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

@c Note:
@c
@c A menu is needed before every deeper *section nesting of @nodes
@c Run M-x texinfo-all-menus-update
@c to automagically fill in these menus
@c before saving changes


@macro refbugs
@unnumberedsubsec Bugs

@end macro


@c .{Reference Manual}

@node Reference Manual
@chapter Reference Manual

This document describes GNU LilyPond and its input format. The last
revision of this document was for LilyPond 1.3.141.


@menu
* Overview::                    
* Note entry::                  
* Staff notation::              
* Polyphony::                   
* Beaming::                     
* Expressive marks::            
* Ornaments::                   
* Repeats::                     
* Rhythmic music::              
* Piano music::                 
* Lyrics::                      
* Chords::                      
* Writing parts::               
* Custodes::                    
* Tuning output::               
* Page layout::                 
* Sound::                       
* Music entry::                 
* Skipping corrected music::    
* Interpretation context::      
* Syntactic details::           
* Lexical details::             
@end menu

@c . {Overview}
@node Overview
@section Overview


The purpose of LilyPond is explained informally by the term `music
typesetter'.  This is not a fully correct name: not only does the
program print musical symbols, it also makes esthetic decisions.
Symbols and their placements are @emph{generated} from a high-level
musical description.  In other words, LilyPond would be best described
by `music compiler' or `music to notation compiler'.

LilyPond is linked to GUILE, GNU's Scheme library for extension. The
Scheme library provides the glue that holds together the low-level
routines and separate modules general, which are C++.

When lilypond is run to typeset sheet music, the following happens:
@itemize @bullet
@item GUILE Initialization: various scheme files are read
@item parsing: first standard @code{ly} initialization  files are read, and
then the user @file{ly} file is read.
@item interpretation: the music in the file is processed ``in playing
order'', i.e. the order that  you  use to read sheet music, or the
order in which notes are played.

@item typesetting:
in this step, the results of the interpretation, a typesetting
specification, is solved.

@item the visible results ("virtual ink") is written to the output file.
@end itemize

During these stages different types of data play the the main role:
during parsing, @strong{Music} objects are created.  During the
interpretation, @strong{context} is constructed, and with this context
af network of @strong{graphical objects} (``grobs'') is created. The
grobs contain unknown variables, and the network forms a set of
equations. After solving the equations and filling in these variables,
the printed output (in the form of @strong{molecules}) is written to an
output file.

These threemanship of tasks (parsing, translating, typesetting) and
data-structures (music, context, graphical objects) permeates the entire
design of the program.  This manual is ordered in terms of user
tasks. With each concept will be explained to which of the three parts
it belongs.


@c . {Note entry}
@node Note entry
@section Note entry
@cindex Note entry

The most basic forms of music are notes. We discuss how you enter them
here.  Notes on their own don't form valid input, but for the sake of
brevity we omit obligotary lint such as @code{\score} blocks and
@code{\paper} declarations.


@menu
* Pitches::                     
* Defining pitch names::        
* Durations::                   
* Notes::                       
* Easy Notation note heads ::   
* Tie::                         
* Tuplets::                     
* Rests::                       
* Skip::                        
* Note mode::                   
@end menu

@c .  {Pitches}
@node Pitches
@subsection Pitches

@cindex Pitch names
@cindex Note specification
@cindex pitches
@cindex entering notes

The verbose syntax for pitch specification is

@cindex @code{\pitch}
@example
  \pitch @var{scmpitch}
@end example

@var{scmpitch} is a pitch scheme object, see @ref{Pitch data type}.

In Note and Chord mode, pitches may be designated by names.  The default
names are the Dutch note names.  The notes are specified by the letters
@code{a} through @code{g} (where the octave is formed by notes ranging
from @code{c}, to @code{b}).  The pitch @code{c} is an octave below
middle C and the letters span the octave above that C.

@cindex note names, Dutch

In Dutch, a sharp is formed by adding @code{-is} to the end of a pitch
name.  A flat is formed by adding @code{-es}. Double sharps and double
flats are obtained by adding @code{-isis} or @code{-eses}.  @code{aes}
and @code{ees} are contracted to @code{as} and @code{es} in Dutch, but
both forms are accepted.

LilyPond has predefined sets of notenames for various other languages.
To use them, simply include the language specific init file.  For
example: @code{\include "english.ly"}.  The available language files and
the names they define are:

@example 
                        Note Names               sharp       flat
nederlands.ly  c   d   e   f   g   a   bes b   -is         -es
english.ly     c   d   e   f   g   a   bf  b   -s/-sharp   -f/-flat
deutsch.ly     c   d   e   f   g   a   b   h   -is         -es
norsk.ly       c   d   e   f   g   a   b   h   -iss/-is    -ess/-es
svenska.ly     c   d   e   f   g   a   b   h   -iss        -ess
italiano.ly    do  re  mi  fa  sol la  sib si  -d          -b
catalan.ly     do  re  mi  fa  sol la  sib si  -d/-s       -b 
@end example 

@cindex @code{'}
@cindex @code{,}



The optional octave specification takes the form of a series of
single quote (`@code{'}') characters or a series of comma
(`@code{,}') characters.  Each @code{'} raises the pitch by one
octave; each @code{,} lowers the pitch by an octave.

@lilypond[fragment,verbatim,center]
  c' c'' es' g' as' gisis' ais'  
@end lilypond

@c .  {Defining pitch names}
@node Defining pitch names
@subsection Defining pitch names

@cindex defining pitch names
@cindex pitch names, defining 

Note names and chord modifiers can be customised for nationalities.  The
syntax is as follows.

@cindex @code{\pitchnames}
@cindex @code{\chordmodifiers}
@example
   \pitchnames @var{scheme-alist}
   \chordmodifiers @var{scheme-alist}
@end example

See @file{ly/nederlands.ly} and @file{ly/chord-modifiers.ly} for
specific examples how to do this.



@c .  {Durations}
@node Durations
@subsection Durations


@cindex duration
@cindex @code{\duration}

The syntax for a verbose duration specification is
@example
 \duration @var{scmduration}
@end example
Here, @var{scmduration} is a Scheme object of type Duration. See
@ref{Duration} for more information.


In Note, Chord, and Lyrics mode, durations may be designated by numbers
and dots: durations are entered as their reciprocal values.  For notes
longer than a whole note, use identifiers.

@quotation

@example 
c'\longa c'\breve  
c'1 c'2 c'4 c'8 c'16 c'32 c'64 c'64 
r\longa r\breve  
r1 r2 r4 r8 r16 r32 r64 r64 
@end example 


@lilypond[]
\score {
  \notes \relative c'' {
    a\longa a\breve  \autoBeamOff
    a1 a2 a4 a8 a16 a32 a64 a64 
    r\longa r\breve  
    r1 r2 r4 r8 r16 r32 r64 r64 
  }
  \paper {
    \translator {
      \StaffContext
        \remove "Clef_engraver";
        \remove "Staff_symbol_engraver";
        \remove "Time_signature_engraver";
        \consists "Pitch_squash_engraver";
    }
  }
}
@end lilypond
@end quotation

As you can see, the longa is not printed. To get a longa note head, you
have to use a mensural note heads. This is done accomplished by setting
the @code{style} property of the NoteHead grob to @code{mensural}.

If the duration is omitted then it is set to the previous duration
entered.  At the start of parsing a quarter note is assumed.  The
duration can be followed by a dot (`@code{.}')  to obtain dotted note
lengths.
@cindex @code{.}

@lilypond[fragment,verbatim,center]
  a'4. b'4.. c'8.
@end lilypond
@cindex @code{r}
@cindex @code{s}

You can alter the length of duration by appending
`@code{*}@var{fraction}'.  This will not affect the appearance of the
notes or rests produced.

@c . {Notes}
@node Notes
@subsection Notes

A note specification has the form

@example
  @var{pitch}[@var{octavespec}][!][?][@var{duration}]
@end example

LilyPond will determine what accidentals to typeset depending on the key
and context. The alteration refers to what note is heard, not to whether
an accidental is printed.  A reminder accidental
@cindex reminder accidental
@cindex @code{?}
can be forced by adding an exclamation mark @code{!} after the pitch.  A
cautionary accidental,
@cindex cautionary accidental
i.e., an accidental within parentheses can be obtained by adding the
question mark `@code{?}' after the pitch.

@lilypond[fragment,verbatim,center]
  cis' d' e' cis'  c'? d' e' c'!
@end lilypond


@node Easy Notation note heads 
@subsection Easy Notation note heads

@cindex easy notation
@cindex Hal Leonard

A entirely different type of note head is the "easyplay" note head: a
note head that includes a note name.  It is used in some publications by
Hal-Leonard Inc.  music publishers.

@lilypond[singleline,verbatim]
\include "paper26.ly"
\score {
        \notes { c'2 e'4 f' | g'1 }
        \paper { \translator { \EasyNotation } } 
}
@end lilypond

Note that @code{EasyNotation} overrides a @code{Score} context.  You
probably will want to print it with magnification to make it better
readable.

@cindex Xdvi
@cindex ghostscript

If you view the result with Xdvi, then staff lines will show through the
letters.  Printing the postscript file obtained either by using dvips or
the @code{-f ps} option of lilypond will produce the desired result.


@node Tie
@subsection Tie

@cindex Tie
@cindex ties
@cindex @code{~}

A tie connects two adjacent note heads of the same pitch.  When used
with chords, it connects all of the note heads whose pitches match.
Ties are indicated using the tilde symbol `@code{~}'.
If you try to tie together chords which have no common pitches, a
warning message will appear and no ties will be created.

@lilypond[fragment,verbatim,center]
  e' ~ e' <c' e' g'> ~ <c' e' g'>
@end lilypond

If you dislike the amount of ties created for a chord, you set
@code{Voice.sparseTies} to true, resulting in  a smaller number of
ties:
@lilypond[fragment,verbatim,center]
  \property Voice.sparseTies = ##t
  <c' e' g'> ~ <c' e' g'>
@end lilypond

In its meaning a tie is just a way of extending a note duration, similar
to the augmentation dot: the following example are three ways of notating
exactly the same concept.
@lilypond[fragment, singleline]
c'2 c'4 ~ c'4
@end lilypond

@refbugs

At present, the tie is implemented as a separate thing, temporally
located in between the notes. There is also no way to convert
between tied notes, dotted notes and plain notes.

Tieing only a subset of the note heads of a chord is not supported in a
simple way.  It can be achieved by moving the tie-engraver into Thread
context and turning off ties per Thread.


@node Tuplets
@subsection Tuplets

@cindex tuplets
@cindex triplets
@cindex @code{\times}

Tuplets are made out of a music expression by multiplying their duration
with a fraction.

@cindex @code{\times}
@example
  \times @var{fraction} @var{musicexpr}
@end example

The duration of @var{musicexpr} will be multiplied by the fraction. 
In print, the fraction's denominator will be printed over the notes,
optionally with a bracket.  The most common tuplet is the triplet in
which 3 notes have the length of 2, so the notes are 2/3 of
their written length:

@lilypond[fragment,verbatim,center]
  g'4 \times 2/3 {c'4 c' c'} d'4 d'4
@end lilypond

The property @code{tupletSpannerDuration} specifies how long brackets
should last.  With this, you can make lots of tuplets while typing
@code{\times} only once. This saves typing work when you must make lots
of tuplets.

@lilypond[fragment,  relative, singleline, verbatim]
\property Voice.tupletSpannerDuration = #(make-moment 1 4)
\times 2/3 { c''8 c c c c c }
@end lilypond

@c .  {Rests}
@node  Rests
@subsection Rests
@cindex Rests

Rests are entered like notes, with note name `@code{r}'.


@c .  {Skip}
@node Skip
@subsection Skip
@cindex Skip


@example
  \skip @var{duration} @code{;}
  s@var{duration}
@end example
@cindex @code{\skip}

Skips the amount of time specified by @var{duration}.  If no other music
is played, a gap will be left for the skipped time with no notes
printed.  The shorthand is only available in Note and Chord mode.



@node Note mode
@subsection Note mode



@cindex note mode
@cindex @code{\notes}

Note mode is the lexical mode generally used for inputting notes. The
syntax is
@example
\notes @var{expr}
@end example

This instructs the tokenizer to interpret @var{expr} in note mode.  If a
a sequence of alfabetical characters, like @code{foobar}, LilyPond first
checks if @code{foobar} is a pitch name.  If it is not a pitch name,
then it is treated as a string.

Numbers and dots indicate durations, so you can enter floating point
numbers in this mode.


@node Staff notation
@section Staff notation

@cindex Staff notation

@menu
* Key signature::               
* Clef::                        
* Time signature::              
* Unmetered music::             
* Bar lines::                   
@end menu

@c .  {Key}
@node Key signature
@subsection Key signature
@cindex Key

@cindex @code{\key}

Changing the key signature is done with the @code{\key} command.
@example
  @code{\key} @var{pitch} @var{type} @code{;}
@end example

@cindex @code{\minor}
@cindex @code{\major}
@cindex @code{\minor}
@cindex @code{\ionian}
@cindex @code{\locrian}
@cindex @code{\aeolian}
@cindex @code{\mixolydian}
@cindex @code{\lydian}
@cindex @code{\phrygian}
@cindex @code{\dorian}

Here, @var{type} should be @code{\major} or @code{\minor} to get
@var{pitch}-major or @var{pitch}-minor, respectively.  The second
argument is optional; the default is major keys.  The @var{\context}
argument can also be given as an integer, which tells the number of
semitones that should be added to the pitch given in the subsequent
@code{\key} commands to get the corresponding major key, e.g.,
@code{\minor} is defined as 3.  The standard mode names @code{\ionian},
@code{\locrian}, @code{\aeolian}, @code{\mixolydian}, @code{\lydian},
@code{\phrygian}, and @code{\dorian} are also defined.

This command sets context property @code{Staff.keySignature}.

@cindex @code{keySignature}

@c .  {Clef}
@node Clef
@subsection Clef
@cindex @code{\clef}
@example
  \clef @var{clefname} @code{;}
@end example

Shortcut for

@example
  \property Staff.clefGlyph = @var{glyph associated with clefname} 
  \property Staff.clefPosition = @var{clef Y-position for clefname}
  \property Staff.clefOctavation = @var{extra pitch of clefname}
@end example

Supported clef-names include 

@itemize @bullet
@item treble, violin, G, G2: G clef on 2nd line
@item french: G clef on 1st line
@item soprano: C clef on 1st line
@item mezzosoprano: C clef on 2nd line
@item alto: C clef on 3rd line
@item tenor: C clef on 4th line
@item baritone: C clef on 5th line
@item varbaritone: F clef on 3rd line
@item bass, F: F clef on 4th line
@item subbass: F clef on 5th line
@item percussion: percussion clef
@end itemize

Supported associated glyphs (for @code{Staff.clefGlyph}) are:

@itemize @bullet
@item clefs-C: modern style C clef
@item clefs-F: modern style F clef
@item clefs-G: modern style G clef
@item clefs-vaticana_do: Editio Vaticana style do clef
@item clefs-vaticana_fa: Editio Vaticana style fa clef
@item clefs-medicaea_do: Editio Medicaea style do clef
@item clefs-medicaea_fa: Editio Medicaea style fa clef
@item clefs-mensural1_c: modern style mensural C clef
@item clefs-mensural2_c: historic style small mensural C clef
@item clefs-mensural3_c: historic style big mensural C clef
@item clefs-mensural1_f: historic style traditional mensural F clef
@item clefs-mensural2_f: historic style new mensural F clef
@item clefs-mensural_g: historic style mensural G clef
@item clefs-hufnagel_do: historic style hufnagel do clef
@item clefs-hufnagel_fa: historic style hufnagel fa clef
@item clefs-hufnagel_do_fa: historic style hufnagel combined do/fa clef
@item clefs-percussion: modern style percussion clef
@end itemize

@emph{Modern style} means ``as is typeset in current editions.''
@emph{Historic style} means ``as was typeset or written in contemporary
historic editions''.  @emph{Editio XXX style} means ``as is/was printed in
Editio XXX.''

