lilypond-book version 2.0 had a fragment option called filename="xyzzy".
Unfortunately, all this option did was to rename the lily-*.ly file to
xyzzy.ly, while the .png, etc files (which I would consider the important
ones, since they are the final results) are still referred to with random
numbers.
I was able to fix this in lilypond-book-2.0.1 with the patch below. It
then used "xyzzy" as the basename for the whole process (.ly .tex .dvi
.eps .png etc). This patch saved my project, since python was choking on
the size of my source file. I had to break it into smaller pieces for
lilypond-book to generate the .png's from, and I was able to keep <img>
references to predictable filenames in the original source file.
I've downloaded lilypond version 2.3 now, and noticed that lilypond-book
has been extensively redesigned. It now tells me this:
lilypond-book: warning: ignoring unknown ly option: filename="baron1"
However the documentation still has "filename" listed as an option:
filename=filename
This names the file for the |printfilename| option. The argument
should be unquoted.
There doesn't appear to be any other references to the "printfilename"
option.
Is there any possibility of getting a filename option which uses the
given filename for the whole process, instead of just leaving a copy
of the .ly file with that name?
~ John Williams
--- lilypond-2.0.1/scripts/out/lilypond-book Sun Jan 11 12:42:20 2004
+++ ./lilypond-book Thu May 20 11:11:13 2004
@@ -732,8 +732,6 @@
}
''' % (optstring, music_size, linewidth, indent, notime) + body
- if orig_name:
- body = '\\renameinput \"%s\"\n%s' % (orig_name, body)
# ughUGH not original options
@@ -1251,7 +1249,7 @@
original_name = os.path.basename (original_name)
- base = unique_file_name (content)
+ base = original_name
outname = base + '.ly'
changed = update_file (content, outname)
_______________________________________________
bug-lilypond mailing list
[EMAIL PROTECTED]
http://lists.gnu.org/mailman/listinfo/bug-lilypond
