@) mv $*.pdf $@ $(outdir)/version.%: $(top-src-dir)/VERSION___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Comand token too long?
Francisco Vila: 2011/8/13 Francisco Vila paconet@gmail.com: Hello, does anybody what this error means? cd Documentation/ cd ./out-www; texi2pdf -I ./out-www -I ../out -I .. -I .. --quiet notation.texi egrep: Invalid range end This is caused by line 1713 of /usr/bin/texi2dvi (GNU Texinfo 4.13) 1.135 in my system, # If the COMMAND_LINE_FILENAME is not absolute (e.g., --debug.tex), # prepend `./' in order to avoid that the tools take it as an option. echo $command_line_filename | $EGREP '^(/|[A-z]:/)' 6 \ || command_line_filename=./$command_line_filename and [A-z] is an invalid range. Works here: $ echo '/usr/' | egrep '^(/|[A-z]:/)' /usr/ $ echo 'B:/' | egrep '^(/|[A-z]:/)' B:/ $ echo 'z:/' | egrep '^(/|[A-z]:/)' z:/ $ echo 'bin/' | egrep '^(/|[A-z]:/)' $ egrep --version | head -1 GNU grep 2.6.3 $ echo $LANG C $ [A-z] is a valid range in the C locale. Maybe you have a locale with a sorting order where z comes before A. Regards, /Karl Hammar --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Postscript printer errors with rounded barlines?
Han-Wen Nienhuys: On Thu, Jul 21, 2011 at 12:31 PM, Karl Hammar k...@aspodata.se wrote: ... After dup there is two blots, gt consumes one and setlinewidth the other. I'm sorry - I misunderstood; I thought you saw a problem with the code rather than the comments. No problem. I see no problem with the code, it was just the comment that didn't match the code. Regards, /Karl Hammar --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Postscript printer errors with rounded barlines?
Han-Wen Nienhuys:
...
Werner, can you have a look at http://codereview.appspot.com/4819041 ?
/draw_round_box % width height x y blot
{
- setlinewidth % w h x y
- 0 setlinecap
- 1 setlinejoin
+dup
+ 0.0 gt {
+ setlinewidth % w h x y
+ 0 setlinecap
+ 1 setlinejoin
+
There is no blot on the stack below (as indicated by the comment),
it was swallowed by setlinewidth.
+ rmoveto % b w h
+ currentpoint % b w h x1 y1
+ 4 2 roll % b x1 y1 w h
+ 4 copy
+ rectfill
+ rectstroke
+ } {
+ pop % w h x y
+ rmoveto % w h
+ currentpoint % w h x1 y1
+ 4 2 roll % x1 y1 w h
+ rectfill
+ } ifelse
+} bind def
You don't seem to use this, why defining it?
+/draw_box % width height x y
+{
rmoveto % w h
currentpoint % w h x1 y1
4 2 roll % x1 y1 w h
4 copy
rectfill
- rectstroke
} bind def
Regards,
/Karl Hammar
---
Aspö Data
Lilla Aspö 148
S-742 94 Östhammar
Sweden
+46 173 140 57
___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Postscript printer errors with rounded barlines?
Han-Wen Nienhuys:
On Thu, Jul 21, 2011 at 8:11 AM, Karl Hammar k...@aspodata.se wrote:
Han-Wen Nienhuys:
Werner, can you have a look at http://codereview.appspot.com/4819041 ?
There is no blot on the stack below (as indicated by the comment),
there is; the dup puts it on the stack. The comment indicate the
state after the call.
/draw_round_box % width height x y blot
{
w h x y b
dup
w h x y b b
0.0 gt
w h x y b a_bool % the a_bool it consumed by the ifelse below
{
w h x y b
setlinewidth % w h x y
w h x y
0 setlinecap
1 setlinejoin
w h x y
rmoveto % b w h
w h
currentpoint % b w h x1 y1
4 2 roll % b x1 y1 w h
4 copy
rectfill
rectstroke
} {
pop % w h x y
rmoveto % w h
currentpoint % w h x1 y1
4 2 roll % x1 y1 w h
rectfill
} ifelse
} bind def
After dup there is two blots, gt consumes one and setlinewidth
the other.
///
Also if there were a blot at rmoveto time due to a dup, it would
be at top of stack, not below w h as indicated by the comment.
///
If it would behave as the comment indicate, then
w h x y 0draw_round_box = -
w h x y blot draw_round_box = blot
e.i. it would leave a lone blot at the top of the stack if it's 0.
Regards,
/Karl Hammar
---
Aspö Data
Lilla Aspö 148
S-742 94 Östhammar
Sweden
+46 173 140 57
___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: GOP-PROP 3 - C++ formatting (probable decision)
Graham: ... ** Eliminate tabs I'm going to make the bold step of assuming that we will eliminate tabs in all C++ files. ... That implies that tabs in strings should be replaced with \t, is that what you want? Regards, /Karl Hammar --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: No more releases: savannah sometimes does not like git fetch --depth
Trevor: I've attached an email which explains a similar problem ... Attached email Path-mtu-discovery is described in rfc-1191 and some problems with it in rfc-2923. The symtoms described in attached email is matches black hole in rfc-2923. It should be wholly agnostic to the git fetch --depth 1 vs. git clone choise. The black hole problem happens only for large packets. ** If it is the black hole: The best fix is to correct your router, setting the mtu in the router might be a solution for that, but your router should return a proper icmp error response (ICMP Destination Unreachable messages with a code meaning fragmentation needed and DF set) if the packet is too large. Make sure the router and your host don't filter out thoose icmp packets. /// The second best is to send smaller packets from your host going through that route. This can be set with route del default gw your router route add default gw your router mss some value where mss = maximum segment size. A good first try is MTU - 80, see rfc-879 for details. Regards, /Karl Hammar --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: No more releases: savannah sometimes does not like git fetch --depth
Graham: ... mkdir new-lilypond-dir cd new-lilypond-dir git init git config core.bare false git remote add -t master origin git://git.sv.gnu.org/lilypond.git git fetch --depth 1 fails every time. Until this is fixed, *** I CANNOT MAKE ANY MORE LILYPOND RELEASES. *** ... Works here (turkos.aspodata.se): $ mkdir new-lilypond-dir $ cd new-lilypond-dir $ git init Initialized empty Git repository in /var/home/karl/new-lilypond-dir/.git/ $ git config core.bare false $ git remote add -t master origin git://git.sv.gnu.org/lilypond.git $ git fetch --depth 1 remote: Counting objects: 4236, done. remote: Compressing objects: 100% (4031/4031), done. remote: Total 4236 (delta 643), reused 1669 (delta 191) Receiving objects: 100% (4236/4236), 15.06 MiB | 206 KiB/s, done. Resolving deltas: 100% (643/643), done. From git://git.sv.gnu.org/lilypond * [new branch] master - origin/master $ Regards, /Karl Hammar --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: GOP-PROP 1: python formatting
Graham: (this proposal will be rushed because nobody will argue against it. Initial discussion 6 June, summary and tentative decision 8 June, implementation 10 June) Having set a policy about policy discussions and then breaking it the first thing you do is not a good policy. If there is no rush, don't break the rules. On the contrary, how can I feel confidence in the rules if they are beeing broken. Proposal: letâs follow PEP-8. http://www.python.org/dev/peps/pep-0008/ * use 4 spaces per indentation level * never max tabs and spaces In the code I presume, in strings this wouldn't apply. * Code indented with a mixture of tabs and spaces should be converted to using spaces exclusively Presenting rules without rationales, gives no ground for decisions. One could the feeling that this is pure nonsense. If you really care about this, provide a pretty printer instead, and stipulate that all code should go through that. I propose instead: . Don't ever provide patches whith white space changes (in the code) unless the patch is only about white space changes Regards, /Karl Hammar --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: GOP-PROP 1: python formatting
Graham: On Mon, Jun 06, 2011 at 10:58:11AM +0200, Karl Hammar wrote: Graham: (this proposal will be rushed because nobody will argue against it. Initial discussion 6 June, summary and tentative decision 8 June, implementation 10 June) Having set a policy about policy discussions and then breaking it the first thing you do is not a good policy. ... I felt a bit of a rush because the mix of indentation is delaying some much-needed improvements to our build system, but I agree ... Well then, say so. I wouldn't argue against that. ... Presenting rules without rationales, gives no ground for decisions. One could the feeling that this is pure nonsense. If you really care about this, provide a pretty printer instead, and stipulate that all code should go through that. I am not aware of any pretty printers for python code -- remember A quick search did not turn up anything useful either, well, sorry for that then. For the simple case of removing tabs, col -b -x could be one. Manually finding/expanding tabs is a pain. that unlike C++ or scheme, indentation in python is the way that one indicates code blocks. (this makes mixing tabs and spaces particularly horrible!) ... There is your rationale. There is a check the python code should pass python -tt, maybe it could be run as part of some make process... Regards, /Karl Hammar --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Openoffice import?
David: Offlist URL:http://www.haumacher.de/svg-import/ has been suggested to me. The JAR file mentioned in there works for converting lilypond-book images placed into PDF, extracted with preview.sty, converted with pdf2svg to SVG to .odg files (OpenOffice Draw Graphics?) as OLE links into an OpenOffice document from where they will export to scalable PDF again. So there is a reasonable hope they might also export to scalable DOC contents. ... I found http://sk1project.org/modules.php?name=Productsproduct=uniconvertor which claims to be able to do eps - svg, can that be of any help... Regards, /Karl Hammar --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Black mensural notation
James Lowe: ... Lukas (although for some reason I am getting bounces from your email address so I hope you are reading this on the lists)! Lukas cannot do anything about this except changing mail provider. The mailing list should be fine. You can do something by making sure the sending mailserver ip-number and dns name points back to each other, ask your mail- and dnsadmin. Is anyone else getting 10.mx.freenet.de rejected your message to the following e-mail addresses: Lukas Pietsch (lukas.piet...@freenet.de)mailto:lukas.piet...@freenet.de 10.mx.freenet.de gave this error: inconsistent or no DNS PTR record for 173.221.96.141 (see RFC 1912 2.1) Yes, freenet.de only accepts mail from mail-servers with a valid reverse pointer. If I like to find out who 172.221.96.141 is, I can do: $ host -t ptr 173.221.96.141 141.96.221.173.in-addr.arpa domain name pointer cas-ftl1.datacore.com. $ But cas-ftl1.datacore.com doesn't poin back to the same number: $ host cas-ftl1.datacore.com. cas-ftl1.datacore.com has address 205.237.192.88 $ Regards, /Karl Hammar --- Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Black mensural notation
Lukas: Hi, I'm quite new to Lilypond programming, but I thought I'd jump in at what seemed to me to be pretty much the deep end, and try if I could get black mensural notation implemented. Here's what I've come up with so far: http://lukas-pietsch.de/Music/blackmensural.ly (source file) http://lukas-pietsch.de/Music/blackmensural.pdf (doc) It's all just scheme functions and embedded postscript, and works for me under the current stable 2.12.3. The biggest stumbling block turned out to be black ligatures, which I found no other way of doing than rewriting from scratch in a rather hackish way, sidestepping the normal ligature engraver completely; and horizontal spacing, for which I had to figure out some rather ugly workarounds (probably far from optimal). If anybody finds it useful, please feel free to use or modify. On line 22 in the ly-file: %% Accidentals are valid only once (same as Shouldn't the accidental be valid for the next note and any same repeted note, this has bothered me with the current white mesural support. E.g. if you say bes bes bes, it should be only one b-flat sign? And the b-flat sign could be displaced horisontally to the left as in gregorian chant? The rests looks a little thick in section 1.4 in the pdf. Wery nice, I'm very happy to see this!!! Regards, /Karl Hammar - Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: critical issues
Graham: On Sat, Jan 01, 2011 at 09:10:49AM +0100, Werner LEMBERG wrote: ... But maybe there is a group of LilyPond philanthropists who can afford this and are willing to do so... I'm not optimistic about that; I think a more realistic opportunity would be to get some grant money from some artistic organization. ... Of course, writing artistic and research grants is a non-trivial amount of work, and it's hardly guaranteed to have any results. But I think that with the right angle -- be that collaborative folk music archival, or high-quality, specialized music notation, or educational software for cheap 3rd-world donated computers, I could imagine getting a grant. In Uppsala there is one person at the music institution who is interested in Schenker analysis. Could that be a lead? Regards, /Karl Hammar - Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: sustainable development in LilyPond
Graham: ... http://percival-music.ca/blog/2010-08-01-sustainable-development.html ... Thank you for the slides, I liked them. Regards, /Karl Hammar - Aspö Data Lilla Aspö 148 S-742 94 Östhammar Sweden +46 173 140 57 ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Issue 881 Arpeggios may collide with laissezVibrer ties (was Re: Critical issues)
- This is a multipart MIME message. Karl Hammar: 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. Sorry to bring this up again, but this conclusion is still unproven, since the test was done with test-redo. As I have found out in the thread for the issue 915, test-redo does not test the things we wanted to test. *** $ git-status # On branch master # Untracked files: # (use git add file... to include in what will be committed) # # issue1195044_1_2.diff # issue1195044_1_3.diff # issue931041_1.diff nothing added to commit but untracked files present (use git add to track) $ git-log | head -1 commit 22d889f4d27469864c31db81445e9de49774ae23 $ cat issue1195044_1_[23].diff | patch -p1 patching file scm/define-grobs.scm patching file scm/output-lib.scm $ make check log 21 ... And I get a distance of 18.406667 in the attached output. The difference seems to be because the c and the b are horizontally closer to each other. Regards, /Karl Hammar attachment: tie-semi-single.compare.jpegattachment: tie-semi-single.png___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Issue 881 Arpeggios may collide with laissezVibrer ties (was Re: Critical issues)
And here comes the test-baseline file. Regards, /Karl Hammar attachment: tie-semi-single.png___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Issue 915 Multi-measure rests dependent on prefatory matter in other staves
Neil Puttock: On 15 May 2010 14:37, Karl Hammar k...@aspodata.se wrote: git-pull wget http://codereview.appspot.com/download/issue931041_1.diff patch -p1 issue931041_1.diff --dry-run patch -p1 issue931041_1.diff make log 21; make test-redo log 21 I very rarely use `make test-redo'. The difference between test-redo and check is small: $ grep -A 6 ^test-redo: GNUmakefile test-redo: for a in `cat $(RESULT_DIR)/changed.txt` ; do \ echo removing $$a* ; \ rm -f $$a* ;\ done $(MAKE) check $ grep RESULT_DIR GNUmakefile | head -1 RESULT_DIR=$(top-build-dir)/out/test-results $ cat out/test-results/changed.txt; echo # missing final newline input/regression/out-test//test-output-distance input/regression/out-test//rest-collision-beam-note input/regression/out-test//tree $ Does it matter, if not, why is in the contributor manual ? I basically do what Carl outlined in the other thread: make test-baseline git apply issue931041_1.diff make check The contributor manual says (3.6.3.1): * Initial test: make [-jX] make test-baseline make [-jX CPU_COUNT=X] check * Edit/compile/test cycle: _## edit source files, then..._ make clean_## only if needed (see below)_ make [-jX]_## only if needed (see below)_ make test-redo_## redo files differing from baseline_ make [-jX CPU_COUNT=X] check _## CPU_COUNT here?_ * Reset: make test-clean It has an make check after the test-baseline which you don't have, is it redundant? git-apply and patch -p1 produces the same result, so that is not the cause of our different output: $ cd .. $ cp -a lilypond/ lilypond.b $ cd lilypond.b/ $ git-apply -v issue931041_1.diff Applied patch lily/bar-line.cc cleanly. Applied patch lily/include/bar-line.hh cleanly. Applied patch lily/include/note-spacing.hh cleanly. Applied patch lily/include/paper-column.hh cleanly. Applied patch lily/multi-measure-rest.cc cleanly. Applied patch lily/note-spacing.cc cleanly. Applied patch lily/paper-column.cc cleanly. Applied patch scm/define-grob-properties.scm cleanly. Applied patch scm/define-grobs.scm cleanly. $ cd ../lilypond $ patch -p1 issue931041_1.diff patching file lily/bar-line.cc patching file lily/include/bar-line.hh patching file lily/include/note-spacing.hh patching file lily/include/paper-column.hh patching file lily/multi-measure-rest.cc patching file lily/note-spacing.cc patching file lily/paper-column.cc patching file scm/define-grob-properties.scm patching file scm/define-grobs.scm $ cd .. $ diff -Naur lilypond* $ Also, it would be much easier to look throuht the patch if it did not contain so many whitespace changes. In the first 100lines, I see: Why would you want to look at the bare diff? Isn't the diff a complete statement of what you want to change? Also you *did* cite the diff. The whole point of uploading the patchset to Rietveld is to make it easier to see the changes in each file. Well, it might help someone else, but not me -- I have mail and git. Putting things on web sites and getting things from them makes it take more time. *** If I take it from the beginning: $ git-log | head -1 commit 22d889f4d27469864c31db81445e9de49774ae23 $ git-status # On branch master # Untracked files: # (use git add file... to include in what will be committed) # # issue1195044_1_2.diff # issue1195044_1_3.diff # issue931041_1.diff # log nothing added to commit but untracked files present (use git add to track) $ make clean /dev/null 2/dev/null $ make log 21 $ make test-baseline log1 21 Processing ./snippet-map-1678077630 ... Processing 41/lily-67666737 Failed files: () $ make check log2 21 Processing ./snippet-map-1678077630 ... Processing 90/lily-015675ea Failed files: () $ patch -p1 issue931041_1.diff $ make log3 21 $ make test-redo log4 21 Processing ./snippet-map--1837524129 ... Failed files: () $ fgrep 'img src=input/regression/out-test/' out/test-results/index.html | grep -v test-output-distance.compare.jpeg img src=input/regression/out-test//rest-collision-beam-note.compare.jpeg style=border-style: none; max-width: 500px; $ # above file attached $ make check log5 21 Processing ./snippet-map-170730255 ... Failed files: () $ # the same output is produced $ ls -1 input/regression/out-test/multi-measure-rest-multi-staff-center* input/regression/out-test/multi-measure-rest-multi-staff-center-1.eps input/regression/out-test/multi-measure-rest-multi-staff-center-1.signature input/regression/out-test/multi-measure-rest-multi-staff-center-systems.count input/regression/out-test/multi-measure-rest-multi-staff-center-systems.tex input/regression/out-test/multi-measure-rest-multi-staff-center-systems.texi input/regression/out-test/multi-measure-rest-multi-staff-center.eps input/regression/out
Re: Issue 915 Multi-measure rests dependent on prefatory matter in other staves
Karl Hammar: Neil Puttock: ... I basically do what Carl outlined in the other thread: make test-baseline git apply issue931041_1.diff make check The contributor manual says (3.6.3.1): * Initial test: make [-jX] make test-baseline make [-jX CPU_COUNT=X] check * Edit/compile/test cycle: _## edit source files, then..._ make clean_## only if needed (see below)_ make [-jX]_## only if needed (see below)_ make test-redo_## redo files differing from baseline_ make [-jX CPU_COUNT=X] check _## CPU_COUNT here?_ * Reset: make test-clean After yet another run: make clean /dev/null patch -p1 issue931041_1.diff make log 21 make check log 21 I get the expected output (i.e. Neil's output). Regards, /Karl Hammar ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
contributors manual
From the debacle with Issue 915 I have learned that . make test-redo does not catch regression due to program changes . rest-collision-beam-note.ly changes is more or less expected . section 8 Regression tests is related to section 3.6.3.1 Does make check after test-baseline serve any purpose ? What is the purpose for test-redo, is it there to test changes in the regression test files themselves, and only that ? Regards, /Karl Hammar ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Issue 881 Arpeggios may collide with laissezVibrer ties (was Re: Critical issues)
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.
