Merge with master

[lilypond.git] / Documentation / user / working.itely

@c -*- coding: utf-8; mode: texinfo; -*-
@c This file is part of lilypond.tely
@ignore
    Translation of GIT committish: FILL-IN-HEAD-COMMITTISH

    When revising a translation, copy the HEAD committish of the
    version that you are working on.  See TRANSLATION for details.
@end ignore

@node Working on LilyPond projects
@chapter Working on LilyPond projects

This section explains how to solve or avoid certain common
problems.  If you have programming experience, many of these
tips may seem obvious, but it is still advisable to read
this chapter.


@menu
* Suggestions for writing LilyPond files::  
* Saving typing with identifiers and functions::  
* Style sheets::                
* Updating old files::          
* Troubleshooting (taking it all apart)::  
* Minimal examples::            
@end menu


@node Suggestions for writing LilyPond files
@section Suggestions for writing LilyPond files

Now you're ready to begin writing larger LilyPond files -- not just the
little examples in the tutorial, but whole pieces.  But how should you
go about doing it?

As long as LilyPond can understand your files and produces the output
that you want, it doesn't matter what your files look like.  However,
there are a few other things to consider when writing lilypond files.

@itemize @bullet
@item What if you make a mistake?  The structure of a lilypond
file can make certain errors easier (or harder) to find.

@item What if you want to share your files with somebody
else?  In fact, what if you want to alter your own files in
a few years?  Some lilypond files are understandable at
first glance; other files may leave you scratching your head
for an hour.

@item What if you want to upgrade your lilypond file for use
with a later version of lilypond?  The input syntax changes
occasionally as lilypond improves.  Most changes can be
done automatically with @code{convert-ly}, but some changes
might require manual assistance.  Lilypond files can be
structured in order to be easier (or header) to update.
@end itemize

@menu
* General suggestions::         
* Typesetting existing music::  
* Large projects::              
@end menu


@node General suggestions
@subsection General suggestions

Here are a few suggestions that can help you to avoid or fix
problems:

@itemize @bullet
@item @strong{Include @code{\version} numbers in every file}.  Note that all
templates contain a @code{\version "2.9.13"} string.  We
highly recommend that you always include the @code{\version}, no matter
how small your file is.  Speaking from personal experience, it's
quite frustrating to try to remember which version of LilyPond you were
using a few years ago.  @code{convert-ly} requires you to declare
which version of LilyPond you used.

@item @strong{Include checks}: @ref{Bar check} and @ref{Octave check}.  If
you
include checks every so often, then if you make a mistake, you can pinpoint
it quicker.  How often is @q{every so often}?  It depends on the complexity
of the music.  For very simple music, perhaps just once or twice.  For
very complex music, perhaps every bar.

@item @strong{One bar per line of text}.  If there is anything complicated,
either in the music
itself or in the output you desire, it's often good to write only one bar
per line.  Saving screen space by cramming eight bars per line just isn't
worth it if you have to @q{debug} your files.

@item @strong{Comment your files}.  Use either bar numbers (every so often)
or
references to musical themes (@q{second theme in violins,} @q{fourth
variation,} etc.).  You may not need comments when you're writing the piece
for the first time, but if you want to go back to change something two or
three years later, or if you pass the source over to a friend, it will
be much more
challenging to determine your intentions or how your file is structured if
you didn't comment the file.

@item @strong{Indent your braces}.  A lot of problems are caused by an
imbalance
in the number of @code{@{} and @code{@}}.

@item @strong{Explicity add durations} at the beginnings of sections
and identifiers.  If you specify @code{c4 d e} at the beginning of a
phrase (instead of just @code{c d e}) you can save yourself some
problems if you rearrange your music later.

@item @strong{Separate tweaks} from music definitions.  See
@ref{Saving typing with identifiers and functions} and
@ref{Style sheets}.

@end itemize


@node Typesetting existing music
@subsection Typesetting existing music

If you are entering music from an existing score (i.e., typesetting a
piece of existing sheet music),

@itemize @bullet

@item Enter one manuscript (the physical copy) system at a time (but still
only one bar per line of text), and
check each system when you finish it.  You may use the
@code{showLastLength} command to speed up processing -- see
@ref{Skipping corrected music}.

