[lilypond.git] / Documentation / notation / input.itely

@c -*- coding: utf-8; mode: texinfo; -*-

@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.  For details, see the Contributors'
    Guide, node Updating translation committishes..
@end ignore

@c \version "2.19.22"

@node General input and output
@chapter General input and output

This section deals with general LilyPond input and output issues,
rather than specific notation.

@menu
* Input structure::
* Titles and headers::
* Working with input files::
* Controlling output::
* Creating MIDI output::
* Extracting musical information::
@end menu


@node Input structure
@section Input structure

The main format of input for LilyPond are text files.  By convention,
these files end with @file{.ly}.

@menu
* Structure of a score::
* Multiple scores in a book::
* Multiple output files from one input file::
* Output file names::
* File structure::
@end menu


@node Structure of a score
@subsection Structure of a score

@funindex \score

A @code{\score} block must contain a single music expression
delimited by curly brackets:

@example
\score @{
  @dots{}
@}
@end example

@warning{There must be @strong{only one} outer music expression in
a @code{\score} block, and it @strong{must} be surrounded by
curly brackets.}

This single music expression may be of any size, and may contain
other music expressions to any complexity.  All of these examples
are music expressions:

@example
@{ c'4 c' c' c' @}
@end example

@lilypond[verbatim,quote]
{
  { c'4 c' c' c' }
  { d'4 d' d' d' }
}
@end lilypond

@lilypond[verbatim,quote]
<<
  \new Staff { c'4 c' c' c' }
  \new Staff { d'4 d' d' d' }
>>
@end lilypond

@example
@{
  \new GrandStaff <<
    \new StaffGroup <<
      \new Staff @{ \flute @}
      \new Staff @{ \oboe @}
    >>
    \new StaffGroup <<
      \new Staff @{ \violinI @}
      \new Staff @{ \violinII @}
    >>
  >>
@}
@end example

Comments are one exception to this general rule.  (For others see
@ref{File structure}.)  Both single-line comments and comments
delimited by @code{%@{ @dots{} %@}} may be placed anywhere within an
input file.  They may be placed inside or outside a @code{\score}
block, and inside or outside the single music expression within a
@code{\score} block.

Remember that even in a file containing only a @code{\score} block, it
is implicitly enclosed in a \book block.  A \book block in a source
file produces at least one output file, and by default the name of the
output file produced is derived from the name of the input file, so
@file{fandangoforelephants.ly} will produce
@file{fandangoforelephants.pdf}.

(For more details  about @code{\book} blocks, see
@ref{Multiple scores in a book},
@ref{Multiple output files from one input file} @ref{File structure}.)

@seealso
Learning Manual:
@rlearning{Working on input files},
@rlearning{Music expressions explained},
@rlearning{Score is a (single) compound musical expression}.


@node Multiple scores in a book
@subsection Multiple scores in a book

@funindex \book
@cindex movements, multiple

A document may contain multiple pieces of music and text.  Examples
of these are an etude book, or an orchestral part with multiple
movements.  Each movement is entered with a @code{\score} block,

@example
\score @{
  @var{@dots{}music@dots{}}
@}
@end example

and texts are entered with a @code{\markup} block,

@example
\markup @{
  @var{@dots{}text@dots{}}
@}
@end example

@funindex \book

All the movements and texts which appear in the same @file{.ly} file
will normally be typeset in the form of a single output file.

@example
\score @{
  @var{@dots{}}
@}
\markup @{
  @var{@dots{}}
@}
\score @{
  @var{@dots{}}
@}
@end example

One important exception is within lilypond-book documents,
where you explicitly have to add a @code{\book} block, otherwise only
the first @code{\score} or @code{\markup} will appear in the output.

The header for each piece of music can be put inside the @code{\score}
block.  The @code{piece} name from the header will be printed before
each movement.  The title for the entire book can be put inside the
@code{\book}, but if it is not present, the @code{\header} which is at
the top of the file is inserted.

@example
\header @{
  title = "Eight miniatures"
  composer = "Igor Stravinsky"
@}
\score @{
  @dots{}
  \header @{ piece = "Romanze" @}
@}
\markup @{
   @dots{}text of second verse@dots{}
@}
\markup @{
   @dots{}text of third verse@dots{}
@}
\score @{
  @dots{}
  \header @{ piece = "Menuetto" @}
@}
@end example

@funindex \bookpart

Pieces of music may be grouped into book parts using @code{\bookpart}
blocks.  Book parts are separated by a page break, and can start with a
title, like the book itself, by specifying a @code{\header} block.

@example
\bookpart @{
  \header @{
    title = "Book title"
    subtitle = "First part"
  @}
  \score @{ @dots{} @}
  @dots{}
@}
\bookpart @{
  \header @{
    subtitle = "Second part"
  @}
  \score @{ @dots{} @}
  @dots{}
@}
@end example

@node Multiple output files from one input file
@subsection Multiple output files from one input file

If you want multiple output files from the same @file{.ly} file,
then you can add multiple @code{\book} blocks, where each
such \book block will result in a separate output file.
If you do not specify any @code{\book} block in the
input file, LilyPond will implicitly treat the whole
file as a single \book block, see
@ref{File structure}.

When producing multiple files from a single source file, Lilypond
ensures that none of the output files from any @code{\book} block
overwrites the output file produced by a preceding @code{\book} from
the same input file.

It does this by adding a suffix to the output name for each
@code{\book} which uses the default output file name derived from the
input source file.

The default behaviour is to append a version-number suffix for each
name which may clash, so

@example
\book @{
  \score @{ @dots{} @}
  \paper @{ @dots{} @}
@}
\book @{
  \score @{ @dots{} @}
  \paper @{ @dots{} @}
@}
\book @{
  \score @{ @dots{} @}
  \paper @{ @dots{} @}
@}
@end example

in source file @file{eightminiatures.ly}
will produce

@itemize
@item
@file{eightminiatures.pdf},
@item
@file{eightminiatures-1.pdf} and
@item
@file{eightminiatures-2.pdf}.
@end itemize

@node Output file names
@subsection Output file names

@funindex \bookOutputSuffix
@funindex \bookOutputName

Lilypond provides facilities to allow you to control what file names
are used by the various back-ends when producing output files.

In the previous section, we saw how Lilypond prevents name-clashes when
producing several outputs from a single source file.  You also have the
ability to specify your own suffixes for each @code{\book} block, so
for example you can produce files called
@file{eightminiatures-Romanze.pdf}, @file{eightminiatures-Menuetto.pdf}
and @file{eightminiatures-Nocturne.pdf} by adding a
@code{\bookOutputSuffix} declaration inside each @code{\book} block.

@example
\book @{
  \bookOutputSuffix "Romanze"
  \score @{ @dots{} @}
  \paper @{ @dots{} @}
@}
\book @{
  \bookOutputSuffix "Menuetto"
  \score @{ @dots{} @}
  \paper @{ @dots{} @}
@}
\book @{
  \bookOutputSuffix "Nocturne"
  \score @{ @dots{} @}
  \paper @{ @dots{} @}
@}
@end example

You can also specify a different output filename for @code{book} block,
by using @code{\bookOutputName} declarations

@example
\book @{
  \bookOutputName "Romanze"
  \score @{ @dots{} @}
  \paper @{ @dots{} @}
@}
\book @{
  \bookOutputName "Menuetto"
  \score @{ @dots{} @}
  \paper @{ @dots{} @}
@}
\book @{
  \bookOutputName "Nocturne"
  \score @{ @dots{} @}
  \paper @{ @dots{} @}
@}
@end example

The file above will produce these output files:

@itemize
@item
@file{Romanze.pdf},
@item
@file{Menuetto.pdf} and
@item
@file{Nocturne.pdf}.
@end itemize


@node File structure
@subsection File structure

@funindex \paper
@funindex \midi
@funindex \layout
@funindex \header
@funindex \score
@funindex \book
@funindex \bookpart

A @file{.ly} file may contain any number of toplevel expressions, where a
toplevel expression is one of the following:

@itemize
@item
An output definition, such as @code{\paper}, @code{\midi}, and
@code{\layout}.  Such a definition at the toplevel changes the default
book-wide settings.  If more than one such definition of the same type
is entered at the top level the definitions are combined, but in
conflicting situations the later definitions take precedence.  For
details of how this affects the @code{\layout} block see
@ref{The layout block,,The @code{@bs{}layout} block}.

@item
A direct scheme expression, such as
@code{#(set-default-paper-size "a7" 'landscape)} or
@code{#(ly:set-option 'point-and-click #f)}.

@item
A @code{\header} block.  This sets the global (i.e. the top of
file) header block.  This is the block containing the default
settings of titling fields like composer, title, etc. for all
books within the file (see @ref{Titles explained}).

@item
A @code{\score} block.  This score will be collected with other
toplevel scores, and combined as a single @code{\book}.
This behavior can be changed by setting the variable
@code{toplevel-score-handler} at toplevel.  The default handler is
defined in the init file @file{../scm/lily.scm}.

@item
A @code{\book} block logically combines multiple movements
(i.e., multiple @code{\score} blocks) in one document.  If there
are a number of @code{\score}s, one output file will be created
for each @code{\book} block, in which all corresponding movements
are concatenated.  The only reason to explicitly specify
@code{\book} blocks in a @file{.ly} file is if you wish to create
multiple output files from a single input file.  One exception is
within lilypond-book documents, where you explicitly have to add
a @code{\book} block if you want more than a single @code{\score}
or @code{\markup} in the same example.  This behavior can be
changed by setting the variable @code{toplevel-book-handler} at
toplevel.  The default handler is defined in the init file
@file{../scm/lily.scm}.

@item
A @code{\bookpart} block.  A book may be divided into several parts,
using @code{\bookpart} blocks, in order to ease the page breaking,
or to use different @code{\paper} settings in different parts.

@item
A compound music expression, such as
@example
@{ c'4 d' e'2 @}
@end example

This will add the piece in a @code{\score} and format it in a
single book together with all other toplevel @code{\score}s and music
expressions.  In other words, a file containing only the above
music expression will be translated into

@example
\book @{
  \score @{
    \new Staff @{
      \new Voice @{
        @{ c'4 d' e'2 @}
      @}
    @}
    \layout @{ @}
  @}
  \paper @{ @}
  \header @{ @}
@}
@end example

This behavior can be changed by setting the variable
@code{toplevel-music-handler} at toplevel.  The default handler is
defined in the init file @file{../scm/lily.scm}.

@item
A markup text, a verse for example
@example
\markup @{
   2.  The first line verse two.
@}
@end example

Markup texts are rendered above, between or below the scores or music
expressions, wherever they appear.

@cindex variables

@item
A variable, such as
@example
foo = @{ c4 d e d @}
@end example

This can be used later on in the file by entering @code{\foo}.  The
name of a variable should have alphabetic characters only; no
numbers, underscores or dashes.

@end itemize

The following example shows three things that may be entered at
toplevel

@example
\layout @{
  % Don't justify the output
  ragged-right = ##t
@}

\header @{
   title = "Do-re-mi"
@}

@{ c'4 d' e2 @}
@end example


At any point in a file, any of the following lexical instructions can
be entered:

@itemize
@item @code{\version}
@item @code{\include}
@item @code{\sourcefilename}
@item @code{\sourcefileline}
@item
A single-line comment, introduced by a leading @code{%} sign.

@item
A multi-line comment delimited by @code{%@{ @dots{} %@}}.

@end itemize

@cindex whitespace

Whitespace between items in the input stream is generally ignored,
and may be freely omitted or extended to enhance readability.
However, whitespace should always be used in the following
circumstances to avoid errors:

@itemize

@item Around every opening and closing curly bracket.

@item After every command or variable, i.e. every item that
begins with a @code{\} sign.

@item After every item that is to be interpreted as a Scheme
expression, i.e. every item that begins with a @code{#}@tie{}sign.

@item To separate all elements of a Scheme expression.

@item In @code{lyricmode} before and after @code{\set} and
@code{\override} commands.

@end itemize

@seealso
Learning Manual:
@rlearning{How LilyPond input files work}.

Notation Reference:
@ref{Titles explained},
@ref{The layout block,,The @code{@bs{}layout} block}.


@node Titles and headers
@section Titles and headers

@cindex titles
@cindex headers
@cindex footers

Almost all printed music includes a title and the composer's name;
some pieces include a lot more information.