***
But I also notice that there are no test for the present problem.
Perhaps one should be added?
The one presented in the report[1] could be a candidate, comments?
\version 2.12.1 % regression: 2.10.33 and 2.8.8 are ok
{
f'\laissezVibrer f a' \arpeggio
f'\laissezVibrer g a' \arpeggio \mark never
f'\laissezVibrer a a' \arpeggio
f'\laissezVibrer b a' \arpeggio
\bar ||
f'\laissezVibrer f g' \arpeggio
f'\laissezVibrer g g' \arpeggio \mark sometimes
f'\laissezVibrer a g' \arpeggio
f'\laissezVibrer b g' \arpeggio
\bar ||
f'\laissezVibrer f f' \arpeggio
f'\laissezVibrer g f' \arpeggio \mark always
f'\laissezVibrer a f' \arpeggio
f'\laissezVibrer b f' \arpeggio
}
***
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 g4.\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?
Regards,
/Karl Hammar
[1] http://code.google.com/p/lilypond/issues/detail?id=881
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Issue 915 Multi-measure rests dependent on prefatory matter in other staves
Patching with [3] (see [2]) still gives something (see attachment) more like rest-2.12.png in [1], i.e. it was not fixed. The comment for the fix: Allow user override of left/right spacing of full-bar rests relative to barlines or prefatory material. implies that one has to do an override. The comment gives no help for what override to use. Since this regression is considered critical, should not the default be for 2.11.27 behaviour instead of 2.11.28 or later ? Regards, /Karl Hammar [1] http://code.google.com/p/lilypond/issues/detail?id=915 [2] http://codereview.appspot.com/931041/show [3] http://codereview.appspot.com/download/issue931041_1.diff 3.pdf Description: 3.pdf ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Issue 915 Multi-measure rests dependent on prefatory matter in other staves
Neil Puttock:
On 15 May 2010 09:58, Karl Hammar k...@aspodata.se wrote:
Patching with [3] (see [2]) still gives something (see attachment)
more like rest-2.12.png in [1], i.e. it was not fixed.
You must've done something wrong when applying the patch/rebuilding.
I wouldn't post a patch on Rietveld without first confirming the
regression is fixed.
I don't think so either, you are more than welcome to point out what I
did wrong:
git-pull
wget http://codereview.appspot.com/download/issue931041_1.diff
patch -p1 issue931041_1.diff --dry-run
patch -p1 issue931041_1.diff
make log 21; make test-redo log 21
 Allow user override of left/right spacing of full-bar rests relative
 to barlines or prefatory material.
implies that one has to do an override. The comment gives no help for
what override to use.
The patched files should make it clear:
+ (spacing-pair ,pair? A pair of booleans which set the spacing for a
+multi-measure rest relative to its left and right @code{BreakAlignment}s.
+If true, the rest is spaced to the bar line, ignoring prefatory items such
+as clefs and time signatures; if false, it is spaced relative to the
+BreakAlignment, which includes prefatory items. )
+ (spacing-pair . (#f . #t))
^ default value for MultiMeasureRest
Yes, I can see the lines (last in the patch) when you point the out for me.
But, nevertheless I got the result I got.
Also, it would be much easier to look throuht the patch if it did not
contain so many whitespace changes. In the first 100lines, I see:
+bool
+Bar_line::non_empty_barline (Grob *me)
+{
+ return has_interface (me) !me-extent (me, X_AXIS).is_empty ();
+}
+
and the rest is whitespace changes. That makes the reader tired.
Regards,
/Karl Hammar
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Critical issues
Carl Sorensen: On 5/13/10 1:11 PM, Karl Hammar k...@aspodata.se wrote: ... make test-baseline ... make check ... Ok, done that. With the guidance from http://code.google.com/p/lilypond/issues/detail?id=881: I can't explain why, but making the print function pure by redefining ly:tie::print just for LaissezVibrerTie seems to fix this (I haven't done a regtest check, so it might end up breaking something else): (define-public (laissez-vibrer::print grob) (ly:tie::print grob)) (then add laissez-vibrer::print to pure-print-callbacks) I tried this: diff --git a/scm/define-grobs.scm b/scm/define-grobs.scm index 176debd..35186f8 100644 --- a/scm/define-grobs.scm +++ b/scm/define-grobs.scm @@ -2345,8 +2345,12 @@ (interval-union '(0 . 0) (cons smaller larger))) '(0 . 0 +(define-public (laissez-vibrer::print grob) + (ly:tie::print grob)) + (define pure-print-callbacks (list + laissez-vibrer::print fret-board::calc-stencil note-head::brew-ez-stencil print-circled-text-callback $ make test-redo out/test-results/index.html that shows the results of the regression tests. ... I got (except the test-output-distance.ly) an distance of 0300030 and HEAD is: 22d889f4d27469864c31db81445e9de49774ae23 to the right and left plus the git-diff to the rigth. So I assume it was the wrong patch try. Does anybody have Neil's proposed fix available ? Regards, /Karl Hammar ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Issue 881 Arpeggios may collide with laissezVibrer ties (was Re: Critical issues)
Carl Sorensen: On 5/14/10 7:01 AM, Karl Hammar k...@aspodata.se wrote: Carl Sorensen: ... You also need to redefine the 'stencil for laissez-vibrez tie in scm/define-grobs.scm. ... I can help with doning the regression test. Second-guessing what Niels patch was about was not included in that offer. Regards, /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Critical issues
Issue 881: Arpeggios may collide with laissezVibrer ties According to the bug tracker, v2.11.19's output is what to aim for. Neil gave the fix: (define-public (laissez-vibrer::print grob) (ly:tie::print grob)) (then add laissez-vibrer::print to pure-print-callbacks) But he has not done a regtest check. /// How do one do a regtest? I remember there were discussion about this a while ago, but doing a quick search (regtest check lilypond) did not turn up anything useful, except [1] and [2]. [1] http://lilypond.org/doc/v2.13/Documentation/contributor/checking-and-verifying-issues [2] http://lilypond.org/test/ Perhaps it should be documented, maybe it already are. Doing info ./Documentation/out/lilypond-contributor.info-1 I find: 8.1 Introduction to regression tests The regression tests are automatically compiled using special `make' targets. The output of the regression tests is also automatically So, what targets? The targets test* seems to have something with the regression directory to do, but I don't understand how to do a comparision to a known good set. . Do we have a known good set? . If so, how do I compare current output with it? Regards, /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Critical issues
Issue 815: Enhancement: AJAX-powered search auto-completion for the online documentation Issue 1038: more technical website items Theese two seem to be related to the web site, not to the released software. I can understand that it can be critical for the official site, but how can that be critical for the release ? Regards, /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Critical issues
Graham: On Thu, May 13, 2010 at 10:06:45AM +0200, Karl Hammar wrote: How do one do a regtest? Regression check; by compiling stuff. 8.1 Introduction to regression tests The regression tests are automatically compiled using special `make' targets. The output of the regression tests is also automatically So, what targets? They might be make baseline-test, followed by applying a patch, followed by make check, but I'm not certain. It's explained somewhere in Contributing 2. Ok, found something in 3.6.3 Testing LilyPond (though nothing in the chapter on regression tests). * Initial test: make [-jX] make test-baseline make [-jX CPU_COUNT=X] check * Edit/compile/test cycle: _## edit source files, then..._ make clean_## only if needed (see below)_ make [-jX]_## only if needed (see below)_ make test-redo_## redo files differing from baseline_ make [-jX CPU_COUNT=X] check _## CPU_COUNT here?_ Hmm, the make check seems redundant since test-redo already does it: $ find . -name GNUmakefile | xargs grep -A 10 test-redo ... ./GNUmakefile:test-redo: ./GNUmakefile- for a in `cat $(RESULT_DIR)/changed.txt` ; do \ ./GNUmakefile- echo removing $$a* ; \ ./GNUmakefile- rm -f $$a* ;\ ./GNUmakefile- done ./GNUmakefile- $(MAKE) check ... ./scripts/build/out/output-distance seems to be the workhorse of the regression tests. I cannot find any useful documentation of it with: find . -type f | xargs grep output-distance except the source code itself. But if I already have a known good result from the code tracker, how do I compare it with the new result? Regards, /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Critical issues
Graham complaind about 10 critical issues in the last lilypond report.
***
Issue 815: Enhancement: AJAX-powered search auto-completion for the online
documentation
Why is this a critical issue for the lilypond release?
***
Issue 1080: Regression: bar lines in double bar are positioned too close
together
pnorcks mentions commit 27a4d9354effb09c696925881ec4df007da8a0db
as a possible cause. Reverting part of that commit:
diff --git a/lily/bar-line.cc b/lily/bar-line.cc
index d3f21e5..85b2e52 100644
--- a/lily/bar-line.cc
+++ b/lily/bar-line.cc
@@ -196,11 +196,9 @@ Bar_line::compound_barline (Grob *me, string str, Real h,
/*
should align to other side? this never appears
on the system-start?
+ */
m.add_at_edge (X_AXIS, RIGHT, thin, 0);
m.add_at_edge (X_AXIS, RIGHT, thin, thinkern);
- */
- m.add_at_edge (X_AXIS, LEFT, thin, thinkern / 2);
- m.add_at_edge (X_AXIS, RIGHT, thin, thinkern / 2);
}
else if (str.find (S) != NPOS || str == |._.|)
{
gives me the attached grace-start result which resembles the 2.13.17
result presented in the bug tracker.
What should we do about it?
***
Issue 989: ensure that no information is only in the regtests
Though Graham complain about this issue in the report,
this seems to be taken by Valentin. He has a list at
http://wiki.lilynet.net/index.php/Regtests
What to do about it?
Shall we discuss individual items on the list?
***
Issue 915: Multi-measure rests dependent on prefatory matter in other staves
There seems to be a fix: http://codereview.appspot.com/931041/show
Do you need help testing the fix?
***
Issue 1031: constantly-changing input/regression/rest-collision-beam-note.ly
If it is changes every time, what is the correct output?
grace-start.pdf
Description: grace-start.pdf
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Where is the bug tracker?
Wouldn't it be nice to have a link to the bug tracker from http://lilypond.org/devel/ Regards /Karl Hammar ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Aligning single systems?
James Bailey:
On 07.05.2010, at 21:53, David Kastrup wrote:
Boris Shingarov b...@shingarov.com writes:
for theoretical work it often is necessary to write several short
systems in one line, interspersed with text.
...
\markuplines {
\justified-lines {
some text
\score { e''' \layout {} }
more text
\score { c \layout {} }
}
}
Nothing lines up. Text is in the sky, music systems are skewed.
While not completely ideal, frequently, markup has to be aligned. So
the original snippet:
[Bailey: attachment, above with \vcenter added]
One can add placeholders to define vertical extent (how do one make
it zero length?):
\version 2.13.0
\paper {
indent = 0\mm
}
range = {
\hideNotes
c e4
\unHideNotes
}
\markuplines {
\justified-lines {
some text
\vcenter
\score { { \range e''' } \layout { } }
more text
\vcenter
\score { { \range c } \layout { } }
}
}
Regards,
/Karl Hammar
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Aligning single systems?
David:
Hi, for theoretical work it often is necessary to write several short
systems in one line, interspersed with text.
I don't manage to have
a) new systems continue aligned to the previous system in a line, like
when doing
...
b) to have interspersed text appear at useful height with relation to
the surrounding score.
An additional problem is when using lilypond-book: there is no baseline
info for the included images, leading to additional nuisances when
images are to be placed in-line.
With latex, I've done (see [1], for my tries):
\newcommand{\INL}[1]{\raisebox{-0.8ex}{\includegraphics[height=3ex]{#1}}}
...
The limit of the \emph{Medium} register in all female voices varies
from \INL{op31_fig4.eps}; as a general rule, however,
\INL{op31_fig5.eps} should be looked as the highest note.
...
But I have to hand-tune the \raisebox and the height.
Would it be possible to have some Staff property, say, baseline-height
that specifies a Staff line to be aligned with the baseline of
surrounding constructs? If it were #f, we'd get the old behavior, and
of course only one Staff's baseline-height could be heeded within a
StaffGroup. This might not be easy to make work in lilypond-book, but
at least for interspersed text and staffs all within Lilypond, it might
help.
Good idéa!
What I would need is the total height, and the amount of space
below the staff,
The values are available in some form, since the notation manual
(for v2.13), sec. 4.6.1 Displaying spacing demonstrates
\paper { annotate-spacing = ##t }, where one can find numbers for
bottom-of-extent and extent-estimate.
As a workaround one could, if possible, set thoose two heights (with
margins) to fixed values, then you'd only have to hand-tune it once.
Regards,
/Karl Hammar
[1] git://turkos.aspodata.se/musik.git
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Why don't we get rid of \chordmode?
Kieren:
It's not obvious to me whether
c\chord #'(1 7)
should produce c b or c bes. Musically speaking, I'd look at
the key signature; if it were c major, I would assume it meant
c b since b is the seventh note of the scale.
This may be a problem for the numeric syntax. I suppose we could
define each number as being a perfect, major, or minor interval;
users can adjust those intervals with + or - as required.
Or... we could use dodecaphonic intervals, i.e.
c\chord #'(1 11) is a minor seventh
c\chord #'(1 12) is a major seventh
Of course, none of these ideas support systems with
other-than-12-tones-per-octave
=(
I propose we skip the number things and go for a local translate,
see below last in mail.
One could misuse floating poing values:
c\chord_float #'(1 6.5) % minor seventh
c\chord_float #'(1 7.0) % major seventh
To explain that eis == f is doable, but how would one explain
that 3.5 == 4 ??
(That flaw was present in my inital proposal c 3 5)
Rationals:
c\chord_rational #'(1 5/4 3/2) % perfect 3rd and 5th
% replace 5/4 etc. with the proper scheme expression
Cents:
c\chord_cent #'(0 386 702) % just intonation
Or other variants.
One could also define a shorthand translating thing, something like
(though I'm not shure what syntax to use):
cis\chord_barre c e g - cis eis gis
//
If we are discussing shortcuts to notate music (chordal notation are
shortcuts, is it not?), consider the similarity of:
d fis a - d\chord #'(1 3 5) % parallell music
{d fis a} - d\seq #'(1 3 5) % sequential music
There are two things here:
. replacing a note name with a number, eg. e - 3
. naively replacing 3 for the interval third, gives use the enharmonic
problem of 3.5 == 4, and what does 7 mean, a minor or major 7th?