@item Define @code{mBreak = @{ \break @}} and insert @code{\mBreak}
in the input file whenever the manuscript has a line break.  This
makes it much easier to compare the LilyPond music to the original
music.  When you are finished proofreading your score, you may
define @code{mBreak = @{ @}} to remove all those line breaks.  This
will allow LilyPond to place line breaks wherever it feels are
best.

@end itemize


@node Large projects
@subsection Large projects

When working on a large project, having a clear structure to your
lilypond files becomes vital.

@itemize @bullet

@item @strong{Use an identifier for each voice}, with a minimum of
structure inside the definition.  The structure of the
@code{\score} section is the most likely thing to change;
the @code{violin} definition is extremely unlikely to change
in a new version of LilyPond.

@example
violin = \relative c'' @{
g4 c'8. e16
@}
...
\score @{
  \new GrandStaff @{
    \new Staff @{
      \violin
    @}
  @}
@}
@end example

@item @strong{Separate tweaks from music definitions}.  This
point was made in @ref{General suggestions}, but for large
projects it is absolutely vital.  We might need to change
the definition of @code{fthenp}, but then we only need
to do this once, and we can still avoid touching anything
inside @code{violin}.

@example
fthenp = _\markup@{
  \dynamic f \italic \small @{ 2nd @} \hspace #0.1 \dynamic p @}
violin = \relative c'' @{
g4\fthenp c'8. e16
@}
@end example

@end itemize


@node Saving typing with identifiers and functions
@section Saving typing with identifiers and functions

@cindex variables
@cindex identifiers

By this point, you've seen this kind of thing:

@lilypond[quote,verbatim,ragged-right]
hornNotes = \relative c'' { c4 b dis c }
\score {
  {
    \hornNotes
  }
}
@end lilypond

You may even realize that this could be useful in minimalist music:

@lilypond[quote,verbatim,ragged-right]
fragA = \relative c'' { a4 a8. b16 }
fragB = \relative c'' { a8. gis16 ees4 }
violin = \new Staff { \fragA \fragA \fragB \fragA }
\score {
  {
    \violin
  }
}
@end lilypond

However, you can also use these identifiers (also known as
variables, macros, or (user-defined) command) for tweaks:

@lilypond[quote,verbatim,ragged-right]
dolce = \markup{ \italic \bold dolce }
padText = { \once \override TextScript #'padding = #5.0 }
fthenp=_\markup{ \dynamic f \italic \small { 2nd } \hspace #0.1 \dynamic p }
violin = \relative c'' {
  \repeat volta 2 {
    c4._\dolce b8 a8 g a b |
    \padText
    c4.^"hi there!" d8 e' f g d |
    c,4.\fthenp b8 c4 c-. |
  }
}
\score {
  {
    \violin
  }
\layout{ragged-right=##t}
}
@end lilypond

These identifiers are obviously useful for saving
typing.  But they're worth considering even if you
only use them once -- they reduce complexity.  Let's
look at the previous example without any
identifiers.  It's a lot harder to read, especially
the last line.

@example
violin = \relative c'' @{
  \repeat volta 2 @{
    c4._\markup@{ \italic \bold dolce @} b8 a8 g a b |
    \once \override TextScript #'padding = #5.0
    c4.^"hi there!" d8 e' f g d |
    c,4.\markup@{ \dynamic f \italic \small @{ 2nd @}
      \hspace #0.1 \dynamic p @} b8 c4 c-. |
  @}
@}
@end example

So far we've seen static substitution -- when LilyPond
sees @code{\padText}, it replaces it with the stuff that
we've defined it to be (ie the stuff to the right of
@code{padtext=}).

LilyPond can handle non-static substitution, too (you
can think of these as functions).

@lilypond[quote,verbatim,ragged-right]
padText =
#(define-music-function (parser location padding) (number?)
  #{
    \once \override TextScript #'padding = #$padding
  #})

\relative c''' {
  c4^"piu mosso" b a b
  \padText #1.8
  c4^"piu mosso" d e f
  \padText #2.6
  c4^"piu mosso" fis a g
}
@end lilypond

Using identifiers is also a good way to reduce work if the
LilyPond input syntax changes (see @ref{Updating old files}).  If
you have a single definition (such as @code{\dolce}) for all your
files (see @ref{Style sheets}), then if the syntax changes, you
only need to update your single @code{\dolce} definition,
instead of making changes throughout every @code{.ly} file.