@menu
* Creating titles headers and footers::
* Custom titles headers and footers::
* Creating PDF metadata::
* Creating footnotes::
* Reference to page numbers::
* Table of contents::
@end menu


@node Creating titles headers and footers
@subsection Creating titles headers and footers

@menu
* Titles explained::
* Default layout of bookpart and score titles::
* Default layout of headers and footers::
@end menu


@node Titles explained
@unnumberedsubsubsec Titles explained

Each @code{\book} block in a single input file produces a separate
output file, see @ref{File structure}.  Within each output file
three types of titling areas are provided: @emph{Book Titles} at the
beginning of each book, @emph{Bookpart Titles} at the beginning of
each bookpart and @emph{Score Titles} at the beginning of each score.

Values of titling fields such as @code{title} and @code{composer}
are set in @code{\header} blocks.  (For the syntax of @code{\header}
blocks and a complete list of the fields available by default see
@ref{Default layout of bookpart and score titles}).  Book Titles,
Bookpart Titles and Score Titles can all contain the same fields,
although by default the fields in Score Titles are limited to
@code{piece} and @code{opus}.

@code{\header} blocks may be placed in four different places to form
a descending hierarchy of @code{\header} blocks:

@itemize

@item
At the top of the input file, before all @code{\book},
@code{\bookpart}, and @code{\score} blocks.

@item
Within a @code{\book} block but outside all the @code{\bookpart} and
@code{\score} blocks within that book.

@item
Within a @code{\bookpart} block but outside all @code{\score} blocks
within that bookpart.

@item
After the music expression in a @code{\score} block.

@end itemize

The values of the fields filter down this hierarchy, with the values
set higher in the hierarchy persisting unless they are over-ridden
by a value set lower in the hierarchy, so:

@itemize

@item
A Book Title is derived from fields set at the top of the input file,
modified by fields set in the @code{\book} block.  The resulting
fields are used to print the Book Title for that book, providing that
there is other material which generates a page at the start of the
book, before the first bookpart.  A single @code{\pageBreak} will
suffice.

@item
A Bookpart Title is derived from fields set at the top of the input
file, modified by fields set in the @code{\book} block, and further
modified by fields set in the @code{\bookpart} block.  The resulting
values are used to print the Bookpart Title for that bookpart.

@item
A Score Title is derived from fields set at the top of the input
file, modified by fields set in the @code{\book} block, further
modified by fields set in the @code{\bookpart} block and finally
modified by fields set in the @code{\score} block.  The resulting
values are used to print the Score Title for that score.  Note,
though, that only @code{piece} and @code{opus} fields are printed
by default in Score Titles unless the @code{\paper} variable,
@code{print-all-headers}, is set to @code{#t}.

@end itemize

@warning{Remember when placing a @bs{}@code{header} block inside a
@bs{}@code{score} block, that the music expression must come before the
@bs{}@code{header} block.}

It is not necessary to provide @code{\header} blocks in all four
places: any or even all of them may be omitted.  Similarly, simple
input files may omit the @code{\book} and @code{\bookpart} blocks,
leaving them to be created implicitly.

If the book has only a single score, the @code{\header} block should
normally be placed at the top of the file so that just a Bookpart
Title is produced, making all the titling fields available for use.

If the book has multiple scores a number of different arrangements
of @code{\header} blocks are possible, corresponding to the various
types of musical publications.  For example, if the publication
contains several pieces by the same composer a @code{\header} block
placed at the top of the file specifying the book title and the
composer with @code{\header} blocks in each @code{\score} block
specifying the @code{piece} and/or @code{opus} would be most
suitable, as here:

@lilypond[papersize=a5,quote,verbatim,noragged-right]
\header {
  title = "SUITE I."
  composer = "J. S. Bach."
}

\score {
  \new Staff \relative {
    \clef bass
    \key g \major
    \repeat unfold 2 { g,16( d' b') a b d, b' d, } |
    \repeat unfold 2 { g,16( e' c') b c e, c' e, } |
  }
  \header {
    piece = "Prélude."
  }
}

\score {
  \new Staff \relative {
    \clef bass
    \key g \major
    \partial 16 b16 |
    <g, d' b'~>4 b'16 a( g fis) g( d e fis) g( a b c) |
    d16( b g fis) g( e d c) b(c d e) fis( g a b) |
  }
  \header {
    piece = "Allemande."
  }
}
@end lilypond

More complicated arrangements are possible.  For example, text
fields from the @code{\header} block in a book can be displayed in
all Score Titles, with some fields over-ridden and some manually
suppressed:

@lilypond[papersize=a5,quote,verbatim,noragged-right]
\book {
  \paper {
    print-all-headers = ##t
  }
  \header {
    title = "DAS WOHLTEMPERIRTE CLAVIER"
    subtitle = "TEIL I"
    % Do not display the default LilyPond footer for this book
    tagline = ##f
  }
  \markup { \vspace #1 }
  \score {
    \new PianoStaff <<
      \new Staff { s1 }
      \new Staff { \clef "bass" s1 }
    >>
    \header {
      title = "PRAELUDIUM I"
      opus = "BWV 846"
      % Do not display the subtitle for this score
      subtitle = ##f
    }
  }
  \score {
    \new PianoStaff <<
      \new Staff { s1 }
      \new Staff { \clef "bass" s1 }
    >>
    \header {
      title = "FUGA I"
      subsubtitle = "A 4 VOCI"
      opus = "BWV 846"
      % Do not display the subtitle for this score
      subtitle = ##f
    }
  }
}
@end lilypond

@seealso
Notation Reference:
@ref{File structure},
@ref{Default layout of bookpart and score titles},
@ref{Custom layout for titles}.


@node Default layout of bookpart and score titles
@unnumberedsubsubsec Default layout of bookpart and score titles

This example demonstrates all printed @code{\header} variables:

@lilypond[papersize=a6landscape,quote,verbatim,noragged-right]
\book {
  \header {
      % The following fields are centered
    dedication = "Dedication"
    title = "Title"
    subtitle = "Subtitle"
    subsubtitle = "Subsubtitle"
      % The following fields are evenly spread on one line
      % the field "instrument" also appears on following pages
    instrument = \markup \with-color #green "Instrument"
    poet = "Poet"
    composer = "Composer"
      % The following fields are placed at opposite ends of the same line
    meter = "Meter"
    arranger = "Arranger"
      % The following fields are centered at the bottom
    tagline = "The tagline goes at the bottom of the last page"
    copyright = "The copyright goes at the bottom of the first page"
  }
  \score {
    { s1 }
    \header {
        % The following fields are placed at opposite ends of the same line
      piece = "Piece 1"
      opus = "Opus 1"
    }
  }
  \score {
    { s1 }
    \header {
        % The following fields are placed at opposite ends of the same line
      piece = "Piece 2 on the same page"
      opus = "Opus 2"
    }
  }
  \pageBreak
  \score {
    { s1 }
    \header {
        % The following fields are placed at opposite ends of the same line
      piece = "Piece 3 on a new page"
      opus = "Opus 3"
    }
  }
}
@end lilypond

Note that

@itemize
@item
The instrument name will be repeated on every page.

@item
Only @code{piece} and @code{opus} are printed in a @code{\score}
when the paper variable @code{print-all-headers} is set to
@code{##f} (the default).

@item
@c Is the bit about \null markups true? -mp
Text fields left unset in a @code{\header} block are replaced with
@code{\null} markups so that the space is not wasted.

@item
The default settings for @code{scoreTitleMarkup} place the @code{piece}
and @code{opus} text fields at opposite ends of the same line.

@end itemize

To change the default layout see @ref{Custom layout for titles}.

@cindex breakbefore

If a @code{\book} block starts immediately with a @code{\bookpart}
block, no Book Title will be printed, as there is no page on which
to print it.  If a Book Title is required, begin the @code{\book}
block with some markup material or a @code{\pageBreak} command.

Use the @code{breakbefore} variable inside a @code{\header} block
that is itself in a @code{\score} block, to make the higher-level
@code{\header} block titles appear on the first page on their own, with
the music (defined in the @code{\score} block) starting on the next.

@lilypond[papersize=c7landscape,verbatim,noragged-right]
\book {
  \header {
    title = "This is my Title"
    subtitle = "This is my Subtitle"
    copyright = "This is the bottom of the first page"
  }
  \score {
    \repeat unfold 4 { e'' e'' e'' e'' }
    \header {
      piece = "This is the Music"
      breakbefore = ##t
    }
  }
}
@end lilypond

@seealso
Learning Manual:
@rlearning{How LilyPond input files work},

Notation Reference:
@ref{Custom layout for titles},
@ref{File structure}.

Installed Files:
@file{ly/titling-init.ly}.


@node Default layout of headers and footers
@unnumberedsubsubsec Default layout of headers and footers

@emph{Headers} and @emph{footers} are lines of text appearing at
the top and bottom of pages, separate from the main text of a book.
They are controlled by the following @code{\paper} variables:

@itemize
@item @code{oddHeaderMarkup}
@item @code{evenHeaderMarkup}
@item @code{oddFooterMarkup}
@item @code{evenFooterMarkup}
@end itemize

These markup variables can only access text fields from top-level
@code{\header} blocks (which apply to all scores in the book) and are
defined in @file{ly/titling-init.ly}.  By default:

@itemize

@item
page numbers are automatically placed on the top far left (if even) or
top far right (if odd), starting from the second page.

@item
the @code{instrument} text field is placed in the center of every
page, starting from the second page.

@item
the @code{copyright} text is centered on the bottom of the first page.

@item
the @code{tagline} is centered on the bottom of the last page, and below
the @code{copyright} text if there is only a single page.

@end itemize

The default LilyPond footer text can be changed by adding a
@code{tagline} in the top-level @code{\header} block.

@lilypond[papersize=a8landscape,verbatim]
\book {
  \header {
    tagline = "... music notation for Everyone"
  }
  \score {
    \relative {
      c'4 d e f
    }
  }
}
@end lilypond

To remove the default LilyPond footer text, the @code{tagline} can be
set to @code{##f}.


@node Custom titles headers and footers
@subsection Custom titles headers and footers

@c TODO: somewhere put a link to header spacing info
@c       (you'll have to explain it more in NR 4).

@menu
* Custom text formatting for titles::
* Custom layout for titles::
* Custom layout for headers and footers::
@end menu


@node Custom text formatting for titles
@unnumberedsubsubsec Custom text formatting for titles

Standard @code{\markup} commands can be used to customize any header,
footer and title text within the @code{\header} block.

@lilypond[quote,verbatim,noragged-right]
\score {
  { s1 }
  \header {
    piece = \markup { \fontsize #4 \bold "PRAELUDIUM I" }
    opus = \markup { \italic "BWV 846" }
  }
}
@end lilypond

@seealso
Notation Reference:
@ref{Formatting text}.


@node Custom layout for titles
@unnumberedsubsubsec Custom layout for titles

@cindex bookTitleMarkup
@cindex scoreTitleMarkup
@funindex bookTitleMarkup
@funindex scoreTitleMarkup

@code{\markup} commands in the @code{\header} block are useful for
simple text formatting, but they do not allow precise control over the
placement of titles.  To customize the placement of the text fields,
change either or both of the following @code{\paper} variables:

@itemize
@item @code{bookTitleMarkup}
@item @code{scoreTitleMarkup}
@end itemize

The placement of titles when using the default values of these
@code{\markup} variables is shown in the examples in
@ref{Default layout of bookpart and score titles}.

The default settings for @code{scoreTitleMarkup} as defined in
@file{ly/titling-init.ly} are:

@example
scoreTitleMarkup = \markup @{ \column @{
  \on-the-fly \print-all-headers @{ \bookTitleMarkup \hspace #1 @}
  \fill-line @{
    \fromproperty #'header:piece
    \fromproperty #'header:opus
  @}
@}
@}
@end example

This places the @code{piece} and @code{opus} text fields at opposite
ends of the same line:

@lilypond[quote,verbatim,noragged-right]
\score {
  { s1 }
  \header {
    piece = "PRAELUDIUM I"
    opus = "BWV 846"
  }
}
@end lilypond

This example redefines @code{scoreTitleMarkup} so that the @code{piece}
text field is centered and in a large, bold font.

@lilypond[papersize=a5,quote,verbatim,noragged-right]
\book {
  \paper {
    indent = 0\mm
    scoreTitleMarkup = \markup {
      \fill-line {
        \null
        \fontsize #4 \bold \fromproperty #'header:piece
        \fromproperty #'header:opus
      }
    }
  }
  \header { tagline = ##f }
  \score {
    { s1 }
    \header {
      piece = "PRAELUDIUM I"
      opus = "BWV 846"
    }
  }
}
@end lilypond

Text fields not normally effective in score @code{\header} blocks
can be printed in the Score Title area if @code{print-all-headers} is
placed inside the @code{\paper} block.  A disadvantage of using this
method is that text fields that are intended specifically for the
Bookpart Title area need to be manually suppressed in every
@code{\score} block.  See @ref{Titles explained}.

