doc fixes, rune pats

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

@c -*-texinfo-*-
@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


@node Advanced Topics
@chapter Advanced Topics


When translating the input to notation, there are number of distinct
phases.  We list them here:

@c todo: moved from refman. 

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 aesthetic 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
programming. The Scheme library provides the glue that holds together
the low-level routines and separate modules which are written in 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. The result of this step is a typesetting
specification.

@item typesetting:
The typesetting specification is solved: positions and formatting is
calculated.

@item the visible results ("virtual ink") are 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{contexts} are constructed, and with these
contexts a network of @strong{graphical objects} (``grobs'') is
created. These grobs contain unknown variables, and the network forms a
set of equations. After solving the equations and filling in these
variables, the printed output 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.



@table @b

@item Parsing:

The LY file is read, and converted to a list of @code{Scores}, which
each contain @code{Music} and paper/midi-definitions. Here @code{Music},
@code{Pitch} and @code{Duration}  objects are created.

@item Interpreting music
@cindex interpreting music

All music events are "read" in the same order as they would be played
(or read from paper). At every step of the interpretation, musical
events are delivered to
interpretation contexts,
@cindex engraver
which use them to build @code{Grob}s (or MIDI objects, for MIDI output).

In this stage @code{Music_iterators} do a traversal of the @code{Music}
structure. The music events thus encountered are reported to
@code{Translator}s, a set of objects that collectively form interpretation
contexts.


@item Prebreaking

@cindex prebreaking

At places where line breaks may occur, clefs and bars are prepared for
a possible line break. 

@item Preprocessing

@cindex preprocessing

In this stage, all information that is needed to determine line breaking
is computed. 

@item Break calculation:

The lines and horizontal positions of the columns are determined.

@item Breaking

Relations between all grobs are modified to reflect line breaks: When a
spanner, e.g. a slur, crosses a line-break, then the spanner is "broken
into pieces", for every line that the spanner is in, a copy of the grob
is made. A substitution process redirects all grob-reference so that
each spanner grob will only reference other grobs in the same line.

@item Outputting:

All vertical dimensions and spanning objects are computed, and all grobs
are output, line by line. The output is encoded in the form of
@code{Molecule}s

@end table

The data types that are mentioned here are all discussed in this
section.



@c FIXME: Note entry vs Music entry at top level menu is confusing.
@c . {Music entry}
@menu
* Interpretation context::      
* Syntactic details::           
* Lexical details::             
@end menu

@node Interpretation context
@section Interpretation context

@menu
* Creating contexts::           
* Default contexts::            
* Context properties::          
* Engravers and performers::    
* Changing context definitions::  
* Defining new contexts::       
@end menu


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

During this interpretation, the interpretation context 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 @internalsref{Voice} context is
contained in a @internalsref{Staff} context (because a staff can contain
multiple voices at any point), a @internalsref{Staff} context is contained in
@internalsref{Score}, @internalsref{StaffGroup}, or @internalsref{ChoirStaff} context.

Contexts associated with sheet music output are called @emph{notation
contexts}, those for sound output are called @emph{performance
contexts}. The default definitions of the standard notation and
performance contexts can be found in @file{ly/engraver-init.ly} and
@file{ly/performer-init.ly}, respectively.

@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 an explicit @code{\context}
declaration: 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, contexts are created automatically to be able to interpret the
music expressions. 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.

@cindex explicit context
@cindex starting with chords
@cindex chords, starting with

@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 have properties. These properties are from
the @file{.ly} file using the following  expression:
@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
@internalsref{Voice} context can be set in the @internalsref{Score} context (for
example) and thus take effect in all @internalsref{Voice} contexts.

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

@cindex properties, unsetting
@cindex @code{\unset} 

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

The syntax of @code{\unset} is asymmetric: @code{\property \unset} is not
the inverse of @code{\property \set}.

@node Engravers and performers
@subsection Engravers and performers

[TODO]

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



@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 and remove engravers using the following syntax:
@example
 \remove @var{engravername}
 \consists @var{engravername}
@end example


Here @var{engravername} is a string, the name of an engraver in the
system.


@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:
@example
 @var{propname} = @var{value}
 @var{propname} \set  @var{grob-propname} = @var{pvalue}
 @var{propname} \override @var{grob-propname} =  @var{pvalue}
 @var{propname} \revert @var{grob-propname} 
@end example
@var{propname} is a string, @var{grob-propname} a symbol, @var{value}
and @code{pvalue} are Scheme expressions. These type of property
assignments happen before interpretation starts, so a @code{\property}
command 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}, see @file{ly/engraver-init.ly}.

@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

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 top level context.

@end table 

Other modifiers   are

@itemize @bullet
  @item @code{\alias} @var{alternate-name}
    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} 
    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 stay there even if a user adds or removes engravers.
End-users generally don't need this command.
    
  @item  @code{\accepts} @var{contextname}
    Add @var{contextname} to the list of contexts this context can
    contain in the context hierarchy. 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} 
    This sets the type name of the context, e.g. @internalsref{Staff},
    @internalsref{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

      
@node Syntactic details
@section Syntactic details
@cindex Syntactic details

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

@menu
* Identifiers::                 
* Music expressions::           
* Manipulating music expressions::  
* Span requests::               
* Assignments::                 
* Lexical modes::               
* Ambiguities::                 
@end menu

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

@ignore
 What has this section got to do with identifiers?
 It seems more appropriate in the introduction to Chapter 4,
 "Internals".

   /MB
@end ignore

All of the information in a LilyPond input file, is internally
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. 
@item Molecule: device-independent page output object,
including dimensions.  Produced by some Grob functions
@item Translator: object that produces audio objects or Grobs. This is
not yet user accessible.
@item Font_metric: object representing a font.
@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 { c'4_"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

More examples are given in the distributed example files in
@code{input/test/}.

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

@node Span requests
@subsection 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 precede 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.

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 in addition to 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 expression
@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 just instruct the parser in what mode
to parse their arguments.

Different input modes may be nested.

@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 
         is interpreted as the string identifier assignment. However,
it can also be interpreted as making a string identifier @code{\foo}
containing @code{"bar"}, or a music identifier @code{\foo} containing
the syllable `bar'.  The former interpretation is chosen.

  @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.

@end itemize

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

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

@menu
* Direct Scheme::               
* Reals::                       
* Strings::                     
@end menu


@node Direct Scheme
@subsection Direct Scheme

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



@cindex GUILE
@cindex Scheme
@cindex accessing Scheme
@cindex evaluating Scheme
@cindex LISP

LilyPond internally uses GUILE, a Scheme-interpreter. Scheme is a
language from the LISP family. You can learn more about Scheme at
@uref{http://www.scheme.org}. It is used to represent data throughout
the whole program. The hash-sign (@code{#}) accesses GUILE directly: the
code following the hash-sign is evaluated as Scheme.  The boolean value
@var{true} is @code{#t} in Scheme, so for LilyPond @var{true} looks like
@code{##t}.

LilyPond contains a Scheme interpreter (the GUILE library) for
internal use. In some places, Scheme expressions also form valid syntax:
wherever it is allowed,
@example
  #@var{scheme}
@end example
evaluates the specified Scheme code.  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).

In-line scheme may be used at the top level. In this case the result is
discarded.

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 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.  Strings can
be concatenated with the @code{+} operator.