@node Style sheets
@section Style sheets

The output that LilyPond produces can be heavily modified; see
@ref{Tweaking output} for details.  But what if you have many
files that you want to apply your tweaks to?  Or what if you
simply want to separate your tweaks from the actual music?  This
is quite easy to do.

Let's look at an example.  Don't worry if you don't understand
the parts with all the #().  This is explained in
@ref{Advanced tweaks with Scheme}.

@lilypond[quote,verbatim,ragged-right]
mpdolce = #(make-dynamic-script (markup #:hspace 1 #:translate (cons 5 0)
  #:line(#:dynamic "mp" #:text #:italic "dolce" )))
tempoMark = #(define-music-function (parser location markp) (string?)
#{
  \once \override Score . RehearsalMark #'self-alignment-X = #left
  \once \override Score . RehearsalMark #'extra-spacing-width = #'(+inf.0 . -inf.0)
  \mark \markup { \bold $markp }
#})

\relative c'' {
  \tempo 4=50
  a4.\mpdolce d8 cis4--\glissando a | b4 bes a2
  \tempoMark "Poco piu mosso"
  cis4.\< d8 e4 fis | g8(\! fis)-. e( d)-. cis2
}
@end lilypond

There are some problems with overlapping output; we'll fix those using
the techniques in @ref{Moving objects}.  But let's also
do something about the @code{mpdolce} and @code{tempoMark}
definitions.  They produce the output we desire, but we might want
to use them in another piece.  We could simply copy-and-paste them
at the top of every file, but that's an annoyance.  It also leaves
those definitions in our music files, and I personally find all
the #() somewhat ugly.  Let's hide them in another file:

@example
%%% save this to a file called "definitions.ly"
mpdolce = #(make-dynamic-script (markup #:hspace 1 #:translate (cons 5 0)
  #:line(#:dynamic "mp" #:text #:italic "dolce" )))
tempoMark = #(define-music-function (parser location markp) (string?)
#@{
  \once \override Score . RehearsalMark #'self-alignment-X = #left
  \once \override Score . RehearsalMark #'extra-spacing-width = #'(+inf.0 . -inf.0)
  \mark \markup @{ \bold $markp @}
#@})
@end example

Now let's modify our music (let's save this file as @file{"music.ly"}).

@c  We have to do this awkward example/lilypond-non-verbatim
@c  because we can't do the \include stuff in the manual.

@example
\include "definitions.ly"

\relative c'' @{
  \tempo 4=50
  a4.\mpdolce d8 cis4--\glissando a | b4 bes a2
  \once \override Score.RehearsalMark #'padding = #2.0
  \tempoMark "Poco piu mosso"
  cis4.\< d8 e4 fis | g8(\! fis)-. e( d)-. cis2
@}
@end example

@lilypond[quote,ragged-right]
mpdolce = #(make-dynamic-script (markup #:hspace 1 #:translate (cons 5 0)
  #:line(#:dynamic "mp" #:text #:italic "dolce" )))
tempoMark = #(define-music-function (parser location markp) (string?)
#{
  \once \override Score . RehearsalMark #'self-alignment-X = #left
  \once \override Score . RehearsalMark #'extra-spacing-width = #'(+inf.0 . -inf.0)
  \mark \markup { \bold $markp }
#})

\relative c'' {
  \tempo 4=50
  a4.\mpdolce d8 cis4--\glissando a | b4 bes a2
  \once \override Score.RehearsalMark #'padding = #2.0
  \tempoMark "Poco piu mosso"
  cis4.\< d8 e4 fis | g8(\! fis)-. e( d)-. cis2
}
@end lilypond

That looks better, but let's make a few changes.  The glissando is hard
to see, so let's make it thicker and closer to the noteheads.  Let's
put the metronome marking above the clef, instead of over the first
note.  And finally, my composition professor hates "C" time signatures,
so we'd better make that "4/4" instead.