To avoid this, add the desired text field to the @code{scoreTitleMarkup}
definition.  In the following example, the @code{composer} text field
(normally associated with @code{bookTitleMarkup}) is added to
@code{scoreTitleMarkup}, allowing each score to list a different
composer:

@lilypond[papersize=a5,quote,verbatim,noragged-right]
\book {
  \paper {
    indent = 0\mm
    scoreTitleMarkup = \markup {
      \fill-line {
        \null
        \fontsize #4 \bold \fromproperty #'header:piece
        \fromproperty #'header:composer
      }
    }
  }
  \header { tagline = ##f }
  \score {
    { s1 }
    \header {
      piece = "MENUET"
      composer = "Christian Petzold"
    }
  }
  \score {
    { s1 }
    \header {
      piece = "RONDEAU"
      composer = "François Couperin"
    }
  }
}
@end lilypond

It is also possible to create your own custom text fields, and refer to
them in the markup definition.

@lilypond[papersize=a5,quote,verbatim,noragged-right]
\book {
  \paper {
    indent = 0\mm
    scoreTitleMarkup = \markup {
      \fill-line {
        \null
        \override #`(direction . ,UP) {
          \dir-column {
            \center-align \fontsize #-1 \bold
              \fromproperty #'header:mycustomtext %% User-defined field
            \center-align \fontsize #4 \bold
              \fromproperty #'header:piece
          }
        }
        \fromproperty #'header:opus
      }
    }
  }
  \header { tagline = ##f }
  \score {
    { s1 }
    \header {
      piece = "FUGA I"
      mycustomtext = "A 4 VOCI" %% User-defined field
      opus = "BWV 846"
    }
  }
}
@end lilypond

@seealso
Notation Reference:
@ref{Titles explained}.


@node Custom layout for headers and footers
@unnumberedsubsubsec Custom layout for headers and footers

@c can make-header and make-footer be removed from
@c paper-defaults-init.ly? -mp

@code{\markup} commands in the @code{\header} block are useful for
simple text formatting, but they do not allow precise control over the
placement of headers and footers.  To customize the placement of
the text fields, use either or both of the following @code{\paper}
variables:

@itemize
@item @code{oddHeaderMarkup}
@item @code{evenHeaderMarkup}
@item @code{oddFooterMarkup}
@item @code{evenFooterMarkup}
@end itemize

@cindex markup, conditional
@cindex on-the-fly
@funindex \on-the-fly

The @code{\markup} command @code{\on-the-fly} can be used to add
markup conditionally to header and footer text defined within the
@code{\paper} block, using the following syntax:

@example
variable = \markup @{
  @dots{}
  \on-the-fly  \@var{procedure}  @var{markup}
  @dots{}
@}
@end example

The @var{procedure} is called each time the @code{\markup} command
in which it appears is evaluated.  The @var{procedure} should test
for a particular condition and interpret (i.e. print) the
@var{markup} argument if and only if the condition is true.

A number of ready-made procedures for testing various conditions are
provided:

@quotation
@multitable {print-page-number-check-first-----} {should this page be printed-----}

@headitem  Procedure name           @tab  Condition tested

@item print-page-number-check-first @tab  should this page number be printed?
@item create-page-number-stencil    @tab  print-page-numbers true?
@item print-all-headers             @tab  print-all-headers true?
@item first-page                    @tab  first page in the book?
@item (on-page nmbr)                @tab  page number = nmbr?
@item last-page                     @tab  last page in the book?
@item not-first-page                @tab  not first page in the book?
@item part-first-page               @tab  first page in the book part?
@item part-last-page                @tab  last page in the book part?
@item not-single-page               @tab  pages in book part > 1?

@end multitable
@end quotation

The following example centers page numbers at the bottom of every
page.  First, the default settings for @code{oddHeaderMarkup} and
@code{evenHeaderMarkup} are removed by defining each as a @emph{null}
markup.  Then, @code{oddFooterMarkup} is redefined with the page
number centered.  Finally, @code{evenFooterMarkup} is given the
same layout by defining it as @code{\oddFooterMarkup}:

@lilypond[papersize=a8,quote,verbatim,noragged-right]
\book {
  \paper {
    print-page-number = ##t
    print-first-page-number = ##t
    oddHeaderMarkup = \markup \null
    evenHeaderMarkup = \markup \null
    oddFooterMarkup = \markup {
      \fill-line {
        \on-the-fly \print-page-number-check-first
        \fromproperty #'page:page-number-string
      }
    }
    evenFooterMarkup = \oddFooterMarkup
  }
  \score {
    \new Staff { s1 \break s1 \break s1 }
  }
}
@end lilypond

Several @code{\on-the-fly} conditions can be combined with an
@q{and} operation, for example,

@example
  \on-the-fly \first-page
  \on-the-fly \last-page
  @code{@{ \markup @dots{} \fromproperty #'header: @dots{} @}}
@end example

determines if the output is a single page.

@seealso
Notation Reference:
@ref{Titles explained},
@ref{Default layout of bookpart and score titles}.

Installed Files:
@file{../ly/titling-init.ly}.

@node Creating PDF metadata
@subsection Creating PDF metadata

@cindex PDF metadata

In addition to being shown in the printed output, @code{\header} variables
are also used to set PDF metadata (the information displayed by PDF readers
as the @code{properties} of the PDF file).  For example, setting the
@code{title} property of the @code{header} block @q{Symphony I} will also give
this title to the PDF document.

@example
  \header@{
    title = "Symphony I"
  @}
@end example

If you want to set the title of the printed output to one value, but have the
title property of the PDF to have a different value, you can use
@code{pdftitle}, as below.

@example
  \header@{
    title = "Symphony I"
    pdftitle = "Symphony I by Beethoven"
  @}
@end example

The variables @code{title}, @code{subject}, @code{keywords},
@code{subtitle}, @code{composer}, @code{arranger}, @code{poet}, @code{author}
and @code{copyright} all set PDF properties and can all be prefixed with
@q{pdf} to set a PDF property to a value different from the printed output.

The PDF property @code{Creator} is automatically set to @q{LilyPond} plus
the current LilyPond version, and @code{CreationDate} and @code{ModDate} are
both set to the current date and time.  @code{ModDate} can be overridden by
setting the header variable @code{moddate} (or @code{pdfmoddate}) to a
valid PDF date string.

@node Creating footnotes
@subsection Creating footnotes

@cindex footnotes

Footnotes may be used in many different situations.  In all cases,
a @q{footnote mark} is placed as a reference in text or music, and
the corresponding @q{footnote text} appears at the bottom of the
same page.

Footnotes within music expressions and footnotes in stand-alone text
outside music expressions are created in different ways.

@menu
* Footnotes in music expressions::
* Footnotes in stand-alone text::
@end menu

@node Footnotes in music expressions
@unnumberedsubsubsec Footnotes in music expressions

@cindex footnotes in music expressions
@funindex \footnote

@subsubsubheading Music footnotes overview

Footnotes in music expressions fall into two categories:

@table @emph
@item Event-based footnotes
are attached to a particular event.  Examples for such events are
single notes, articulations (like fingering indications, accents,
dynamics), and post-events (like slurs and manual beams).  The
general form for event-based footnotes is as follows:

@example
[@var{direction}] \footnote [@var{mark}] @var{offset} @var{footnote} @var{music}
@end example

@item Time-based footnotes
are bound to a particular point of time in a musical context.  Some
commands like @code{\time} and @code{\clef} don't actually use events
for creating objects like time signatures and clefs.  Neither does a
chord create an event of its own: its stem or flag is created at the
end of a time step (nominally through one of the note events inside).
Exactly which of a chord's multiple note events will be deemed the
root cause of a stem or flag is undefined.  So for annotating those,
time-based footnotes are preferable as well.

A time-based footnote allows such layout objects to be annotated
without referring to an event.  The general form for Time-based
footnotes is:

@example
\footnote [@var{mark}] @var{offset} @var{footnote} [@var{Context}].@var{GrobName}
@end example

@end table

The elements for both forms are:

@table @var

@item direction
If (and only if) the @code{\footnote} is being applied to a
post-event or articulation, it must be preceded with a direction
indicator (@code{-, _, ^}) in order to attach @var{music} (with
a footnote mark) to the preceding note or rest.

@item mark
is a markup or string specifying the footnote mark which is used for
marking both the reference point and the footnote itself at the
bottom of the page.  It may be omitted (or equivalently replaced with
@code{\default}) in which case a number in sequence will be generated
automatically.  Such numerical sequences restart on each page
containing a footnote.

@item offset
is a number pair such as @samp{#(2 . 1)} specifying the X and
Y@tie{}offsets in units of staff-spaces from the boundary of the
object where the mark should be placed.  Positive values of the
offsets are taken from the right/top edge, negative values from the
left/bottom edge and zero implies the mark is centered on the edge.

@item Context
is the context in which the grob being footnoted is created.  It
may be omitted if the grob is in a bottom context, e.g. a
@code{Voice} context.

@item GrobName
specifies a type of grob to mark (like @samp{Flag}).  If it is
specified, the footnote is not attached to a music expression in
particular, but rather to all grobs of the type specified which
occur at that moment of musical time.

@item footnote
is the markup or string specifying the footnote text to use at the
bottom of the page.

@item music
is the music event or post-event or articulation
that is being annotated.

@end table

@subsubsubheading Event-based footnotes

@cindex footnotes, event-based

A footnote may be attached to a layout object directly caused
by the event corresponding to @var{music} with the syntax:

@example
\footnote [@var{mark}] @var{offset} @var{footnote} @var{music}
@end example

@lilypond[quote,verbatim,papersize=a8landscape]
\book {
  \header { tagline = ##f }
  \relative c'' {
    \footnote #'(-1 . 3) "A note" a4
    a4
    \footnote #'(2 . 2) "A rest" r4
    a4
  }
}
@end lilypond

Marking a @emph{whole} chord with an event-based footnote is not
possible: a chord, even one containing just a single note, does
not produce an actual event of its own.  However, individual
notes @emph{inside} of the chord can be marked:

@lilypond[quote,verbatim,papersize=a8landscape]
\book {
  \header { tagline = ##f }
  \relative c'' {
    \footnote #'(2 . 3) "Does not work" <a-3>2
    <\footnote #'(-2 . -3) "Does work" a-3>4
    <a-3 \footnote #'(3 . 1/2) "Also works" c-5>4
  }
}
@end lilypond

If the footnote is to be attached to a post-event or articulation
the @code{\footnote} command @emph{must} be preceded by a direction
indicator, @code{-, _, ^}, and followed by the post-event or
articulation to be annotated as the @var{music} argument.  In this
form the @code{\footnote} can be considered to be simply a copy of
its last argument with a footnote mark attached to it.  The syntax
is:

@example
@var{direction} \footnote [@var{mark}] @var{offset} @var{footnote} @var{music}
@end example

@lilypond[quote,verbatim,papersize=a8landscape]
\book {
  \header { tagline = ##f }
  \relative {
    a'4_\footnote #'(0 . -1) "A slur forced down" (
    b8^\footnote #'(1 . 0.5) "A manual beam forced up" [
    b8 ]
    c4 )
    c-\footnote #'(1 . 1) "Tenuto" --
  }
}
@end lilypond

@subsubsubheading Time-based footnotes

@cindex footnotes, time-based

If the layout object being footmarked is @emph{indirectly} caused by
an event (like an @code{Accidental} or @code{Stem} caused by a
@code{NoteHead} event), the @var{GrobName} of the layout object
is required after the footnote text instead of @var{music}:

@lilypond[quote,verbatim,papersize=a8landscape]
\book {
  \header { tagline = ##f }
  \relative c'' {
    \footnote #'(-1 . -3) "A flat" Accidental
    aes4 c
    \footnote #'(-1 . 0.5) "Another flat" Accidental
    ees
    \footnote #'(1 . -2) "A stem" Stem
    aes
  }
}
@end lilypond

Note, however, that when a GrobName is specified, a footnote
will be attached to all grobs of that type at the current time step:

@lilypond[quote,verbatim,papersize=a8landscape]
\book {
  \header { tagline = ##f }
  \relative c' {
    \footnote #'(-1 . 3) "A flat" Accidental
    <ees ges bes>4
    \footnote #'(2 . 0.5) "Articulation" Script
    c'->-.
  }
}
@end lilypond

A note inside of a chord can be given an individual (event-based)
footnote.  A @samp{NoteHead} is the only grob directly caused
from a chord note, so an event-based footnote command is
@emph{only} suitable for adding a footnote to the @samp{NoteHead}
within a chord.  All other chord note grobs are indirectly caused.
The @code{\footnote} command itself offers no syntax for
specifying @emph{both} a particular grob type @emph{as well as} a
particular event to attach to.  However, one can use a time-based
@code{\footnote} command for specifying the grob type, and then
prefix this command with @code{\single} in order to have it
applied to just the following event:

@lilypond[quote,verbatim,papersize=a8landscape]
\book {
  \header { tagline = ##f }
  \relative c'' {
    < \footnote #'(1 . -2) "An A" a
      \single \footnote #'(-1 . -1) "A sharp" Accidental
      cis
      \single \footnote #'(0.5 . 0.5) "A flat" Accidental
      ees fis
    >2
  }
}
@end lilypond

@warning {When footnotes are attached to several musical elements at
the same musical moment, as they are in the example above, the
footnotes are numbered from the higher to the lower elements as they
appear in the printed output, not in the order in which they are
written in the input stream.}

Layout objects like clefs and key-change signatures are mostly caused
as a consequence of changed properties rather than actual events.
Others, like bar lines and bar numbers, are a direct consequence of
timing.  For this reason, footnotes on such objects have to be based
on their musical timing.  Time-based footnotes are also preferable
when marking features like stems and beams on @emph{chords}: while
such per-chord features are nominally assigned to @emph{one} event
inside the chord, relying on a particular choice would be imprudent.