@cindex Vaticana, Editio
@cindex Medicaea, Editio
@cindex hufnagel clefs


@c .  {Time signature}
@node Time signature
@subsection Time signature
@cindex Time signature
@cindex meter
@cindex @code{\time}

The time signature is changed by the @code{\time} command. Syntax:
@example
  \time @var{numerator}@code{/}@var{denominator} @code{;}
@end example
Internally, this is a shortcut for doing
@example
     \property Score.timeSignatureFraction = #'(@var{numerator} . @var{denominator})
@end example

[TODO: discuss options for layout]

@c .   {Partial}
@subsection Partial
@cindex Partial
@cindex anacrusis
@cindex upstep
@cindex partial measure
@cindex measure, partial
@cindex shorten measures
@cindex @code{\partial}

Partial measures are entered using the @code{\partial} command:
@example
  \partial @var{duration} @code{;}
@end example

Internally,  this is a shortcut for 

@example
  \property Score.measurePosition = -@var{length of duration}
@end example
@cindex @code{|}


@node Unmetered music
@subsection Unmetered music

Bar lines and bar numbers are calculated automatically. For unmetered
music (e.g. cadenzas), this is not desirable. The property
@code{Score.timing} can be used to switch off this automatic timing

@lilypond[fragment,relative,singleline,verbatim]
c'2.
\property Score.timing = ##f
c4 c4 c4  
\property Score.timing = ##t
c4 c4 c4 
@end lilypond

The identifiers @code{\cadenzaOn} and @code{\cadenzaOff} can be used to
achieve the same effect.



@c .   {Bar lines}
@node Bar lines
@subsection Bar lines
@cindex Bar lines

@cindex @code{\bar}
@cindex measure lines
@cindex repeat bars

@example
  \bar @var{bartype};
@end example

This is a shortcut for doing
@example
  \property Score.whichBar = @var{bartype} 
@end example

You are encouraged to use @code{\repeat} for repetitions.  See
@ref{Repeats}, and the documentation of @code{whichBar} in the generated
documentation.


@cindex Bar_line_engraver
@cindex whichBar
@cindex repeatCommands
@cindex defaultBarType

Bar lines are created by the @code{Bar_line_engraver}. That engraver examines
@code{whichBar} at every moment. Whenever it is set to a string, it will
create a bar with that type.  @code{whichBar} is usually set
automatically: at the start of a measure it is set to
@code{defaultBarType}. The contents of @code{repeatCommands} is used to
override default measure bars. 

@code{whichBar} can also be set directly, using @code{\property} or
@code{\bar ; }.  These settings take precedence over automatic @code{whichBar}
settings. 


@c .   {Polyphony}
@node Polyphony
@section Polyphony
@cindex polyphony

Polyphonic parts, i.e. parts with more than one voice on a staff can be
typeset with LilyPond.   To use this, instantiate a separate Voice
context for each part, and assign a stem direction to each part. 
@lilypond[fragment,verbatim]
\context Staff
< \context Voice = VA { \stemUp b'4 a' g' f' e' }
  \context Voice = VB { \stemDown g'4 g' g' g' g' } >
@end lilypond

When there are more than two voices on a staff, you must also indicate
which voice should moved horizontally in case of a collision. This can
be done with the identifiers @code{\shiftOff}, @code{\shiftOn},
@code{\shiftOnn}, etc. (which sets grob property @code{horizontal-shift}
in @code{NoteColumn}).

@lilypond[fragment, verbatim]
  \context Staff \notes\relative c''<
       \context Voice=one {
       \shiftOff \stemUp e4
       }
       \context Voice=two {
          \shiftOn \stemUp cis
       }
       \context Voice=three {
         \shiftOnn \stemUp ais
       }
       \context Voice=four {
          \shiftOnnn \stemUp fis-2
       }
  >
@end lilypond

The most convenient way is to use the identifiers @code{\voiceOne}
through @code{\voiceFour}, which also set slur and tie directions in the
correct manner.

@lilypond[singleline, verbatim]
\relative c''
\context Staff < \context Voice = VA { \voiceOne cis2 b  }
  \context Voice = VB { \voiceThree b4 ais ~ ais4 gis4 } 
  \context Voice = VC { \voiceTwo fis4~  fis4 f ~ f  } >
@end lilypond


LilyPond also vertically shifts rests that are opposite of a stem. 

@lilypond[singleline,verbatim]
\context Staff <
\context Voice {  \stemUp c''4  }
\context Voice =VB { r4 }
>
@end lilypond

Note head collisions (horizontal shifting of note heads) are handled by
the @code{NoteCollision} grob. @code{RestCollision} handles vertical
shifting of rests.

@cindex @code{NoteCollision}
@cindex @code{RestCollision}


@refbugs

Resolving collisions is a very intricate subject, and LilyPond only
handles a few situations. When it can not cope, you are advised to use
@code{force-hshift} of the NoteColumn grob and @code{staff-position} of
the Rest grob to override typesetting decisions.

[TODO: doc merge-differently-dotted]

@node Beaming
@section Beaming

Beams are used to group short notes into chunks that are aligned with
the metrum.  LilyPond guesses where beams should be inserted, but if
you're not satisfied with the automatic beaming, you can either instruct
lilypond which patterns to beam automatically. In specific cases, you
can also specify explicitly what to beam and what not.


@c .    {Automatic beams}
@subsection Automatic beams

@cindex @code{Voice.autoBeamSettings}
@cindex @code{(end * * * *)}
@cindex @code{(begin * * * *)}

A large number of Voice properties are used to decide how to generate
beams.  Their default values appear in @file{scm/auto-beam.scm}.

By default, automatic beams can start on any note@footnote{In exotic
time signatures such as 1/8 and 1/16 this is not true} but can only end
in a few positions within the measure: they can end on a beat, or at
durations specified by the properties in
@code{Voice.autoBeamSettings}. The defaults for @code{autoBeamSettings}
are defined in @file{scm/auto-beam.scm}.

The syntax for  changing the value @code{autoBeamSettings} is set using
@code{\override} and unset using @code{\revert}:
@example
\property Voice.autoBeamSettings \override #'(@var{BE} @var{N} @var{M} @var{P} @var{Q}) = @var{dur}
\property Voice.autoBeamSettings \revert #'(@var{BE} @var{N} @var{M} @var{P} @var{Q})
@end example
Here, @var{BE} is the symbol @code{begin} or @code{end}. It determines
whether the rule applies to begin or end-points.  The quantity
@var{N}/@var{M} refers to a time signature (@code{* *} may be entered to
designate all time signatures), @var{P}/@var{Q} refers to the length of
the beamed notes (@code{* *} designate notes of any length).

If you want automatic beams to end on every  quarter note, you can
use the following:
@example
\property Voice.autoBeamSettings \override
    #'(end * * * *) = #(make-moment 1 4)
@end example
The duration a quarter note is 1/4 of a whole note. It is entered as
@code{(make-moment 1 4)}. 

The same syntax can be used to specify beam starting points. In this
example, you automatic beams can only end on a dotted quarter note. 
@example
\property Voice.autoBeamSettings \override
    #'(begin * * * *) = #(make-moment 3 8)
@end example
In 4/4 time signature, this means that automatic beams could end only on
3/8 and on the fourth beat of the measure (after 3/4, that is 2 times
3/8 has passed within the measure).

You can also restrict rules to specific time signatures. A rule that
should only be applied in @var{N}/@var{M} time signature is formed by
replacing the first asterisks by @var{N} and @var{M}. For example, a
rule for 6/8 time exclusively looks like
@example
\property Voice.autoBeamSettings \override
    #'(begin 6 8 * *) =  ... 
@end example

If you want a rule to apply to certain types of beams, you can use the
second pair of asterisks. Beams are classified according to the shortest
note they contain. For a beam ending rule that only applies to beams
with 32nd notes (and no shorter notes), you would use @code{(end * * 1
32)}.

[say something about irregular meters. eg 5/8 = 2+3/8, 3+2/8] 

Automatic beams can not be put on the last note in a score.

@cindex automatic beam generation
@cindex autobeam
@cindex @code{Voice.noAutoBeaming}

Automatic beaming is on by default, but it can switched off by setting
@code{Voice.noAutoBeaming} to true.  You you may find this necessary for
a melody that goes with lyrics.

@refbugs

It is not possible to specify beaming for beams with mixed durations,
that differs from the beaming of all separate durations, ie, you'll
have to specify manual beams to get:
@lilypond[fragment,singleline,relative]
  \property Voice.autoBeamSettings
    \override #'(end * * * *) = #(make-moment 3 8)
  \time 12/8; c'8 c c  c16 c c c c c  [c c c c] c8 c  c4
@end lilypond


@c .    {Manual beams}
@cindex Automatic beams
@subsection Manual beams
@cindex beams, manual
@cindex @code{]}
@cindex @code{[}

In some cases it may be necessary to override LilyPond's automatic
beaming algorithm.  For example, the auto beamer will not beam over
rests or bar lines, so if you want that, specify the begin and end point
manually using @code{[} and @code{]}:

@lilypond[fragment,relative,verbatim]
  \context Staff {
    r4 [r8 g'' a r8] r8 [g | a] r8
  }
@end lilypond
Whenever an manual beam is busy, the auto beam will not produce
anything.

@cindex @code{stemLeftBeamCount}

If you have specific wishes for the number of beams, you can fully
control the number of beams through the properties
@code{Voice.stemLeftBeamCount} and @code{Voice.stemRightBeamCount}.

@lilypond[fragment,relative,verbatim]
  \context Staff {
    [f'8 r16 f g a]
    [f8 r16 \property Voice.stemLeftBeamCount = #1 f g a]
  }
@end lilypond
@cindex @code{stemRightBeamCount}

The beam symbol can be tweaked through @code{Voice.Beam}'s
grob-properties @code{height} and @code{staff-position},
in staff-spaces.

Set @code{height} to zero, to get horizontal beams:

@lilypond[fragment,relative,verbatim]
  \property Voice.Beam \set #'direction = #1
  \property Voice.Beam \set #'height = #0
  [a''8 e' d c]
@end lilypond

Here's how you'd specify a weird looking beam that instead of being
horizontal, falls two staff spaces:

@lilypond[fragment,relative,verbatim]
  \property Voice.Beam \set #'staff-position = #2
  \property Voice.Beam \set #'height = #-2
  [c'8 c] 
@end lilypond
@cindex @code{default-neutral-direction}

@node Expressive marks
@section Expressive marks

@c .   {Slur}
@menu
* Slur ::                       
* Phrasing slur::               
* Breath marks::                
* Tempo::                       
* Text spanner::                
@end menu

@node Slur 
@subsection Slur
@cindex slur

A slur indicates that notes are to be played bound or @emph{legato}.  In
lilypond, they are entered using parentheses:
@lilypond[fragment,verbatim,center]
  f'()g'()a' [a'8 b'(] a'4 g'2 )f'4
@end lilypond


Slurs avoid crossing stems, and are attached to note heads whenever
possible.  In some instances involving beams slurs may be attached to a
stem end.  If you want to override this layout you can do this through
@code{Voice.Slur}'s grob-property @code{attachment}:

Maybe reinclude other slur features and move back to tricks?  Esp. the
second example, how to fix, can be very helpful.

@lilypond[fragment,relative,verbatim]
  \property Voice.Slur \set #'direction = #1
  \property Voice.Stem \set #'length = #5.5
  g''8(g)g4
  \property Voice.Slur \set #'attachment = #'(stem . stem)
  g8(g)g4
@end lilypond

If a slur would strike through a stem or beam, the slur will be moved
away upward or downward. If this happens, attaching the slur to the
stems might look better:

@lilypond[fragment,relative,verbatim]
  \property Voice.Stem \set #'direction = #1
  \property Voice.Slur \set #'direction = #1
  d'32( d'4 )d8..
  \property Voice.Slur \set #'attachment = #'(stem . stem)
  d,32( d'4 )d8..
@end lilypond


Similarly, the curvature of a slur is adjusted to stay clear of note
heads and stems.  When that would increase the curvature too much, the
slur is reverted to its default shape.  The threshold for this decision
is in @code{Voice.Slur}'s grob-property @code{beautiful}.  It is loosely
related to the enclosed area between the slur and the notes.  Usually,
the default setting works well, but in some cases you may prefer a
curved slur when LilyPond decides for a vertically moved one.  You can
express this by increasing the @code{beautiful} value:

@lilypond[verbatim,singleline,relative]
  \property Voice.Beam \override #'direction = #-1
  \property Voice.Slur \override #'direction = #1
  c'16( a' f' a a f a, )c,
  c( a' f' a a f d, )c
  \property Voice.Slur \override #'beautiful = #5.0
  c( a' f' a a f d, )c
@end lilypond

@refbugs

The definition for @code{beautiful} is vague, the default setting is
experimental computer science.

@cindex Adusting slurs

@node Phrasing slur
@subsection Phrasing slur

@cindex phrasing slur
@cindex phrasing mark

A phrasing slur (or phrasing mark) connects chords and is used to
indicate a musical sentence. It is entered using @code{\(} and
@code{\)}.

@lilypond[fragment,verbatim,center,relative]
  \time 6/4; c''\((d)e f(e)\)d
@end lilypond

Typographically, the phrasing slur behaves almost exactly like a normal
slur. The grob associated with it is @code{Voice.PhrasingSlur}.

@node Breath marks
@subsection Breath marks

Breath marks are entered using @code{\breathe}:

@lilypond[fragment,relative]
c'4 \breathe d4
@end lilypond

Currently, only tick marks are supported, comma style breath marks are
not. The grob for this object is called @code{Voice.BreathingSign}.


@refbugs

  Currently, only tick marks are supported, comma style breath marks are
not.


@c .  {Tempo}
@node Tempo
@subsection Tempo
@cindex Tempo
@cindex beats per minute
@cindex metronome marking

@cindex @code{\tempo}
@example
  \tempo @var{duration} = @var{perminute} @code{;}
@end example

Used to specify the tempo.  For example, @code{\tempo 4 = 76;} requests
output with 76 quarter notes per minute.
  