. using dodecaphonic, cents, retionals or the like somehow misses the
shortcut-idéa
. a local transposition, eg. d\chord... == \translate c d { c\chord... }
this would be a better \translate, since it would put the music
in the current octave
The chords are easily defined, we don't need any special
syntax for that (like \chordmode):
chord_major = \relative c { c e g }
\translate below does not work since the d is not relative the
previous music:
... music ...
\translate c d \chord_major
... music ...
Could something like this (or some other syntax) be useful:
... music ...
d\tr\chord_major % -- d fis a in the current octave
... music ...
Going this route, one could define whatever complex chord expressable
in an ordinary ...
Regards,
/Karl Hammar
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Why don't we get rid of \chordmode?
David Kastrup:
Graham Percival gra...@percival-music.ca writes:
...
And what if you see \chordmode { c,4:1/c c g,:1/g c } in the input
(which is basically how you put bass notes in now if you really must)?
...
For the simple chords c\maj c\dim c\maj7 etc. could suffice.
For more complex chords one could extend the c e g syntax to c 3 5.
Regards,
/Karl Hammar
-
Aspö Data
Lilla Aspö 148
S-742 94 Östhammar
Sweden
+46 173 140 57
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Why don't we get rid of \chordmode?
Graham: ... OTOH, what about doing something like this: c\maj c\dim c\chord #'(1 4 5) c\chord #'(1 3 5 7 11) I like this. Regards, /Karl Hammar ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Solution for including a file only once
Reinhold Kainhofer: ... In my view there are two different kinds of include uses: 1) To include package-like .ily files, which define additional functionality. There many utility functions really might/should be in their own namespace, but many definitions (and in particular all \header, \paper or \layout blocks) need to be at toplevel. I would be interested in conditionally including files with \paper and \header blocks, so can with a command line flag get output for choirbook booklets or plain a4 sheets. The common thing with this thread would be the ability to \include thoose blocks. 2) ... ... I think it should be as easy and straightforward to use to a non-programmer as possible, while allowing more complex use without making easy things harder than necessary for easy things. Ack. Regards, /Karl --- Karl HammarAspö Data k...@aspodata.se Lilla Aspö 148 Networks S-742 94 Östhammar +46 173 140 57 Computers Sweden +46 70 511 97 84 Consulting --- ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: The \\ construct for simultaneous voices
...
or when there are lyrics to assign.
What do you mean here?
Do you mean lyrics to assign to the *second voice* (since the first
voice assignment would be automagic)?
Try:
\version 2.13.0
\score {
\new Staff {
\time 4/4
\relative g' { g4 g \\ d g2 }
}
\addlyrics{ a b c }
}
No voice in the \\ section gets any lyrics here.
Regards,
/Karl
---
Karl HammarAspö Data k...@aspodata.se
Lilla Aspö 148 Networks
S-742 94 Östhammar +46 173 140 57 Computers
Sweden +46 70 511 97 84 Consulting
---
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: ly files as a MIMEtype?
Valentin wrote: sometimes I wonder what it would take to register .ly files as an official text/lilypond MIMEtype (for instance at IANA, on http://www.iana.org/cgi-bin/mediatypes.pl ) Very nice idéa! But it seems you have to read a lot of rfc's for that. An easier way might be to file a bug report in debian like http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=379838 and other distributions. Or the lilypond installation might add it. ... (if you don't know which icon I'm referring to, have a look at http://git.savannah.gnu.org/gitweb/?p=lilypond.git;a=blob_plain;f=Documentation/pictures/ly.xpm ) Therefore, I have taken the liberty to draw some icons (attached here), based on the note icon. - one follows the Apple human interface guidelines - one follows the MS Vista guidelines (yeah, no comment) - one follows the Tango guidelines (GNOME, XFCE and GTK) - one follows the Oxygen guidelines (the official theme for KDE4) Wery nice examples, but I must say my first reaction is the it looks like a stand, a claw, or something, but not a note. I don't know what to do with these, maybe you guys will have an idea. It would be just great if GNOME or KDE included a default lilypond MIMEtype Icon for .ly files (the Windows/Apple ones could just be provided with the binary program). Great work. /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Build requires t1asm
It's a bit of a halfbaked solution. If you don't want to compile lily, why not use the pre-rolled binary. Also, it would be nice if some of these perverse hackers could just add an autoconf check for said program. Hmm, something like [1] (s/inkscape/t1asm/) ? [1] http://lists.gnu.org/archive/html/lilypond-devel/2007-10/msg2.html On Fri, Jul 18, 2008 at 8:37 AM, Werner LEMBERG [EMAIL PROTECTED] wrote: FWIW I also like the idea of having less dependencies, but not at the price of being unable to hack the source. Even if that is the source to a font. Well, adding pre-compiled fonts to the tarball just remove a dependency. As soon as you remove the fonts, they are rebuilt. Where's the problem? Regards, /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: \once \macro
Graham:
[EMAIL PROTECTED] (Karl Hammar) wrote:
1) Does anybody else remember if this worked in the past? (ie am
I going crazy?)
{
c'4
\once \override Stem #'direction = #DOWN
%\stemDown
c'4
c'
}
Yes, I know that works. I'm asking a general question about the
way we handle tweaks, for documentation purposes. If I had an
actual question about actually writing music, I'd sent it to the
-user list.
I forgot to answer the question, I see. I tested 2.6.5 and 2.11.33 and
it did not work with them.
Regards,
/Karl
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: \once \macro
I could have sworn that
{
c'4
\once \stemDown
c'4
}
worked, but it clearly doesn't.
1) Does anybody else remember if this worked in the past? (ie am
I going crazy?)
...
Try
{
c'4
\once \override Stem #'direction = #DOWN
%\stemDown
c'4
c'
}
Regards,
/Karl
---
Karl HammarAspö Data [EMAIL PROTECTED]
Lilla Aspö 2340Networks
S-742 94 Östhammar +46 173 140 57 Computers
Sweden +46 70 511 97 84 Consulting
---
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Incipits
Till:
Juergen Reuter schrieb:
as far as I understand, all problems with incipits boil down to the
following two major issues:
(1) horizontal spacing, and
I did not focus on that, it is an ancient notation issue, it is not
specific to incipits.
(2) the system start delimiter.
Karl's solution unfortunately leaves horizontal space between the
incipit and the actual score; hence it does *not* solve issue (2).
Rather, the score lines should not stop inbetween.
If you don't like the space, then remove the \stopStaff \startStaff
thing. But then, whats wrong with the \bar || variant as Till suggests
below. (cont. below)
Well, I am not so sure about that. I agree, the space is a bit too big
here, but I personally like Nocolas' solution where the space mainly
covers the distance between the choirstaff brace and the beginning of
the staff. I know that much incipits are engraved with the double line
method that is now used in the templates, but it feels logical to me
that the real staff with the brace start only after the incipit. I
somehow recall this would be also hand engraved convention, but I
might be wrong. There is an example pdf on the list sent by Stefan
Slapeta that shows a broad range of incipits which show mainly the staff
lines going through all the way. So it might be my personal opinion only
that it would be nice to have the incipit seperated by a little white space.
Getting less space is simple, us less space between \stop-\startStaff.
See attached pdf.
$ diff -u c.ly.old c.ly
--- c.ly.old2008-02-11 19:41:54.923054408 +0100
+++ c.ly2008-02-11 19:42:34.417301808 +0100
@@ -34,8 +34,8 @@
\stopStaff
\once \override Score.TimeSignature #'print-function = ##f
\nor
- \time 3/8
- s4.
+ \time 1/16
+ s16
\revert Score.SpanBar #'print-function
\revert Score.BarLine #'print-function
\startStaff
@@ -74,7 +74,7 @@
\score {
\new ChoirStaff \with {
% a pain to set? Thats Right!!!
-\override SystemStartBracket #'extra-offset = #'( 38.8 . 0 )
+\override SystemStartBracket #'extra-offset = #'( 30.2 . 0 )
} {
\new Staff {
Till
IMHO, all other features have been quite well demonstrated in template
A.5.1 for a while now. Hence, you probably should concentrate on
solving these two remaining issues.
Regarding issue (2), maybe the right solution is to extend the
system start delimiter implementation in such a way, that
(a) automatic printing at the system start may be suppressed for a
specific range of the score with kind of
\turnSystemStartDelimiterOff and \turnSystemStartDelimiterOn
commands, and
(b) a kind of \bar #'system-start-delimiter command prints an
additional system start delimiter whereever you want. (But if this
bar should happen to appear at the system start, the delimiter
should not be printed twice.)
That would be nice, but you can use the workaround with the
extra-offset as shown above.
Regarding issue (1), I am not sure what the right solution is.
Please note that the note durations in the incipit are typically much
larger than in the actually score. Karl and Nicolas, did you check
what happens with the horizontal spacing in your incipit solutions, if
you double or quadruple the note durations in the incipit (and only
there)? In your examples, in the incipts you are unfortunately using
the same durations as in the actual score. I may be wrong, but I
suspect that your solutions do *not* solve issue (1).
No I did not try. Try this (\men \nor as in previous mail)
cold = { \clef petrucci-c4 }
cnew = { \clef treble_8 }
\new Voice \relative c {
\men\cold d\breve*1/8 e\longa*1/16 f\maxima*1/32 g\breve*1/8
\nor\cnew | d4 e2 f4~ | f2. g4 |
}
I tried with
B = #(ly:make-duration -1 0 1 8)
...
d\B e\L f\M
but it didn't work.
Maybe someone have know how to write a function \foo:
breve - breve*1/8
...
I tried with
% as a first test
#(define-public (nop music)
(if (ly:duration? music)
(let music (ly:make-duration -1 0 1 8
foo = #(define-music-function (parser location music)
(ly:music?)
(let ((new-music (ly:music-deep-copy music)))
(music-map nop new-music)
new-music))
but it didn't work.
%%%
You can use the lsr snippet 305 to scale down the note values (and a
modern lilypond version). There is the problem with/without ligatures
with solutions like A.5.1 and the one I provided. And it would
probably be som problems with incipits in vaticana style.
%%%
What would be nice is to be able to glue to \scores together, without a
newline. Then one could have independent engraving on the historical
and the modern part, or to have multiple parts stacked together,
continuing on the same line.
%%%
Also it would be nice to have an
#(set-accidental-style 'renaissance)
which would be like 'forget, except it remebers the
Re: Incipits
Han-Wen:
Sorry for looking at this so late.
General comments:
- you're adding an enormous bunch of code to the instrument name
engraver. That's a bad idea. If this idea merits an engraver, it
should be a separate one.
...
- If I'm reading this correctly, this is all a grand hack to add a
little staff to an instrument name. I don't see the essential
difference between this and
\set instrumentName = \markup { \score { .. } the-real-name }
except that this is longer, more error prone and less flexible. Again,
what problem are you trying to solve?
2008/2/9, Nicolas Sceaux [EMAIL PROTECTED]:
Here is draft patch: the Instrument_name engraver is used to create an
Incipit grob before the first line when an incipit has been specified.
There are several problems in it:
...
Both these problems are solved in the following patch.
Do core developpers see a problem with it?
Inicipits can be done with the code below.
Regards,
/Karl
%---
\version 2.6.4
men = {
\override Accidental #'style = #'mensural
\override NoteHead #'style = #'petrucci
\override Rest #'style = #'neomensural
\override Staff.TimeSignature #'style = #'mensural
\override Stem #'flag-style = #'mensural
\override Stem #'thickness = #0.8
% \set Score.timing = ##f
\set Staff.defaultBarType =
#(set-accidental-style 'forget)
}
nor = {
\override Accidental #'style = #'default
\override NoteHead #'style = #'default
\override Rest #'style = #'default
\override Staff.TimeSignature #'style = #'default
\override NoteHead #'font-size = #'0
\override Stem #'flag-style = #'default
\override Stem #'thickness = #1.9
\override TimeSignature #'style = #'default
#(set-accidental-style 'default)
% \set Score.timing = ##t
}
staa = {
\override Score.SpanBar #'print-function = ##f
\override Score.BarLine #'print-function = ##f
\men
}
jump = {
\stopStaff
\once \override Score.TimeSignature #'print-function = ##f
\nor
\time 3/8
s4.
\revert Score.SpanBar #'print-function
\revert Score.BarLine #'print-function
\startStaff
\once \override Score.TimeSignature #'print-function = ##f
\time 1/16
s16
\override Score.SpanBar #'print-function = ##f
\once \override Staff.Clef #'full-size-change = ##t
\set Staff.forceClef = ##t
\clef bass
\time 4/4
\bar
}
%
% Music and lyrics
clefic = { \clef petrucci-c1 }
vic = \relative c' { c1 c }
lic = \lyricmode { A B }
clefmc = { \clef soprano }
vmc = \relative c' { \vic d }
lmc = \lyricmode { \lic c d e f g }
clefib = { \clef petrucci-f }
vib = \relative c { a1 a }
lib = \lyricmode { A B }
clefmb = { \clef bass }
vmb = \relative c { \vib g }
lmb = \lyricmode { \lib c d e f g }
%
\score {
\new ChoirStaff \with {
% a pain to set? Thats Right!!!
\override SystemStartBracket #'extra-offset = #'( 38.8 . 0 )
} {
\new Staff {
\set Staff.instrument = \markup { Cantus }
\context Voice = vc { \clefic \staa \vic \jump \clefmc \vmc }
}
\context Lyrics = lc \lyricsto vc { \lic \lmc }
\new Staff {
\set Staff.instrument = \markup { Bassus }
\context Voice = vb { \clefib \staa \vib \jump \clefmb \vmb }
}
\context Lyrics = lb \lyricsto vb { \lib \lmb }
}
\layout {
\context {
\Score
\consists Span_bar_engraver
\remove System_start_delimiter_engraver
}
}
}
c.pdf
Description: c.pdf
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: [patch] first-clef property
Nicolas Sceaux:
...
\tag does not solve the first-clef detection problem, but the two
editions problem, about which the proposed patch is not about.
I know about \tag,
...
I also find \tag messy and there is no else part either.
Anyway, wouldn't it be nicer to have some kind of scheme macro that
expands to code that prints an incipit? Your first clef could
then be just part of the incipit that the macro creates. And maybe
the clef's name either could passed as argument to the scheme
macro. Or, alternatively, you set an, say, original-clef property
that the macro recognizes and accordingly acts upon.
As I understand it, incipits are hackingly achieved using the
instrument name. I want instrument names to be defined separately
from incipit (they are not the same thing, there is no serious
reason to bind them, beside purely technical ones):
\new Staff
\global
\set Staff . instrumentName = \markup { The instrument name }
\clef xyz %% automagically set the incipit clef
{ ... the notes ... }
How could you make the mix of the two?
Now, seeing the incipt examples, I realize that my patch is a hack
too, for it would be nice to have also the time and key signatures,
not only the ancient clef.
For the time and key signatures you can have to varialbles like
mensural = {
\override Accidental #'style = #'mensural
\override NoteHead #'style = #'petrucci
\override Rest #'style = #'neomensural
\override Staff.TimeSignature #'style = #'mensural
\override Stem #'flag-style = #'mensural
\override Stem #'thickness = #0.8
\set timing = ##f
\set Staff.defaultBarType =
#(set-accidental-style 'forget)
}
normal = {
\override Accidental #'style = #'default
\override NoteHead #'style = #'default
\override Rest #'style = #'default
\override TimeSignature #'style = #'default
\override NoteHead #'font-size = #'0
\override Stem #'flag-style = #'default
\override Stem #'thickness = #1.9 %
default
\override TimeSignature #'style = #'default
#(set-accidental-style 'default)
}
and use them like
\new Staff { \normal \relative c' { ... } }
\new Staff { \mensural \relative c' { ... } }
For the clefs I use
%clefcs = { \clef petrucci-c2 }
%clefct = { \clef petrucci-c4 }
%clefcb = { \clef petrucci-f }
clefcs = { \clef treble }
clefct = { \clef treble_8 }
clefcb = { \clef bass }
What I think would be useful is either a setting
\override Clef = #'style = #'petrucci % e.g alto - petrucci-c3
\override Clef = #'style = #'modern % e.g petrucci-c3 - alto
\override Clef = #'style = #'lazysinger % e.g petrucci-c3 - treble
so I can include them in the variable above
or a function like
% always keep as close to source as possible
aa = \relative c { \clef petrucci-f \time 2/2 c\breve d\longa ... }
\makeThisMensural { \aa } % no change?
\makeThisModern { \aa } % petrucci-f - bass, c\breve - c1 ...
\makeThisIncipit { \aa } % mensural but only first breve or semiminima
Maybe creating an incipit engraver, reading new context properties,
like incipitKeySignature, incipitTimeSignature, etc, and creating a
In some instances it would be useful to be able to append \score's on
the same line, somewhat like if you used { \startStaff s1 \stopStaff }.