The layout object in question must always be explicitly specified
for time-based footnotes, and the appropriate context must be
specified if the grob is created in a context other than the bottom
context.

@lilypond[quote,verbatim,papersize=a8landscape]
\book {
  \header { tagline = ##f }
  \relative c'' {
    r1 |
    \footnote #'(-0.5 . -1) "Meter change" Staff.TimeSignature
    \time 3/4
    \footnote #'(1 . -1) "Chord stem" Stem
    <c e g>4 q q
    \footnote #'(-0.5 . 1) "Bar line" Staff.BarLine
    q q
    \footnote #'(0.5 . -1) "Key change" Staff.KeySignature
    \key c \minor
    q
  }
}
@end lilypond

Custom marks can be used as alternatives to numerical marks, and the
annotation line joining the marked object to the mark can be
suppressed:

@lilypond[quote,verbatim,papersize=a8landscape]
\book {
  \header { tagline = ##f }
  \relative c' {
    \footnote "*" #'(0.5 . -2) \markup { \italic "* The first note" } a'4
    b8
    \footnote \markup { \super "$" } #'(0.5 . 1)
      \markup { \super "$" \italic " The second note" } e
    c4
    \once \override Score.FootnoteItem.annotation-line = ##f
    b-\footnote \markup \tiny "+" #'(0.1 . 0.1)
      \markup { \super "+" \italic " Editorial" } \p
  }
}
@end lilypond

More examples of custom marks are shown in
@ref{Footnotes in stand-alone text}.


@node Footnotes in stand-alone text
@unnumberedsubsubsec Footnotes in stand-alone text

@cindex footnotes in stand-alone text

These are for use in markup outside of music expressions.  They do
not have a line drawn to their point of reference: their marks simply
follow the referenced markup.  Marks can be inserted automatically,
in which case they are numerical.  Alternatively, custom marks can be
provided manually.

Footnotes to stand-alone text with automatic and custom marks are
created in different ways.

@subsubsubheading Footnotes in stand-alone text with automatic marks

The syntax of a footnote in stand-alone text with automatic marks is

@example
\markup @{ @dots{} \auto-footnote @var{text} @var{footnote} @dots{} @}
@end example

The elements are:

@table @var

@item text
is the markup or string to be marked.

@item footnote
is the markup or string specifying the footnote text to use at the bottom
of the page.

@end table

For example:

@lilypond[verbatim,quote,ragged-right,papersize=a8]
\book {
  \header { tagline = ##f }
  \markup {
    "A simple"
    \auto-footnote "tune" \italic " By me"
    "is shown below.  It is a"
    \auto-footnote "recent" \italic " Aug 2012"
    "composition."
  }
  \relative {
    a'4 b8 e c4 d
  }
}
@end lilypond

@subsubsubheading Footnotes in stand-alone text with custom marks

The syntax of a footnote in stand-alone text with custom marks is

@example
\markup @{ @dots{} \footnote @var{mark} @var{footnote} @dots{} @}
@end example

The elements are:

@table @var

@item mark
is a markup or string specifying the footnote mark which is used for
marking the reference point.  Note that this mark is @emph{not}
inserted automatically before the footnote itself.

@item footnote
is the markup or string specifying the footnote text to use at the
bottom of the page, preceded by the @var{mark}.

@end table

Any easy-to-type character such as * or + may be used as a mark, as
shown in @ref{Footnotes in music expressions}.  Alteratively, ASCII
aliases may be used (see @ref{ASCII aliases}):

@lilypond[verbatim,quote,ragged-right,papersize=a8]
\book {
  \paper { #(include-special-characters) }
  \header { tagline = ##f }
  \markup {
    "A simple tune"
    \footnote "*" \italic "* By me"
    "is shown below.  It is a recent"
    \footnote \super &dagger; \concat {
      \super &dagger; \italic " Aug 2012"
    }
    "composition."
  }
  \relative {
    a'4 b8 e c4 d
  }
}
@end lilypond

Unicode character codes may also be used to specify marks
(see @ref{Unicode}):

@lilypond[verbatim,quote,ragged-right,papersize=a8]
\book {
  \header { tagline = ##f }
  \markup {
    "A simple tune"
    \footnote \super \char##x00a7 \concat {
      \super \char##x00a7 \italic " By me"
    }
    "is shown below.  It is a recent"
    \footnote \super \char##x00b6 \concat {
      \super \char##x00b6 \italic " Aug 2012"
    }
    "composition."
  }
  \relative {
    a'4 b8 e c4 d
  }
}
@end lilypond

@seealso
Learning Manual:
@rlearning{Objects and interfaces}.

Notation Reference:
@ref{ASCII aliases},
@ref{Balloon help},
@ref{List of special characters},
@ref{Text marks},
@ref{Text scripts},
@ref{Unicode}.

Internals Reference:
@rinternals{FootnoteEvent},
@rinternals{FootnoteItem},
@rinternals{FootnoteSpanner},
@rinternals{Footnote_engraver}.

@knownissues
Multiple footnotes for the same page can only be stacked, one above
the other; they cannot be printed on the same line.

Footnotes cannot be attached to @code{MultiMeasureRests} or
automatic beams or lyrics.

Footnote marks may collide with staves, @code{\markup} objects, other
footnote marks and annotation lines.


@node Reference to page numbers
@subsection Reference to page numbers

A particular place of a score can be marked using the @code{\label}
command, either at top-level or inside music.  This label can then be
referred to in a markup, to get the number of the page where the marked
point is placed, using the @code{\page-ref} markup command.

@lilypond[verbatim,papersize=a8landscape]
\header { tagline = ##f }
\book {
  \label #'firstScore
  \score {
    {
      c'1
      \pageBreak \mark A \label #'markA
      c'1
    }
  }
  \markup { The first score begins on page \page-ref #'firstScore "0" "?" }
  \markup { Mark A is on page \page-ref #'markA "0" "?" }
}
@end lilypond

The @code{\page-ref} markup command takes three arguments:
@enumerate
@item the label, a scheme symbol, eg. @code{#'firstScore};
@item a markup that will be used as a gauge to estimate the dimensions
of the markup;
@item a markup that will be used in place of the page number if the label
is not known;
@end enumerate

The reason why a gauge is needed is that, at the time markups are
interpreted, the page breaking has not yet occurred, so the page numbers
are not yet known.  To work around this issue, the actual markup
interpretation is delayed to a later time; however, the dimensions of
the markup have to be known before, so a gauge is used to decide these
dimensions.  If the book has between 10 and 99 pages, it may be "00",
ie. a two digit number.


@predefined
@funindex \label
@code{\label},
@funindex \page-ref
@code{\page-ref}.
@endpredefined


@node Table of contents
@subsection Table of contents
A table of contents is included using the
@code{\markuplist \table-of-contents} command.  The elements which
should appear in the table of contents are entered with the
@code{\tocItem} command, which may be used either at top-level, or
inside a music expression.

@verbatim
\markuplist \table-of-contents
\pageBreak

\tocItem \markup "First score"
\score {
  {
    c'4  % ...
    \tocItem \markup "Some particular point in the first score"
    d'4  % ...
  }
}

\tocItem \markup "Second score"
\score {
  {
    e'4 % ...
  }
}
@end verbatim

Markups used for formatting the table of contents are defined in the
@code{\paper} block.  There are two @q{pre-defined} markups already
available;

@itemize

@item
@code{tocTitleMarkup}

@noindent
Used for formatting the title of the table of contents.

@verbatim
tocTitleMarkup = \markup \huge \column {
  \fill-line { \null "Table of Contents" \null }
  \null
}
@end verbatim

@item
@code{tocItemMarkup}

@noindent
Used for formatting the elements within the table of contents.

@verbatim
tocItemMarkup = \markup \fill-line {
  \fromproperty #'toc:text \fromproperty #'toc:page
}
@end verbatim

@end itemize

@noindent
Both of these variables can be changed.

Here is an example changing the table of contents' title into French;

@verbatim
\paper {
  tocTitleMarkup = \markup \huge \column {
    \fill-line { \null "Table des matières" \null }
    \hspace #1
  }
@end verbatim

Here is an example changing the font-size of the elements in the table
of contents;

@verbatim
tocItemMarkup = \markup \large \fill-line {
  \fromproperty #'toc:text \fromproperty #'toc:page
}
@end verbatim

Note how the element text and page numbers are referred to in the
@code{tocItemMarkup} definition.

The @code{\tocItemWithDotsMarkup} command can be included within the
@code{tocItemMarkup} to fill the line, between a table of contents item
and its corresponding page number, with dots;

@lilypond[verbatim,line-width=10.0\cm]
\header { tagline = ##f }
\paper {
  tocItemMarkup = \tocItemWithDotsMarkup
}

\book {
  \markuplist \table-of-contents
  \tocItem \markup { Allegro }
  \tocItem \markup { Largo }
  \markup \null
}
@end lilypond

Custom commands with their own markups can also be defined to build a
more complex table of contents.  In the following example, a new style
is defined for entering act names in a table of contents of an opera;

@noindent
A new markup variable (called @code{tocActMarkup}) is defined in the
@code{\paper} block;

@verbatim
\paper {
  tocActMarkup = \markup \large \column {
    \hspace #1
    \fill-line { \null \italic \fromproperty #'toc:text \null }
    \hspace #1
  }
}
@end verbatim

@noindent
A custom music function (@code{tocAct}) is then created -- which uses
the new @code{tocActMarkup} markup definition.

@verbatim
tocAct =
  #(define-music-function (text) (markup?)
     (add-toc-item! 'tocActMarkup text))
@end verbatim

@noindent
A LilyPond input file, using these customer definitions, could look
something like this;

@lilypond[line-width=10.0\cm]
\header { tagline = ##f }
\paper {
  tocActMarkup = \markup \large \column {
    \hspace #1
    \fill-line { \null \italic \fromproperty #'toc:text \null }
    \hspace #1
  }
}

tocAct =
#(define-music-function (text) (markup?)
   (add-toc-item! 'tocActMarkup text))

\book {
  \markuplist \table-of-contents
  \tocAct \markup { Atto Primo }
  \tocItem \markup { Coro. Viva il nostro Alcide }
  \tocItem \markup { Cesare. Presti omai l'Egizia terra }
  \tocAct \markup { Atto Secondo }
  \tocItem \markup { Sinfonia }
  \tocItem \markup { Cleopatra. V'adoro, pupille, saette d'Amore }
  \markup \null
}
@end lilypond


Here is an example of the @code{\fill-with-pattern} command used within
the context of a table of contents;

@verbatim
\paper {
  tocItemMarkup = \markup { \fill-line {
    \override #'(line-width . 70)
    \fill-with-pattern #1.5 #CENTER . \fromproperty #'toc:text \fromproperty #'toc:page
    }
  }
}
@end verbatim

@seealso
Installed Files:
@file{ly/toc-init.ly}.