@refbugs
  
The tempo setting is not printed, but is currently only used in the MIDI
output.
  


@node Text spanner
@subsection Text spanner
@cindex Text spanner

Some textual indications, e.g. rallentando, accelerando, often extend
over a many measures. This is indicated by following the text with a
dotted line.   You can create such texts in LilyPond using 
text spanners. The syntax is as follows: 
@example
\spanrequest \start "text"
\spanrequest \stop "text"
@end example
LilyPond will respond by creating a @code{Voice.TextSpanner} grob.  The
string to be printed, as well as the style is set through grob
properties.

An application---or rather, a hack---is to fake octavation indications.
@lilypond[fragment,relative,verbatim]
 \relative c' {  a'''' b c a
  \property Voice.TextSpanner \set #'type = #'dotted-line
  \property Voice.TextSpanner \set #'edge-height = #'(0 . 1.5)
  \property Voice.TextSpanner \set #'edge-text = #'("8va " . "")
  \property Staff.centralCPosition = #-13
  a\spanrequest \start "text" b c a \spanrequest \stop "text" }
@end lilypond


@c .  {Ornaments}
@node Ornaments
@section Ornaments
@cindex Ornaments
@menu
* Articulation::                
* Text scripts::                
* Grace notes::                 
* Glissando ::                  
* Dynamics::                    
@end menu

@c .   {Articulation}
@node Articulation
@subsection Articulation
@cindex Articulation

@cindex articulations
@cindex scripts
@cindex ornaments

A variety of symbols can appear above and below notes to indicate
different characteristics of the performance.  These symbols can be
added to a note with `@var{note}@code{-\}@var{name}'.  Numerous symbols
are defined in @file{script.ly} and @file{script.scm}.  Symbols can be
forced to appear above or below the note by writing
`@var{note}@code{^\}@var{name}' and `@var{note}@code{_\}@var{name}'
respectively.  Here is a chart showing symbols above notes, with the
name of the corresponding symbol appearing underneath.

@lilypond[]
  \score {
    < \notes {
        \property Score.LyricSyllable \override #'font-family =
#'typewriter
        \property Score.LyricSyllable \override #'font-shape = #'upright
        c''-\accent      c''-\marcato      c''-\staccatissimo c''-\fermata 
        c''-\stopped     c''-\staccato     c''-\tenuto        c''-\upbow
        c''-\downbow     c''^\lheel        c''-\rheel         c''^\ltoe
        c''-\rtoe        c''-\turn         c''-\open          c''-\flageolet
        c''-\reverseturn c''-\trill        c''-\prall         c''-\mordent
        c''-\prallprall  c''-\prallmordent c''-\upprall       c''-\downprall
        c''-\thumb       c''-\segno        c''-\coda
      }
      \context Lyrics \lyrics {
        accent__      marcato__      staccatissimo__ fermata
        stopped__     staccato__     tenuto__        upbow
        downbow__     lheel__        rheel__         ltoe
        rtoe__        turn__         open__          flageolet
        reverseturn__ trill__        prall__         mordent
        prallprall__  prallmordent__ uprall__        downprall
        thumb__       segno__        coda
      }
    >
    \paper {
      linewidth = 5.875\in;          
      indent    = 0.0;
    }
  }
@end lilypond

To save typing work, some shorthands are available:
@lilypond[singleline]
  \score {
    \notes \context Voice {
      \property Voice.TextScript \set #'font-family = #'typewriter
      \property Voice.TextScript \set #'font-shape = #'upright
      c''4-._"c-."      s4
      c''4--_"c-{}-"    s4
      c''4-+_"c-+"      s4
      c''4-|_"c-|"      s4
      c''4->_"c->"      s4
      c''4-^_"c-\\^{ }" s4
    }
  }
@end lilypond

@cindex fingering

Fingering instructions can also be entered in  this shorthand.
@lilypond[verbatim, singleline, fragment]
      c'4-1 c'4-2 c'4-3 c'4-4
@end lilypond


@cindex @code{\script}
@cindex scripts
@cindex superscript
@cindex subscript

@example
  \script @var{alias}
@end example

Defines a script printing request.  The argument is a string which
points into the script-alias table defined in @file{scm/script.scm}.
Usually the @code{\script} keyword is not used directly.  Various
helpful identifier definitions appear in @file{script.ly}.

For information on how to add scripts, consult @file{scm/script.scm}.


@refbugs

All of these note ornaments appear in the printed output but have no
effect on the MIDI rendering of the music.

Unfortunately, there is no support adding fingering instructions or
ornaments to individual note heads. Some hacks exist, though. See
@file{input/test/script-horizontal.ly}.


@c .  {Text scripts}
@node Text scripts
@subsection Text scripts
@cindex Text scripts

In addition, it is possible to place arbitrary strings of text or markup
text (see @ref{Text markup}) above or below notes by using a string:
@code{c^"text"}.  The text is typeset in italic by default.

The amount of space taken by these indications by default does not
influence, spacing, but setting @code{Voice.textNonEmpty} to true will
take the widths into account.  The identifier @code{\fattext} is defined
in the standard  includes.
@lilypond[fragment,singleline,verbatim]
\relative c' { c4^"longtext" \fatText c4_"longlongtext" c4 }
@end lilypond

Text scripts are created in form of @code{Voice.TextScript} grobs.

For purposes of defining identifiers, a more verbose form also exists:

@example
  \textscript @var{text} 
@end example

Defines a text to be printed over or under a note.  @var{text} is a
string or  a markup text.
@quotation

@example 
foo = \textscript #'(finger "6")
  [..]
c4-\foo
@end example 

@end quotation

This is equivalent to @code{c4-6 c4-"foo"}.  


@c .   {Grace notes}
@node Grace notes
@subsection Grace notes







@cindex Grace music
@cindex @code{\grace}
@cindex ornaments
@cindex grace notes
@cindex @code{graceAlignPosition}

Grace notes are ornaments that are written out, but do not take up  any
logical time in a measure. LilyPond has limited support for grace notes.
The syntax is as follows. 
@example
  \grace @var{musicexpr}
@end example

When grace music is interpreted, a score-within-a-score is set up:
@var{musicexpr} has its own time bookkeeping, and you could (for
example) have a separate time signature within grace notes.  While in
this score-within-a-score, you can create notes, beams, slurs, etc.
Unbeamed eighth notes and shorter by default have a slash through the
stem.  This behavior can be controlled with the
@code{Stem}.@code{flag-style} property.

@lilypond[fragment,verbatim]
\relative c'' {
  \grace c8 c4 \grace { [c16 c16] } c4
  \grace { \property Grace.Stem \override #'flag-style = #'() c16 } c4
}
@end lilypond


At present, nesting @code{\grace} notes is not supported. The following
may cause run-time errors:
@example
  @code{\grace @{ \grace c32 c16 @} c4}
@end example
Since the meaning of such a construct is unclear, we don't consider this
a loss.  Similarly, juxtaposing two @code{\grace} sections is
syntactically valid, but makes no sense and may cause runtime errors.
Ending a staff or score with grace notes may also generate a run-time
error, since there will be no main note to attach the grace notes to.


A grace note expression has duration 0; the next real note is assumed to
be the main note. If you want the note to appear after the main note,
set @code{Voice.graceAlignPosition} to @code{1}.

@refbugs

The present implementation of grace notes is not robust and generally
kludgy. We expect it to change after LilyPond 1.4. Syntax changes might
also be implemented.



@menu
* Glissando ::                  
* Dynamics::                    
@end menu



@c .   {Glissando}
@node Glissando 
@subsection Glissando
@cindex Glissando 

@cindex @code{\glissando}

A glissando line can be requested by attaching a @code{\glissando} to a
note:

@lilypond[fragment,relative,verbatim]
  c'' \glissando c'
@end lilypond

@refbugs

Printing of an additional text (such as @emph{gliss.}) must be done
manually.



@c .   {Dynamics}
@node Dynamics
@subsection Dynamics
@cindex Dynamics



@cindex @code{\ppp}
@cindex @code{\pp}
@cindex @code{\p}
@cindex @code{\mp}
@cindex @code{\mf}
@cindex @code{\f}
@cindex @code{\ff}
@cindex @code{\fff}
@cindex @code{\ffff}
@cindex @code{\fp}
@cindex @code{\sf}
@cindex @code{\sff}
@cindex @code{\sp}
@cindex @code{\spp}
@cindex @code{\sfz}
@cindex @code{\rfz}


Absolute dynamic marks are specified by using an identifier after a
note: @code{c4-\ff}.  The available dynamic marks are: @code{\ppp},
@code{\pp}, @code{\p}, @code{\mp}, @code{\mf}, @code{\f}, @code{\ff},
@code{\fff}, @code{\fff}, @code{\fp}, @code{\sf}, @code{\sff},
@code{\sp}, @code{\spp}, @code{\sfz}, and @code{\rfz}.

@lilypond[verbatim,singleline,fragment,relative]
  c''\ppp c\pp c \p c\mp c\mf c\f c\ff c\fff
  c2\sf c\rfz
@end lilypond

@cindex Crescendo and Decrescendo
@cindex crescendo
@cindex @code{\cr}
@cindex @code{\rc}
@cindex @code{\decr}
@cindex @code{\rced}
@cindex @code{\<}
@cindex @code{\>}
@cindex @code{\"!}


A crescendo mark is started with @code{\cr} and terminated with
@code{\rc} (the textual reverse of @code{cr}).  A decrescendo mark is
started with @code{\decr} and terminated with @code{\rced}.  There are
also shorthands for these marks.  A crescendo can be started with
@code{\<} and a decrescendo can be started with @code{\>}.  Either one
can be terminated with @code{\!}.  Note that @code{\!}  must go before
the last note of the dynamic mark whereas @code{\rc} and @code{\rced} go
after the last note.  Because these marks are bound to notes, if you
want to get several marks during one note, you must use spacer notes.

@lilypond[fragment,verbatim,center]
  c'' \< \! c''   d'' \decr e'' \rced 
  < f''1 { s4 s4 \< \! s4 \> \! s4 } >
@end lilypond

You can also use a text saying @emph{cresc.} instead of hairpins. Here
is an example how to do it:

@lilypond[fragment,relative,verbatim]
  \context Voice {
    \property Voice.crescendoText = "cresc."
    \property Voice.crescendoSpanner = #'dashed-line
    a''2\mf\< a a \!a 
  }
@end lilypond


@refbugs

When using spacer notes to subdivide note dynamics and @code{linewidth =
-1}, starting a hairpin on the first spacer note (the one coinciding
with the real note) exposes an embarassing bug.



@c .  {Repeats}
@node Repeats
@section Repeats


@cindex repeats
@cindex @code{\repeat}

To specify repeats, use the @code{\repeat} keyword.  Since repeats
should work differently when played or printed, there are a few
different variants of repeats.

@table @asis
@item unfolded  
Repeated music is fully written (played) out.  Useful for MIDI
output.

@item volta  
This is the normal notation: Repeats are not written out, but
alternative endings (voltas) are printed, left to right.

@item folded  
Alternative endings are written stacked.  Which is unfortunately not
practical for anything right now.

@item tremolo
Make tremolo beams.

@item percent
Make  measure repeats. These look like percent signs.

@end table  

@menu
* Repeat syntax::               
* Manual repeat commands::      
* Tremolo repeats::             
* Tremolo subdivision::         
* Measure repeats::             
@end menu

@node Repeat syntax
@subsection Repeat syntax

The syntax for repeats is

@example
  \repeat @var{variant} @var{repeatcount} @var{repeatbody}
@end example

If you have alternative endings, you may add
@cindex @code{\alternative}
@example
 \alternative @code{@{} @var{alternative1}
            @var{alternative2}
            @var{alternative3} @dots{} @code{@}}
@end example
where each @var{alternative} is a music expression.

Normal notation repeats are used like this:
@lilypond[fragment,verbatim]
  c'1
  \repeat volta 2 { c'4 d' e' f' }
  \repeat volta 2 { f' e' d' c' }
@end lilypond

With alternative endings:
@lilypond[fragment,verbatim]
  c'1
  \repeat volta 2 {c'4 d' e' f'} 
  \alternative { {d'2 d'} {f' f} }
@end lilypond

Folded repeats look like this:@footnote{Folded repeats offer little
more over simultaneous music.  However, it is to be expected that
more functionality -- especially for the MIDI backend -- will be
implemented at some point in the future.}

@lilypond[fragment,verbatim]
  c'1
  \repeat fold 2 {c'4 d' e' f'} 
  \alternative { {d'2 d'} {f' f} }

@end lilypond

If you don't give enough alternatives for all of the repeats, then
the first alternative is assumed to be repeated often enough to equal
the specified number of repeats.

@lilypond[fragment,verbatim]
\context Staff {
  \relative c' {
    \partial 4;
    \repeat volta 3 { e | c2 d2 | e2 f2 | }
    \alternative { { g4 g g } { a | a a a a | b2. } }
  }
}
@end lilypond

@refbugs

As you can see, LilyPond doesn't remember the timing information, nor
are slurs or ties repeated, so you have to reset timing information
after a repeat, e.g. using a bar-check (See @ref{Bar check}),
@code{Score.measurePosition} or @code{\partial}. We hope to fix this
after 1.4.

It is possible to nest @code{\repeat}, although it probably is only
meaningful for unfolded repeats.

@node Manual repeat commands
@subsection Manual repeat commands

@cindex @code{repeatCommands}

The property @code{repeatCommands} can be used to control the layout of
repeats. Its value is a Scheme list of repeat commands, where each repeat
command can be

@table @code
@item 'start-repeat
 Print a |: bar line
@item 'stop-repeat
 Print a :| bar line
@item (volta . @var{text})
  Print a volta bracket saying @var{text}.
@item (volta . #f)
  Stop a running volta bracket
@end table

@lilypond[verbatim, fragment]
 c''4
    \property Score.repeatCommands = #'((volta "93") end-repeat)
 c4 c4
    \property Score.repeatCommands = #'((volta #f))
 c4 c4
@end lilypond


@node Tremolo repeats
@subsection Tremolo repeats
@cindex tremolo beams

To place tremolo marks between notes, use @code{\repeat} with tremolo
style.  
@lilypond[verbatim,center,singleline]
\score { 
  \context Voice \notes\relative c' {
    \repeat "tremolo" 8 { c16 d16 }
    \repeat "tremolo" 4 { c16 d16 }    
    \repeat "tremolo" 2 { c16 d16 }
    \repeat "tremolo" 4 c16
  }
}
@end lilypond

@refbugs


At present, the spacing between tremolo beams is not regular, since the
spacing engine does not notice that not all notes are printed.