Don't change @file{music.ly}, though.  Replace our @file{definitions.ly}
with this:

@example
%%%  definitions.ly
mpdolce = #(make-dynamic-script (markup #:hspace 1 #:translate (cons 5 0)
  #:line( #:dynamic "mp" #:text #:italic "dolce" )))
tempoMark = #(define-music-function (parser location markp) (string?)
#@{
  \once \override Score . RehearsalMark #'self-alignment-X = #left
  \once \override Score . RehearsalMark #'extra-spacing-width = #'(+inf.0 . -inf.0)
  \mark \markup @{ \bold $markp @}
#@})

\layout@{
  \context @{ \Score
    \override MetronomeMark #'extra-offset = #'(-9 . 0)
    \override MetronomeMark #'padding = #'3
  @}
  \context @{ \Staff
    \override TimeSignature #'style = #'numbered
  @}
  \context @{ \Voice
    \override Glissando #'thickness = #3
    \override Glissando #'gap = #0.1
  @}
@}
@end example

@lilypond[quote,ragged-right]
mpdolce = #(make-dynamic-script (markup #:hspace 1 #:translate (cons 5 0)
  #:line( #:dynamic "mp" #:text #:italic "dolce" )))
tempoMark = #(define-music-function (parser location markp) (string?)
#{
  \once \override Score . RehearsalMark #'self-alignment-X = #left
  \once \override Score . RehearsalMark #'extra-spacing-width = #'(+inf.0 . -inf.0)
  \mark \markup { \bold $markp }
#})

\layout{
  \context { \Score
    \override MetronomeMark #'extra-offset = #'(-9 . 0)
    \override MetronomeMark #'padding = #'3
  }
  \context { \Staff
    \override TimeSignature #'style = #'numbered
  }
  \context { \Voice
    \override Glissando #'thickness = #3
    \override Glissando #'gap = #0.1
  }
}

\relative c'' {
  \tempo 4=50
  a4.\mpdolce d8 cis4--\glissando a | b4 bes a2
  \once \override Score.RehearsalMark #'padding = #2.0
  \tempoMark "Poco piu mosso"
  cis4.\< d8 e4 fis | g8(\! fis)-. e( d)-. cis2
}
@end lilypond

That looks nicer!  But now suppose that I want to publish this
piece.  My composition professor doesn't like "C" time
signatures, but I'm somewhat fond of them.  Let's copy the
current @file{definitions.ly} to @file{web-publish.ly} and
modify that.  Since this music is aimed at producing a pdf which
will be displayed on the screen, we'll also increase the
overall size of the output.

@example
%%%  definitions.ly
mpdolce = #(make-dynamic-script (markup #:hspace 1 #:translate (cons 5 0)
  #:line( #:dynamic "mp" #:text #:italic "dolce" )))
tempoMark = #(define-music-function (parser location markp) (string?)
#@{
  \once \override Score . RehearsalMark #'self-alignment-X = #left
  \once \override Score . RehearsalMark #'extra-spacing-width = #'(+inf.0 . -inf.0)
  \mark \markup @{ \bold $markp @}
#@})

#(set-global-staff-size 23)
\layout@{
  \context @{ \Score
    \override MetronomeMark #'extra-offset = #'(-9 . 0)
    \override MetronomeMark #'padding = #'3
  @}
  \context @{ \Staff
  @}
  \context @{ \Voice
    \override Glissando #'thickness = #3
    \override Glissando #'gap = #0.1
  @}
@}
@end example

@lilypond[quote,ragged-right]
mpdolce = #(make-dynamic-script (markup #:hspace 1 #:translate (cons 5 0)
  #:line( #:dynamic "mp" #:text #:italic "dolce" )))
tempoMark = #(define-music-function (parser location markp) (string?)
#{
  \once \override Score . RehearsalMark #'self-alignment-X = #left
  \once \override Score . RehearsalMark #'extra-spacing-width = #'(+inf.0 . -inf.0)
  \mark \markup { \bold $markp }
#})

#(set-global-staff-size 23)
\layout{
  \context { \Score
    \override MetronomeMark #'extra-offset = #'(-9 . 0)
    \override MetronomeMark #'padding = #'3
  }
  \context { \Voice
    \override Glissando #'thickness = #3
    \override Glissando #'gap = #0.1
  }
}

\relative c'' {
  \tempo 4=50
  a4.\mpdolce d8 cis4--\glissando a | b4 bes a2
  \once \override Score.RehearsalMark #'padding = #2.0
  \tempoMark "Poco piu mosso"
  cis4.\< d8 e4 fis | g8(\! fis)-. e( d)-. cis2
}
@end lilypond

Now in our music, I simply replace
@code{\include "definitions.ly"} with
@code{\include "web-publish.ly"}.  Of course, we could make this
even more convenient.  We could make a @file{definitions.ly} file which
contains only the definitions of @code{mpdolce} and @code{tempoMark}, a
@file{web-publish.ly} file which contains only the @code{\layout}
section listed above, and a @file{university.ly} file which
contains only the tweaks to produce the output that my professor
prefers.  The top of @file{music.ly} would then look like this:

@example
\include "definitions.ly"

%%%  Only uncomment one of these two lines!
\include "web-publish.ly"
%\include "university.ly"
@end example

This approach can be useful even if you are only producing
one set of parts.  I use half a dozen different
@q{style sheet} files for my projects.  I begin every music
file with @code{\include "../global.ly"}, which contains

@example
%%%   global.ly
\version "2.9.13"
#(ly:set-option 'point-and-click #f)
\include "../init/init-defs.ly"
\include "../init/init-layout.ly"
\include "../init/init-headers.ly"
\include "../init/init-paper.ly"
@end example


@node Updating old files
@section Updating old files

The LilyPond input syntax occasionally changes.  As LilyPond itself
improves, the syntax (input language) is modified accordingly.  Sometimes
these changes are made to make the input easier to read and write or
sometimes the changes are made to accomodate new features of LilyPond.

LilyPond comes with a file that makes this updating easier:
@code{convert-ly}.  For details about how to run this program, see
@ref{Updating files with convert-ly}.

Unforunately, @code{convert-ly} cannot handle all input changes.  It
takes care of simple search-and-replace changes (such as @code{raggedright}
becoming @code{ragged-right}), but some changes are too
complicated.  The syntax changes that @code{convert-ly} cannot handle
are listed in @ref{Updating files with convert-ly}.

For example, in LilyPond 2.4 and earlier, accents and non-English
letters were entered using LaTeX -- for example,
@samp{No\"el} (this would print the French word for
@q{Christmas}).  In LilyPond 2.6 and above, the special
@samp{ë} must be entered directly into the LilyPond file as an
UTF-8 character.  @code{convert-ly} cannot change all the LaTeX
special characters into UTF-8 characters; you must manually update
your old LilyPond files.


@node Troubleshooting (taking it all apart)
@section Troubleshooting (taking it all apart)

Sooner or later, you will write a file that LilyPond cannot
compile.  The messages that LilyPond gives may help
you find the error, but in many cases you need to do some
investigation to determine the source of the problem.

The most powerful tools for this purpose are the
single line comment (indicated by @code{%}) and the block
comment (indicated by @code{%@{ ... %@}}).  If you don't
know where a problem is, start commenting out huge portions
of your input file.  After you comment out a section, try
compiling the file again.  If it works, then the problem
must exist in the portion you just commented.  If it doesn't
work, then keep on commenting out material until you have
something that works.

In an extreme case, you might end up with only

@example
\score @{
  <<
    % \melody
    % \harmony
    % \bass
  >>
  \layout@{@}
@}
@end example

@noindent
(in other words, a file without any music)

If that happens, don't give up.  Uncomment a bit -- say,
the bass part -- and see if it works.  If it doesn't work,
then comment out all of the bass music (but leave
@code{\bass} in the @code{\score} uncommented.

@example
bass = \relative c' @{
%@{
  c4 c c c
  d d d d
%@}
@}
@end example

Now start slowly uncommenting more and more of the
@code{bass} part until you find the problem line.

Another very useful debugging technique is constructing
@ref{Minimal examples}.


@node Minimal examples
@section Minimal examples

A minimal example is an example which is as small as possible.  These
examples are much easier to understand than long examples.  Minimal
examples are used for

@itemize
@item Bug reports
@item Sending a help request to mailists
@item Adding an example to the @uref{http://lsr@/.dsi@/.unimi/2.it/,LilyPond
Snippet Repository}
@end itemize

To construct an example which is as small as possible, the rule is
quite simple: remove anything which is not necessary.  When trying to
remove unnecessary parts of a file, it is a very good idea to comment
out lines instead of deleting them.  That way, if you discover that you
actually @emph{do} need some lines, you can un-comment them, instead of
typing them in from scratch.

There are two exceptions to the @qq{as small as possible} rule:

@itemize
@item Include the @code{\version} number.
@item If possible, use @code{\paper@{ ragged-right=##t @}} at the
top of your example.
@end itemize

The whole point of a minimal example is to make it easy to read:

@itemize
@item Avoid using complicated notes, keys, or time signatures, unless you
wish to demonstrate something is about the behavior of those items.
@item Do not use \override commands unless that is the point of the
example.
@end itemize