lilypond-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: lilybuntu confusion


From: Trevor Daniels
Subject: Re: lilybuntu confusion
Date: Fri, 3 Jul 2009 21:37:53 +0100


Jonathan Kulp wrote Friday, July 03, 2009 6:41 PM


Graham Percival wrote:
On Wed, Jul 01, 2009 at 10:50:07PM +0100, Trevor Daniels wrote:
Graham Percival wrote Wednesday, July 01, 2009 9:32 PM
Subject: Re: lilybuntu confusion

On Wed, Jul 01, 2009 at 10:24:24AM -0700, Mark Polesky wrote:

Not to discourage this whole endeavor, but this can be kind-of
done already -- you use the last binary rather than compiling
lilyond from source, but as long as there's no syntax updates, it
works just fine.  I've done this for years (yes, even during
GDP!), and Trevor still does this.
Sort of.  I never actually compile the whole documentation.

Ah, sorry.

All I need to do is to check that any edits I make are not
going to break compilation, so all I need is to pass
the single file I've just edited through lilypond-book,
run a simple script to check refs, and then run texi2html
so I can check the formatting and appearance in a browser.
I have a script to do this and it usually works fine.  It's
far quicker than building the whole set of documents - typically 10-30 secs depending on the number of examples and snippets to be compiled.

Does this still compile the lilypond examples?  That would take
take much longer than 10-30 seconds. If there's a way to generate
the docs without compiling lilypond examples (i.e. just show the
verbatim code, without actually generating the images), that would be awesome. 99% of doc updates don't need to be image-checked, so
that could save me/whoever a lot of time.  :)


I'm able to copy to different location and compile certain files by themselves as long as they don't require a bunch of @include files. They compile very quickly this way. I wonder if it might help to define some new Make targets. For example "make lilypond-learning.pdf" would only compile the LM, and I would assume save quite a bit of time in compiling over "make doc".

Here's the script I use.  It's a bit of a hack,
especially getting the macros right, but it works
well for me.  I never did get around to tidying it
up :(

Trevor

#!/bin/sh

NAME="rhythms"
FROMDIR="C:/Users/Trevor/LilyPond-git"
DOCDIR="C:/Users/Trevor/LilyPond-docs"
TODIR=$DOCDIR/$NAME

if test ! -d $TODIR; then
 mkdir $TODIR
fi
if test ! -d $TODIR/out; then
 mkdir $TODIR/out
fi

cp $FROMDIR/Documentation/user/macros.itexi $TODIR/out/macros.itexi

cp $FROMDIR/input/manual/$NAME-headword.ly $TODIR/$NAME-headword.ly

echo Running lilypond-book
cd $TODIR
python "c:/program files/lilypond/usr/bin/lilypond-book.py" -f texi-html -I $FROMDIR/input/lsr -o out $FROMDIR/Documentation/user/$NAME.itely

echo Running RefCheck
cd d:/users/trevor/leisure/computing/python/refcheck
python refcheck.py
cd $DOCDIR

echo Running texi2html
cat $DOCDIR/macros.texi $TODIR/out/$NAME.texi > $TODIR/$NAME.texi
c:/programs/texi2html/texi2html --no-validate --output=$TODIR/out/$NAME.html $TODIR/$NAME.texi

echo Running IE
"C:\Program Files\Internet Explorer\iexplore.exe" $TODIR/out/$NAME.html





reply via email to

[Prev in Thread] Current Thread [Next in Thread]