@node Tremolo subdivision
@subsection Tremolo subdivision
@cindex tremolo marks
@cindex @code{tremoloFlags}

Tremolo marks can be printed on a single note by adding
`@code{:}[@var{length}]' after the note.  The length must be at least 8.
A @var{length} value of 8 gives one line across the note stem.  If the
length is omitted, then then the last value (stored in
@code{Voice.tremoloFlags}) is used.

@lilypond[verbatim,fragment,center]
  c'2:8 c':32
@end lilypond
Using this mechanism pays off when you entering many tremolos, since the
default argument saves a lot of typing.

@refbugs


Tremolos in this style do not carry over into the MIDI output.


@node Measure repeats
@subsection Measure repeats

@cindex percent repeats
@cindex measure repeats

In the @code{percent} style, a note pattern can be repeated. It is
printed once, and then the pattern is replaced with a special sign.

@lilypond[verbatim,singleline]
 \context Voice { \repeat  "percent" 4  { c'4 }
    \repeat "percent" 2 { c'2 es'2 f'4 fis'4 g'4 c''4 }
}
@end lilypond

@refbugs

You can not nest percent repeats, filling in the first measure with
slashes, and repeating that measure with percents.

@node Rhythmic music
@section Rhythmic music


@menu
* Rhythmic staffs::             
@end menu

@node Rhythmic staffs
@subsection Rhythmic staffs

Some times you might want to show only the rhythm of a melody.  This can
be done  with the rhythmic staff. All pitches of notes on such a staff
are squashed, and the  staff itself  looks has  a single staff line:

@lilypond[fragment,relative ]
  \context RhythmicStaff {
      \time 4/4; 
      c4 e8 f  g2 | r4 g r2 | g1:32 | r1 |
  }
@end lilypond


@c . {Piano music}
@node Piano music
@section Piano music

Piano music is an odd type of notation: two staffs are largely
independent, but sometimes voices can cross between the two staffs.  The
@code{PianoStaff} is especially built to handle this cross-staffing
behavior.  In this section we discuss the @code{PianoStaff} and some
other pianistic peculiarities.

@menu
* Automatic staff changes::     
* Manual staff switches::       
* Pedals::                      
* Arpeggio::                    
* VoiceFollower::               
@end menu 


@c .   {Automatic staff changes}
@node Automatic staff changes
@subsection Automatic staff changes
@cindex Automatic staff changes

Voices can be switched from top to bottom staff automatically. The
syntax for this is
@example
        \autochange @var{contexttype} @var{musicexp}
@end example
This will switch notation context of @var{musicexp} between a
@var{contexttype} named @code{up} and @code{down}. Typically, you use
@code{Staff} for @var{contexttype}.  The autochanger switches on basis
of pitch (central C is the turning point), and it looks ahead skipping
over rests to switch rests in advance.
        
@lilypond[verbatim,singleline]
\score { \notes \context PianoStaff <
  \context Staff = "up" {
    \autochange Staff \context Voice = VA < \relative c' {
       g4 a  b c d r4 a g } > }
  \context Staff = "down" {
       \clef bass; 
       s1*2
} > }
@end lilypond

Note how spacer rests are used to prevent the bottom staff from
terminating too soon.


@node Manual staff switches
@subsection Manual staff switches

@cindex manual staff switches
@cindex staff switch, manual

Voices can be switched between staffs manually, using the following command:
@example
  \translator Staff = @var{which} @var{music}
@end example
The string @var{which} is the name of the staff. Typically it is
@code{"up"} or @code{"down"}. 

Formally, this construct is a music expression indicating that the
context which is a direct child of the a context of type
@var{contexttype} should be shifted to a context of type
@var{contexttype} and the specified name.

@cindex @code{\translator}
@example
  \translator @var{contexttype} = @var{name}
@end example


@c .   {Pedals}
@node Pedals
@subsection Pedals
@cindex Pedals

Piano pedal instruction can be expressed using 
@code{\sustainDown}, @code{\sustainUp}, @code{\unaChorda},
@code{\treChorde}, @code{\sostenutoDown} and @code{\sostenutoUp}.

These identifiers are shorthands for spanner commands of the types
@code{Sustain}, @code{UnaChorda} and @code{Sostenuto}:

@lilypond[fragment,verbatim]
c''4 \spanrequest \start "Sustain" c''4 c''4 \spanrequest \stop "Sustain"
@end lilypond

The symbols that are printed can be modified by setting
@code{pedal@var{X}Strings}, where @var{X} is one of the pedal
types. Refer to the generated documentation for more information.

@refbugs


Currently, brackets are not supported, only text markings (ie. *Ped
style).


@c .   {Arpeggio}
@node Arpeggio
@subsection Arpeggio
@cindex Arpeggio

@cindex broken arpeggio
@cindex @code{\arpeggio}

You can specify an arpeggio sign on a chord by attaching an
@code{\arpeggio} to a note of the chord.


@lilypond[fragment,relative,verbatim]
  \context Voice <c'\arpeggio e g c>
@end lilypond

When an arpeggio crosses staffs in piano music, you attach an arpeggio
to the chords in both staffs, and set
@code{PianoStaff.connectArpeggios}.

@lilypond[fragment,relative,verbatim]
  \context PianoStaff <
    \property PianoStaff.connectArpeggios = ##t
    \context Voice = one  { <c''\arpeggio e g c> }
    \context Voice = other { \clef bass;  <c,,\arpeggio e g>}
  >  
@end lilypond

This command creates @code{Arpeggio} grobs. 

@refbugs

 It is not possible to mix
connected arpeggios and unconnected arpeggios at the same time.


@c .    {VoiceFollower}
@node  VoiceFollower
@subsection VoiceFollower

@cindex follow voice
@cindex staff switching
@cindex cross staff

@cindex @code{followVoice}

Whenever a voice switches to another staff a line connecting the notes
can be printed automatically. This is enabled if the property
@code{PianoStaff.followVoice} is set to true:

@lilypond[fragment,relative,verbatim]
  \context PianoStaff <
    \property PianoStaff.followVoice = ##t
    \context Staff \context Voice {
      c'1
      \translator Staff=two
      b2 a
    }
    \context Staff=two {\clef bass; \skip 1*2;}
  >  
@end lilypond


@c . {Lyrics}
@node Lyrics
@section Lyrics


@menu
* Lyrics mode::                 
* Printing lyrics::             
* Automatic syllable durations::  
* More stanzas::                
@end menu

@c .  {Lyrics mode}
@node Lyrics mode
@subsection Lyrics mode
@cindex Lyrics mode

To print lyrics in LilyPond, you must first make a music expression from
the lyric text. When they're in a music expression, that music
expression can be printed by selecting an appropriate context.  We shall
discuss lyric printing in this order.


@cindex lyric mode
@cindex @code{\lyrics}

You can enter lyrics in a special input mode of LilyPond. This mode is
called Lyrics mode, and it is introduced by the keyword @code{\lyrics}.
The purpose of this mode is that you can enter lyrics as plain text,
punctuation and accents without any hassle.

The precise definition of this mode is in @ref{Lyrics mode
definition}. The definition itself is ludicrous, and this will remain so
until the authors of LilyPond acquire a deeper understanding of
character encoding, or someone else steps up to fix this.

Syllables are entered like notes, with pitches replaced by text.  For
example, @code{Twin- kle twin- kle} enters four syllables.  Note that
the hyphen has no special meaning for lyrics, and does not introduce
special symbols.

Spaces can be introduced into a lyric either by using quotes
(@code{"}) or by using an underscore without quotes: @code{He_could4
not4}.  All unquoted underscores are converted to spaces.

@c .  {Printing lyrics}
@node Printing lyrics
@subsection Printing lyrics
@cindex lyrics

Normally, you will want to have notes and syllables matched
automatically. This is accomplished using @code{\addlyrics}, which is
documented in @ref{Automatic syllable durations}. Setting
@code{automaticMelismata} in the melody staff, will cause slurs to be
interpreted as melismata. Lyric syllables must be interpreted within a
@code{Lyrics} context in order to printing them.

@lilypond[verbatim,singleline]
\addlyrics \notes \relative c' {
        \time 7/4;
        \property Staff.automaticMelismata = ##t
        d'2 c4 b2 a2
        b2 c4 b4 ()  a4 g2 }
    \context Lyrics \lyrics { 
       Join us now and
       share the so -- ftware; }
@end lilypond

@cindex extender
@cindex lyric extender
@cindex melisma

As you can see, extender lines are entered as @code{__}.  This will
create an extender, a line that extends over the entire duration of the
lyric.  This line will run all the way to the start of the next lyric,
so you may want to shorten it by using a blank lyric (using @code{_}).

@cindex hyphen

If you want to have hyphens centered between syllables (rather than
attached to the end of the first syllable) you can use the special
`@code{-}@code{-}' lyric as a separate word between syllables.  This
will result in a hyphen which length varies depending on the space
between syllables, and which will be centered between the syllables. 

@cindex Lyric hyphen

@node Automatic syllable durations
@subsection Automatic syllable durations
@cindex Automatic syllable durations

@cindex automatic lyric durations
@cindex @code{\addlyrics}

If you have lyrics that are set to a melody, you can import the rhythm
of that melody into the lyrics using @code{\addlyrics}.  The syntax for
this is
@example
  \addlyrics @var{musicexpr1 musicexpr2}
@end example

This means that both @var{musicexpr1} and @var{musicexpr2} are
interpreted, but that every non-command atomic music expression
(``every syllable'') in @var{musicexpr2} is interpreted using timing
of @var{musicexpr1}.
@cindex @code{automaticMelismata}

If the property @code{automaticMelismata} is set in the
context of @var{musicexpr1}, no lyrics will be put on slurred or tied
notes.

@lilypond[verbatim,fragment]
\addlyrics
\transpose c'' {
  \property Voice.automaticMelismata = ##t
  c8 () cis d8. e16 f2
}
\context Lyrics \lyrics {
 do4 re mi fa }
@end lilypond

If you want the lyric lines to be above the melody staff, or in some
other, more complex configuration, then build that configuration first
using simultaneous music, and use @code{\addlyrics} after that.

@lilypond[verbatim, singleline]
\notes <
  \context Lyrics = LA { s1 }
  \context Staff = SA { s1 }
  \addlyrics
        \context Staff = SA \relative c' { c4 e g g }
        \context Lyrics  = LA \lyrics { geen ge -- don -- der } >
@end lilypond

For @code{\addlyrics} you should use a single rhythm melody, and single
rhythm lyrics (a constant duration is the obvious choice).  If you do
not, you can get undesired effects when using multiple stanzas:

@lilypond[verbatim,fragment]
\addlyrics
\transpose c'' {
  c8 () cis d8. e16 f2
}
\context Lyrics \lyrics
< { do4 re mi fa }
  { do8 re mi fa } >
@end lilypond

It is valid (but probably not very useful) to use notes instead of
lyrics for @var{musicexpr2}.

@node More stanzas
@subsection More stanzas

@cindex phrasing

If you have multiple stanzas printed underneath each other, the separate
syllables should be aligned around punctuation. LilyPond can do this if
you explain it which lyric lines belong to which melody.

To this end, give the Voice context an identity, and set the LyricsVoice
to name starting with that identity. In the following example, the Voice
identity is @code{duet}, and the identities of the LyricsVoices are
@code{duet-1} and @code{duet-2}.


@lilypond[singleline,verbatim]
\score {
\addlyrics
  \notes \relative c'' \context Voice = duet { \time 3/4;
     g2 e4 a2 f4 g2.  }
  \lyrics \context Lyrics <
  \context LyricsVoice = "duet-1" {
    \property LyricsVoice . stanza = "Bert"
    Hi, my name is bert.    }
  \context LyricsVoice = "duet-2" {
    \property LyricsVoice . stanza = "Ernie" 
    Ooooo, ch\'e -- ri, je t'aime. }
  >
}
@end lilypond

You can add stanza numbers by setting @code{LyricsVoice.Stanza} (for the
first system) and @code{LyricsVoice.stz} for the following systems.

@cindex stanza numbering


@c . {Chords}
@node Chords
@section Chords
@cindex Chords

LilyPond has support for both entering and printing chords.  Chords are
a harmonic device that is characterized by a set of pitches. It is
something different from simultaneous music, although you can express a
chord using simultaneous music. In fact, chords are internally stored as
simultaneous music expressions. This means you can enter chords by name,
and print them as note head, or enter as notes and print as chord names:


@lilypond[verbatim,singleline]
twoWays = \notes \transpose c'' {
  \chords {
    c1 f:sus4 bes/f
  }
  <c e g>
  <f bes c'>
  <f bes d'>
  }

\score {
   < \context ChordNames \twoWays
     \context Staff \twoWays > }
@end lilypond

Note that this example also shows that the LilyPond chord does not
attempt to be intelligent, if you enter @code{f bes d}, it does no
attempt to find out whether it this is an inversion.

@menu
* Chords mode::                 
* Printing named chords::       
@end menu

@c .  {Chords mode}
@node Chords mode
@subsection Chords mode
@cindex Chords mode

Chord mode is a mode where you can input sets of pitches using common
names.  It is introduced by the keyword @code{\chords}.  It is similar
to note mode, but words are also looked up in a chord modifier table
(containing @code{maj}, @code{dim}, etc).

Dashes and carets are used to indicate chord additions and subtractions,
so articulation scripts can not be entered in Chord mode.

The syntax for named chords is as follows:
@example

  @var{tonic}[@var{duration}][@code{-}@var{modifiers}][@code{^}@var{subtractions}][@code{/}@var{inversion}][@code{/+}@var{bass}].
@end example

@var{tonic} should be the tonic note of the chord, and @var{duration}
is the chord duration in the usual notation.  There are two kinds of
modifiers.  One type is @emph{chord additions}, which are obtained by
listing intervals separated by dots.  An interval is written by its
number with an optional @code{+} or @code{-} to indicate raising or
lowering by half a step.  Chord additions has two effects: It adds
the specified interval and all lower odd numbered intervals to the
chord, and it may lower or raise the specified interval.  Intervals
must be separated by a dot (@code{.}).

Throughout these examples, chords have been shifted around the staff
using @code{\transpose}.

@lilypond[fragment,verbatim]
\transpose c'' {
  \chords {
    c1  c:3-       c:7     c:8
    c:9 c:9-.5+.7+ c:3-.5- c:4.6.8
  }
}
@end lilypond

@cindex @code{aug}
@cindex @code{dim}
@cindex @code{maj}
@cindex @code{sus}

The second type of modifier that may appear after the @code{:} is a
named modifier.  Named modifiers are listed in the file
@file{chord-modifiers.ly}.  The available modifiers are @code{m} and
@code{min} which lower the 3rd half a step, `@code{aug}' which
raises the 5th, `@code{dim}' which lowers the 5th,
`@code{maj}' which adds a raised 7th, and `@code{sus}'
which replaces the 5th with a 4th.

@lilypond[fragment,verbatim]
\transpose c'' {
  \chords {
    c1:m c:min7 c:maj c:aug c:dim c:sus
  }
}
@end lilypond
 

Chord subtractions are used to eliminate notes from a chord.  The
notes to be subtracted are listed after a @code{^} character,
separated by dots.