@predefined
@funindex \table-of-contents
@code{\table-of-contents},
@funindex \tocItem
@code{\tocItem}.
@endpredefined


@node Working with input files
@section Working with input files

@menu
* Including LilyPond files::
* Different editions from one source::
* Special characters::
@end menu


@node Including LilyPond files
@subsection Including LilyPond files

@funindex \include
@cindex including files

A large project may be split up into separate files.  To refer to
another file, use

@example
\include "otherfile.ly"
@end example

The line @code{\include "otherfile.ly"} is equivalent to pasting the
contents of @file{otherfile.ly} into the current file at the place
where the @code{\include} appears.  For example, in a large
project you might write separate files for each instrument part
and create a @qq{full score} file which brings together the
individual instrument files.  Normally the included file will
define a number of variables which then become available
for use in the full score file.  Tagged sections can be
marked in included files to assist in making them usable in
different places in a score, see @ref{Different editions from
one source}.

Files in the current working directory may be referenced by
specifying just the file name after the @code{\include} command.
Files in other locations may be included by giving either a full
path reference or a relative path reference (but use the UNIX
forward slash, /, rather than the DOS/Windows back slash, \, as the
directory separator.)  For example, if @file{stuff.ly} is located
one directory higher than the current working directory, use

@example
\include "../stuff.ly"
@end example

@noindent
or if the included orchestral parts files are all located in a
subdirectory called @file{parts} within the current directory, use

@example
\include "parts/VI.ly"
\include "parts/VII.ly"
@dots{} etc
@end example

Files which are to be included can also contain @code{\include}
statements of their own.  By default, these second-level
@code{\include} statements are not interpreted until they have
been brought into the main file, so the file names they specify
must all be relative to the directory containing the main file,
not the directory containing the included file.  However,
this behavior can be changed globally by passing the option
@option{-drelative-includes} option at the command line
(or by adding @code{#(ly:set-option 'relative-includes #t)}
at the top of the main input file).

When @code{relative-includes} is set to @code{#t}, the path for each
@code{\include} command will be taken relative to the file containing
that command.  This behavior is recommended and it will become the
default behavior in a future version of lilypond.

Files relative to the main directory and files relative to some other
directory may both be @code{\include}d by setting
@code{relative-includes} to @code{#t} or @code{#f} at appropriate
places in the files.  For example, if a general library, libA, has
been created which itself uses sub-files which are @code{\include}d
by the entry file of that library, those @code{\include} statements
will need to be preceded by
@code{#(ly:set-option #relative-includes #t)} so they are interpreted
correctly when brought into the main @code{.ly} file, like this:

@example
libA/
  libA.ly
  A1.ly
  A2.ly
  @dots{}
@end example

@noindent
then the entry file, @code{libA.ly}, will contain

@example
#(ly:set-option 'relative-includes #t)
\include "A1.ly"
\include "A2.ly"
@dots{}
% return to default setting
#(ly:set-option 'relative-includes #f)
@end example

Any @file{.ly} file can then include the entire library simply with

@example
\include "~/libA/libA.ly"
@end example

More complex file structures may be devised by switching at
appropriate places.

Files can also be included from a directory in a search path
specified as an option when invoking LilyPond from the command
line.  The included files are then specified using just their
file name.  For example, to compile @file{main.ly} which includes
files located in a subdirectory called @file{parts} by this method,
cd to the directory containing @file{main.ly} and enter

@example
lilypond --include=parts main.ly
@end example

and in main.ly write

@example
\include "VI.ly"
\include "VII.ly"
@dots{} etc
@end example

Files which are to be included in many scores may be placed in
the LilyPond directory @file{../ly}.  (The location of this
directory is installation-dependent - see
@rlearning{Other sources of information}).  These files can then
be included simply by naming them on an @code{\include} statement.
This is how the language-dependent files like @file{english.ly} are
included.

LilyPond includes a number of files by default when you start
the program.  These includes are not apparent to the user, but the
files may be identified by running @code{lilypond --verbose} from
the command line.  This will display a list of paths and files that
LilyPond uses, along with much other information.  Alternatively,
the more important of these files are discussed in
@rlearning{Other sources of information}.  These files may be
edited, but changes to them will be lost on installing a new
version of LilyPond.

Some simple examples of using @code{\include} are shown in
@rlearning{Scores and parts}.

@seealso
Learning Manual:
@rlearning{Other sources of information},
@rlearning{Scores and parts}.

@knownissues
If an included file is given a name which is the same as one in
LilyPond's installation files, LilyPond's file from the
installation files takes precedence.


@node Different editions from one source
@subsection Different editions from one source

Several methods can be used to generate different versions of a score
from the same music source.  Variables are perhaps the most useful for
combining lengthy sections of music and/or annotation.  Tags are more
useful for selecting one section from several alternative shorter
sections of music, and can also be used for splicing pieces of music
together at different points.

Whichever method is used, separating the notation from the structure of
the score will make it easier to change the structure while leaving the
notation untouched.

@menu
* Using variables::
* Using tags::
* Using global settings::
@end menu

@node Using variables
@unnumberedsubsubsec Using variables

@cindex variables, use of

If sections of the music are defined in variables they can be
reused in different parts of the score, see @rlearning{Organizing
pieces with variables}.  For example, an @notation{a cappella}
vocal score frequently includes a piano reduction of the parts
for rehearsal purposes which is identical to the vocal music, so
the music need be entered only once.  Music from two variables
may be combined on one staff, see @ref{Automatic part combining}.
Here is an example:

@lilypond[verbatim,quote]
sopranoMusic = \relative { a'4 b c b8( a) }
altoMusic = \relative { e'4 e e f }
tenorMusic = \relative { c'4 b e d8( c) }
bassMusic = \relative { a4 gis a d, }
allLyrics = \lyricmode { King of glo -- ry }
<<
  \new Staff = "Soprano" \sopranoMusic
  \new Lyrics \allLyrics
  \new Staff = "Alto" \altoMusic
  \new Lyrics \allLyrics
  \new Staff = "Tenor" {
    \clef "treble_8"
    \tenorMusic
  }
  \new Lyrics \allLyrics
  \new Staff = "Bass" {
    \clef "bass"
    \bassMusic
  }
  \new Lyrics \allLyrics
  \new PianoStaff <<
    \new Staff = "RH" {
      \partcombine \sopranoMusic \altoMusic
    }
    \new Staff = "LH" {
      \clef "bass"
      \partcombine \tenorMusic \bassMusic
    }
  >>
>>
@end lilypond

Separate scores showing just the vocal parts or just the piano
part can be produced by changing just the structural statements,
leaving the musical notation unchanged.

For lengthy scores, the variable definitions may be placed in
separate files which are then included, see @ref{Including
LilyPond files}.

@node Using tags
@unnumberedsubsubsec Using tags

@funindex \tag
@funindex \keepWithTag
@funindex \removeWithTag
@cindex tag
@cindex keep tagged music
@cindex remove tagged music

The @code{\tag #'@var{partA}} command marks a music expression
with the name @var{partA}.
Expressions tagged in this way can be selected or filtered out by
name later, using either @code{\keepWithTag #'@var{name}} or
@code{\removeWithTag #'@var{name}}.  The result of applying these filters
to tagged music is as follows:
@multitable @columnfractions .5 .5
@headitem Filter
  @tab Result
@item
Tagged music preceded by @code{\keepWithTag #'@var{name}} or
       @code{\keepWithTag #'(@var{name1} @var{name2}@dots{})}
@tab Untagged music and music tagged with any of the given tag
     names is included;
     music tagged with any other tag name is excluded.
@item
Tagged music preceded by @code{\removeWithTag #'@var{name}} or
       @code{\removeWithTag #'(@var{name1} @var{name2}@dots{})}
@tab Untagged music and music not tagged with any of the given tag names
     is included; music tagged with any of the given tag names is
     excluded.
@item
Tagged music not preceded by either @code{\keepWithTag} or
@code{\removeWithTag}
@tab All tagged and untagged music is included.
@end multitable

The arguments of the @code{\tag}, @code{\keepWithTag} and
@code{\removeWithTag} commands should be a symbol or list of
symbols (such as @code{#'score} or @code{#'(violinI violinII}),
followed by a music expression.  If @emph{and only if} the symbols
are valid LilyPond identifiers (alphabetic characters only, no
numbers, underscores, or dashes) which cannot be confused with notes,
the @code{#'} may be omitted and, as a shorthand, a list of symbols
can use the dot separator: i.e. @code{\tag #'(violinI violinII)} can
be written @code{\tag violinI.violinII}.  The same applies to
@code{\keepWithTag} and @code{\removeWithTag}.

In the following example, we see two versions of a piece of music,
one showing trills with the usual notation, and one with trills
explicitly expanded:

@lilypond[verbatim,quote]
music = \relative {
  g'8. c32 d
  \tag #'trills { d8.\trill }
  \tag #'expand { \repeat unfold 3 { e32 d } }
  c32 d
 }

\score {
  \keepWithTag #'trills \music
}
\score {
  \keepWithTag #'expand \music
}
@end lilypond

@noindent
Alternatively, it is sometimes easier to exclude sections of music:

@lilypond[verbatim,quote]
music = \relative {
  g'8. c32 d
  \tag #'trills { d8.\trill }
  \tag #'expand {\repeat unfold 3 { e32 d } }
  c32 d
 }

\score {
  \removeWithTag #'expand
  \music
}
\score {
  \removeWithTag #'trills
  \music
}
@end lilypond

Tagged filtering can be applied to articulations, texts, etc. by
prepending

@example
-\tag #'@var{your-tag}
@end example

to an articulation.  For example, this would define a note with a
conditional fingering indication and a note with a conditional
annotation:

@example
c1-\tag #'finger ^4
c1-\tag #'warn ^"Watch!"
@end example

Multiple tags may be placed on expressions with multiple
@code{\tag} entries, or by combining multiple tags into one symbol
list:

@lilypond[quote,verbatim]
music = \relative c'' {
  \tag #'a \tag #'both { a4 a a a }
  \tag #'(b both) { b4 b b b }
}
<<
\keepWithTag #'a \music
\keepWithTag #'b \music
\keepWithTag #'both \music
>>
@end lilypond

Multiple @code{\removeWithTag} filters may be applied to a single
music expression to remove several differently named tagged
sections.  Alternatively, you can use a single @code{\removeWithTag}
with a list of tags.

@lilypond[verbatim,quote]
music = \relative c'' {
  \tag #'A { a4 a a a }
  \tag #'B { b4 b b b }
  \tag #'C { c4 c c c }
  \tag #'D { d4 d d d }
}
\new Voice {
  \removeWithTag #'B
  \removeWithTag #'C
  \music
  \removeWithTag #'(B C)
  \music
}
@end lilypond

Using two or more @code{\keepWithTag} filters on a single music
expression will cause @emph{all} of the tagged sections to be removed.
The first filter will remove all except the one named and any subsequent
filters will remove the rest.  Using one @code{\keepWithTag} command
with a list of multiple tags will only remove tagged sections that are
not specified in that list.

@lilypond[verbatim,quote]
music = \relative c'' {
  \tag #'violinI { a4 a a a }
  \tag #'violinII { b4 b b b }
  \tag #'viola { c4 c c c }
  \tag #'cello { d4 d d d }
}

\new Staff {
  \keepWithTag #'(violinI violinII)
  \music
}
@end lilypond

@noindent
will print @code{\tag}s @var{violinI} and @var{violinII} but not
@var{viola} or @var{cello}.

@cindex tag groups
@funindex \tagGroup

While @code{\keepWithTag} is convenient when dealing with @emph{one} set
of alternatives, the removal of music tagged with @emph{unrelated} tags
is problematic when using them for more than one purpose.  In that case
@q{groups} of tags can be declared:

@example
\tagGroup #'(violinI violinII viola cello)
@end example

@noindent
Now all the different tags belong to a single @q{tag group}.  Note that
individual tags cannot be members of more than one @emph{tag group}.

@example
\keepWithTag #'violinI @dots{}
@end example

@noindent
will now only show music tagged from @code{violinI}'s tag group and any
music tagged with one of the @emph{other} tags will removed.