Regards,
/Karl
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: [patch] first-clef property
Reinhold Kainhofer:
Am Freitag, 1. Februar 2008 schrieb Nicolas Sceaux:
When typesetting ancient music, one may want to produce two editions:
eventually one with original clefs, as found in the manuscripts, and an
other one with new fashioned clefs. It is also custom in the later case
...
Actually, we're already two ;-)
I'm transcribing some old pieces by Schubert, who wrote the vocal voices
using
soprano/alto/tenor clef, while modern typesetting practice uses only the
treble clef for these voices. Now, I want to make two editions, one with the
original clefs, the other with modern clefs (but the original clefs as
incipits). So, you see, I'm having exactly the same problem.
...
There is two problems here,
1, should the clef look ancient or modern, e.g.
{ \clef petrucci-f } or { \clef bass }
2, should we exchange a C clef for a G clef
For problem 1 one can look at what theese do
\displayMusic { \clef petrucci-f c1 c }
\displayMusic { \clef bass c1 c }
(make-music
'SequentialMusic
'elements
(list (make-music
'ContextSpeccedMusic
'context-type
'Staff
'element
(make-music
'SequentialMusic
'elements
(list (make-music
'PropertySet
'value
clefs.petrucci.f
'symbol
'clefGlyph)
(make-music
'EventChord
...
(ly:make-pitch -1 0 0
(make-music
'EventChord
...
(ly:make-pitch -1 0 0))
Where the difference is clefs.petrucci.f vs. clefs.F.
So an \ancientToModernClef should traverce the list and exchange
ancient clefs.xx to a modern variant.
===
Problem two is similar
\displayMusic { \clef petrucci-c2 c1 c }
\displayMusic { \clef petrucci-c3 c1 c }
gives this diff
$ diff -u b.log b2
--- b.log 2008-02-02 20:46:38.888477652 +0100
+++ b2 2008-02-02 20:47:08.884108561 +0100
@@ -12,19 +12,19 @@
(list (make-music
'PropertySet
'value
-clefs.petrucci.c2
+clefs.petrucci.c3
'symbol
'clefGlyph)
(make-music
'PropertySet
'value
--2
+0
'symbol
'middleCPosition)
(make-music
'PropertySet
'value
--2
+0
'symbol
'clefPosition)
(make-music
and
\displayMusic { \clef petrucci-c2 c1 c }
\displayMusic { \clef petrucci-g c1 c }
this
$ diff -u b.log b2
--- b.log 2008-02-02 20:48:23.373258918 +0100
+++ b2 2008-02-02 20:48:44.370200812 +0100
@@ -12,13 +12,13 @@
(list (make-music
'PropertySet
'value
-clefs.petrucci.c2
+clefs.petrucci.g
'symbol
'clefGlyph)
(make-music
'PropertySet
'value
--2
+-6
'symbol
'middleCPosition)
(make-music
To make a generic \changeClef one has to find the part between a-b below (a
list), and exchange it for something else
(make-music
'SequentialMusic
'elements
(list (make-music
'ContextSpeccedMusic
'context-type
'Staff
'element
(make-music
'SequentialMusic
'elements
a
(list (make-music
'PropertySet
'value
clefs.petrucci.g
'symbol
'clefGlyph)
(make-music
'PropertySet
'value
-6
'symbol
'middleCPosition)
(make-music
'PropertySet
'value
-2
'symbol
'clefPosition)
(make-music
'PropertySet
'value
0
'symbol
'clefOctavation
b
...
(ly:make-pitch -1 0 0))
=
Or what do you think.
/Karl
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Ancient notation: spacing, ligatures, bar lines: some ideas...
Till:
Hi all, who are interested in Ancient notation.
I am.
I started now some trial how to improve the spacing of the notes. The
goal is to have an even spacing where each note head regardless of their
length takes up just the same space. Visible signs in the score as
accidentals, clefs and lyrics should influence the spacing, though, as
well as alignment of more than one system (even though this is quite
rare in mensural music, it should be possible).
If the notes are note are not aligned to each other (syncroniced on the
time axis) then we must more manually specify where to put them,
time-wise. I think it is good to skip such layouts for the moment.
To acchieve this I
started removing the spaces that get added to a notehead according to
its duration. I came up with those properties:
\override SpacingSpanner #'spacing-increment = #0
\override SpacingSpanner #'common-shortest-duration =
#(ly:make-moment 1 1)
\override SpacingSpanner #'base-shortest-duration =
#(ly:make-moment 4 1)
\override SpacingSpanner #'shortest-duration-space = #1
I have used
\override Score.SpacingSpanner #'base-shortest-duration = #(ly:make-moment 1
1)
\override Score.SpacingSpanner #'common-shortest-duration = #(ly:make-moment
1 4)
but it did not help much.
The first means, that longer notes than the smallest won't get more
spacing around them (ie it gets increased by 0).
The next three properties, though, I was not so sure about what they do.
As I understand I get quite close spacing when assigning the
common-shortest-duration to a long note. But it seems a whole is the
longes note it can handle,
Seems so.
I don't see any changes in the layout when
assigning it for instance to 4/1. The next property didn't seem to have
much influence on spacing in this case -- can it just be ommitted, or
did I understand this wrong?
The last, I understand, sets the space the shortest note takes, to 1,
that is to one notehead-width plus some amount left and right. I would
understand that if my common-shortest-duration is a whole that this
whole gets assigned the 1. But, as you can see from the attached file:
The first semibreves (wholes) are much wider spaced than the smaller
note values further on. When switching on the bar lines they reveal
quite a lot of space they take up. This might be the reason why the
semibreves get spaced so widely.
But there are also bigger and smaller gaps between notes of different
duration. They show better in the example whith normal note heads, but
are still present in the mensural examples as well. Yet they become more
obvious when using a barline instead of the empty barline. Is this
some space added to each bar? Something about centering notes in bars? I
Yes, the empty bar line seems to take some space. In
http://aspodata.se/noter/victoria/o_quam_gloriosum/motet/
it does so anyhow, but that might be because of the barnumbers.
don't understand where this space comes from.
Speaking of barlines: There are three ways avoiding bar lines, as I see it.
1. Remove the engraver. This is not a good solution, because sometimes
there have to be inserted bar lines, for instance at the end of a piece.
2. Switch on Cadenza. There are two problems with this approach: First
we don't have any breaks anymore. I thought of a funktion that enquires
the last bar number of the engraved piece and sets then in regular
spaces invisible bar lines, for instance repeating the pattern {\skip
1*4 \bar } n/4 times, if the measures are a whole long. The other,
bigger problem, are ligatures: they get really ugly squeezed when
switching on Cadenza. What can we do there?
3. The approach from now. In addition the space bar lines take should be
reduced. This again interferes with the cases when we actually need
barlines. They should then get (automatically) added additional space.
\cadensaOn is not a good solution, I use \set Score.defaultBarType =
One way to do it would be to have loong measures without bars:
% get rid of the bars
\set Score.defaultBarType =
% set the mensuration sign manually
\set Staff.timeSignatureFraction = #'(2 . 2)
% use what ever long measures you want
\time 2/1
Another problem are ligatures (and to some extent also ligature
brackets). If I understand the underlying techniques right, ligatures
get produced from legal note durations. Their duration is known to the
spacing engine and they get accordingly space around them. But since
ligatures are drawn as one graph of closely realted heads, the
additional space that is counted to fit normallly between noteheads gets
accumulated at the end of the ligature.
I see no other way if you use it in a score, what would you otherwise
do (i.e. if you want the notes to line up)?
In a single voice you could write it like
music = \relative c' {
d\breve*1/8 d\longa*1/16 b\breve*1/8 b\longa*1/16
Inkscape needed for make web
Suggests attached patch. Regards, /Karl diff --git a/configure.in b/configure.in index 15a41e9..8c98e9a 100644 --- a/configure.in +++ b/configure.in @@ -176,6 +176,9 @@ STEPMAKE_GUILE(OPTIONAL) # perl for help2man. STEPMAKE_PERL(OPTIONAL) +# inkscape for make web, to generate svg-files +STEPMAKE_INKSCAPE(OPTIONAL) + # New makeinfo for multi-page website docs Simply always require new # makeinfo should avoid bug reports; if people want to build lily # without the docs, they can find this comment and use an older diff --git a/stepmake/aclocal.m4 b/stepmake/aclocal.m4 index b69dc9c..5afdc76 100644 --- a/stepmake/aclocal.m4 +++ b/stepmake/aclocal.m4 @@ -1305,3 +1305,7 @@ AC_DEFUN(STEPMAKE_WINDOWS, [ STEPMAKE_PROGS(WINDRES, $target-windres windres, x) AC_SUBST(WINDRES) ]) + +AC_DEFUN(STEPMAKE_INKSCAPE, [ +STEPMAKE_PROGS(INKSCAPE, inkscape, $1) +])___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
texi2dvi, @appendixsubsec and utf-8
When running $ make web log 21 i get: ./fdl.pdftexi:371: Undefined control sequence. argument SUPPLÃ89 MENT : comment utiliser cette licence pour vos documents when it tries to run texi2pdf --batch lilypond-program.pdftexi in Documentation/fr/user/out-www The lines in fdl.pdftexi is: @page @appendixsubsec SUPPLÃ89MENT : comment utiliser cette licence pour vos documents Copying SUPPLÉMENT to two lines below makes the error message go away... Soo, what is going on, cannot you have utf-8 chars in arguments to @appendixsubsec, is my texi2pdf broken, or what? $ texi2pdf --version | head -1 texi2dvi (GNU Texinfo 4.8) 1.34 Regards, /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Bash completion strings
[EMAIL PROTECTED]: ... In order to handle the .ly-extension correctly, on needs to install bash completion string in file /etc/bash_completion.d/lilypond : # LilyPond completion. complete -f -X '!*.@(ly|LY)' lilypond Nice tip, but you have to have the shell option extglob to be able to use the @(...) construct: $ complete -f -X '!*.@(ly|LY)' lilypond $ lilypond TABTAB no reaction $ shopt -s extglob $ complete -f -X '!*.@(ly|LY)' lilypond $ lilypond TABTAB Mliv.ly Sliv.ly ... $ lilypond Why not just ignore uppercase LY and add lytex complete -f -X '!*.ly' lilypond complete -f -X '!*.lytex' lilypond-book Regards, /Karl --- Karl HammarAspö Data [EMAIL PROTECTED] Lilla Aspö 2340Networks S-742 94 Östhammar +46 173 140 57 Computers Sweden +46 70 511 97 84 Consulting --- ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Saying hello and considering my first contribution
[EMAIL PROTECTED]:
I am currently working with a cataloging project at the British
Library which is recording all musical manuscripts between the years
1600 and 1800, I have developed an application for them to data input
and manage the manuscripts. The information supplied with each
manuscript sometimes comes with an incipit for the beginning of the
music. Here is an example of an entire incipit pulled from the
record for Sonata 1 by Johann Sebastian Bach:
''4G{3G FbEDC'BA5 BG}{8GxF+}{3FEDEFGA''5C'B}/''8{C+3C5DC'3B''C}
{6.Ct'5B''C6D'A}'' 8G'3-{G5AB''CD}{8b E'B}/{6BA''bED}8-{BxF}
{6D3DbEDC'B''5C'A}{3BxFGDbE''5DC'B''CC'B''C}/
I would like to convert this to Lily textual notation so that Lily
can then reproduce it in musical notation so regular users of the web
catalogue can just click a link and run this up into Lilypond. My
...
If you provide the syntax rules for the record I might be able to help
you. Will the archive be publically searchable or is just for thoose
who have a subcription?
Regards,
/Karl
---
Karl HammarAspö Data [EMAIL PROTECTED]
Lilla Aspö 2340Networks
S-742 94 Östhammar +46 173 140 57 Computers
Sweden +46 70 511 97 84 Consulting
---
___
lilypond-devel mailing list
[EMAIL PROTECTED]
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: GIT pull error
[EMAIL PROTECTED]: Sorry. Now I actually checked: the url is http://git.sv.gnu.org/r/lilypond.git (Note that not only http://; is new, but also /r/). Ok, so I update this file: $ grep URL .git/remotes/origin URL: http://git.sv.gnu.org/r/lilypond.git $ in my local copy, $ git-pull Fetching refs/heads/master from http://git.sv.gnu.org/r/lilypond.git using http got 6ba19b9f48c1ce10adb30a69557ce38b8a073785 walk 6ba19b9f48c1ce10adb30a69557ce38b8a073785 ... and things works again. Regards /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
git problem
Does anybody have a solution for this problem? $ git-pull fatal: protocol error: bad line length character Fetch failure: git://git.sv.gnu.org/lilypond.git fatal: protocol error: bad line length character Failed to find remote refs No changes. Regards /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Melisma_translator obsolete?
From Documentation/misc/ChangeLog-2.10:
2006-08-02 Erik Sandberg [EMAIL PROTECTED]
...
* ly/declarations-init.ly: change melisma/melismaEnd. Eliminates
ManualMelismaEvent, and obsoletes Melisma_translator.
So what should be done with:
$ lily_git -I input/manual input/test/engraver-one-by-one.ly
GNU LilyPond 2.11.12
Processing `input/test/engraver-one-by-one.ly'
Parsing...
error: unknown translator: `Melisma_translator'
$
and
$ grep Melisma_translator python/convertrules.py
str = re.sub (r'Melisma_engraver', 'Melisma_translator', str)
$
???
Regards
/Karl
export LILY_SRC_DIR=$HOME/most/music/lilypond/git/lilypond
lily_git() {
$LILY_SRC_DIR/out/bin/lilypond --relocate $@
}
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Styles [Was Re: Barline scans]
[EMAIL PROTECTED]: Karl Hammar wrote: [moving this from bug-lilypond] [EMAIL PROTECTED]: Karl Hammar wrote: Maybe we should have a my style section in the doc. Like section 4.3 Style sheets ? There you tell us how to do it. But perhaps more like the snippet respository, examples of different styles. Right now we only have one style, the default one. Plus tweaks, where the tweaks are more something you do when lilypond don't cope with it, not a different style. I have no objection to this; see http://lilypond.org/web/devel/participating/documentation-adding I guess it would go to 9.4 Style examples. Before I can present any new style I have to make one though. Regards /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: makeinfo with utf-8 support
[EMAIL PROTECTED]: Le jeudi 04 janvier 2007 à 00:33 +0100, Karl Hammar a écrit : Hmm, when building I get: out/lilypond.nexi:62: warning: unrecognized encoding name `utf-8'. ... Does anybody know where the makeinfo with utf-8 support is, a patch ? AFAIK Texinfo does no character encoding conversion for the Info and HTML output, so the LilyPond devels have decided to use the modern character set UTF-8, not Latin1. This will really become a problem for the PDF (outputted by Texinfo/TeX) when we decide to make PDF of Lily docs translation. Well, you can safely ignore these warnings. See also http://lists.gnu.org/archive/html/lilypond-devel/2007-01/msg00047.html Tried cvs version [1] of texinfo/makeinfo $ makeinfo --version makeinfo (GNU texinfo) 4.8.90 But the warning persist. Well, never mind. Regards /Karl [1] http://savannah.gnu.org/cvs/?group=texinfo This is fixed in latest texinfo cvs: Karl Berry: I just added UTF-8 to the list of encoding names recognized by makeinfo (from CVS). It just passes through the bytes as they are received, which hopefully will be better than nothing. I can't undertake to actually parse UTF-8 at this point (I added it to the TODO file :). Thanks, Karl Regards /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Styles [Was Re: Barline scans]
[moving this from bug-lilypond] [EMAIL PROTECTED]: Karl Hammar wrote: Maybe we should have a my style section in the doc. Like section 4.3 Style sheets ? There you tell us how to do it. But perhaps more like the snippet respository, examples of different styles. Right now we only have one style, the default one. Plus tweaks, where the tweaks are more something you do when lilypond don't cope with it, not a different style. Regards, /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: LilyPond suggestions
[EMAIL PROTECTED]:
Jesse Engle wrote:
4.) There is no way to create an entire page of exercises (a feature
that would be incredibly useful for music teachers such as myself)
like this:
http://jengle.waferbaby.com/tmp/lp/exercises.png
Actually, you could do this too within LilyPond, since you can use
\markup commands to create several columns and you can embed scores
into a \markup command. Here's a simple example:
...
Whu not use \startStaff and \stopStaff, like:
\score {
\new Staff {
\relative c' {
\clef alto c1 c \stopStaff s1
\startStaff b1 b \stopStaff s1
\startStaff a1 a \stopStaff
}
}
\new Staff {
\relative c' {
\clef alto c1 c \stopStaff s1
\startStaff b1 b \stopStaff s1
\startStaff a1 a \stopStaff
}
}
\new Staff {
\relative c' {
\clef alto c1 c \stopStaff s1
\startStaff b1 b \stopStaff s1
\startStaff a1 a \stopStaff
}
}
}
See staff-halfway.ly [1] in the regression tests.