@lilypond[fragment,verbatim,center]
  \transpose c'' {
    \chords {
      c1^3 c:7^5.3 c:8^7
    }
  }
@end lilypond 
@cindex @code{/}

Chord inversions can be specified by appending `@code{/}' and the name
of a single note to a chord.  In a chord inversion, the inverted note is
transposed down until it is the lowest note in the chord.  If the
specified note is not in the chord, a warning will be printed.

@lilypond[fragment,verbatim,center]
  \transpose c''' {
    \chords {
      c1 c/e c/g c:7/e
    }
  }

@end lilypond 
@cindex @code{/+}

Bass notes can be added by `@code{/+}' and
the name of a single note to a chord.  This has the effect of
adding the specified note to the chord, lowered by an octave,
so it becomes the lowest note in the chord.

@lilypond[fragment,verbatim,center]
  \transpose c''' {
    \chords {
      c1 c/+c c/+g c:7/+b
    }
  }

@end lilypond 




@c .  {Printing named chords}
@node Printing named chords
@subsection Printing named chords

@cindex printing chord names
@cindex chord names
@cindex chords
@cindex @code{ChordNames}


For displaying printed chord names, use the @code{ChordNames} context.
The chords may be entered either using the notation described above, or
directly using simultaneous music.

@lilypond[verbatim,singleline]
scheme = \notes {
  \chords {a1 b c} <d f g>  <e g b>
}
\score {
  \notes<
    \context ChordNames \scheme
    \context Staff \transpose c'' \scheme
  >
}
@end lilypond

You can make the chord changes stand out by setting property
@code{ChordNames.chordChanges} to true.  This will only display chord
names when there's a change in the chords scheme and at the start of the
line.

@lilypond[verbatim]
scheme = \chords {
  c1:m \break c:m c:m c:m d
}

\score {
  \notes <
    \context ChordNames {
        \property ChordNames.chordChanges = ##t
        \scheme }
    \context Staff \transpose c'' \scheme
  > }
@end lilypond

LilyPond examines chords specified as lists of notes to determine a name
to give the chord. LilyPond will not try to identify chord inversions or
added base, which may result in strange chord names when chords are
entered as a list of pitches:

[base vs. bass ?]

@lilypond[verbatim,center,singleline]
scheme = \notes {
  <c'1 e' g'>
  <e' g' c''>
  <e e' g' c''>
}

\score {
  <
    \context ChordNames \scheme
    \context Staff \scheme
  >
}
@end lilypond


By default LilyPond uses chord name system proposed by Harald Banter
(See @ref{Literature}). The system is is unambiguous and has a logical
structure.  Typical American style chord names may be selected by
setting the @code{style} property of the @code{ChordNames.ChordName}
grob to @code{'american}. Similarly @code{'jazz} selects Jazz
chordnames.

Routines that determine the names to be printed are written in Scheme,
and may be customized by the user.  The code can be found in
@file{scm/chord-name.scm}.

[3 short examples showing differences between american, banter and jazz]

@node Writing parts
@section Writing parts

Orchestral music involves some special notation, both in the full score,
as in the individual parts. This section explains how to tackle common
problems in orchestral music.


@c .  {Transpose}
@menu
* Rehearsal marks::             
* Bar numbers::                 
* Instrument names::            
* Transpose::                   
* Sound output for transposing instruments::  
* Multi measure rests::         
* Automatic part combining::    
* Hara-kiri staffs::            
@end menu

@c .   {Rehearsal marks}
@node Rehearsal marks
@subsection Rehearsal marks
@cindex Rehearsal marks
@cindex mark
@cindex @code{\mark}
@cindex @code{Mark_engraver}

@example
  \mark @var{unsigned};
  \mark @var{string};
  \mark ; 
@end example

With this command, you can print a rehearsal mark above the system. You
can provide a number, a string or a markup text as argument. If there is
no argument, the property @code{rehearsalMark} is used and automatically
incremented.

@lilypond[fragment,verbatim]
\relative c'' {
  c1 \mark "A2";
  c1 \mark ; 
  c1 \mark ; 
  c1 \mark "12";
  c1 \mark #'(music "scripts-segno") ;
  c1
}
@end lilypond

@node Bar numbers
@subsection Bar numbers

Bar numbers are printed at the start of the line by default. This is
done by the @code{Bar_number_engraver} in the Score context.

@refbugs

It is currently not possible to make boxed bar numbers, or print them at
regular intervals.


@node Instrument names
@subsection Instrument names

You can specify an instrument name for a staff by setting
@code{Staff.instrument} and @code{Staff.instr}. This will print a string
before the start of the staff. For the first start, @code{instrument} is
used, for the next ones @code{instr} is used.

@lilypond[verbatim,singleline]
\score { \notes {
  \property Staff.instrument = "ploink " { c''4 } }
  \paper { 
    \translator { \StaffContext
    \consists "Instrument_name_engraver"; } } }
@end lilypond

This requires  that you add the @code{Instrument_name_engraver} to the
staff context. You can also  use markup texts:


@lilypond[verbatim,singleline]
#(define text-flat
  '((font-relative-size . -2 ) (music "accidentals--1")))

\score { \notes {
  \property Staff.instrument = #`((kern . 0.5) (lines
    "2 Clarinetti" (rows "     (B" ,text-flat ")")))
    c'' 4 }
  \paper { 
    \translator { \StaffContext
    \consists "Instrument_name_engraver"; } } }
@end lilypond


@node Transpose
@subsection Transpose
@cindex Transpose
@cindex transposition of pitches
@cindex @code{\transpose}

A music expression can be transposed with @code{\transpose}.  The syntax
is
@example
  \transpose @var{pitch} @var{musicexpr}
@end example

This means that middle C in @var{musicexpr} is transposed to
@var{pitch}.

@code{\transpose} distinguishes between enharmonic pitches: both
@code{\transpose cis'} or @code{\transpose des'} will transpose up half
a tone.  The first version will print sharps and the second version
will print flats.

@lilypond[fragment,verbatim]
\context Staff {
  \clef "F";
  { \key e \major; c d e f }
  \clef "G";
  \transpose des'' { \key e \major; c d e f }
  \transpose cis'' { \key e \major; c d e f }
}
@end lilypond

If you want to use both @code{\transpose} and @code{\relative}, then
you must use @code{\transpose} first.  @code{\relative} will have no
effect music that appears inside a @code{\transpose}.

@node Sound output for transposing instruments
@subsection Sound output transposing instruments

When you want to play a score containing transposed and untransposed
instruments, you have to instruct LilyPond the pitch offset (in
semitones) for the transposed instruments. This is done using the
@code{transposing} property. It does not affect printed output.

@cindex @code{transposing}

@example
        \property Staff.instrument = #"Cl. in B-flat"
        \property Staff.transposing = #-2
@end example


@c .  {Multi measure rests}
@node  Multi measure rests
@subsection Multi measure rests
@cindex Multi measure rests

@cindex @code{R}

Multi measure rests are entered using `@code{R}'. It is specifically
meant for entering parts: the rest can expand to fill a score with
rests, or it can be printed as a single multimeasure rest This expansion
is controlled by the property @code{Score.skipBars}. If this is set to true,
Lily will not expand empty measures, and the appropriate number is added
automatically.

@lilypond[fragment,verbatim]
 \time 3/4; R2.*2 \property Score.skipBars = ##t R2.*17  R2.*4
@end lilypond

Currently, there is  no way to condense multiple rests into a single
multimeasure rest.

@cindex condensing rests

@node Automatic part combining
@subsection Automatic part combining
@cindex automatic part combining
@cindex part combiner

Automatic part combining is used to merge two parts of music onto on
staff in an intelligent way.  It is aimed primarily at typesetting Hymns
and orchestral scores.  When the two parts are identical for a period of
time, only one is shown.  In places where the two parts differ, stem
directions are set automatically.  Also, soli and @emph{a due} parts can be
identified and marke.

The syntax for part combining is

@example
  \partcombine @var{context} @var{musicexpr1} @var{musicexpr2}
@end example

where the pieces of music @var{musicexpr1} and @var{musicexpr2} will be
combined into one context @var{context}.  The names of the music
expressions must start with the prefixes @code{one} and @code{two}.

[Name of music expressions?  is that context name? ]

The most useful function of the part combiner to combining threads into
one voice, as common for wind parts in orchestral scores:

@lilypond[verbatim,singleline,fragment]
  \context Staff <
    \context Voice=one \partcombine Voice
      \context Thread=one \relative c'' {
        g a b r
      }
      \context Thread=two \relative c'' {
        g r2 f4
      }
  >
@end lilypond

Notice that the first @code{g} appears only once, although it was
specified twice (once in each Thread). Also note that stem, slur and tie
directions are set automatically, depending whether there is a solo or
unisono. The Thread called @code{one} always gets up stems, and "solo",
while @code{two} always gets down stems and "Solo II".

If you just want the splitting of Threads and setting of directions, and
not the textual markings, you may set the property @var{soloADue} to
false.  This mode can be used to set hymns:

@lilypond[verbatim,singleline,fragment]
  \context Staff <
    \property Staff.soloADue = ##f
    \context Voice=one \partcombine Voice
      \context Thread=one \relative c'' {
        b4 a c g
      }
      \context Thread=two \relative c'' {
        d,2 a4 g'
      }
  >
@end lilypond

There are a number of other properties that you can use to tweak
the behavior of part combining, refer to the automatically generated
documentation. Look for @code{Thread_devnull_engraver}
@code{Voice_devnull_engraver} and @code{A2_engraver}. 

@refbugs

In @code{soloADue} mode, when the two voices play the same notes on and
off, the part combiner may typeset @code{a2} more than once in a
measure.

@lilypond[fragment,singleline]
  \context Staff <
    \context Voice=one \partcombine Voice
      \context Thread=one \relative c'' {
        c b c b c a c a
      }
      \context Thread=two \relative c'' {
        b b b b f a f a
      }
  >
@end lilypond

@cindex @code{Thread_devnull_engraver}
@cindex @code{Voice_engraver}
@cindex @code{A2_engraver}

@node Hara-kiri staffs
@subsection Hara-kiri staffs

In orchestral scores, staffs that only have rests are usually removed.
This saves some space.  LilyPond also supports this through the
hara-kiri@footnote{Hara kiri, also called Seppuku, is the ritual suicide
of the Japanese Samourai warriors.} staff. This staff commits suicide
when it finds itself to be empty after the line-breaking process---note
that it will not disappear when it contains normal rests, you must use
multi measure rests.

The hara kiri staff is specialized version of the Staff context. It is
available as the context identifier @code{\HaraKiriStaffContext}.
Observe how the second staff in this example disappears in the second
line.

@lilypond[verbatim]
\score  {
  \notes \relative c' <
    \context Staff = SA { e4 f g a \break c1 }
    \context Staff = SB { c4 d e f \break R1 }
  >
  \paper {
    linewidth = 6.\cm ; 
    \translator { \HaraKiriStaffContext }
  }
}
@end lilypond



@c . {Custodes}
@node Custodes
@section Custodes
@cindex Custos
@cindex Custodes

A @emph{custos} (plural: @emph{custodes}; latin word for "guard") is a
staff context symbol that appears at the end of a staff line.  It
anticipates the pitch of the first note(s) of the following line and
thus helps the player or singer to manage line breaks during
performance, thus enhancing readability of a score.

@lilypond[verbatim]
\score {
  \notes { c'1 d' e' d' \break c' d' e' d' }
  \paper {
    \translator {
      \StaffContext
      \consists Custos_engraver;
      Custos \override #'style = #'mensural;
    }
  }
}
@end lilypond

Custodes were frequently used in music notation until the 16th century.
There were different appearences for different notation styles.
Nowadays, they have survived only in special forms of musical notation
such as via the @emph{editio vaticana} dating back to the beginning of
the 20th century.

For typesetting custodes, just put a @code{Custos_engraver} into the
@code{StaffContext} when declaring the @code{\paper} block.  In this
block, you can also globally control the appearance of the custos symbol
by setting the custos @code{style} property.  Currently supported styles
are @code{vaticana}, @code{medicaea}, @code{hufnagel} and
@code{mensural}.

@example
\paper @{
  \translator @{
      \StaffContext
      \consists Custos_engraver;
      Custos \override #'style = #'mensural;
  @}
@}
@end example

The property can also be set locally, for example in a @code{\notes}
block:

@example
\notes @{
  \property Staff.Custos \override #'style = #'vaticana
  c'1 d' e' d' \break c' d' e' d'
@}
@end example

@c . {Tuning output}
@node Tuning output
@section Tuning output

LilyPond tries to take as much formatting as possible out of your
hands. Nevertheless, there are situations where it needs some help, or
where you want to override its decisions. In this section we discuss
ways to do just that.

Notation output is specified in so called grobs (graphic objects). Each
grob carries with it a set of properties (grob properties) specific to
that object.  For example, a stem grob has properties that specify its
direction, length and thickness.


The most common way of tuning the output is to alter the values of these
properties. There are two ways of doing that: first, you can temporarily
change the definition of a certain type of grob, thus affecting a whole
set of objects.  Second, you can select one specific object, and set a 
grob property.

@menu
* Tuning groups of grobs ::     
* Tuning per grob ::            
* What to tune?::               
* Font selection::              
* Text markup::                 
@end menu

@node Tuning groups of grobs 
@subsection Tuning groups of grobs 

@cindex grob description

A grob definition is a Scheme association list, that is stored in a context
property.  By assigning to that property (using plain @code{\property}),
you can change the resulting grobs.

@lilypond[verbatim, fragment]
c'4 \property Voice.Stem  = #'((meta . ((interfaces . ())))) c'4
@end lilypond

The @code{\property} assignment effectively empties the definition of
the Stem object. One of the effects is that property specifying how it
should be printed is erased, with the effect of rendering it invisible.
The above assignment is available as a standard identifier, lest you
find this useful:
@example
  \property Voice.Stem = \turnOff
@end example

@cindex \override
@cindex \revert
@cindex \set

This mechanism is fairly crude, since you can only set, but not modify,
the definition of a grob. For this reason, there is a more advanced
mechanism: you can add a property on top of an existing definition, or
remove a property: @code{\override} adds a settings, @code{\revert}
removes that setting.

@lilypond[verbatim]
c'4 \property Voice.Stem \override #'thickness = #4.0
c'4 \property Voice.Stem \revert #'thickness
c'4
@end lilypond

For the digirati, the grob description is an Scheme association
list. Since it is singly linked, we can treat it as a stack, and
@code{\override} and @code{\revert} are just push and pop
operations. This pushing and popping is also used for overriding automatic
beaming settings.

If you revert a setting which was not set in the first place, then it
has no effect. However, if the setting was set as a system default, it
may remove the default value, and this may give surprising results,
including crashes.  In other words, if you use @code{\override} and
@code{\revert}, be sure to balance the overrides and reverts.