@lilypond[verbatim,quote]
music = \relative {
  \tagGroup #'(violinI violinII viola cello)
  \tag #'violinI { c''4^"violinI" c c c }
  \tag #'violinII { a2 a }
  \tag #'viola { e8 e e2. }
  \tag #'cello { d'2 d4 d }
  R1^"untagged"
}

\new Voice {
  \keepWithTag #'violinI
  \music
}
@end lilypond

When using the @code{\keepWithTag} command, only tags from the tag
groups of the tags given in the command are visible.

@funindex \pushToTag
@funindex \appendToTag
@cindex splice into tagged music

Sometimes you want to splice some music at a particular place in an
existing music expression.  You can use @code{\pushToTag} and
@code{\appendToTag} for adding material at the front or end of the
@code{elements} of an existing music construct.  Not every music
construct has @code{elements}, but sequential and simultaneous music are
safe bets:

@lilypond[verbatim,quote]
music = { \tag #'here { \tag #'here <<c''>> } }

{
  \pushToTag #'here c'
  \pushToTag #'here e'
  \pushToTag #'here g' \music
  \appendToTag #'here c'
  \appendToTag #'here e'
  \appendToTag #'here g' \music
}
@end lilypond

Both commands get a tag, the material to splice in at every occurence of
the tag, and the tagged expression.

@seealso
Learning Manual:
@rlearning{Organizing pieces with variables}.

Notation Reference:
@ref{Automatic part combining},
@ref{Including LilyPond files}.

@knownissues
Calling @code{\relative} on a music expression obtained by filtering
music through @code{\keepWithTag} or @code{\removeWithTag} might cause
the octave relations to change, as only the pitches actually
remaining in the filtered expression will be considered.  Applying
@code{\relative} first, before @code{\keepWithTag} or
@code{\removeWithTag}, avoids this danger as @code{\relative} then
acts on all the pitches as-input.


@node Using global settings
@unnumberedsubsubsec Using global settings

@cindex include-settings

Global settings can be included from a separate file:

@example
lilypond -dinclude-settings=MY_SETTINGS.ly MY_SCORE.ly
@end example

Groups of settings such as page size, font or type face can be stored
in separate files. This allows different editions from the same score
as well as standard settings to be applied to many scores, simply by
specifying the proper settings file.

This technique also works well with the use of style sheets, as
discussed in @rlearning{Style sheets}.

@seealso
Learning Manual:
@rlearning{Organizing pieces with variables},
@rlearning{Style sheets}.

Notation Reference:
@ref{Including LilyPond files}.


@node Special characters
@subsection Special characters

@cindex special characters
@cindex non-ASCII characters

@menu
* Text encoding::
* Unicode::
* ASCII aliases::
@end menu


@node Text encoding
@unnumberedsubsubsec Text encoding

@cindex UTF-8

LilyPond uses the character repertoire defined by the Unicode
consortium and ISO/IEC 10646.  This defines a unique name and
code point for the character sets used in virtually all modern
languages and many others too.  Unicode can be implemented using
several different encodings.  LilyPond uses the UTF-8 encoding
(UTF stands for Unicode Transformation Format) which represents
all common Latin characters in one byte, and represents other
characters using a variable length format of up to four bytes.

The actual appearance of the characters is determined by the
glyphs defined in the particular fonts available - a font defines
the mapping of a subset of the Unicode code points to glyphs.
LilyPond uses the Pango library to layout and render multi-lingual
texts.

LilyPond does not perform any input-encoding conversions.  This
means that any text, be it title, lyric text, or musical
instruction containing non-ASCII characters, must be encoded in
UTF-8.  The easiest way to enter such text is by using a
Unicode-aware editor and saving the file with UTF-8 encoding.  Most
popular modern editors have UTF-8 support, for example, vim, Emacs,
jEdit, and Gedit do.  All MS Windows systems later than NT use
Unicode as their native character encoding, so even Notepad can
edit and save a file in UTF-8 format.  A more functional
alternative for Windows is BabelPad.

If a LilyPond input file containing a non-ASCII character is not
saved in UTF-8 format the error message

@example
FT_Get_Glyph_Name () error: invalid argument
@end example

will be generated.

Here is an example showing Cyrillic, Hebrew and Portuguese
text:

@c NOTE: No verbatim in the following example as the code does not
@c display correctly in PDF Font settings for Cyrillic and Hebrew

@lilypond[quote]
% Linux Libertine fonts contain Cyrillic and Hebrew glyphs.
\paper {
  #(define fonts
    (set-global-fonts
     #:roman "Linux Libertine O,serif"
     #:sans "Linux Biolinum O,sans-serif"
     #:typewriter "Linux Libertine Mono O,monospace"
   ))
}

% Cyrillic
bulgarian = \lyricmode {
  Жълтата дюля беше щастлива, че пухът, който цъфна, замръзна като гьон.
}

% Hebrew
hebrew = \lyricmode {
  זה כיף סתם לשמוע איך תנצח קרפד עץ טוב בגן.
}

% Portuguese
portuguese = \lyricmode {
  à vo -- cê uma can -- ção legal
}

\relative {
  c'2 d e f g f e
}
\addlyrics { \bulgarian }
\addlyrics { \hebrew }
\addlyrics { \portuguese }
@end lilypond


@node Unicode
@unnumberedsubsubsec Unicode

@cindex Unicode

To enter a single character for which the Unicode code point is
known but which is not available in the editor being used, use
either @code{\char ##xhhhh} or @code{\char #dddd} within a
@code{\markup} block, where @code{hhhh} is the hexadecimal code for
the character required and @code{dddd} is the corresponding decimal
value.  Leading zeroes may be omitted, but it is usual to specify
all four characters in the hexadecimal representation.  (Note that
the UTF-8 encoding of the code point should @emph{not} be used
after @code{\char}, as UTF-8 encodings contain extra bits indicating
the number of octets.)  Unicode code charts and a character name
index giving the code point in hexadecimal for any character can be
found on the Unicode Consortium website,
@uref{http://www.unicode.org/}.

For example, @code{\char ##x03BE} and @code{\char #958} would both
enter the Unicode U+03BE character, which has the Unicode name
@qq{Greek Small Letter Xi}.

Any Unicode code point may be entered in this way and if all special
characters are entered in this format it is not necessary to save
the input file in UTF-8 format.  Of course, a font containing all
such encoded characters must be installed and available to LilyPond.

The following example shows Unicode hexadecimal values being entered
in four places -- in a rehearsal mark, as articulation text, in
lyrics and as stand-alone text below the score:

@lilypond[quote,verbatim]
\score {
  \relative {
    c''1 \mark \markup { \char ##x03EE }
    c1_\markup { \tiny { \char ##x03B1 " to " \char ##x03C9 } }
  }
  \addlyrics { O \markup { \concat { Ph \char ##x0153 be! } } }
}
\markup { "Copyright 2008--2015" \char ##x00A9 }
@end lilypond

@cindex copyright sign

To enter the copyright sign in the copyright notice use:

@example
\header @{
  copyright = \markup @{ \char ##x00A9 "2008" @}
@}
@end example


@node ASCII aliases
@unnumberedsubsubsec ASCII aliases

A list of ASCII aliases for special characters can be included:

@lilypond[quote,verbatim]
\paper {
  #(include-special-characters)
}

\markup "&flqq; &ndash; &OE;uvre incomplète&hellip; &frqq;"

\score {
  \new Staff { \repeat unfold 9 a'4 }
  \addlyrics {
    This is al -- so wor -- kin'~in ly -- rics: &ndash;_&OE;&hellip;
  }
}

\markup \column {
  "The replacement can be disabled:"
  "&ndash; &OE; &hellip;"
  \override #'(replacement-alist . ()) "&ndash; &OE; &hellip;"
}
@end lilypond

You can also make your own aliases, either globally:

@lilypond[quote,verbatim]
\paper {
  #(add-text-replacements!
    '(("100" . "hundred")
      ("dpi" . "dots per inch")))
}
\markup "A 100 dpi."
@end lilypond

or locally:

@lilypond[quote,verbatim]
\markup \replace #'(("100" . "hundred")
                    ("dpi" . "dots per inch")) "A 100 dpi."
@end lilypond

@seealso
Notation Reference:
@ref{List of special characters}.

Installed Files:
@file{ly/text-replacements.ly}.


@node Controlling output
@section Controlling output

@menu
* Extracting fragments of music::
* Skipping corrected music::
* Alternative output formats::
* Replacing the notation font::
@end menu

@funindex clip-regions
@cindex Fragments, music
@cindex Music fragments

@node Extracting fragments of music
@subsection Extracting fragments of music

It is possible to output one or more fragments of a score by defining
the explicit location of the music to be extracted within the
@code{\layout} block of the input file using the @code{clip-regions}
function, and then running LilyPond with the @option{-dclip-systems}
option;

@example
\layout @{
  clip-regions
  = #(list
      (cons
       (make-rhythmic-location 5 1 2)
       (make-rhythmic-location 7 3 4)))
@}
@end example

@noindent
This example will extract a single fragment of the input file
@emph{starting} after a half-note duration in fifth measure
(@code{5 1 2}) and @emph{ending} after the third quarter-note in the
seventh measure (@code{7 3 4}).

Additional fragments can be extracted by adding more pairs of
@code{make-rhythmic-location} entries to the @code{clip-regions} list in
the @code{\layout} block.

By default, each music fragment will be output as a separate @code{EPS}
file, but other formats such as @code{PDF} or @code{PNG} can also be
created if required.  The extracted music is output as if had been
literally @q{cut} from the original printed score so if a fragment runs
over one or more lines, a separate output file for each line will be
generated.

@seealso
Notation Reference:
@ref{The layout block}.

Application Usage:
@rprogram{Command-line usage}.



@node Skipping corrected music
@subsection Skipping corrected music


@funindex skipTypesetting
@funindex showFirstLength
@funindex showLastLength

When entering or copying music, usually only the music near the end (where
you
are adding notes) is interesting to view and correct.  To speed up
this correction process, it is possible to skip typesetting of all but
the last few measures.  This is achieved by putting

@example
showLastLength = R1*5
\score @{ @dots{} @}
@end example

@noindent
in your source file.  This will render only the last 5 measures
(assuming 4/4 time signature) of every @code{\score} in the input
file.  For longer pieces, rendering only a small part is often an order
of magnitude quicker than rendering it completely.  When working on the
beginning of a score you have already typeset (e.g. to add a new part),
the @code{showFirstLength} property may be useful as well.

Skipping parts of a score can be controlled in a more fine-grained
fashion with the property @code{Score.skipTypesetting}.  When it is
set, no typesetting is performed at all.

This property is also used to control output to the MIDI file.  Note that
it skips all events, including tempo and instrument changes.  You have
been warned.

@lilypond[quote,ragged-right,verbatim]
\relative c' {
  c1
  \set Score.skipTypesetting = ##t
  \tempo 4 = 80
  c4 c c c
  \set Score.skipTypesetting = ##f
  d4 d d d
}
@end lilypond

In polyphonic music, @code{Score.skipTypesetting} will affect all
voices and staves, saving even more time.

@node Alternative output formats
@subsection Alternative output formats

@cindex scalable vector graphics output
@cindex SVG output
@cindex encapsulated postscript output
@cindex EPS output

The default output formats for the printed score are Portable
Document Format (PDF) and PostScript (PS).  Portable
Network Graphics (PNG), Scalable Vector Graphics (SVG) and Encapsulated
PostScript (EPS) output is available through the command line option,
see @rprogram{Basic command line options for LilyPond}.


@node Replacing the notation font
@subsection Replacing the notation font

Gonville is an alternative to the Feta font used in LilyPond and can
be downloaded from:
@example
@uref{http://www.chiark.greenend.org.uk/~sgtatham/gonville/ ,http://www.chiark.greenend.org.uk/~sgtatham/gonville/}
@end example

Here are a few sample bars of music set in Gonville:

@c NOTE: these images are a bit big, but that's important
@c       for the font comparison.  -gp
@sourceimage{Gonville_after,15cm,,}

Here are a few sample bars of music set in LilyPond's Feta font:

@sourceimage{Gonville_before,15cm,,}

@subsubheading Installation Instructions for MacOS

Download and extract the zip file.  Copy the @code{lilyfonts}
directory to @file{@var{SHARE_DIR}/lilypond/current}; for more
information, see @rlearning{Other sources of information}.  Rename the
existing @code{fonts} directory to @code{fonts_orig} and the
@code{lilyfonts} directory to @code{fonts}.  To revert back to Feta,
reverse the process.

@seealso
Learning Manual:
@rlearning{Other sources of information}.

@knownissues
Gonville cannot be used to typeset @q{Ancient Music} notation and it is
likely newer glyphs in later releases of LilyPond may not exist in the
Gonville font family.  Please refer to the author's website for more
information on these and other specifics, including licensing of
Gonville.


@node Creating MIDI output
@section Creating MIDI output

@cindex Sound
@cindex MIDI

LilyPond can produce files that conform to the MIDI (Musical Instrument
Digital Interface) standard and so allow for the checking of the music
output aurally (with the help of an application or device that
understands MIDI).  Listening to MIDI output may also help in spotting
errors such as notes that have been entered incorrectly or are missing
accidentals and so on.

MIDI files do not contain sound (like AAC, MP3 or Vorbis files) but
require additional software to produce sound from them.

@menu
* Supported notation for MIDI::
* Unsupported notation for MIDI::
* The MIDI block::
* Controlling MIDI dynamics::
* Using MIDI instruments::
* Using repeats with MIDI::
* MIDI channel mapping::
* Context properties for MIDI effects::
* Enhancing MIDI output::
@end menu

@cindex MIDI, Supported notation

@node Supported notation for MIDI
@subsection Supported notation for MIDI

The following musical notation can be used with LilyPond's default
capabilities to produce MIDI output;

@itemize
@item Breath marks
@item Chords entered as chord names
@item Crescendi, decrescendi over multiple notes.  The volume is altered
linearly between the two extremes
@item Dynamic markings from @code{ppppp} to @code{fffff}, including
@code{mp}, @code{mf} and @code{sf}
@item Microtones but @emph{not} microtonal chords.  A MIDI player that
supports pitch bending will also be required.
@item Lyrics
@item Pitches
@item Rhythms entered as note durations, including tuplets
@item @q{Simple} articulations; staccato, staccatissimo, accent, marcato
and portato
@item Tempo changes using the @code{\tempo} function
@item Ties
@item Tremolos that are @emph{not} entered with a
@q{@code{:}[@var{number}]} value
@end itemize

Panning, balance, expression, reverb and chorus effects can also be
controlled by setting context properties,
see @ref{Context properties for MIDI effects}.

When combined with the @file{articulate} script the following,
additional musical notation can be output to MIDI;

@itemize
@item Appoggiaturas.  These are made to take half the value of the note
following (without taking dots into account).  For example;

@example
\appoggiatura c8 d2.
@end example

@noindent
The c will take the value of a crotchet.

@item Ornaments (i.e. mordents, trills and turns et al.)
@item Rallentando, accelerando, ritardando and a tempo
@item Slurs, including phrasing slurs
@item Tenuto
@end itemize

@noindent
See @ref{Enhancing MIDI output}.

@cindex MIDI, Unsupported notation

@node Unsupported notation for MIDI
@subsection Unsupported notation for MIDI

The following items of musical notation cannot be output to MIDI;

@itemize
@item Articulations other than staccato, staccatissimo, accent, marcato
and portato
@item Crescendi and decrescendi over a @emph{single} note
@item Fermata
@item Figured bass
@item Glissandi
@item Falls and doits
@item Microtonal chords
@item Rhythms entered as annotations, e.g. swing
@item Tempo changes without @code{\tempo} (e.g. entered as annotations)
@item Tremolos that @emph{are} entered with a @q{@code{:}[@var{number}]}
value
@end itemize


@node The MIDI block
@subsection The MIDI block

@cindex MIDI block

To create a MIDI output file from a LilyPond input file, insert a
@code{\midi} block, which can be empty, within the @code{\score} block;

@example
\score @{
  @var{@dots{} music @dots{}}
  \layout @{ @}
  \midi @{ @}
@}
@end example

@warning{ A @code{@bs{}score} block that, as well as the music, contains
only a @code{@bs{}midi} block (i.e. @emph{without} the @code{@bs{}layout}
block), will only produce MIDI output files.  No notation will be
printed.}

The default output file extension (@code{.midi}) can be changed by using
the @code{-dmidi-extension} option with the @code{lilypond} command:

@example
lilypond -dmidi-extension=mid MyFile.ly
@end example

Alternatively, add the following Scheme expression before the start of
either the @code{\book}, @code{\bookpart} or @code{\score} blocks.  See
@ref{File structure}.

@example
#(ly:set-option 'midi-extension "mid")
@end example

@seealso
Notation Reference:
@ref{File structure}.

Installed Files:
@file{scm/midi.scm}.

@knownissues
There are fifteen MIDI channels available and one additional channel
(#10) for drums.  Staves are assigned to channels in sequence, so a
score that contains more than fifteen staves will result in the extra
staves sharing (but not overwriting) the same MIDI channel.  This may be
a problem if the sharing staves have conflicting, channel-based, MIDI
properties -- such as different MIDI instruments -- set.


@node Controlling MIDI dynamics
@subsection Controlling MIDI dynamics

It is possible to control the overall MIDI volume, the relative volume
of dynamic markings and the relative volume of different instruments.

Dynamic marks translate automatically into volume levels in the
available MIDI volume range whereas crescendi and decrescendi vary the
volume linearly between their two extremes.  It is possible to control
the relative volume of dynamic markings, and the overall volume levels
of different instruments.

@menu
* Dynamic marks in MIDI::
* Setting MIDI volume::
* Setting MIDI block properties::
@end menu

@cindex MIDI volume
@cindex MIDI equalization
@cindex MIDI dynamics
@cindex Dynamics in MIDI


@node Dynamic marks in MIDI
@unnumberedsubsubsec Dynamic marks in MIDI

Only the dynamic markings from @code{ppppp} to @code{fffff}, including
@code{mp}, @code{mf} and @code{sf} have values assigned to them.  This
value is then applied to the value of the overall MIDI volume range to
obtain the final volume included in the MIDI output for that particular
dynamic marking.  The default fractions range from 0.25 for
@notation{ppppp} to 0.95 for @notation{fffff}. The complete set of
dynamic marks and their associated fractions can be found in
@file{scm/midi.scm}.


@snippets
@lilypondfile[verbatim,quote,ragged-right,texidoc,doctitle]
{creating-custom-dynamics-in-midi-output.ly}

Installed Files:
@file{ly/script-init.ly}
@file{scm/midi.scm}.

Snippets:
@rlsr{MIDI}.

Internals Reference:
@rinternals{Dynamic_performer}.


@node Setting MIDI volume
@unnumberedsubsubsec Setting MIDI volume

The minimum and maximum overall volume of MIDI dynamic markings is
controlled by setting the properties @code{midiMinimumVolume} and
@code{midiMaximumVolume} at the @code{Score} level.  These properties
have an effect only at the start of a voice and on dynamic marks.  The
fraction corresponding to each dynamic mark is modified with this
formula

@example
midiMinimumVolume + (midiMaximumVolume - midiMinimumVolume) * fraction
@end example

In the following example the dynamic range of the overall MIDI
volume is limited to the range @code{0.2} - @code{0.5}.

@example
\score @{
  <<
    \new Staff @{
      \set Staff.midiInstrument = #"flute"
      @var{@dots{} music @dots{}}
    @}
    \new Staff @{
      \set Staff.midiInstrument = #"clarinet"
      @var{@dots{} music @dots{}}
    @}
  >>
  \midi @{
    \context @{
      \Score
      midiMinimumVolume = #0.2
      midiMaximumVolume = #0.5
    @}
  @}
@}
@end example

Simple MIDI instrument equalization can be achieved by setting
@code{midiMinimumVolume} and @code{midiMaximumVolume} properties within
the @code{Staff} context.

@example
\score @{
  \new Staff @{
    \set Staff.midiInstrument = #"flute"
    \set Staff.midiMinimumVolume = #0.7
    \set Staff.midiMaximumVolume = #0.9
    @var{@dots{} music @dots{}}
  @}
  \midi @{ @}
@}
@end example

For scores with multiple staves and multiple MIDI instruments, the
relative volumes of each instrument can be set individually;

@example
\score @{
  <<
    \new Staff @{
      \set Staff.midiInstrument = #"flute"
      \set Staff.midiMinimumVolume = #0.7
      \set Staff.midiMaximumVolume = #0.9
      @var{@dots{} music @dots{}}
    @}
    \new Staff @{
      \set Staff.midiInstrument = #"clarinet"
      \set Staff.midiMinimumVolume = #0.3
      \set Staff.midiMaximumVolume = #0.6
      @var{@dots{} music @dots{}}
    @}
  >>
  \midi @{ @}
@}
@end example

In this example the volume of the clarinet is reduced relative to the
volume of the flute.

If these volumes properties are not set then LilyPond still applies a
@q{small degree} of equalization to certain instruments.  See
@file{scm/midi.scm}.

Installed Files:
@file{scm/midi.scm}.

@seealso
Notation Reference:
@ref{Score layout}.

Internals Reference:
@rinternals{Dynamic_performer}.

@snippets
@lilypondfile[verbatim,quote,ragged-right,texidoc,doctitle]
{replacing-default-midi-instrument-equalization.ly}

@knownissues
Changes in the MIDI volume take place only on starting a note, so
crescendi and decrescendi cannot affect the volume of a single note.


@node Setting MIDI block properties
@unnumberedsubsubsec Setting MIDI block properties

The @code{\midi} block can contain context rearrangements, new context
definitions or code that sets the values of certain properties.

@example
\score @{
  @var{@dots{} music @dots{}}
  \midi @{
    \tempo 4 = 72
  @}
@}
@end example

Here the tempo is set to 72 quarter-note beats per minute.  The tempo
mark in the @code{\midi} block will not appear in the printed score.
Although any other @code{\tempo} indications specified within the
@code{\score} block will also be reflected in the MIDI output.

In a @code{\midi} block the @code{\tempo} command is setting properties
during the interpretation of the music and in the context of output
definitions; so it is interpreted @emph{as if} it were a context
modification.

@cindex MIDI context definitions
@cindex context definitions with MIDI

Context definitions follow the same syntax as those in a @code{\layout}
block;

@example
\score @{
  @var{@dots{} music @dots{}}
  \midi @{
    \context @{
      \Voice
      \remove "Dynamic_performer"
    @}
  @}
@}
@end example

This example removes the effect of dynamics from the MIDI output.  Note:
LilyPond's translation modules used for sound are called @q{performers}.

@seealso
Learning Manual:
@rlearning{Other sources of information}.

Notation Reference:
@ref{Expressive marks},
@ref{Score layout}.

Installed Files:
@file{ly/performer-init.ly}.

Snippets:
@rlsr{MIDI}.

Internals Reference:
@rinternals{Dynamic_performer}.

@knownissues
Some MIDI players do not always correctly handle tempo changes in the
midi output.

Changes to the @code{midiInstrument}, as well as some MIDI options, at
the @emph{beginning} of a staff may appear twice in the MIDI output.



@node Using MIDI instruments
@subsection Using MIDI instruments

MIDI instruments are set using the @code{midiInstrument} property within
a @code{Staff} context.

@example
\score @{
  \new Staff @{
    \set Staff.midiInstrument = #"glockenspiel"
    @var{@dots{} music @dots{}}
  @}
  \midi @{ @}
@}
@end example

or

@example
\score @{
  \new Staff \with @{midiInstrument = #"cello"@} @{
    @var{@dots{} music @dots{}}
  @}
  \midi @{ @}
@}
@end example

If the instrument name does not match any of the instruments listed in
the @q{MIDI instruments} section, the @code{acoustic grand} instrument
will be used instead. See @ref{MIDI instruments}.

@seealso
Learning Manual:
@rlearning{Other sources of information}.

Notation Reference:
@ref{MIDI instruments},
@ref{Score layout}.

Installed Files:
@file{scm/midi.scm}.

@knownissues
Percussion instruments that are notated in a @code{DrumStaff}
context will be output, correctly, to MIDI channel@tie{}10 but some
pitched, percussion instruments like the xylophone, marimba, vibraphone
or timpani, are treated as @qq{normal} instruments so the music for
these should be entered in a @code{Staff} (not @code{DrumStaff}) context
to obtain correct MIDI output.  A full list of
@code{channel 10 drum-kits} entries can be found in @file{scm/midi.scm}.
See @rlearning{Other sources of information}.


@node Using repeats with MIDI
@subsection Using repeats with MIDI

@cindex repeats in MIDI
@cindex MIDI using repeats
@funindex \unfoldRepeats

Repeats can be represented in the MIDI output by applying the
@code{\unfoldRepeats} command.