[1] http://lilypond.org/doc/v2.11/input/regression/lily-1a72868935.ly
Regards
/Karl
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: makeinfo with utf-8 support
[EMAIL PROTECTED]: Le jeudi 04 janvier 2007 à 00:33 +0100, Karl Hammar a écrit : Hmm, when building I get: out/lilypond.nexi:62: warning: unrecognized encoding name `utf-8'. ... Does anybody know where the makeinfo with utf-8 support is, a patch ? AFAIK Texinfo does no character encoding conversion for the Info and HTML output, so the LilyPond devels have decided to use the modern character set UTF-8, not Latin1. This will really become a problem for the PDF (outputted by Texinfo/TeX) when we decide to make PDF of Lily docs translation. Well, you can safely ignore these warnings. See also http://lists.gnu.org/archive/html/lilypond-devel/2007-01/msg00047.html Tried cvs version [1] of texinfo/makeinfo $ makeinfo --version makeinfo (GNU texinfo) 4.8.90 But the warning persist. Well, never mind. Regards /Karl [1] http://savannah.gnu.org/cvs/?group=texinfo ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: makeinfo with utf-8 support
[EMAIL PROTECTED]: Tried cvs version [1] of texinfo/makeinfo $ makeinfo --version makeinfo (GNU texinfo) 4.8.90 But the warning persist. Well, never mind. Please report this to bug-texinfo, together with a small example (if possible). Done. Regards /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: GUILE-with-rational-bugfix
[EMAIL PROTECTED]:
Rune Zedeler escreveu:
Where can the famous patch be found?
(Unable to compile latest guile-cvs).
see install.txt
There is no install.txt, and the INSTALL.txt I found does not mention it:
$ git-pull
remote: Generating pack...
...
rename input/{no-notation/parse5.ly =
regression/warn-conflicting-key-signatures.ly} (100%)
$ find . -name install.txt
$ find . | grep -i install.txt
./Documentation/topdocs/out/INSTALL.txt
$ grep -C2 -i guile Documentation/topdocs/out/INSTALL.txt
work.
* GUILE (http://www.gnu.org/software/guile/guile.html) (version
1.6.7 or newer). If you are installing binary packages, you may
need to install guile-devel or guile-dev or libguile-dev too.
* Texinfo (ftp://ftp.gnu.org/gnu/texinfo/) (version 4.8 or newer).
--
* Pango (http://www.pango.org/) (version 1.12 or newer).
* GUILE (http://www.gnu.org/software/guile/guile.html) (version
1.8.0 or newer).
--
export GUILE=guile-1.6
export GUILE_CONFIG=guile-1.6-config
export
PKG_CONFIG_PATH=/sw/lib/freetype219/lib/pkgconfig/:/sw/lib/fontconfig2/lib/pkgconfig/
./configure
$
Regards
/Karl
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: GUILE-with-rational-bugfix
Rune Zedeler [EMAIL PROTECTED] writes: Where can the famous patch be found? ... http://lilypond.org/vc/gub.darcs/patches/guile-1.8-rational.patch ... Got the patch and build debian guile debs for i386. If anybody would need them, they are available at: http://kalle.csb.ki.se/pub/lilypond/ You need debian unstable to use them. To build them do: # apt-get install -t unstable debhelper ... # apt-get build-dep guile-1.8 ... $ wget http://lilypond.org/vc/gub.darcs/patches/guile-1.8-rational.patch $ apt-get source guile-1.8 Reading package lists... Done Building dependency tree... Done Need to get 0B/2829kB of source archives. dpkg-source: extracting guile-1.8 in guile-1.8-1.8.1+1 $ cd guile-1.8-1.8.1+1/libguile $ patch -p1 ../../guile-1.8-rational.patch patching file libguile/numbers.c patching file libguile/numbers.h $ cd .. $ fakeroot ./debian/rules # this goes first so we clean using patched code rm -f debian/stamp/build ... dh_builddeb dpkg-deb: building package `guile-1.8' in `../guile-1.8_1.8.1+1-4_i386.deb'. dpkg-deb: building package `guile-1.8-dev' in `../guile-1.8-dev_1.8.1+1-4_i386.deb'. dpkg-deb: building package `guile-1.8-libs' in `../guile-1.8-libs_1.8.1+1-4_i386.deb'. make[1]: Leaving directory `/home/karl/most/music/lilypond/guile-1.8-1.8.1+1' $ ./autogen.sh works for me now. Regards /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: GUILE-with-rational-bugfix
Karl Hammar wrote: [EMAIL PROTECTED]: Rune Zedeler escreveu: Where can the famous patch be found? (Unable to compile latest guile-cvs). see install.txt There is no install.txt, and the INSTALL.txt I found does not mention it: At www.lilypond.org, click on the documentation link for version 2.11 and then on INSTALL. /Mats Ohh, I see, a catch-22 situation. I could have seen it if I peeked inside Documentation/topdocs/INSTALL.texi, but was mislead by the configure.in text. Regards /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
makeinfo with utf-8 support
Hmm, when building I get: out/lilypond.nexi:62: warning: unrecognized encoding name `utf-8'. out/lilypond-internals.nexi:10: warning: unrecognized encoding name `utf-8'. out/music-glossary.nexi:6: warning: unrecognized encoding name `utf-8'. out/INSTALL.texi:6: warning: unrecognized encoding name `utf-8'. out/README.texi:6: warning: unrecognized encoding name `utf-8'. I've got: $ makeinfo --version makeinfo (GNU texinfo) 4.8 The latest texinfo at http://ftp.gnu.org/gnu/texinfo/ seems to be too old. Does anybody know where the makeinfo with utf-8 support is, a patch ? Regards /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: `Vera Bold' replaced by `AlbanyAMT-Bold'
Werner: Hmm. What do you get with fc-match? Just for comparison purposes. [EMAIL PROTECTED] lilypond]$ fc-match Vera DejaVuLGCSans.ttf: DejaVu LGC Sans Book Ah, ok, so you also don't get `Vera'. This is good to know. ... Don't be too happy: $ fc-match Vera Bitstream-Vera-Sans.ttf: Bitstream Vera Sans Roman $ fc-match Albany Bitstream-Vera-Sans.ttf: Bitstream Vera Sans Roman $ fc-match DejaVu Bitstream-Vera-Sans.ttf: Bitstream Vera Sans Roman $ fc-match courie Bitstream-Vera-Sans.ttf: Bitstream Vera Sans Roman $ fc-match courier cour.pfa: Courier Regular $ fc-match --version fontconfig version 2.4.1 Seems that vera is something of a default over here. Regards, /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: Some fixes to musicxml2ly
[EMAIL PROTECTED]: On Fri, 2006-09-15 at 19:33 +0200, Han-Wen Nienhuys wrote: else: - return 0 + tolerance_exp = -4 # means that notes that are 1/32 or less shorter than 1/4 are quarter notes etc. + return int (math.ceil (-math.log(self._duration) / math.log(2) - 2**tolerance_exp)) Since music notation uses exact rational numbers, using math.log () is usually forbidden, and it's better to use exact arithmetic. Can you doublecheck whether or not exact arithmetic applies to your case, please add an explanation comment why you're using floating point numbers? In this case inexactness doesn't matter. The term 2**tolerance_exp makes sure that notes that are slightly shorter than the next whole duration are rounded to the whole duration instead of somethin like c8.. In some cases floating point arithmetic can cause some randomness regarding whether the note becomes c8... or c4 (if the duration is exactly 1/32 from the next whole duration), but I don't think that matters -- such note length are weird enough to start with. When determining the number of dots the use of floating point arithmetic doesn't obviously harm, because the result is rounded anyway. I can't think of any simple way to achieve the same result without using math.log(). The attached diff for musicxml.py now has an explanatory comment. I don't understand this: tolerance_exp = -4 a = - math.log(self._duration) / math.log(2) b = 2**tolerance_exp c = a - b 1: if tolerance_exp is a constant, why don't you simply say b = 1/16 (another constant), which is clearly a duration(?) value. 2: why are you taking the difference between the log of a duration and a duration? In physics there is something which is called a dimension test: you simply cannot do 1 meter - 1 second. In math, you don't easily agree to do log(entity) - entity. And in this case I read your code as -2log( seconds ) - seconds which is clearly wrong. Now b seems to be an implementation detail caused by using floating point values, the same with math.ceil, is it so? If you simply want to map a duration to its -2log, why don't you use something like arr = [ 1, 2, 4, ...]; f(x) = arr[x]; which is a clear statement about what you are doing (although incomplete, I sense that self._duration is a rational number). 3: I don't understand what you are trying to do and your comment does not match the code 4: using floating point values for exact things usually open up enough nasty bugs and is very hard to verify due to rounding errors. Unless you have VERY good reason, just don't do it, it is not worth the trouble. === Later in your patch: +factor = self._duration / 2**(-self.get_duration_log()) +return int (round (-math.log(2 - factor) / math.log(2))) If: a = 1/2**(-self.get_duration_log()) then: a = 2**self.get_duration_log() assuming that get_duration_log() returns a 2log and that get_duration() exists, then a = self.get_duration() get_duration() look suspiciously like _duration(), so my guess is that factor = self._duration ** 2 by quickly reading the code as a math guy. Now, if duration usually is less than 1 (we don't see many breves and longas today), then -math.log(2 - factor) / math.log(2) more or less reduces to a near constant -1. Is that your intention, or am I reading your code false? I assume that duration is the lenght of a note, like whole note is 1, a quater note is 1/4, ... Regards, /Karl --- Karl HammarAspö Data [EMAIL PROTECTED] Lilla Aspö 2340Networks S-742 94 Östhammar +46 173 140 57 Computers Sweden +46 70 511 97 84 Consulting --- ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: configure find wrong python path
[EMAIL PROTECTED]: ... Unfortunately, configure completely bails out then using something like $ PYTHON=/usr/local/bin/python2.4 ./configure ... The problem is in STEPMAKE_GETVERSION, since it takes the first line that looks like it contains a version number of the output from [Path stripped to take the line shorter]: $ python2.4 --version 21 || python2.4 -V 21 which is Unknown option: -- usage: python2.4 [option] ... [-c cmd | -m mod | file | -] [arg] ... Try `python -h' for more information. Python 2.4.3 The net result is that, for whatever reason, configure first detects 2.4.3, but later 2.3.5. I think the attached patch fixes this. (Works at least when configuring on OpenBSD). Seen same problem, your patch fixes this for me also. Note that the patch contains another quick hack to let the version check work on OpenBSD and other systems where the shell has 32 bit integer arithmetic (without it, you'll get too large version numbers for fontforge, letting configure think that 20060413 is smaller than the requested version). Ciao, Kili ps: do we really still have to use backticks instead of $(...)? Anyone running LilyPond on old systems with rotten shells? configure is about portability, why not keep the backticks? The same goes for shells with 32bit int arith. pps: I *only* tested the configure step yet. No idea wether it will actually build and run. ... It builds and runs ok (using debian v3.1). Regards, /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: configure: readlink problem
Karl Hammar schreef: Since configure does not test for readlink it is perhaps a bad choise, the alternative would be to use a shell or python script or if someone comes up with a good way to find the afm files. we rely on python anyway, so I'm all for a Python based solution. Patch? -- Han-Wen Nienhuys - [EMAIL PROTECTED] - http://www.xs4all.nl/~hanwen LilyPond Software Design -- Code for Music Notation http://www.lilypond-design.com Ack, later this week. Regards, /Karl --- Karl HammarAspö Data [EMAIL PROTECTED] Lilla Aspö 2340Networks S-742 94 Östhammar +46 173 140 57 Computers Sweden +46 70 511 97 84 Consulting --- ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: configure: readlink problem
[EMAIL PROTECTED]: Karl Hammar schreef: Since configure does not test for readlink it is perhaps a bad choise, the alternative would be to use a shell or python script or if someone comes up with a good way to find the afm files. we rely on python anyway, so I'm all for a Python based solution. Patch? ... Ack, later this week. ... Attached (diff) is a possible patch: * configure.in: readlink -f is system dependent, replaced with python version 1. One could maybe write it as an one-liner: $ ls -l /usr/bin/emacs lrwxrwxrwx 1 root root 23 Nov 12 2005 /usr/bin/emacs - /etc/alternatives/emacs $ ls -l /etc/alternatives/emacs lrwxrwxrwx 1 root root 18 Nov 24 2005 /etc/alternatives/emacs - /usr/bin/emacs21-x $ python -c 'import os.path, sys; print os.path.realpath(sys.argv.pop(1))' /usr/bin/emacs /usr/bin/emacs21-x $ 2. It requires python = 2.2. One could put the python version check elsewhere. 3. the manual doesn't indicate wheter sys.argv is ro or rw --- so sys.argv.pop(1) could fail for others, the alternative is to use changeqoute's and sys.argv[1] instead, as in diff2 4. one could put the python code in a seperate file, as in diff3, and put readlink.py in python/ or elsewhere Regards, /Karl import os import sys for i in sys.argv[1:]: print os.path.realpath(i) Index: configure.in === RCS file: /sources/lilypond/lilypond/configure.in,v retrieving revision 1.194 diff -u -r1.194 configure.in --- configure.in24 May 2006 09:11:18 - 1.194 +++ configure.in5 Jun 2006 21:33:30 - @@ -53,6 +53,8 @@ # must come before any header checks STEPMAKE_COMPILE +# os.path.realpath() requires python 2.2 and unix +STEPMAKE_PYTHON(REQUIRED, 2.2) AC_CHECK_PROG(FCMATCH, fc-match, fc-match) AC_MSG_CHECKING([New Century Schoolbook PFB files]) AC_SUBST(NCSB_SOURCE_FILES) @@ -71,7 +73,7 @@ NCSB_FILE=`$FCMATCH --verbose Century Schoolbook L:style=$style | grep 'file:'` NCSB_FILE=`echo $NCSB_FILE | sed 's/^.*\(.*\).*$/\1/g'` - NCSB_FILE=`readlink -f $NCSB_FILE` + NCSB_FILE=`$PYTHON python/readlink.py $NCSB_FILE` NCSB_SOURCE_FILES=$NCSB_FILE $NCSB_SOURCE_FILES done elseIndex: configure.in === RCS file: /sources/lilypond/lilypond/configure.in,v retrieving revision 1.194 diff -u -r1.194 configure.in --- configure.in24 May 2006 09:11:18 - 1.194 +++ configure.in5 Jun 2006 21:19:05 - @@ -53,6 +53,8 @@ # must come before any header checks STEPMAKE_COMPILE +# os.path.realpath() requires python 2.2 and unix +STEPMAKE_PYTHON(REQUIRED, 2.2) AC_CHECK_PROG(FCMATCH, fc-match, fc-match) AC_MSG_CHECKING([New Century Schoolbook PFB files]) AC_SUBST(NCSB_SOURCE_FILES) @@ -71,7 +73,13 @@ NCSB_FILE=`$FCMATCH --verbose Century Schoolbook L:style=$style | grep 'file:'` NCSB_FILE=`echo $NCSB_FILE | sed 's/^.*\(.*\).*$/\1/g'` - NCSB_FILE=`readlink -f $NCSB_FILE` + # resolv symbolic links + changequote(, )#dnl + NCSB_FILE=`cat EOF | $PYTHON - $NCSB_FILE +import os.path, sys +print os.path.realpath(sys.argv[1]) +EOF` + changequote([, ])#dnl NCSB_SOURCE_FILES=$NCSB_FILE $NCSB_SOURCE_FILES done else Index: configure.in === RCS file: /sources/lilypond/lilypond/configure.in,v retrieving revision 1.194 diff -u -r1.194 configure.in --- configure.in24 May 2006 09:11:18 - 1.194 +++ configure.in5 Jun 2006 20:53:18 - @@ -53,6 +53,8 @@ # must come before any header checks STEPMAKE_COMPILE +# os.path.realpath() requires python 2.2 and unix +STEPMAKE_PYTHON(REQUIRED, 2.2) AC_CHECK_PROG(FCMATCH, fc-match, fc-match) AC_MSG_CHECKING([New Century Schoolbook PFB files]) AC_SUBST(NCSB_SOURCE_FILES) @@ -71,7 +73,11 @@ NCSB_FILE=`$FCMATCH --verbose Century Schoolbook L:style=$style | grep 'file:'` NCSB_FILE=`echo $NCSB_FILE | sed 's/^.*\(.*\).*$/\1/g'` - NCSB_FILE=`readlink -f $NCSB_FILE` + # resolv symbolic links + NCSB_FILE=`cat EOF | $PYTHON - $NCSB_FILE +import os.path, sys +print os.path.realpath(sys.argv.pop(1)) +EOF` NCSB_SOURCE_FILES=$NCSB_FILE $NCSB_SOURCE_FILES done else ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: configure: readlink problem
Hi, I'm trying to build lilypond on freebsd 6.1, mostly on a whim. ./configure complains that ... readlink: illegal option -- f usage: readlink [-n] [file ...] ... This is called in only one place, if test $FCMATCH != ; then for style in Roman Italic Bold Italic Bold; do NCSB_FILE=`$iiFCMATCH --verbose Century Schoolbook L:style=$style | grep 'file:'` NCSB_FILE=`echo $NCSB_FILE | sed 's/^.*\(.*\).*$/\1/g'` NCSB_FILE=`readlink -f $NCSB_FILE` NCSB_SOURCE_FILES=$NCSB_FILE $NCSB_SOURCE_FILES done I'm not certain what it does, so I can't suggest an alternative. If it's a big deal to fix, don't bother -- especially since GUB is available for freebsd. I've actually already downloaded GUB; I'm running ./configure so that I get the makefiles so I can make web. Cheers, - Graham The -f option tells readlink to rekursively resolv all links and find the file which the link finally points to. Without the -f readlink just just gives you what this link points to. On my system the file pointed to by fc-match is a link to the real file, so for me it would suffice to use readlink without the -f. Also we don't need the readlink at all for finding the pfb file, the problem comes later, somewhere else when we want the afm file. Currently lilypond assumes it is in the same directory as the pfb file. Since configure does not test for readlink it is perhaps a bad choise, the alternative would be to use a shell or python script or if someone comes up with a good way to find the afm files. Regards, /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
out/bin/*.ly ?