If balancing them is too much work, use the @code{\set} shorthand. It
performs a revert followed by an override:
@example
\property Voice.Stem \set #'thickness = #2.0
@end example

Formally the syntax for these constructions is
@example
\property @var{context}.@var{grobname} \override @var{symbol} = @var{value}
\property @var{context}.@var{grobname} \set @var{symbol} = @var{value}
\property @var{context}.@var{grobname} \revert @var{symbol}
@end example
Here @var{symbol} is a Scheme expression of symbol type, @var{context}
and @var{grobname} are strings and @var{value} is a Scheme expression.

If you want to be
Correct nesting of @code{\override}, @code{\set}, @code{\revert} is as
follows

@example 
\override \set \set \set \set
\revert
@end example

This is always correct, but if you know the default value, you can also use 
@example
\set \set \set \set
\set @var{to default value}
@end example

If there is no default (i.e. by default, the grob property is unset),
then you can use
@example
\set \set \set \set \set
\revert
@end example


@refbugs

LilyPond will hang or crash if @var{value} contains cyclic references.
The backend is not very strict in type-checking grob properties. If you
@code{\revert} properties that are expected to be set by default,
LilyPond may crash.



@node Tuning per grob 
@subsection Tuning per grob 

@cindex \outputproperty

A second way of tuning grobs is the more arcane @code{\outputproperty}
feature.
Syntax is as follows
@example
\outputproperty @var{predicate} @var{symbol} = @var{value}
@end example
Here @code{predicate} is a Scheme function taking a grob argument, and
returning a boolean.  This statement is processed by the
@code{Output_property_engraver}.  It instructs the engraver to feed all
grobs that it sees to @var{predicate}. Whenever the predicate returns
true, the grob property @var{symbol} will be set to @var{value}.

You will need to combine this statement with @code{\context} to select
the appropriate context to apply this to.

If possible, avoid this feature: the semantics are not very clean, and
the syntax and semantics are up for rewrite.

Here are some random examples:

@lilypond[fragment,verbatim,singleline]
\relative c'' { c4
  \context Staff \outputproperty
  #(make-type-checker 'note-head-interface)
  #'extra-offset = #'(0.5 . 0.75)
  <c8 e g> }
@end lilypond

@cindex @code{extra-offset}

This selects all note heads occurring at current staff level, and sets
the @code{extra-offset} of those heads to @code{(0.5,0.75)}, shifting
them up and right.

Move the text "m.d.", but not the fingering instruction "2".
@lilypond[verbatim,singleline]
#(define (make-text-checker text)
   (lambda (grob) (equal? text (ly-get-grob-property grob 'text))))

\score {    
  \notes\relative c''' {
    \property Voice.Stem \set #'direction = #1
    \outputproperty #(make-text-checker "m.d.")
      #'extra-offset = #'(-3.5 . -4.5)
    a^2^"m.d."    
  }
}
@end lilypond




@node What to tune?
@subsection What to tune?

This all tells you how to tune grobs, but what variables are there? The
question is not  answered in this manual (although you may encounter
some examples.).

Grob properties are tied directly to the implementation of LilyPond, and
they are thus a moving target. Documentation of such variables are part
of the generated documentation: this documentation is generated from the
sourcecode of lily for each version, so it is usually mch more up to
date  than this manual. It should be available from the same place where
you got this manual.

To decide how to tune a grob, you need to find the following information
@itemize @bullet
@item
which grob to modify
@item
which property to modify
@item
which context the grob comes from.
@end itemize

Included with the automatically generated documentation is a master list
of grobs. Each one can be clicked, taking you to a overview of the
available properties.

There is also a master list of contexts. Clicking each takes you to an
overview of the context, listing which grob types are created there.


@node Font selection
@subsection Font selection

Most graphics in LilyPond are composed of characters of fonts.  You can
alter the characteristics of the font by setting certain grob
properties. The mechanism that is used for this resembles LaTeX's New
Font Selection Scheme. Within this scheme, a font is entirely
characterized by its font name.

For each grob that uses fonts (in other words, each grob that supports
@code{font-interface}) a font-name must be selected before it can be
printed.  The font name is selected by looking at a number of grob
properties:

@table @code
@item font-family
  The general class of the typeface.  Supported are roman (Computer
Modern), braces (for piano staff braces), music (the standard music
font), dynamic (font for dynamic signs) and typewriter

@item font-shape
  A symbol indicating the shape of the font, a finer gradation than
  font-family. Choices are italic and upright
@item font-series
  Symbol indicating the serie of the font.  Series form a finer gradation
  than font-shape. Choices are medium and bold.

@item font-relative-size
  A number indicating the size relative the standard size.  For example,
  with 20pt staff height, relative size -1  corresponds to 16pt staff
  height, and relative size +1 corresponds to 23 pt staff height.

@item font-design-size
A number indicating  the design size of the font. 

This is a feature of the Computer Modern Font: each point size has a
slightly different design. Smaller design sizes are relatively wider,
which enhances readability. Scalable type faces such TrueType and Adobe
Type1 usually come as ``one design fits all sizes''.

@item font-name
  The name of the font, without the design size, eg. @code{cmr},
@code{cmti}, etc. Setting this overrides font-family, font-shape and
font-series.

@end table

The font is selected by taking the first font that satisfies all
qualifiers specified. You can override any of these fields through
@code{\override} and @code{\revert}. The special value @code{*} matches
any value for that qualifier.

@example
  \property Lyrics.LyricText \override #'font-series = #'bold
  \property Lyrics.LyricText \override #'font-shape = #'*
@end example

@cindex @code{font-style}

There are also pre-cooked font selection qualifiers. These are selected
through the grob property @code{font-style}.  For example, the style
@code{finger} selects family @code{number} and relative size @code{-3}.
Styles available include: volta, finger, tuplet, timesig, mmrest,
script, large, Large and dynamic.

The style sheets and tables for selecting fonts are located in
@file{scm/font.scm}. Refer to this file for more information.

@refbugs

Relative size is not linked to any real size.  There is no mechanism to
select magnifications of fonts, meaning that you can not scale fonts
continuoussly. There is no style sheet provided for other fonts besides
the @TeX{} family.

@cindex font selection
@cindex font magnification
@cindex @code{font-interface}


@node Text markup
@subsection Text markup
@cindex text markup
@cindex markup text

LilyPond has an internal mechanism to typeset texts. You can
form text markup expressions by composing scheme expressions
in the following way.

@lilypond[verbatim, singleline]
 \relative c' {
    \fatText
    a^#"upright"
    b_#'(bold "bold")
    c^#'(italic "italic")
    d_#'((bold italic) "ff")
    e^#'(dynamic "ff")
    f_#'(lines "one" (bold "two"))
    g^#'(music "noteheads-2" "flags-u3")
  }
@end lilypond

Normally, the Scheme markup text is stored in the @code{text} property
of a grob.  Formally, it is defined as follows:

@example
text: string | (head? text+)
head: markup | (markup+)
markup-item: property | abbrev
property: (@var{key} . @var{value})
abbrev: @code{rows lines roman music bold italic named super sub text}
        @code{finger volta timesig mmrest mark script large Large dynamic}
@end example

The markup is broken down and converted into a list of grob properties,
which are prepended to the property list.  The @var{key}-@var{value}
pair is a grob property.

The following abbreviations are currently defined:

@table @code
@item rows
horizontal mode: set all text on one line (default)
@item lines
 vertical mode: set every text on new line
@item roman
 select roman font
@item music
 select feta font
@item bold
 select bold series
@item italic
 select italic shape
@item named
 lookup by character name
@item text
 plain text lookup (by character value)
@item super
 superscript
@item sub
 subscript
@item finger
 select fingering number fontstyle
@item volta
 select volta number fontstyle
@item timesig
 select time signature number fontstyle
@item mmrest
 select multi measure rest number fontstyle
@item mark
 select mark number fontstyle
@item script
 select scriptsize roman fontstyle
@item large
 select large roman fontstyle
@item Large
 select Large roman fontstyle
@item dynamic
 select dynamics fontstyle
@end table

It is possible to use @TeX{} commands in the strings, but this should be
avoided because this makes it impossible for LilyPond to compute the
exact length of the string, which may lead to collisions.  Also, @TeX{}
commands won't work with direct postscript output.

@cindex metronome mark

One practical application of complicated markup is to fake a metronome
marking:

@lilypond[verbatim]
#(define note '(rows
  (music "noteheads-2" ((kern . -0.1) "flags-stem"))))
#(define eight-note `(rows ,note ((kern . -0.1)
  (music ((raise . 3.5) "flags-u3")))))
#(define dotted-eight-note
  `(rows ,eight-note (music "dots-dot")))

\score {
  \notes\relative c'' {
    a1^#`((rows (font-relative-size . -1)) ,dotted-eight-note " = 64")
  }
  \paper {
    linewidth = -1.;
    \translator{
      \ScoreContext
      TextScript \override #'font-shape = #'upright
    }
  }
}
@end lilypond



@c . {Page layout}
@node Page layout
@section Page layout
@cindex Page layout

@menu
* Paper block::                 
* Paper variables::             
* Font Size::                   
* Paper size::                  
* Line break::                  
* Page break::                  
@end menu

@c .  {Paper block}
@node Paper block
@subsection Paper block
@cindex Paper block

The most important output definition is the @code{\paper} block, for
music notation.  The syntax is

@example
  @code{\paper @{} [@var{paperidentifier}] @var{items} @code{@}}
@end example

where each of the items is one of

@itemize @bullet
  @item  An assignment.  The assignment must be terminated by a
       semicolon.  

  @item  A context definition.  See @ref{Notation Contexts} for
       more information on context definitions.

  @item  \stylesheet  declaration.  Its syntax is
       @example
                \stylesheet @var{alist}
       @end example

        See @file{scm/font.scm} for details of @var{alist}.
  @item an \elementdescriptions declaration.
        @example
                \elementdescriptions @var{alist}
        @end example
        See @file{scm/grob-description.scm} for details of @var{alist}.

@end itemize

@c .  {Paper variables}
@node Paper variables
@subsection Paper variables 
@cindex Paper variables

The paper block has some variables you may want to use or change:

@table @code
@cindex @code{indent}
  @item @code{indent}  
    The indentation of the first line of music.
@cindex @code{staffspace}

  @item @code{staffspace}
    The distance between two staff lines, calculated from the center
    of the lines.  If you want scale independent output, then you should
use either this or @code{stafflinethickness}
    as a unit for distances you modify.
  
@cindex @code{linewidth}
  @item @code{linewidth}  
    Sets the width of the lines.

If set to a negative value, a single unjustified line is produced.
@c rename to singleLinePaper ?
The shorthand @code{\singleLine} defines a default paper block that
produces a single line.

@cindex @code{textheight}

  @item @code{textheight}  
    Sets the total height of the music on each page. Only used by
@code{ly2dvi}.

@cindex @code{interscoreline}

  @item @code{interscoreline}  
    Sets the spacing between systems.
Not set by default.
@cindex @code{interscorelinefill}



  @item @code{interscorelinefill}  
    If set to a positive number, the distance between the score 
    lines will stretch in order to fill the full page. In that
    case @code{interscoreline} specifies the minimum spacing.

        Not set by default.


@cindex @code{stafflinethickness}

  @item @code{stafflinethickness}  
    Determines the thickness of staff lines, and also acts as a scaling
    parameter for other line thicknesses.
@end table



@c .  {Font size}
@node Font Size
@subsection Font size
@cindex font size

The Feta font provides musical symbols at six different sizes.  These
fonts are 11 point, 13 point, 16 point, 20 point,
23 point, and 26 point.  The point size of a font is the
height of the five lines in a staff when displayed in the font.

Definitions for these sizes are the files @file{paperSZ.ly}, where
@code{SZ} is one of 11, 13, 16, 20, 23 and 26.  If you include any of
these files, the identifiers @code{paperEleven}, @code{paperThirteen},
@code{paperSixteen}, @code{paperTwenty}, @code{paperTwentythree}, and
@code{paperTwentysix} are defined respectively.  The default
@code{\paper} block is also set.

The font definitions are generated using a Scheme function. For more
details, see the file @file{font.scm}.



@c .  {Paper size}
@node Paper size
@subsection Paper size
@cindex Paper size

@cindex paper size
@cindex page size
@cindex @code{papersize}

To change the paper size, you must first set the
@code{papersize} variable at top level.  Set it to
the strings @code{a4}, @code{letter}, or @code{legal}.  After this
specification, you must set the font as described above.  If you want
the default font, then use the 20 point font.  The new paper size will
not take effect if the font is not loaded and selected afterwards.

@example
        papersize = "a4"
        \include "paper16.ly"

        \score @{
                ...
                \paper @{ \paperSixteen @}
        @}
@end example

The file @code{paper16.ly}  will now include a file named @file{a4.ly}, which
will set the paper variables @code{hsize} and @code{vsize} (used by
@code{ly2dvi})







@c .  {Line break}
@node Line break
@subsection Line break

@cindex line breaks
@cindex breaking lines

Line breaks are normally computed automatically. They are chosen such
that the resulting spacing has low variation, and looks neither cramped
nor loose.

Occasionally you might want to override the automatic breaks; you can do
this by specifying @code{\break}. This will force a line break at this
point. Do remember that line breaks can only occur at places where there
are barlines.  If you want to have a line break where there is no
barline, you can force an invisible barline by entering @code{\bar "";}.

Similarly, @code{\noBreak} forbids a  line break at a certain point.

@cindex @code{\penalty}

The @code{\break} and @code{\noBreak} commands are defined in terms of
the penalty command:
@example
  \penalty @var{int} @code{;}
@end example

This encourages or discourages LilyPond to make a line break at this
point.

@refbugs

The scaling of the @code{\penalty} argument is not well-defined.  The
command is rather kludgy, and slated for rewriting.

@c .  {Page break}
@node Page break
@subsection Page break

@cindex page breaks
@cindex breaking pages


Page breaks are normally computed by @TeX{}, so they are not under
direct control of LilyPond.  However, you can insert a commands into the
@file{.tex} output to instruct @TeX{} where to break pages. For more
details, see the example file @file{input/test/between-systems.ly}





@c . {Sound}
@node Sound
@section Sound
@cindex Sound

LilyPond can produce MIDI output.  The performance lacks lots of
interesting effects, such as swing, articulation, slurring, tieing,
etc., but it is good enough for proof-hearing the music you enter.

Dynamics and tempo changes are interpreted.

[TODO: mention volume control/Instrument Equaliser]


@refbugs

It is currently not possible to use the percussion channel (generally
channel 10 of a MIDI file).

@menu
* MIDI block::                  
* MIDI instrument names::       
@end menu

@c .  {MIDI block}
@node MIDI block
@subsection MIDI block
@cindex MIDI block


The MIDI block is analogous to the paper block, but it is somewhat
simpler.  The @code{\midi} block can contain:
@cindex MIDI block

@itemize @bullet
  @item  a @code{\tempo} definition
  @item  context definitions
@end itemize

Assignments in the @code{\midi} block are not allowed.