@example
\score @{
  \unfoldRepeats @{
    \repeat tremolo 8 @{ c'32 e' @}
    \repeat percent 2 @{ c''8 d'' @}
    \repeat volta 2 @{ c'4 d' e' f' @}
    \alternative @{
      @{ g' a' a' g' @}
      @{ f' e' d' c' @}
    @}
  @}
  \midi @{ @}
@}
@end example

In order to restrict the effect of @code{\unfoldRepeats} to the MIDI
output only, while also generating printable scores, it is necessary to
make @emph{two} @code{\score} blocks; one for MIDI (with unfolded
repeats) and one for the notation (with volta, tremolo, and percent
repeats);

@example
\score @{
  @var{@dots{} music @dots{}}
  \layout @{ @}
@}
\score @{
  \unfoldRepeats @{
    @var{@dots{} music @dots{}}
  @}
  \midi @{ @}
@}
@end example

When using multiple voices, each of the voices must contain completely
unfolded repeats for correct MIDI output.

@seealso
Notation Reference:
@ref{Repeats}.


@node MIDI channel mapping
@subsection MIDI channel mapping

@cindex MIDI Channels
@cindex MIDI Tracks
@funindex midiChannelMapping

When generating a MIDI file from a score, LilyPond will automatically
assign every note in the score to a MIDI channel, the one on which it
should be played when it is sent to a MIDI device.  A MIDI channel has
a number of controls available to select, for example, the instrument
to be used to play the notes on that channel, or to request the MIDI
device to apply various effects to the sound produced on the channel.
At all times, every control on a MIDI channel can have only a single
value assigned to it (which can be modified, however, for example,
to switch to another instrument in the middle of a score).

The MIDI standard supports only 16 channels per MIDI device.  This
limit on the number of channels also limits the number of different
instruments which can be played at the same time.

LilyPond creates separate MIDI tracks for each staff, (or discrete
instrument or voice, depending on the value of
@code{Score.midiChannelMapping}), and also for each lyrics context.
There is no limit to the number of tracks.

To work around the limited number of MIDI channels, LilyPond supports
a number of different modes for MIDI channel allocation, selected using
the @code{Score.midiChannelMapping} context property.  In each case,
if more MIDI channels than the limit are required, the allocated
channel numbers wrap around back to 0, possibly causing the incorrect
assignment of instruments to some notes.  This context property can be
set to one of the following values:

@table @var

@item @code{'staff}

Allocate a separate MIDI channel to each staff in the score (this is
the default).  All notes in all voices contained within each staff will
share the MIDI channel of their enclosing staff, and all are encoded
in the same MIDI track.

The limit of 16 channels is applied to the total number of staff and
lyrics contexts, even though MIDI lyrics do not take up a MIDI channel.

@item @code{'instrument}

Allocate a separate MIDI channel to each distinct MIDI instrument
specified in the score.  This means that all the notes played with the
same MIDI instrument will share the same MIDI channel (and track), even
if the notes come from different voices or staves.

In this case the lyrics contexts do not count towards the MIDI channel
limit of 16 (as they will not be assigned to a MIDI instrument), so
this setting may allow a better allocation of MIDI channels when the
number of staves and lyrics contexts in a score exceeds 16.

@item @code{'voice}

Allocate a separate MIDI channel to each voice in the score that has a
unique name among the voices in its enclosing staff.  Voices in
different staves are always assigned separate MIDI channels, but any two
voices contained within the same staff will share the same MIDI channel
if they have the same name.  Because @code{midiInstrument} and the
several MIDI controls for effects are properties of the staff context,
they cannot be set separately for each voice.  The first voice will be
played with the instrument and effects specified for the staff, and
voices with a different name from the first will be assigned the default
instrument and effects.

Note: different instruments and/or effects can be assigned to several
voices on the same staff by moving the @code{Staff_performer} from the
@code{Staff} to the @code{Voice} context, and leaving
@code{midiChannelMapping} to default to @code{'staff} or set to
@code{'instrument}; see the snippet below.

@end table

For example, the default MIDI channel mapping of a score can be changed
to the @code{'instrument} setting as shown:

@example
\score @{
  ...music...
  \midi @{
    \context @{
      \Score
      midiChannelMapping = #'instrument
    @}
  @}
@}
@end example

@snippets
@lilypondfile[verbatim,quote,ragged-right,texidoc,doctitle]
{changing-midi-output-to-one-channel-per-voice.ly}


@node Context properties for MIDI effects
@subsection Context properties for MIDI effects

@cindex Effects in MIDI
@cindex Pan position in MIDI
@cindex Stereo balance in MIDI
@cindex Balance in MIDI
@cindex Expression in MIDI
@cindex Reverb in MIDI
@cindex Chorus level in MIDI
@funindex midiPanPosition
@funindex midiBalance
@funindex midiExpression
@funindex midiReverbLevel
@funindex midiChorusLevel

The following context properties can be used to apply various MIDI
effects to notes played on the MIDI channel associated with the
current staff, MIDI instrument or voice (depending on the value of the
@code{Score.midiChannelMapping} context property and the context in
which the @code{Staff_performer} is located; see
@ref{MIDI channel mapping}).

Changing these context properties will affect all notes played on the
channel after the change, however some of the effects may even apply
also to notes which are already playing (depending on the
implementation of the MIDI output device).

The following context properties are supported:

@table @var

@item @code{Staff.midiPanPosition}

The pan position controls how the sound on a MIDI channel is
distributed between left and right stereo outputs.  The context
property accepts a number between -1.0 (@code{#LEFT}) and 1.0
(@code{#RIGHT}); the value -1.0 will put all sound power to the left
stereo output (keeping the right output silent), the value 0.0
(@code{#CENTER}) will distribute the sound evenly between the left and
right stereo outputs, and the value 1.0 will move all sound to the
right stereo output.  Values between -1.0 and 1.0 can be used to obtain
mixed distributions between left and right stereo outputs.

@item @code{Staff.midiBalance}

The stereo balance of a MIDI channel.  Similarly to the pan position,
this context property accepts a number between -1.0 (@code{#LEFT}) and
1.0 (@code{#RIGHT}).  It varies the relative volume sent to the two
stereo speakers without affecting the distribution of the stereo
signals.

@item @code{Staff.midiExpression}

Expression level (as a fraction of the maximum available level) to
apply to a MIDI channel.  A MIDI device combines the MIDI channel's
expression level with a voice's current dynamic level (controlled using
constructs such as @code{\p} or @code{\ff}) to obtain the total volume
of each note within the voice.  The expression control could be used, for
example, to implement crescendo or decrescendo effects over single
sustained notes (not supported automatically by LilyPond).

@c Issue 4059 contains an attached snippet which shows how this might
@c be done, but this is too large and complex for the NR, even as a
@c referenced snippet.  It could be added to the LSR.

The expression level ranges from 0.0 (no expression, meaning zero
volume) to 1.0 (full expression).

@item @code{Staff.midiReverbLevel}

Reverb level (as a fraction of the maximum available level) to apply
to a MIDI channel.  This property accepts numbers between 0.0 (no
reverb) and 1.0 (full effect).

@item @code{Staff.midiChorusLevel}

Chorus level (as a fraction of the maximum available level) to apply to
a MIDI channel.  This property accepts numbers between 0.0 (no chorus
effect) and 1.0 (full effect).

@end table


@knownissues

As MIDI files do not contain any actual audio data, changes in these
context properties translate only to requests for changing MIDI channel
controls in the outputted MIDI files.  Whether a particular MIDI device
(such as a software MIDI player) can actually handle any of these
requests in a MIDI file is entirely up to the implementation of the
device: a device may choose to ignore some or all of these requests.
Also, how a MIDI device will interpret different values for these
controls (generally, the MIDI standard fixes the behavior only at the
endpoints of the value range available for each control), and whether a
change in the value of a control will affect notes already playing on
that MIDI channel or not, is also specific to the MIDI device
implementation.

When generating MIDI files, LilyPond will simply transform the
fractional values within each range linearly into values in a
corresponding (7-bit, or 14-bit for MIDI channel controls which support
fine resolution) integer range (0-127 or 0-32767, respectively),
rounding fractional values towards the nearest integer away from zero.
The converted integer values are stored as-is in the generated MIDI
file.  Please consult the documentation of your MIDI device for
information about how the device interprets these values.


@node Enhancing MIDI output
@subsection Enhancing MIDI output

@menu
* The articulate script::
@end menu

The default MIDI output is basic but can be improved by setting MIDI
instruments, @code{\midi} block properties and/or using the
@file{articulate} script.

@cindex instrument names
@cindex MIDI, instruments
@cindex articulate script
@cindex articulate.ly
@funindex Staff.midiInstrument


@node The articulate script
@unnumberedsubsubsec The @file{articulate} script

To use the @file{articulate} script add the appropriate @code{\include}
command at the top of the input file;

@example
\include "articulate.ly"
@end example

The script creates MIDI output into appropriately @q{time-scaled} notes
to match many articulation and tempo indications.  Engraved output
however, will also be altered to literally match the MIDI output.

@example
\score @{
  \articulate <<
    @var{@dots{} music @dots{}}
  >>
  \midi @{ @}
@}
@end example

The @code{\articulate} command enables abbreviatures (such as trills and
turns) to be processed.  A full list of supported items can be found in
the script itself.  See @file{ly/articulate.ly}.

@seealso
Learning Manual:
@rlearning{Other sources of information}.

Notation Reference:
@ref{Score layout}.

Installed Files:
@file{ly/articulate.ly}.

@warning{The @file{articulate} script may shorten chords, which might
not be appropriate for some types of instrument, such as organ music.
Notes that do not have any articulations attached to them may also be
shortened; so to allow for this, restrict the use of the
@code{\articulate} function to shorter segments of music, or modify the
values of the variables defined in the @file{articulate} script to
compensate for the note-shortening behavior.}



@node Extracting musical information
@section Extracting musical information

In addition to creating graphical output and MIDI, LilyPond can
display musical information as text.

@menu
* Displaying LilyPond notation::
* Displaying scheme music expressions::
* Saving music events to a file::
@end menu

@node Displaying LilyPond notation
@subsection Displaying LilyPond notation

@funindex \displayLilyMusic
Displaying a music expression in LilyPond notation can be
done with the music function @code{\displayLilyMusic}.  To see the
output, you will typically want to call LilyPond using the command
line.  For example,

@example
@{
  \displayLilyMusic \transpose c a, @{ c4 e g a bes @}
@}
@end example

will display

@example
@{ a,4 cis4 e4 fis4 g4 @}
@end example

By default, LilyPond will print these messages to the console
along with all the other LilyPond compilation messages.  To split
up these messages and save the results of @code{\displayLilyMusic},
redirect the output to a file.

@example
lilypond file.ly >display.txt
@end example

@funindex \void
Note that Lilypond does not just display the music expression, but
also interprets it (since @code{\displayLilyMusic} returns it in
addition to displaying it).  Just insert @code{\displayLilyMusic} into
the existing music in order to get information about it.

To interpret and display a music section in the console but, at the same
time, remove it from the output file use the @code{\void} command.

@example
@{
  \void \displayLilyMusic \transpose c a, @{ c4 e g a bes @}
  c1
@}
@end example


@node Displaying scheme music expressions
@subsection Displaying scheme music expressions

See @rextend{Displaying music expressions}.


@node Saving music events to a file
@subsection Saving music events to a file

Music events can be saved to a file on a per-staff basis by
including a file in your main score.

@example
\include "event-listener.ly"
@end example

This will create file(s) called @file{FILENAME-STAFFNAME.notes} or
@file{FILENAME-unnamed-staff.notes} for each staff.  Note that if
you have multiple unnamed staves, the events for all staves will
be mixed together in the same file.  The output looks like this:

@example
0.000   note     57       4   p-c 2 12
0.000   dynamic  f
0.250   note     62       4   p-c 7 12
0.500   note     66       8   p-c 9 12
0.625   note     69       8   p-c 14 12
0.750   rest     4
0.750   breathe
@end example

The syntax is a tab-delimited line, with two fixed fields on each
line followed by optional parameters.

@example
@var{time}  @var{type}  @var{@dots{}params@dots{}}
@end example

This information can easily be read into other programs such as
python scripts, and can be very useful for researchers wishing to
perform musical analysis or playback experiments with LilyPond.


@knownissues

Not all lilypond music events are supported by
@file{event-listener.ly}.  It is intended to be a well-crafted
@qq{proof of concept}.  If some events that you want to see are
not included, copy @file{event-listener.ly} into your lilypond
directory and modify the file so that it outputs the information
you want.