What is the meaning of *.ly which point to nothing below ? $ ls -l out/bin/ total 0 lrwxrwxrwx 1 karl users 22 May 22 17:57 *.ly - ../../scripts/out/*.ly lrwxrwxrwx 1 karl users 28 May 22 17:57 convert-ly - ../../scripts/out/convert-ly lrwxrwxrwx 1 karl users 23 May 22 17:57 lilypond - ../../lily/out/lilypond lrwxrwxrwx 1 karl users 31 May 22 17:57 lilypond-book - ../../scripts/out/lilypond-book $ Maybe it should read ln -s .../*2ly . instead of *.ly ? If so, use attached patch. Also note that target $(tree-share-prefix)/lilypond-force in toplevel GNUmakefile is run before the generation of the files in scripts/out/ so any wildcard are useless in the line above, and will generate the file *.ly instead of expanding to future files. * GNUmakefile.in: wildcard in ln -s .../*.ly . don't work when newly checkout'ed Regards, /Karl Index: GNUmakefile.in === RCS file: /sources/lilypond/lilypond/GNUmakefile.in,v retrieving revision 1.188 diff -u -r1.188 GNUmakefile.in --- GNUmakefile.in 25 Feb 2006 00:20:51 - 1.188 +++ GNUmakefile.in 24 May 2006 12:09:13 - @@ -132,9 +132,8 @@ mkdir -p $(tree-share-prefix)/tex cd $(tree-bin) \ ln -sf ../../lily/$(outconfbase)/lilypond . \ - ln -sf ../../scripts/$(outconfbase)/convert-ly . \ - ln -sf ../../scripts/$(outconfbase)/lilypond-book . \ - ln -sf ../../scripts/$(outconfbase)/*.ly . + for i in abc2ly convert-ly etf2ly lilypond-book lilypond-invoke-editor midi2ly musicxml2ly; \ + do ln -sf ../../scripts/$(outconfbase)/$$i . ; done cd $(tree-lib-prefix) \ ln -s ../../../../python/$(outconfbase) python cd $(tree-share-prefix) \ ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
configure find wrong python path
Given: $ ls -l /usr/bin/python* lrwxrwxrwx 1 root root 9 May 24 15:51 /usr/bin/python - python2.3 -rwxr-xr-x 1 root root 958764 Mar 6 11:32 /usr/bin/python2.3 -rwxr-xr-x 1 root root 1024460 Apr 23 01:34 /usr/bin/python2.4 configure will report wrong path of python: $ ./configure ... checking for bash... /bin/sh checking for python... python checking python version... 2.3.5 checking for python... /usr/bin/python checking for gcc... gcc ... checking New Century Schoolbook PFB files... /usr/share/fonts/type... checking for python... /usr/bin/python checking /usr/bin/python version... 2.3.5 python checking python version... 2.3.5 python2 python2.4 checking python2.4 version... 2.4 checking for python2.4... (cached) /usr/bin/python checking gcc version... 4.0.3 ... $ make ... import subprocess ImportError: No module named subprocess make[2]: *** [out/lilypond.nexi] Error 1 make[2]: Leaving directory `/home/karl/most/music/lilypond/head/Documentation/user' make[1]: *** [all] Error 2 make[1]: Leaving directory `/home/karl/most/music/lilypond/head/Documentation' make: *** [all] Error 2 $ STEPMAKE_PYTHON is run twice during configure: STEPMAKE_PYTHON(REQUIRED, 1.5) configure.in line 15: STEPMAKE_INIT() - stepmake/aclocal.m4 line 755 and STEPMAKE_PYTHON(REQUIRED, 2.4) configure.in line 87. STEPMAKE_PYTHON() calls AC_PATH_PROG() which calls AC_CACHE_VAL([ac_cv_path_$1], ... ) The second time AC_PATH_PROG() takes path from cache as seen in configure trace above. Since STEPMAKE_PYTHON() have a second argument (req. version) and the cache don't know of versions, one should clear the cache when seraching for a new python path. Attached patch corrects the error above. * stepmake/aclocal.m4: STEPMAKE_PYTHON(): clear cached value since arg 2 might point us to a new binary Regards /Karl Index: stepmake/aclocal.m4 === RCS file: /sources/lilypond/lilypond/stepmake/aclocal.m4,v retrieving revision 1.168 diff -u -r1.168 aclocal.m4 --- stepmake/aclocal.m4 3 May 2006 20:30:59 - 1.168 +++ stepmake/aclocal.m4 24 May 2006 21:18:50 - @@ -994,6 +994,8 @@ if test -n $pv; then STEPMAKE_ADD_ENTRY($1, $pv) fi +# clear cached value since arg 2 might point us to a new binary +unset ac_cv_path_PYTHON AC_PATH_PROG(PYTHON, $PYTHON) AC_SUBST(PYTHON) ]) ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
lilypond 2.8 make fails
Patch in attachment.
Building lilypond 2.8 (from todays cvs), make fails:
$ ./configure
checking build system type... i686-pc-linux-gnu
checking Package... LILYPOND
...
checking for mftrace... mftrace
checking mftrace version... 1.2.4
...
$ make
cd /home/karl/most/music/lilypond/2.8/out rm -rf bin lib share
mkdir -p ./out/bin
...
make[1]: Entering directory `/home/karl/most/music/lilypond/2.8/mf'
ERROR: mftrace not found
For obtaining PFA/SVG fonts, either install mftrace
(see http://www.xs4all.nl/~hanwen/mftrace/ ),
or try one of the following commands in this directory:
make get-fonts
make get-rpm-fonts
make get-deb-fonts
These commands will try to download the files from internet.
make[1]: Leaving directory `/home/karl/most/music/lilypond/2.8/mf'
$
Error is in mf/GNUmakefile:
$ grep -B1 -A3 '(MFTRACE)' mf/GNUmakefile
pfa_warning:
ifneq ($(shell $(MFTRACE) --version | sed 's/ .*//'),mftrace)
@echo
@echo ERROR: mftrace not found
@echo
$
Running the shell line:
$ mftrace --version | sed 's/ .*//'
mftrace
This
License
under
information.
Copyright
$
It becomes mftrace This License under information. Copyright in make,
which is not eq to mftrace.
It worked fine with older mftrace since they produced one line versions:
$ /usr/bin/mftrace --version
mftrace 1.1.17
$
Regards
/Karl
Index: mf/GNUmakefile
===
RCS file: /sources/lilypond/lilypond/mf/GNUmakefile,v
retrieving revision 1.193.2.1
diff -u -r1.193.2.1 GNUmakefile
--- mf/GNUmakefile 29 Mar 2006 16:19:38 - 1.193.2.1
+++ mf/GNUmakefile 22 May 2006 14:22:08 -
@@ -138,7 +138,7 @@
${MAKE} -C $(top-build-dir) link-mf-tree
pfa_warning:
-ifneq ($(shell $(MFTRACE) --version | sed 's/ .*//'),mftrace)
+ifneq ($(shell $(MFTRACE) --version | sed 's/ .*//;1q'),mftrace)
@echo
@echo ERROR: mftrace not found
@echo
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: lilypond 2.8 make fails
Karl Hammar schreef:
Patch in attachment.
Building lilypond 2.8 (from todays cvs), make fails:
Index: mf/GNUmakefile
===
RCS file: /sources/lilypond/lilypond/mf/GNUmakefile,v
retrieving revision 1.193.2.1
diff -u -r1.193.2.1 GNUmakefile
--- mf/GNUmakefile 29 Mar 2006 16:19:38 - 1.193.2.1
+++ mf/GNUmakefile 22 May 2006 14:22:08 -
@@ -138,7 +138,7 @@
${MAKE} -C $(top-build-dir) link-mf-tree
pfa_warning:
-ifneq ($(shell $(MFTRACE) --version | sed 's/ .*//'),mftrace)
+ifneq ($(shell $(MFTRACE) --version | sed 's/ .*//;1q'),mftrace)
I don't have this in todays CVS of lilypond 2.8.
...
Oops, wrong tag, sorry for the noise.
Regards
/Karl
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
lilypond 2.8, make fails, NCSB_SOURCE_FILES
$ ./configure; make ... make[1]: Entering directory `/home/karl/most/music/lilypond/2.8/mf' make[1]: *** No rule to make target [EMAIL PROTECTED]@', needed by `out/CenturySchL-Ital.otf'. Stop. make[1]: Leaving directory `/home/karl/most/music/lilypond/2.8/mf' make: *** [all] Error 2 $ Does not resolve symlinks when looking for gs fonts. Patch 'cvs diff -u -r lilypond_2_8 configure.in' attached. Regards /Karl Index: configure.in === RCS file: /sources/lilypond/lilypond/configure.in,v retrieving revision 1.184.2.2 diff -u -r1.184.2.2 configure.in --- configure.in9 May 2006 08:21:57 - 1.184.2.2 +++ configure.in22 May 2006 16:15:11 - @@ -75,6 +75,7 @@ NCSB_FILE=`$FCMATCH --verbose 'Century Schoolbook L:style=$style' | grep 'file:'` NCSB_FILE=`echo $NCSB_FILE | sed 's/^.*\(.*\).*$/\1/g'` + NCSB_FILE=`readlink -f $NCSB_FILE` NCSB_SOURCE_FILES=$NCSB_FILE $NCSB_SOURCE_FILES done else ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: configure does not check for python version =2.4
Hi, On Fri, 12 May 2006, Karl Hammar wrote: make: *** [all] Error 2 Which fails due to missing module subprocess. It was introduced recently: And it was discussed recently: http://article.gmane.org/gmane.comp.gnu.lilypond.devel/8996 Yes, but without any resolution or action. Hm, configure.in seems better suited for that change than stepmake/aclocal.m4. And distributing subprocess.py even better! Why not. Ciao, Dscho Now you have two choises, choose one. Regards, /Karl --- Karl HammarAspö Data [EMAIL PROTECTED] Lilla Aspö 2340Networks S-742 94 Östhammar +46 173 140 57 Computers Sweden +46 70 511 97 84 Consulting --- ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
configure does not check for python version =2.4
Running whith e.g python 2.3.5 gives me: $ make ... make[1]: Entering directory `/home/karl/most/music/lilypond/head/scripts' cat convert-ly.py | sed -e '#' -e '[EMAIL PROTECTED]@!/bin/sh!g' -e '[EMAIL PROTECTED] chmod 755 out/convert-ly /usr/bin/perl /home/karl/most/music/lilypond/head/buildscripts/out/help2man out/convert-ly out/convert-ly.1 help2man: can't get `--help' info from out/convert-ly make[1]: *** [out/convert-ly.1] Error 1 make[1]: Leaving directory `/home/karl/most/music/lilypond/head/scripts' make: *** [all] Error 2 Which fails due to missing module subprocess. It was introduced recently: $ grep -B7 -A1 subprocess ChangeLog 2006-03-12 Han-Wen Nienhuys [EMAIL PROTECTED] * scripts/lilypond-book.py (get_latex_textwidth): explicitly close tmphandle. * python/lilylib.py (system): rewrite system() using subprocess. Remove redirection trickery. Which makes python ver. 2.4 a requirement: http://www.python.org/doc/2.4.2/lib/module-subprocess.html Attached patch seems to solve that, though it gives strange error message: $./configure ... config.status: creating config.hh ERROR: Please install required programs: python2.3 = 2.4 (installed: 2.3) Python.h (python-devel, python-dev or libpython-dev package) See INSTALL.txt for more information on how to build LilyPond Regards, /Karl --- Karl HammarAspö Data [EMAIL PROTECTED] Lilla Aspö 2340Networks S-742 94 Östhammar +46 173 140 57 Computers Sweden +46 70 511 97 84 Consulting --- Index: ChangeLog === RCS file: /sources/lilypond/lilypond/ChangeLog,v retrieving revision 1.4948 diff -u -r1.4948 ChangeLog --- ChangeLog 11 May 2006 13:00:10 - 1.4948 +++ ChangeLog 11 May 2006 22:58:37 - @@ -1,3 +1,7 @@ +2006-05-12 Karl Hammar [EMAIL PROTECTED] + + * stepmake/aclocal.m4: bump python requirement to version 2.4 + 2006-05-11 Graham Percival [EMAIL PROTECTED] * Documentation/user/programming-interface.itely: first round of Index: stepmake/aclocal.m4 === RCS file: /sources/lilypond/lilypond/stepmake/aclocal.m4,v retrieving revision 1.168 diff -u -r1.168 aclocal.m4 --- stepmake/aclocal.m4 3 May 2006 20:30:59 - 1.168 +++ stepmake/aclocal.m4 11 May 2006 22:58:37 - @@ -752,7 +752,7 @@ fi AC_SUBST(SHELL) -STEPMAKE_PYTHON(REQUIRED, 1.5) +STEPMAKE_PYTHON(REQUIRED, 2.4) if expr $MAKE : '.*\(echo\)' /dev/null; then $MAKE -v 2 /dev/null | grep GNU /dev/null ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: configure does not check for python version =2.4
Running whith e.g python 2.3.5 gives me: $ make ... make[1]: Entering directory `/home/karl/most/music/lilypond/head/scripts' cat convert-ly.py | sed -e '#' -e '[EMAIL PROTECTED]@!/bin/sh!g' -e '[EMAIL PROTECTED] chmod 755 out/convert-ly /usr/bin/perl /home/karl/most/music/lilypond/head/buildscripts/out/help2man out/convert-ly out/convert-ly.1 help2man: can't get `--help' info from out/convert-ly make[1]: *** [out/convert-ly.1] Error 1 make[1]: Leaving directory `/home/karl/most/music/lilypond/head/scripts' make: *** [all] Error 2 Which fails due to missing module subprocess. It was introduced recently: $ grep -B7 -A1 subprocess ChangeLog 2006-03-12 Han-Wen Nienhuys [EMAIL PROTECTED] * scripts/lilypond-book.py (get_latex_textwidth): explicitly close tmphandle. * python/lilylib.py (system): rewrite system() using subprocess. Remove redirection trickery. Which makes python ver. 2.4 a requirement: http://www.python.org/doc/2.4.2/lib/module-subprocess.html Attached patch seems to solve that, though it gives strange error message: $./configure ... config.status: creating config.hh ERROR: Please install required programs: python2.3 = 2.4 (installed: 2.3) Python.h (python-devel, python-dev or libpython-dev package) See INSTALL.txt for more information on how to build LilyPond Hm, configure.in seems better suited for that change than stepmake/aclocal.m4. Changelog: 2006-05-12 Karl Hammar [EMAIL PROTECTED] * configure.in: bump python requirement to version 2.4 Regards, /Karl --- Karl HammarAspö Data [EMAIL PROTECTED] Lilla Aspö 2340Networks S-742 94 Östhammar +46 173 140 57 Computers Sweden +46 70 511 97 84 Consulting --- Index: configure.in === RCS file: /sources/lilypond/lilypond/configure.in,v retrieving revision 1.191 diff -u -r1.191 configure.in --- configure.in9 May 2006 08:21:17 - 1.191 +++ configure.in12 May 2006 00:01:44 - @@ -83,7 +83,7 @@ -STEPMAKE_PYTHON(REQUIRED, 2.3) +STEPMAKE_PYTHON(REQUIRED, 2.4) STEPMAKE_GCC(REQUIRED, 4.0) STEPMAKE_CXX(REQUIRED) STEPMAKE_GXX(REQUIRED, 4.0) ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
MergeKern: Failed to find kern info in file
(Current cvs) $ make ... make[1]: Entering directory `/home/karl/most/music/lilypond/head/mf' /usr/bin/fontforge -script ../buildscripts/pfx2ttf.fontforge /usr/X11R6/l... Copyright (c) 2000-2005 by George Williams. Executable based on sources from 12:08 5-Dec-2005. MergeKern: Failed to find kern info in file Called from... ../buildscripts/pfx2ttf.fontforge: line 5 make[1]: *** [out/CenturySchL-Ital.otf] Error 1 make[1]: Leaving directory `/home/karl/most/music/lilypond/head/mf' make: *** [all] Error 2 configure checks where the pfb files for Century Schoolbook are, and assumes that the afm files are in the same directory. Which fails if the given file is a link the real pfb file (in some other direcotory): $ fc-match --verbose 'Century Schoolbook L:style=Italic' | grep file: file: /usr/X11R6/lib/X11/fonts/Type1/c059033l.pfb(s) $ ls -l /usr/X11R6/lib/X11/fonts/Type1/c059033l.* lrwxrwxrwx 1 root root 53 Nov 29 12:20 /usr/X11R6/lib/X11/fonts/Type1/c059033l.pfb - ../../../../../share/fonts/type1/gsfonts/c059033l.pfb $ ls -l /usr/share/fonts/type1/gsfonts/c059033* -rw-r--r-- 1 root root 43546 Apr 3 2005 /usr/share/fonts/type1/gsfonts/c059033l.afm -rw-r--r-- 1 root root 113732 Apr 3 2005 /usr/share/fonts/type1/gsfonts/c059033l.pfb Attached patch solves that (still assuming that the afm is in the same directory as the real pfb file). One could possible do the same for the ./configure --with-ncsb-dir=... case. 2006-05-12 Karl Hammar [EMAIL PROTECTED] * configure.in: resolve symlinks for pfb-files, to be able to find its afm file. Regards, /Karl --- Karl HammarAspö Data [EMAIL PROTECTED] Lilla Aspö 2340Networks S-742 94 Östhammar +46 173 140 57 Computers Sweden +46 70 511 97 84 Consulting --- Index: configure.in === RCS file: /sources/lilypond/lilypond/configure.in,v retrieving revision 1.191 diff -u -r1.191 configure.in --- configure.in9 May 2006 08:21:17 - 1.191 +++ configure.in12 May 2006 00:41:35 - @@ -71,6 +71,7 @@ NCSB_FILE=`$FCMATCH --verbose 'Century Schoolbook L:style=$style' | grep 'file:'` NCSB_FILE=`echo $NCSB_FILE | sed 's/^.*\(.*\).*$/\1/g'` + NCSB_FILE=`readlink -f $NCSB_FILE` NCSB_SOURCE_FILES=$NCSB_FILE $NCSB_SOURCE_FILES done else ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: removing unwanted accidentals
Attatched is a version or act1.ly, act1_.ly, that puts the final movement in
its own score block. I know you may not want to put each movement in its own
score, but lilypond expects this by default and notice how all the problems
associated with doing it as you have it set up disappear.
Stephen
Thanks for the suggestion, the code below will work nice as a
workaround.
I'll settle with it for engraving, and using the former for midi.
\score {
\ss
\new GrandStaff
\new Staff {
\set Staff.instrument = \markup { V1 }
\set Staff.instr = \markup { V1 }
{ \keyAM \clefva \vaAM }
}
\new Staff {
\set Staff.instrument = \markup { V2 }
\set Staff.instr = \markup { V2 }
{ \keyAM \clefvb \vbAM }
}
\new Staff {
\set Staff.instrument = \markup { Va }
\set Staff.instr = \markup { Va }
{ \keyAM \clefvc \vcAM }
}
\new Staff {
\set Staff.instrument = \markup { BC }
\set Staff.instr = \markup { BC }
{ \keyAM \clefvd \vdAM }
}
\context FiguredBass \fbAM
\layout {
indent = 0.0\mm
}
}
...
I don't want them to appear as seperate scores, because they flow into
each other.
Except the different acts, of cause.
...
*No*, the forget does not work for:
act 1, aria 13, bar 6 Violin II: the b is shown with natural
act 1, aria 13, bar 11 Violin I: the f is shown with natural
Elsewhere yes, but not for theese two notes.
...
Regards
/Karl
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: removing unwanted accidentals
Citerar Karl Hammar [EMAIL PROTECTED]:
...
\version 2.4.0
#(define-public (print-+100-text-callback grob)
(let* ( ;(text (ly:grob-property grob 'text))
(text (number-string (+ 100 (string-number (ly:grob-property
grob
'text)
(layout (ly:grob-layout grob))
(defs (ly:output-def-lookup layout 'text-font-defaults))
(props (ly:grob-alist-chain grob defs))
(text-stencil (Text_interface::interpret_markup layout props
text)))
(ly:stencil-add (centered-stencil text-stencil
\new Staff {
\set Score.barNumberVisibility = #(every-nth-bar-number-visible 1)
\override Score.BarNumber #'print-function =
#print-+100-text-callback
\repeat unfold 9 { c1 \break } \bar |.
}
HTH,
Erik
I think you misunderstand me, I wanted smallish bar numbers since that
is easier
to say at a rehersal, not something that is +100.
Yes, of course you can substitute -100 for 100 then, or alternatively do
(- (...) 100) iso. (+ 100 (...))
Erik
I see, since I have trouble with lilyponds barnumbering, I sort of do
it myself.
Then I have to:
. keep track of number of bars per movement (= offset for next movement)
. add up the offsets per movements
. make sure it woun't print a 1?
Ok, that is a way to do it.
Can I in the same time tell the print function to not to align to a
breath mark
just before the bar line, detect if there is a volta spanner, staff
group bracket,
note or slur to avoid?
Regards,
/Karl
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: removing unwanted accidentals
Karl, The problem appears when you put all the movements together **under the same score**. I consider this to be a midi bug. No, the unwanted accidentals are still there if I remove the midi block. If you put each movement in its own score, all your engraving problems would disapear, the movements would appear as they do in the separate scores. I don't want them to appear as seperate scores, because they flow into each other. Except the different acts, of cause. The midi files would overwrite each other as the movements were processed leaving the midi file only representing the final movement. Yes, that is what happens e.g. in my all.ly, only the last act is in the midi file. It would be nice if midi worked the sames as engraving, just as the separate scores are concatenated into the same book with a default amount of space separating each one, might not the midi files from each score be concatenated into the same midi file with a default amount of silence between each score? Sounds, reasonable, but midi is not my top priority, engraving is. ... In addition, it would be nice if you added different midi instruments to the voices: \set Staff.midiInstrument = violin etc. I once did that with another piece, and that voice was silent in timidity. I could easily add that, but I'd rather hear it. Any suggestions? I haven't been very helpful to you so far, but I think this is all useful advice, germane to your project. Stephen - Original Message - From: Karl Hammar [EMAIL PROTECTED] To: lilypond-devel@gnu.org Sent: Friday, April 15, 2005 12:59 PM Subject: Re: removing unwanted accidentals *No*, the forget does not work for: act 1, aria 13, bar 6 Violin II: the b is shown with natural act 1, aria 13, bar 11 Violin I: the f is shown with natural Elsewhere yes, but not for theese two notes. ... Regards /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: removing unwanted accidentals
Citerar Stephen [EMAIL PROTECTED]:
I think we want Lilypond to continue to follow standard practive. If we give
users the ability to violate those rules then total novices will start
making mistakes.
We don't have to document the feature in a way such that novices will be able
to
use it. I just htink it's a good idea to have the feature for completeness. If
there is a bug somewhere that makes an unwanted accidental pop up, it's nice
if
there is a possiblilty to just write a quick workaround.
I think you know that in the example you gave, a person reading the score
would just play an f sharp where you want an f natural if you suppressed the
natural. Why supply a way to suppress that natural if we also have to say
never actually do if you do not want them to play f sharp?
What I meant, was related to Karl's example. I don't have lily where I am now,
but imagine a combo of his and my example; something like
\key d\major fis4 r r r | f2 f | \set Score.barNumber = #1 r2 f |
Here someone might want to avoid the natural before the last f (there might
exist better examples), and I think it's impossible with current lily due to
the bug Karl found.
It's just an example, and it's constructed, but shows that there is a
potential
need for this kind of setting. I can recall earlier discussions with similar
content.
The main reason I want it, however, is the philosophy that if you want to
tweak
something, you often want to express exactly what you are tweaking, no more,
no
less. All Karl wanted to express in his code, was that for a particular note,
the accidental should be turned off. He didn't want to express that this note
belongs to a part of the music which uses an alternative practise for
displaying accidentals.
I don't want the singer/instrumentalist to believe that the composer or
me
demands that this note to have this accidental, since it is not in the
original score.
And since the 'forget did not work for two notes, it would nice to have
something to force it away, so I don't have to delay till the bug hunt
is over.
Another application of a \noAccidental feature, is a mail that was sent some
days ago to lily-user iirc. There was a sharp before the cis2 in {cis1~\break
cis2}, and this person (Thomas iirc) wanted to suppress this.
You could do that like
{cis1~\break #(set-accidental-style 'no-forget) cis2 #(set-accidental-style
'default) }
but IMHO this would be more logical, easier to read and write:
{cis1~\break cis2\noAccidental}
Erik
Regards,
/Karl
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: removing unwanted accidentals
[EMAIL PROTECTED]:
On Thursday 14 April 2005 21.31, Stephen wrote:
...
In the example, '\set Score.currentBarNumber = #1' creates the problem. I
Correct.
don't see why the numbering is reset there other than to remove the number
from that bar because it collides with the mark D. There must be another
way to suppress the bar number at that bar.
It also changes the numbering of all subsequent bars. I don't know the
author's intent in this case (I didn't write that snippet), but I guess it
could be something like an attaca movement.
Yes. The bigger picture is H. Purcells opera Dido Æneas:
http://aspodata.se/motett/noter/act*
I did not like idea of big bar number, like 543.
I think it is better with aria/movement 23, bar 25.
However, I fixed it by setting the accidental style to 'forget'. See? there
is already a way to do it.
*No*, the forget does not work for:
act 1, aria 13, bar 6 Violin II: the b is shown with natural
act 1, aria 13, bar 11 Violin I: the f is shown with natural
Elsewhere yes, but not for theese two notes.
I don't know how to solv theese.
If I put another f before bar 11 for V1, that f gets the natural.
It seems that the first f, in this case, gets the natural, and I
tried to insert a \hideNote f \unHideNote, but it was still visible.
Btw, I solved the issue for the other places by starting each aria like:
\set Score.explicitKeySignatureVisibility = #begin-of-line-visible
\set Score.printKeyCancellation = ##f
\key c \minor
when the previous part already was in c minor.
It would be nice if I could write:
\key c \minor \music02
\key c \minor \music03
\key c \minor \music04
\key c \minor \music05
and the key-engraver only shows the key signature when it *changes*.
I tried hack around in lily/key-engraver.cc and doing things like
if (scm_is_eq (last, key)) return, found no solution.
...
Erik
- Original Message -
From: Erik Sandberg [EMAIL PROTECTED]
To: lilypond-devel@gnu.org
Sent: Monday, April 11, 2005 3:53 PM
Subject: removing unwanted accidentals
Hi,
There seems to be a need for a way to notate that a single note should
_not_
have an accidental displayed, something like a 'force-no-accidental'
property.
Possible, but for this case I would prefer:
1, fixing the bug
2, make key-engraver only show changes
The reason for 2, is that I prefer to put the workaround in one place,
instead of littering down the .ly with bug-corrections, that must be
refound when the bug is removed.
I think there was a discussion about this over a year ago, without a
conclusion. Now there have been 2 or 3 (implicit) requests for it in a
short
time, and IMHO it's a reasonable feature.
Also it looks pretty trivial to implement; may I give it a try?
Erik
On Sunday 10 April 2005 14.32, Karl Hammar wrote:
How do I remove the unwanted explicit flats in the next to last
measure in:
\version 2.4.2
\relative c'' {
\override Score.BarNumber #'break-visibility = #end-of-line-invisible
\mark #03
\clef treble
\key c \minor
a2 b |
d1 |
\mark #04
\set Score.currentBarNumber = #1
d2 es8. as,16 bes8 c |
% ^ ^ unwanted explicit flats here
d1
}
Regards
/Karl
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: removing unwanted accidentals
*No*, the forget does not work for: act 1, aria 13, bar 6 Violin II: the b is shown with natural act 1, aria 13, bar 11 Violin I: the f is shown with natural Elsewhere yes, but not for theese two notes. When I print out Score13, neither of these errors are present, even if I add the line: No one of the ScoreXX has theese problems. I used them to speed up proofreading. The problem appears when I put all the movements together and resets the bar number, as in act1.ly. #(set-accidental-style 'forget), which should suppress it in context of the total score. Don't forget you can set that anywhere in the score and reverse it with: #(set-accidental-style 'default). Stephen Regards /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: lilypond book issues version 2.5.19
... Its not a matter circumventing any limitations of Lilypond, but of being able to use the full power of LaTeX. It would not be practical to add anything to LilyPond itself to get this functionality as using it would involve (at least) two calls to lilypond. Actually I already have this functionality (and more) as I call Lilypond from LyX, using a LyX converter that I am developing and this can do some post-processing on the file produced by lilypond-book. The code for the converter at the moment is a bit of a mess and is really only usable by me. I will release it sometime, but I will probably wait till the next version of LyX is released as that will change the way converters are handled. However life would be easier if it could be done with a macro. I'm just trying to help the LilyPond hackers to judge if your request is of sufficiently general interest to be worth to implement and support. So far, I have to say that your request looks fairly exotic to me and I can't see that there would be a large group of users that would use it. If you prove me wrong, I'd be happy to change my mind (can't promise about the main developers, though). /Mats I would appreciate if I in latex could get the first and last \piece, \mark number, and bar number on a page to be able to put it in page headers/footers, or add some footnotes from within the music, as some editors do. Regards /Karl ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: removing unwanted accidentals
On Friday 15 April 2005 10.56, Karl Hammar wrote:
[EMAIL PROTECTED]:
It also changes the numbering of all subsequent bars. I don't know the
author's intent in this case (I didn't write that snippet), but I guess
it could be something like an attaca movement.
Yes. The bigger picture is H. Purcells opera Dido Æneas:
http://aspodata.se/motett/noter/act*
nice! (will this end up on mutopia?)
Anyone is welcome to use it (it is still being worked on).
The code is gpl (cc-sa if you insist) and Purcells music is public
domain.
I did not like idea of big bar number, like 543.
I think it is better with aria/movement 23, bar 25.
You could use something like this (based on print-circled-text-callback in
scm/output-lib.scm) to achieve what you want more cleanly. It works with
2.5.19, but not with 2.4 for some reason.
\version 2.4.0
#(define-public (print-+100-text-callback grob)
(let* ( ;(text (ly:grob-property grob 'text))
(text (number-string (+ 100 (string-number (ly:grob-property grob
'text)
(layout (ly:grob-layout grob))
(defs (ly:output-def-lookup layout 'text-font-defaults))
(props (ly:grob-alist-chain grob defs))
(text-stencil (Text_interface::interpret_markup layout props text)))
(ly:stencil-add (centered-stencil text-stencil
\new Staff {
\set Score.barNumberVisibility = #(every-nth-bar-number-visible 1)
\override Score.BarNumber #'print-function =
#print-+100-text-callback
\repeat unfold 9 { c1 \break } \bar |.
}
HTH,
Erik
I think you misunderstand me, I wanted smallish bar numbers since that
is easier
to say at a rehersal, not something that is +100.
Regards
/Karl
___
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: 'make web' hung?
Just compiled LilyPond 2.5.15 (ChangeLog 1.3261) an hour ago or so. I tried 'make web' and it has seemingly frozen at: lily-51095895-1.epslily-659133186-1.eps] [253lily-563569828-1.eps lily-2030054610-1.epslily-993345859-1.epslily-288871224-1.eps] [254] [255] [256] [257] [258] [259] [260] [261] [262] [263] [264] [265] [266] [267] [268] ps2pdf -sPAPERSIZE=a4 out-www/lilypond.pdf.pdfps out-www/lilypond.pdf It has stopped there for quite some time now. Close to an hour, I think. At first I thought it was taking some time to create a large PDF file or something, but it has been stuck for so long that I think something else is going on (or more to the point *not* going on). It might be a gs version problem. I have gs version 7.07.1. From Changelog: 2004-12-21 Jan Nieuwenhuizen [EMAIL PROTECTED] * python/lilylib.py (setup_environment): Set GS_FONTPATH, GS_LIB to empty. Fixes make web, using gs-gpl-8.01. In fact, gs-afpl-8.50 has more trouble with make web (ie, uses *a lot* more resourses; 3x MAXSIZE, 4x MAXRSS, 1.5x user). * mf/GNUmakefile (pfa_warning): Use less broken check. * python/lilylib.py (setup_environment): Remove cruft from GS_FONTPATH. * configure.in: Bump mftrace requirement to 1.1.1. I didn't mind not building pdf's for the moment so I did: # cd /usr/bin # mv ps2pdf ps2pdf.org # ln -s /bin/true ps2pdf And I got past that problem. Also, I'm still getting the extra large sans serif font for time signatures and multimeasure rests (reported last night on bug-lilypond). I'm no longer getting a seemingly spurious error message about a programming error, and some stencil being moved to the edge. It did not seem to indicate any visible problem with the output. As I said, I'm not getting this error any more. -David ___ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
make web fails
$ make web ... /usr/bin/python ./buildscripts/mutopia-index.py -o /home/karl/lilypond/examples.html ./ generating HTML list /home/karl/lilypond/examples.html cd /home/karl/lilypond [EMAIL PROTECTED] /usr/bin/python /home/karl/lilypond/stepmake/bin/add-html-footer.py --name LilyPond --version 2.5.0 `find . -name '*.html' -print` /bin/sh: /usr/bin/python: Argument list too long make[1]: *** [local-WWW-post] Error 126 make[1]: Leaving directory `/home/karl/lilypond' make: *** [web] Error 2 Patch, this patch assumes that xargs is available and that .../add-html-footer.py a .../add-html-footer.py b is equivalent to .../add-html-footer.py a b $ grep ChangeLog CVS/Entries /ChangeLog/1.2835/Fri Nov 12 06:21:36 2004// $ head -5 ChangeLog 2004-11-12 Karl Hammar [EMAIL PROTECTED] * GNUmakefile.in (local-WWW-post): `find ...` overflows the cmdline, use xargs instead Regards /Karl Index: GNUmakefile.in === RCS file: /cvsroot/lilypond/lilypond/GNUmakefile.in,v retrieving revision 1.128 diff -u -r1.128 GNUmakefile.in --- GNUmakefile.in 6 Nov 2004 10:54:57 - 1.128 +++ GNUmakefile.in 12 Nov 2004 17:41:44 - @@ -67,12 +67,13 @@ web-ext = gz html midi pdf png txt ly -footify = [EMAIL PROTECTED] $(PYTHON) $(step-bindir)/add-html-footer.py --name $(PACKAGE_NAME) --version $(TOPLEVEL_VERSION) +footify = $(PYTHON) $(step-bindir)/add-html-footer.py --name $(PACKAGE_NAME) --version $(TOPLEVEL_VERSION) +footifymail = [EMAIL PROTECTED] local-WWW-post: $(PYTHON) $(buildscript-dir)/mutopia-index.py -o $(builddir)/examples.html ./ - cd $(builddir) $(footify) `$(FIND) . -name '*.html' -print` - cd $(builddir) rm -f `find . -name \*.html~ -print` + cd $(builddir) $(FIND) . -name '*.html' -print | $(footifymail) xargs $(footify) + cd $(builddir) find . -name \*.html~ -print | xargs rm -f cd $(builddir) find Documentation input \ $(web-ext:%=-path '*/out-www/*.%' -or) -false \ $(outdir)/weblist Index: ChangeLog === RCS file: /cvsroot/lilypond/lilypond/ChangeLog,v retrieving revision 1.2835 diff -u -r1.2835 ChangeLog --- ChangeLog 12 Nov 2004 01:44:28 - 1.2835 +++ ChangeLog 12 Nov 2004 17:42:37 - @@ -1,3 +1,8 @@ +2004-11-12 Karl Hammar [EMAIL PROTECTED] + + * GNUmakefile.in (local-WWW-post): `find ...` overflows the + cmdline, use xargs instead + 2004-11-11 Graham Percival [EMAIL PROTECTED] * Documentation/user/lilypond-book.itely: add short warning about ___ lilypond-devel mailing list [EMAIL PROTECTED] http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: programming error: Grob `NoteHead' has no interface for property ...
Juergen Reuter [EMAIL PROTECTED] wrote: On Mon, 8 Nov 2004, Karl Hammar wrote: ... I have found a way to get rid of the errors. Previous run, lots of errors from make web: $ grep -i error log | wc -l 1359 $ New idea: $ (./autogen.sh --prefix=$HOME; make clean; make all install; . buildscripts/out/lilypond-profile; make web) log 21 $ grep -i error log | uniq -c 3 make[1]: [out/parser.hh] Error 1 (ignored) 1 make[1]: [out/parser.cc] Error 1 (ignored) 40 Do not worry about the @ signs in the output, they are not errors. 1 file:line:error style messages enabled. $ So the cause of the programming errors is a missing path in TEXMF. Are you sure that the .ly snippets were actually rebuilt? make clean does not delete them, IIRC. Please also note that, during make web, lilypond is called with the option: -e '(ly:set-option (quote internal-type-checking) #t)' Without explicitly adding this option (e.g. when calling lilypond manually), lilypond will not issue any Grob 'xyz' has no interface for property ... warning. Greetings, Jürgen My bad, youre right. $ unset TEXMF $ (./autogen.sh --prefix=$HOME; make clean; make all install; make web) log 21 $ grep -i error log | uniq -c 3 make[1]: [out/parser.hh] Error 1 (ignored) 1 make[1]: [out/parser.cc] Error 1 (ignored) 40 Do not worry about the @ signs in the output, they are not errors. 1 make[1]: [out/parser.cc] Error 1 (ignored) 1 file:line:error style messages enabled. $ (./autogen.sh --prefix=$HOME; make clean; make web-clean; make all install; make web) log 21 $ grep -i error log | wc -l 1332 $ (./autogen.sh --prefix=$HOME; make clean; make web-clean; make all install; . buildscripts/out/lilypond-profile; make web) log 21 $ grep -i error log | wc -l 1332 It was too god to be true. (And yes, make web-clean fails with rm -f ./out-www/* /bin/sh: /bin/rm: Argument list too long I suggest find ./out-www -name lily-1\* | xargs rm -f or rm -rf ./out-www; mkdir out-www ) Regards /Karl ___ lilypond-devel mailing list [EMAIL PROTECTED] http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: make fails.
Erik Sandberg [EMAIL PROTECTED] wrote: Using debian/testing. make output with 2.4 cvs of make clean; ./autogen.sh; make: [] FontForge used to be named PfaEdit. mv feta-braces-g.pfa ./out mftrace --encoding feta-braces-h.enc -I ./out/ --pfa --simplify feta-braces-h sh: line 1: potrace: command not found mftrace 1.0.34 Font `feta-braces-h'... Using encoding file: `./out/feta-braces-h.enc' Running Metafont... Tracing bitmaps... [0][1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16] [17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35] [36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51][52][53][54] [55][56][57][58][59][60][61][62][63] Assembling font to `feta-braces-h.pfa'... Simplifying font... Copyright (c) 2000-2004 by George Williams. Executable based on sources from 07:34 12-Oct-2004. FontForge used to be named PfaEdit. mv feta-braces-h.pfa ./out mftrace --encoding feta-braces-i.enc -I ./out/ --pfa --simplify feta-braces-i sh: line 1: potrace: command not found mftrace 1.0.34 Font `feta-braces-i'... Using encoding file: `./out/feta-braces-i.enc' Running Metafont... Tracing bitmaps... [0][1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16] [17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35] [36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51][52][53][54] [55][56][57][58][59][60][61][62][63] Assembling font to `feta-braces-i.pfa'... Simplifying font... Copyright (c) 2000-2004 by George Williams. Executable based on sources from 07:34 12-Oct-2004. FontForge used to be named PfaEdit. mv feta-braces-i.pfa ./out mftrace --encoding feta-din10.enc -I ./out/ --pfa --simplify feta-din10 sh: line 1: potrace: command not found mftrace 1.0.34 Font `feta-din10'... Using encoding file: `./out/feta-din10.enc' Running Metafont... Tracing bitmaps... [33][102][109][112][114][115][122] Assembling font to `feta-din10.pfa'... Simplifying font... Copyright (c) 2000-2004 by George Williams. Executable based on sources from 07:34 12-Oct-2004. FontForge used to be named PfaEdit. error: fontforge: command exited with value 11 Traceback (most recent call last): File /usr/bin/mftrace, line 1188, in ? cleanup_font (basename + '.pfa') File /usr/bin/mftrace, line 841, in cleanup_font system (%s -script simplify.pe %s %s % (fontforge_cmd, file, tfmfile)) File /usr/bin/mftrace, line 246, in system error (msg) File /usr/bin/mftrace, line 137, in error raise _ (Exiting ... ) Exiting ... make[1]: *** [out/feta-din10.pfa] Error 1 make[1]: Leaving directory `/home/erik/lily/lilypond24/mf' make: *** [all] Error 2 I have been able to reproduce your problem and I suggest you install potrace: # apt-get install potrace I suggest that potrace is added to the Build-Depends: in debian/control. How to reproduce: $ cat run (./autogen.sh --prefix=$HOME; make clean; make $@) log 21 $ ./run all $ tail log make[4]: Leaving directory `/home/karl/lilypond/input/mutopia/W.A.Mozart' make[4]: Entering directory `/home/karl/lilypond/input/mutopia/R.Schumann' true make[4]: Leaving directory `/home/karl/lilypond/input/mutopia/R.Schumann' make[3]: Leaving directory `/home/karl/lilypond/input/mutopia' make[2]: Leaving directory `/home/karl/lilypond/input' make[2]: Entering directory `/home/karl/lilypond/vim' true make[2]: Leaving directory `/home/karl/lilypond/vim' make[1]: Leaving directory `/home/karl/lilypond' # su Password: # dpkg -P potrace # exit $ ./run all $ tail -30 log Simplifying font... Copyright (c) 2000-2004 by George Williams. Executable based on sources from 07:34 12-Oct-2004. FontForge used to be named PfaEdit. mv feta-braces-i.pfa ./out mftrace --encoding feta-din10.enc -I ./out/ --pfa --simplify feta-din10 sh: potrace: command not found mftrace 1.0.34 Font `feta-din10'... Using encoding file: `./out/feta-din10.enc' Running Metafont... Tracing bitmaps... [33][102][109][112][114][115][122] Assembling font to `feta-din10.pfa'... Simplifying font... Copyright (c) 2000-2004 by George Williams. Executable based on sources from 07:34 12-Oct-2004. FontForge used to be named PfaEdit. error: fontforge: command exited with value 11 Traceback (most recent call last): File /usr/bin/mftrace, line 1188, in ? cleanup_font (basename + '.pfa') File /usr/bin/mftrace, line 841, in cleanup_font system (%s -script simplify.pe %s %s % (fontforge_cmd, file, tfmfile)) File /usr/bin/mftrace, line 246, in system error (msg) File /usr/bin/mftrace, line 137, in error raise _ (Exiting ... ) Exiting ... make[1]: *** [out/feta-din10.pfa] Error 1 make[1]: Leaving directory `/home/karl/lilypond/mf' make: *** [all] Error 2 $ Regards /Karl ___ lilypond-devel mailing list [EMAIL PROTECTED]
Re: programming error: Grob `NoteHead' has no interface for property ...
Juergen Reuter [EMAIL PROTECTED] wrote: This is a known annoying problem (not only) with ancient notation, but harmless. Just consider it as a reminder for the developers; see the thread starting at: http://lists.gnu.org/archive/html/lilypond-devel/2004-09/msg00073.html If it is because of internal variables, how come I got rid of the errors below? Greetings, Jürgen On Sat, 6 Nov 2004, Karl Hammar wrote: $ grep ChangeLog CVS/Entries /ChangeLog/1.2805/Fri Nov 5 17:00:22 2004// $ cat run (./autogen.sh --prefix=$HOME; make clean; make $@) log 21 $ ./run all install web $ grep 'programming error: Grob `NoteHead'\'' has no interface for property' log | wc 1253 12626 98715 $ grep 'programming error: Grob `NoteHead'\'' has no interface for property' log | sed -e 's/^.*property//;' | sort | uniq -c 24 `add-cauda' 52 `add-join' ... 9 `virga' 138 `x-offset' $ ... I have found a way to get rid of the errors. Previous run, lots of errors from make web: $ grep -i error log | wc -l 1359 $ New idea: $ (./autogen.sh --prefix=$HOME; make clean; make all install; . buildscripts/out/lilypond-profile; make web) log 21 $ grep -i error log | uniq -c 3 make[1]: [out/parser.hh] Error 1 (ignored) 1 make[1]: [out/parser.cc] Error 1 (ignored) 40 Do not worry about the @ signs in the output, they are not errors. 1 file:line:error style messages enabled. $ So the cause of the programming errors is a missing path in TEXMF. I somehow lost track among all makefiles, which makefile (make/lilypond-vars.make I guess) should be changed and which directory should be added to TEXMF if lilypond is not installed? The installed share/ and the build share/ seems to differ: $ ls share/lilypond/tex/ mf-out source tex-out $ ls $HOME/share/lilypond/2.5.0.hwn1/tex/ feta-braces-a.tex feta-din8.tex feta18.tex parmesan13.tex feta-braces-b.tex feta-nummer10.tex feta20.tex parmesan14.tex ... maybe the share/.../tex directory doesn't matter. Regards /Karl ___ lilypond-devel mailing list [EMAIL PROTECTED] http://lists.gnu.org/mailman/listinfo/lilypond-devel
Re: 2.4.0 on cygwin - Bad fd number
Han-Wen Nienhuys [EMAIL PROTECTED] wrote: [EMAIL PROTECTED] writes: ... Error invoking `latex \\nonstopmode \\input simple.tex 21 1 /dev/null '. Return value 512Error invoking `dvips -t a4 -u+ec-mftrace.map -u+lilypond.map -Ppdf simple 21 1 /dev/null '. Return value 512 Ah wait, the shell is struggling with the redirection syntax. are there any unix gurus that know the correct syntax for redirecting all output to /dev/null? Short story: $ ... /dev/null 21 or $ ... /dev/null 2 /dev/null Long story: . echo writes to stdout (file descriptor 1) . redirection $cmd a file is: (common case: a==1 || a==2) close(a); open(file,O_WRONLY) /* == a */; . to make it write to stderr (fd 2), dup2(b,a) it by $cmd ab . you can omit the 1 in 1... below, because it is the default number $ (echo output; echo error 12 ) output error $ Example A = Example below works since: 1 close(1); open(a,O_WRONLY) /* == 1 */; 2 fork() and exec() subshell /* remember file descriptor table is inherited */ 3 run echo sequentially $ (echo output; echo error 12 ) 1 a error $ cat a output $ Example B = $ (echo output; echo error 12 ) 2 b output $ cat b error $ Example C = close(1); open(c,O_WRONLY) /* == 1 */; dup2(1,2); $ (echo output; echo error 12 ) 1 c 21 $ cat c output error $ Example D = This one is equivalent to the above, but less to type it is a special syntax to redirect both stdout and stderr (you could also use d) $ (echo output; echo error 12 ) d $ cat d output error $ The drawback is that not all shells accepts it, e.g. ash treats it like . a (echo output; echo error 12 ) command . followed by a commandless redirection d, which empties the file $ ash $ (echo output; echo error 12 ) d $ output error press return to get a new prompt [1] + Done (echo output; echo error 12) $ cat d $ Example E = Redirection is order dependant, this one does: dup2(1,2); close(1); open(e,O_WRONLY) /* == 1 */; $ (echo output; echo error 12 ) 21 1 e error $ cat e output $ Example F = And yes, you can bypass it (/dev/tty is current tty) $ (echo output; (echo wall /dev/tty; echo error) 12 ) f wall $ cat f output error $ Example G = So: $ (echo output; echo error 12 ) 21 1 g error $ cat g output $ will become: dup2(1,2); /* all's well, stderr will go to current stdout (/dev/tty) */ close(1); open(g,O_WRONLY) /* == 1 */; Example H = And lastly, don't use 1, bash accepts it and treats it like , but ash says it's a syntax error $ (echo output; echo error 12 ) 21 1 h $ cat h output error $ ash $ (echo output; echo error 12 ) 21 1 h ash: Syntax error: Bad fd number $ Regards /Karl ___ lilypond-devel mailing list [EMAIL PROTECTED] http://lists.gnu.org/mailman/listinfo/lilypond-devel
programming error: Grob `NoteHead' has no interface for property ...
$ grep ChangeLog CVS/Entries /ChangeLog/1.2805/Fri Nov 5 17:00:22 2004// $ cat run (./autogen.sh --prefix=$HOME; make clean; make $@) log 21 $ ./run all install web $ grep 'programming error: Grob `NoteHead'\'' has no interface for property' log | wc 1253 12626 98715 $ grep 'programming error: Grob `NoteHead'\'' has no interface for property' log | sed -e 's/^.*property//;' | sort | uniq -c 24 `add-cauda' 52 `add-join' 9 `add-stem' 3 `ascendens' 18 `auctum' 2 `balloon-original-callback' 2 `balloon-text' 2 `balloon-text-offset' 2 `balloon-text-props' 157 `context-info' 169 `delta-pitch' 15 `deminutum' 12 `descendens' 5 `flexa-height' 17 `flexa-width' 138 `glyph-name' 31 `head-width' 13 `inclinatum' 15 `join-left-amount' 5 `oriscus' 57 `pes-or-flexa' 138 `prefix-set' 31 `primitive' 5 `quilisma' 5 `stropha' 10 `text' 169 `thickness' 9 `virga' 138 `x-offset' $ Is this a problem? My file:///home/karl/lilypond/input/regression/out-www/collated-files.html looks good though even if I don't know if it is complete. Regards /Karl ___ lilypond-devel mailing list [EMAIL PROTECTED] http://lists.gnu.org/mailman/listinfo/lilypond-devel