@cindex context definition

Context definitions follow precisely the same syntax as within the
\paper block.  Translation modules for sound are called performers.
The contexts for MIDI output are defined in @file{ly/performer.ly}.


@node MIDI instrument names
@subsection MIDI instrument names

@cindex instrument names
@cindex @code{Staff.midiInstrument}
@cindex @code{Staff.instrument}

The MIDI instrument name is set by the @code{Staff.midiInstrument}
property or, if that property is not set, the @code{Staff.instrument}
property.  The instrument name should be chosen from the list in
@ref{MIDI instruments}.

@refbugs

If the selected string does not exactly match, then LilyPond uses the
default piano. It is not possible to select an instrument by number.









@c . {Music entry}
@node Music entry
@section Music entry
@cindex Music entry
@menu
* Relative::                    
* Bar check::                   
* Point and click::             
@end menu

One of the applications of LilyPond is to enter music from existing
written or printed material. When you're doing this kind of copying
work, you can easily make mistakes.  This section deals with tricks and
features that help you enter music, and find and correct mistakes.

@c .  {Relative}
@node Relative
@subsection Relative
@cindex Relative
@cindex relative octave specification

Octaves are specified by adding @code{'} and @code{,} to pitch names.
When you copy existing music, it is easy to accidentally put a pitch in
the wrong octave and hard to find such an error.  To prevent these
errors, LilyPond features octave entry.

@cindex @code{\relative}
@example
  \relative @var{startpitch} @var{musicexpr}
@end example

The octave of notes that appear in @var{musicexpr} are calculated as
follows: If no octave changing marks are used, the basic interval
between this and the last note is always taken to be a fourth or less.
  The octave changing marks @code{'} and @code{,} can then
be added to raise or lower the pitch by an extra octave.  Upon entering
relative mode, an absolute starting pitch must be specified that will
act as the predecessor of the first note of @var{musicexpr}.

This distance is determined without regarding accidentals: a
@code{fisis} following a @code{ceses} will be put above the
@code{ceses}.

Entering music that changes octave frequently  is easy in relative mode.
@lilypond[fragment,singleline,verbatim,center]
  \relative c'' {
    b c d c b c bes a 
  }
@end lilypond

And octave changing marks are used for intervals greater than a fourth.
@lilypond[fragment,verbatim,center]
  \relative c'' {
    c g c f, c' a, e'' }
@end lilypond

If the preceding item is a chord, the first note of the chord is used
to determine the first note of the next chord.  But other notes
within the second chord are determined by looking at the immediately
preceding note.

@lilypond[fragment,verbatim,center]
  \relative c' {
    c <c e g> 
    <c' e g>
    <c, e' g>
  }
@end lilypond 
@cindex @code{\notes}

The pitch after the @code{\relative} contains a notename.  To parse
the pitch as a notename, you have to be in note mode, so there must
be a surrounding @code{\notes} keyword (which is not
shown here).

The relative conversion will not affect @code{\transpose},
@code{\chords} or @code{\relative} sections in its argument.  If you
want to use relative within transposed music, you must place an
additional @code{\relative} inside the @code{\transpose}.


@c .  {Bar check}
@node Bar check
@subsection Bar check
@cindex Bar check

@cindex bar check
@cindex @code{barCheckNoSynchronize}
@cindex @code{|}


Whenever a bar check is encountered during interpretation, a warning
message is issued if it doesn't fall at a measure boundary.  This can
help you find errors in the input.  Depending on the value of
@code{barCheckNoSynchronize}, the beginning of the measure will be
relocated, so this can also be used to shorten measures.

A bar check is entered using the bar symbol, @code{|}

@c .  {Point and click}
@node Point and click
@subsection Point and click

Point and click lets you find notes in the input by clicking on them in
the Xdvi window. This makes it very easy to find input that causes some
error in the sheet music.

To use it, you need the following software

@itemize
@item 
@uref{ftp://ftp.math.berkeley.edu/pub/Software/TeX/xdvi.tar.gz,plain
Xdvi} version 22.36 or newer.

  Note that most @TeX{} distributions ship with xdvik, which is a
  different and less well maintained program. To find out which xdvi you
  are running, try @code{xdvi --version} or @code{xdvi.bin --version}.
@item emacs
@end itemize


Add one these lines to the top of your .ly file. The first one is for
line location only. The second one is more convenient, but requires
patching @code{emacsclient}.

@example
#(set! point-and-click line-location)
#(set! point-and-click line-column-location)
@end example

In the emacs startup file (usually @file{~/.emacs}), add the following
@example
(server-start)
@end example

If you want emacs to jump to the exact spot (and not just the line) on a
click, you must enable column positioning. To do so, you need to patch
emacsclient. Apply @file{emacsclient.patch} (included with the source
package) to @file{emacsclient.c} and @file{server.el} from the emacs
source code. Recompile and stick the recompiled emacsclient into a bin
directory, and put @file{server.el} into a elisp directory
(eg. @file{~/usr/share/emacs/}). Add the following to your @file{.emacs}
init file, before invoking server-start.

@example
 (setq load-path (cons "~/usr/share/emacs" load-path))
@end example


Xdvi must be configured to use the emacs editor.  Before starting, set
the environment variable @code{XEDITOR} to
@example
emacsclient --no-wait +%c:%l %f
@end example
Xdvi also must be configured to find the fonts. Refer to the
xdvi documentation for more information.

When viewing, control-mousebutton 1 will take you to the originating
line and column. Control-mousebutton 2 will show all clickable boxes.

@refbugs

When you convert the TeX file to PostScript using dvips, dvips
will complain about not finding @code{src:X:Y} files. Those complaints are
harmless, and can be ignored.


@node Skipping corrected music
@section Skipping corrected music

The property @code{Score.skipTypesetting} can be used to switch on and
off typesetting completely during the interpretation phase. When
typesetting is switched off, the music is processed much more quickly.
You can use this to skip over the parts of a score that you have already
checked for errors. 

@lilypond[fragment,singleline,verbatim]
\relative c'' { c8 d
\property Score.skipTypesetting = ##t
  e f g a g c, f e d
\property Score.skipTypesetting = ##f
c d b bes a g c2 } 
@end lilypond


@node Interpretation context
@section Interpretation context

@menu
* Notation Contexts::           
* Creating contexts::           
* Default contexts::            
* Context properties::          
* Changing context definitions::  
* Defining new contexts::       
@end menu


@c .  {Notation Contexts}
@node Notation Contexts
@subsection Notation Contexts

@cindex notation contexts

Notation contexts are objects that only exist during a run of LilyPond.
During the interpretation phase of LilyPond (when it prints
"interpreting music"), the music expresiion in a @code{\score} block is
interpreted in time order. This is the same order that humans hear and
play music.

During this interpretation, the notation context is holds the state for
the current point within the music. It contains information like

@itemize @bullet
  @item What notes are playing at this point?
  @item What symbols will be printed at this point?
  @item What is the current key signature, time signature, point within
       the measure, etc.?
@end itemize

Contexts are grouped hierarchically: A @code{Voice} context is
contained in a @code{Staff} context (because a staff can contain
multiple voices at any point), a @code{Staff} context is contained in
a @code{Score}, @code{StaffGroup}, or @code{ChoirStaff} context (because
these can all contain multiple staffs).


Contexts associated with sheet music output are called @emph{notation
contexts}, those for sound output are called performance contexts.


@node Creating contexts
@subsection Creating contexts

@cindex @code{\context}
@cindex context selection

Contexts for a music expression can be selected manually, using the
following music expression.

@example
  \context @var{contexttype} [= @var{contextname}] @var{musicexpr}
@end example

This instructs lilypond to interpret @var{musicexpr} within the context
 of type @var{contexttype} and with name @var{contextname}.  If this
context does not exist, it will be created.  

@lilypond[verbatim,singleline]
\score {
  \notes \relative c'' {
    c4 <d4 \context Staff = "another" e4> f
  }
}

@end lilypond

In this example, the @code{c} and @code{d} are printed on the
default staff.  For the @code{e}, a context Staff called
@code{another} is specified; since that does not exist, a new
context is created.  Within @code{another}, a (default) Voice context
is created for the @code{e4}.  When all music referring to a
context is finished, the context is ended as well.  So after the
third quarter, @code{another} is removed.



@node Default contexts
@subsection Default contexts

Most music expressions don't need @code{\context}: they inherit the
notation context from their parent. Each note is a music expression, and
as you can see in the following example, only the sequential music
enclosing the three notes has an explicit context. 

@lilypond[verbatim,singleline]
\score { \notes \context Voice = goUp { c'4 d' e' } } 
@end lilypond

There are some quirks that you must keep in mind when dealing with
defaults:

First, every top-level music is interpreted by the Score context, in other
words, you may think of @code{\score} working like
@example
        \score @{
                \context Score @var{music}
        @}
@end example

Second, sequential music follows the contexts of its
``children''. Consider the following  example.

@lilypond[verbatim, singleline]
\score { \context Score \notes { c'4 (  d' )e' } }
@end lilypond

The sequential music is interpreted by the Score context initially
(notice that the @code{\context} specification is redundant), but when a
note is encountered, contexts are setup to accept that note. In this
case, a Thread, Voice and Staff are created. The rest of the sequential
music is also interpreted with the same Thread, Voice and Staff context,
putting the notes on the same staff, in the same voice.

This is a convenient mechanism, but do not expect opening chords to work
without @code{\context}. For every note, a separate staff is
instantiated.

@lilypond[verbatim, singleline]
\score { \notes <c'4 es'> } 
@end lilypond

Of course, if the chord is preceded by a normal note in sequential
music, the chord will be interpreted by the Thread of the preceding
note:
@lilypond[verbatim,singleline]
\score { \notes { c'4 <c'4 es'> }  }
@end lilypond



@node Context properties
@subsection Context properties

Notation contexts can be modified from within the @file{.ly} file. The
following music expression does that job:

@cindex @code{\property}
@example
  \property @var{contextname}.@var{propname} =  @var{value}
@end example

Sets the @var{propname} property of the context @var{contextname} to the
specified Scheme expression @var{value}.  All @var{propname} and
@var{contextname} are strings, which are typically unquoted.

Properties that are set in one context are inherited by all of the
contained contexts.  This means that a property valid for the
@code{Voice} context can be set in the @code{Score} context (for
example) and thus take effect in all @code{Voice} contexts.

Properties can be unset using the following expression:
@example
  \property @var{contextname}.@var{propname} \unset
@end example

This removes the definition of @var{propname} in @var{contextname}. If
@var{propname} was not defined in @var{contextname} (but was inherited
from a higher context), then this has no effect.


@refbugs

@code{\property \unset} is not the inverse of @code{\property \set}




@c .  {Context definitions}
@node Changing context definitions
@subsection Changing context definitions

@cindex context definition
@cindex translator definition

The most common way to define a context is by extending an existing
context.  You can change an existing context from the paper block, by
first initializing a translator with an existing context identifier:
@example
\paper @{
  \translator @{
    @var{context-identifier}
  @} @}
@end example
Then you can add engravers, remove engravers.
The syntax for these operations are respectively
@example
 \remove @var{engravername}
 \consists @var{engravername}
@end example


Here @var{engravername} is a string, the name of an engraver in the
system.
@example
 @var{propname} = @var{value} 
@end example


@lilypond[verbatim,singleline]
\score {  \notes {
        c'4 c'4 }
  \paper {
    \translator  { \StaffContext
        \remove Clef_engraver;
       } } }
@end lilypond

@cindex engraver

You can also set properties in a translator definition. The syntax is as
follows:

@var{propname} is a string and @var{value} is a Scheme
expression.
@example
 @var{propname} = @var{value}
 @var{propname} \set  @var{symbol} = @var{value}
 @var{propname} \override @var{symbol} =  @var{value}
 @var{propname} \revert @var{symbol} 

@end example

These type of property assignments happen before interpretation starts,
so a @code{\property} expression will override any predefined settings.


 To simplify editing translators, all standard contexts have standard
identifiers called @var{name}@code{Context}, e.g. @code{StaffContext},
@code{VoiceContext}.

@node Defining new contexts
@subsection Defining new contexts

If you want to build a context from scratch, you must also supply the
following extra information:
@itemize @bullet
  @item  A name, specified by @code{\name @var{contextname};}.

  @item A cooperation module. This is specified by   @code{\type
@var{typename};}.
@end itemize

This is an example:
@example
\translator @code{
  \type "Engraver_group_engraver";
  \name "SimpleStaff";
  \alias "Staff";
  \consists "Staff_symbol_engraver";
  \consists "Note_head_engraver";
  \consistsend "Axis_group_engraver";
}@
@end example

Basic building blocks of translation are called engravers; they are
special C++ classes.

The argument of @code{\type} is the name for a special engraver that
handles cooperation between simple engravers such as
@code{Note_head_engraver} and @code{Staff_symbol_engraver}. Alternatives
for this engraver are the following:
@table @code
@cindex @code{Engraver_group_engraver}
  @item @code{Engraver_group_engraver}  
    The standard cooperation engraver.

@cindex @code{Score_engraver}

  @item @code{Score_engraver}  
    This is cooperation module that should be in the top level context,
and only the toplevel context.

@cindex @code{Grace_engraver_group}

  @item @code{Grace_engraver_group}  
    This is a special cooperation module (resembling
    @code{Score_engraver}) that is used to created an embedded
    `miniscore'.
@end table 

Other modifiers   are

@itemize @bullet
  @item @code{\alias} @var{alternate-name} @code{;}
    This specifies a different name. In the above example,
@code{\property Staff.X = Y} will also work on @code{SimpleStaff}s

  @item  @code{\consistsend} @var{engravername} @code{;}  
    Analogous to @code{\consists}, but makes sure that
    @var{engravername} is always added to the end of the list of
    engravers.

    Some engraver types need to be at the end of the list; this
    insures they are put there, and stay there, if a user adds or
    removes engravers.  This command is usually not needed for
    end-users.
    
  @item  @code{\accepts} @var{contextname} @code{;}  
    Add @var{contextname} to the list of  context this context can
    contain.  The first listed context is the context to create by
    default.

  @item @code{\denies}. The opposite of @code{\accepts}. Added for
completeness, but is never used in practice.
 
  
  @item  @code{\name} @var{contextname} @code{;}  
    This sets name of the context, e.g. @code{Staff}, @code{Voice}.  If
    the name is not specified, the translator won't do anything.
@end itemize

In the @code{\paper} block, it is also possible to define translator
identifiers.  Like other block identifiers, the identifier can only
be used as the very first item of a translator.  In order to define
such an identifier outside of @code{\score}, you must do

@quotation
@example 
\paper @{
  foo = \translator @{ @dots{} @}
@}
\score @{
  \notes @{
    @dots{}
  @}
  \paper @{
    \translator @{ \foo @dots{} @}
  @}
@} 
@end example 

@end quotation


@cindex paper types, engravers, and pre-defined translators

      




@c . {Syntactic details}
@node Syntactic details
@section Syntactic details
@cindex Syntactic details

This section describes details that were too boring to be put elsewhere.

@menu
* Top level::                   
* Identifiers::                 
* Music expressions::           
* Manipulating music expressions::  
* Assignments::                 
* Lexical modes::               
* Ambiguities::                 
@end menu

@c .  {Top level}
@node Top level
@subsection Top level
@cindex Top level

This section describes what you may enter at top level.


@c .   {Score}
@subsubsection Score
@cindex Score

@cindex score definition

The output is generated combining a music expression with an output
definition.  A score block has the following syntax:

@example
  \score @{ @var{musicexpr} @var{outputdefs} @}
@end example

@var{outputdefs} are zero or more output definitions.  If none is
supplied, the default @code{\paper} block will be added.



@c .   {Default output}
@subsubsection Default output

Default values for the @code{\paper} and @code{\midi} block are set by
entering such a block at top-level.

@c .   {Header}
@subsubsection Header
@cindex Header
@cindex @code{\header}


A header describes bibilographic information of the file's contents.  It
can also appear in a @code{\score} block.  Tools like @code{ly2dvi} can
use this information for generating titles.  Key values that are used by
@code{ly2dvi} are: title, subtitle, composer, opus, poet, instrument,
metre, arranger, piece and tagline.

@cindex @code{ly2dvi}

The syntax is
@example
  \header @{ @var{key1} = @var{val1};
             @var{key2} = @var{val2}; @dots{} @}
@end example

It is customary to put the @code{\header} at the top of the file.

@subsubsection Default output

A @code{\midi} or @code{\paper} block at top-level sets the default

paper block for all scores that lack an explicit paper block.

@c .  {Identifiers}
@node Identifiers
@subsection Identifiers
@cindex  Identifiers

All of the information in a LilyPond input file, is represented as a
Scheme value. In addition to normal Scheme data types (such as pair,
number, boolean, etc.), LilyPond has a number of specialized data types,

@itemize @bullet
@item Input
@item c++-function
@item Music
@item Identifier
@item Translator_def
@item Duration
@item Pitch
@item Score
@item Music_output_def
@item Moment (rational number)
@end itemize

LilyPond also includes some transient object types. Objects of these
types are built during a LilyPond run, and do not `exist' per se within
your input file. These objects are created as a result of your input
file, so you can include commands in the input to manipulate them,
during a lilypond run.

@itemize @bullet
@item Grob: short for Graphical object. See @ref{Grobs}. 
@item Molecule: device-independent page output object,
including dimensions.  Produced by some Grob functions
See @ref{Molecules}
@item Translator: object that produces audio objects or Grobs. This is
not yet user accessible.
@item Font_metric: object representing a font. (See @ref{Font metrics})

@end itemize


@node Music expressions
@subsection Music expressions

@cindex music expressions

Music in LilyPond is entered as a music expression.  Notes, rests, lyric
syllables are music expressions, and you can combine music expressions
to form new ones, for example by enclosing a list of expressions in
@code{\sequential @{ @}} or @code{< >}.  In the following example, a
compound expression is formed out of the quarter note @code{c} and a
quarter note @code{d}:

@example 
\sequential @{ c4 d4 @} 
@end example 

@cindex Sequential music
@cindex @code{\sequential}
@cindex sequential music
@cindex @code{<}
@cindex @code{>}
@cindex Simultaneous music
@cindex @code{\simultaneous}

The two basic compound  music expressions are simultaneous  and
sequential music.

@example
  \sequential @code{@{} @var{musicexprlist} @code{@}}
  \simultaneous @code{@{} @var{musicexprlist} @code{@}}
@end example
For both, there is a shorthand:
@example
  @code{@{} @var{musicexprlist} @code{@}}
@end example
for sequential and
@example
  @code{<} @var{musicexprlist} @code{>}
@end example
for simultaneous music.
In principle, the way in which you nest sequential and simultaneous to
produce music is not relevant.  In the following example, three chords
are expressed in two different ways:

@lilypond[fragment,verbatim,center]
  \notes \context Voice {
    <a c'> <b  d' > <c' e'>
    < { a b  c' } { c' d' e' } >
  }
@end lilypond


Other compound music expressions include
@example
 \repeat @var{expr}
 \transpose @var{pitch} @var{expr}
 \apply @var{func} @var{expr}
 \context @var{type} = @var{id} @var{expr}
 \times @var{fraction} @var{expr}
@end example


@c . {Manipulating music expressions}
@node Manipulating music expressions
@subsection  Manipulating music expressions

The @code{\apply} mechanism gives you access to the internal
representation of music. You can write Scheme-functions that operate
directly on it. The syntax is 
@example
        \apply #@var{func} @var{music}
@end example
This means that @var{func} is applied to @var{music}.  The function
@var{func} should return a music expression.

This example replaces the text string of a script. It also shows a dump
of the music it processes, which is useful if you want to know more
about how music is stored.

@lilypond[verbatim,singleline]
#(define (testfunc x)
        (if (equal? (ly-get-mus-property x 'text) "foo")
                (ly-set-mus-property x 'text "bar"))
        ;; recurse
        (ly-set-mus-property x 'elements
          (map testfunc (ly-get-mus-property x 'elements)))
        (display x)
        x        
)
\score { \notes
  \apply #testfunc { c4_"foo" }
} 
@end lilypond

For more information on what is possible, see the automatically
generated documentation.


Directly accessing internal representations is dangerous: the
implementation is subject to changes, so you should avoid this feature
if possible.

A final example is a function that reverses a piece of music in time:

@lilypond[verbatim,singleline]
#(define (reverse-music music)
  (let* ((elements (ly-get-mus-property music 'elements))
         (reversed (reverse elements))
         (span-dir (ly-get-mus-property music 'span-direction)))
    (ly-set-mus-property music 'elements reversed)
    (if (dir? span-dir)
        (ly-set-mus-property music 'span-direction (- span-dir)))
    (map reverse-music reversed)
    music))

music = \notes { c'4 d'4( e'4 f'4 }

\score { \context Voice {
    \music
    \apply #reverse-music \music
  }
}
@end lilypond


@c .   {Span requests}
@menu
* Span requests::               
@end menu

@node Span requests
@subsubsection Span requests
@cindex Span requests

Notational constructs that start and end on different notes can be
entered using span requests. The syntax is as follows:


@example
  \spanrequest @var{startstop} @var{type}
@end example


@cindex @code{\start}
@cindex @code{\stop}

This defines a spanning request. The @var{startstop} parameter is either
-1 (@code{\start}) or 1 (@code{\stop}) and @var{type} is a string that
describes what should be started.  Much of the syntactic sugar is a
shorthand for @code{\spanrequest}, for example,

@lilypond[fragment,verbatim,center]
  c'4-\spanrequest \start "slur"
  c'4-\spanrequest \stop "slur"
@end lilypond

Among the supported types are @code{crescendo}, @code{decrescendo},
@code{beam}, @code{slur}. This is an internal command.  Users are
encouraged to use the shorthands which are defined in the initialization
file @file{spanners.ly}.


@c .   {Assignments}
@node Assignments
@subsection Assignments
@cindex Assignments

Identifiers allow objects to be assigned to names during the parse
stage.  To assign an identifier, you use @var{name}@code{=}@var{value}
and to refer to an identifier, you preceed its name with a backslash:
`@code{\}@var{name}'.  @var{value} is any valid Scheme value or any of
the input-types listed above.  Identifier assignments can appear at top
level in the LilyPond file, but also in @code{\paper} blocks.

Semicolons are forbidden after top level assignments, but mandatory in
other places. The rules about semicolons and assignments are very
confusing, but when LilyPond input evolves more towards Scheme, we hope
that this problem will grow smaller.

An identifier can be created with any string for its name, but you will
only be able to refer to identifiers whose names begin with a letter,
being entirely alphabetical.  It is impossible to refer to an identifier
whose name is the same as the name of a keyword.

The right hand side of an identifier assignment is parsed completely
before the assignment is done, so it is allowed to redefine an
identifier in terms of its old value, e.g.

@example
foo = \foo * 2.0
@end example

When an identifier is referenced, the information it points to is
copied.  For this reason, an identifier reference must always be the
first item in a block.
@example
\paper  @{
  foo = 1.0
  \paperIdent % wrong and invalid
@}

\paper @{
  \paperIdent % correct
  foo = 1.0 @}
@end example


@c .  {Lexical modes}
@node Lexical modes
@subsection Lexical modes
@cindex Lexical modes
@cindex input mode
@cindex mode, input 
@cindex @code{\notes}
@cindex @code{\chords}
@cindex @code{\lyrics}

To simplify entering notes, lyrics, and chords, LilyPond has three
special input modes on top of the default mode: note, lyrics and chords
mode.  These input modes change the way that normal, unquoted words are
interpreted: for example, the word @code{cis} may be interpreted as a
C-sharp, as a lyric syllable `cis' or as a C-sharp major triad
respectively.

A mode switch is entered as a compound music expressions
@example
@code{\notes} @var{musicexpr}
@code{\chords} @var{musicexpr}
@code{\lyrics}  @var{musicexpr}.
@end example

In each of these cases, these expressions do not add anything to the
meaning of their arguments.  They are just a way to indicate that the
arguments should be parsed in indicated mode.  The modes are treated in
more detail in the @ref{Note entry}, @ref{Lyrics} and
@ref{Chords}.

You may nest different input modes.

@c .  {Ambiguities}
@node Ambiguities
@subsection Ambiguities
@cindex ambiguities
@cindex grammar


The grammar contains a number of ambiguities. We hope to resolve them at
some time.

@itemize @bullet
  @item  The assignment

         @example 
foo = bar 
@end example 

       can be interpreted as making a string identifier @code{\foo}
       containing @code{"bar"}, or a music identifier @code{\foo}
       containing the syllable `bar'.

  @item  The assignment

         @example 
foo = -6 
@end example 

       can be interpreted as making an integer identifier
       containing -6, or a Request identifier containing the
       fingering `6' (with neutral direction).

  @item  If you do a nested repeat like

       @quotation

@example 
\repeat @dots{}
\repeat @dots{}
\alternative 
@end example 

       @end quotation

       then it is ambiguous to which @code{\repeat} the
       @code{\alternative} belongs.  This is the classic if-then-else
       dilemma.  It may be solved by using braces.

  @item  (an as yet unidentified ambiguity :-)
@end itemize


@c .  {Lexical details}
@node Lexical details
@section Lexical details

Even more boring details, now on lexical side of the input parser.

@menu
* Comments::                    
* Direct Scheme::               
* Keywords::                    
* Integers::                    
* Reals::                       
* Strings::                     
* Main input::                  
* File inclusion::              
* Version information::         
@end menu


@node Comments
@subsection Comments

@cindex comments
@cindex block comment
@cindex line comment

@cindex @code{%}

A one line comment is introduced by a @code{%} character. 
Block comments are started by @code{%@{} and ended by @code{%@}}. 
They cannot be nested.

@node Direct Scheme
@subsection Direct Scheme

@cindex Scheme
@cindex GUILE
@cindex Scheme, in-line code


LilyPond contains a Scheme interpreter (the GUILE library) for
internal use. In some places Scheme expressions also form valid syntax:
whereever it is allowed,
@example
  #@var{scheme}
@end example
evaluates the specified Scheme code. If this is used at toplevel, then
the result is discarded. Example:
@example
  \property Staff.TestObject \override #'foobar =  #(+ 1 2)
@end example

@code{\override} expects two Scheme expressions, so there are two Scheme
expressions. The first one is a symbol (@code{foobar}), the second one
an integer (namely, 3).

Scheme is a full-blown programming language, and a full discussion is
outside the scope of this document. Interested readers are referred to
the website @uref{http://www.schemers.org/} for more information on
Scheme.


@node Keywords
@subsection Keywords
@cindex Keywords


Keywords start with a backslash, followed by a number of lower case
alphabetic characters.  These are all the keywords.

@example
apply arpeggio autochange spanrequest commandspanrequest
simultaneous sequential accepts alternative bar breathe
char chordmodifiers chords clef cm consists consistsend
context denies duration dynamicscript elementdescriptions
font grace header in lyrics key mark pitch
time times midi mm name pitchnames notes outputproperty
override set revert partial paper penalty property pt
relative remove repeat addlyrics partcombine score
script stylesheet skip textscript tempo translator
transpose type
@end example

@node Integers
@subsection Integers

@cindex integers
@cindex @code{+}
@cindex @code{-}
@cindex @code{*}
@cindex @code{/}

Formed from an optional minus sign followed by digits.  Arithmetic
operations cannot be done with integers, and integers cannot be mixed
with reals.

@node Reals
@subsection Reals
@cindex real numbers





Formed from an optional minus sign and a sequence of digits followed
by a @emph{required} decimal point and an optional exponent such as
@code{-1.2e3}.  Reals can be built up using the usual operations:
`@code{+}', `@code{-}', `@code{*}', and
`@code{/}', with parentheses for grouping.

@cindex @code{\mm},
@cindex @code{\in}
@cindex @code{\cm}
@cindex @code{\pt}
@cindex dimensions

A real constant can be followed by one of the dimension keywords:
@code{\mm} @code{\pt}, @code{\in}, or @code{\cm}, for millimeters,
points, inches and centimeters, respectively.  This converts the number
a number that is the internal representation of that dimension.


@node Strings
@subsection Strings
@cindex string
@cindex concatenate

Begins and ends with the @code{"} character.  To include a @code{"}
character in a string write @code{\"}.  Various other backslash
sequences have special interpretations as in the C language.  A string
that contains no spaces can be written without the quotes.  See
@ref{Lexical modes} for details on unquoted strings; their
interpretation varies depending on the situation.  Strings can be
concatenated with the @code{+} operator.

The tokenizer accepts the following commands. They have no grammatical
function, hence they can appear anywhere in the input.


@node Main input
@subsection Main input
@cindex Main input

@cindex @code{\maininput}

The @code{\maininput} command is used in init files to signal that the
user file must be read. This command cannot be used in a user file.

@node File inclusion
@subsection File inclusion
@cindex @code{\include}
@example
  \include @var{filename}
@end example

Include @var{filename}.  The argument @var{filename} may be a quoted string (an
unquoted string will not work here!) or a string identifier.  The full
filename including the @file{.ly} extension must be given,


@node Version information
@subsection Version information 
@cindex @code{\version}
@example
  \version @var{string} ;
@end example

Specify the version of LilyPond that a file was written for.  The
argument is a version string in quotes, for example @code{"1.2.0"}. 
This is used to detect invalid input, and to aid
@code{convert-ly}  a tool that automatically upgrades input files. See
See @ref{convert-ly} for more information on @code{convert-ly}.

@cindex convert-ly






@c .{Local emacs vars}
@c Local variables:
@c mode: texinfo
@c minor-mode: font-lock
@c minor-mode: outline
@c outline-layout: (-1 : 0)
@c outline-use-mode-specific-leader: "@c \."
@c outline-primary-bullet: "{"
@c outline-stylish-prefixes: nil
@c outline-override-protect: t
@c End: