1 @c -*- coding: utf-8; mode: texinfo; -*-
4 Translation of GIT committish: FILL-IN-HEAD-COMMITTISH
6 When revising a translation, copy the HEAD committish of the
7 version that you are working on. For details, see the Contributors'
8 Guide, node Updating translation committishes..
14 @chapter Scheme tutorial
18 @cindex Scheme, in-line code
19 @cindex accessing Scheme
20 @cindex evaluating Scheme
23 LilyPond uses the Scheme programming language, both as part of the
24 input syntax, and as internal mechanism to glue modules of the program
25 together. This section is a very brief overview of entering data in
26 Scheme. If you want to know more about Scheme, see
27 @uref{http://@/www@/.schemers@/.org}.
29 LilyPond uses the GNU Guile implementation of Scheme, which is
30 based on the Scheme @qq{R5RS} standard. If you are learning Scheme
31 to use with LilyPond, working with a different implementation (or
32 referring to a different standard) is not recommended. Information
33 on guile can be found at @uref{http://www.gnu.org/software/guile/}.
34 The @qq{R5RS} Scheme standard is located at
35 @uref{http://www.schemers.org/Documents/Standards/R5RS/}.
38 * Introduction to Scheme::
39 * Scheme in LilyPond::
40 * Building complicated functions::
43 @node Introduction to Scheme
44 @section Introduction to Scheme
46 We begin with an introduction to Scheme. For this brief introduction,
47 we will use the GUILE interpreter to explore how the language works.
48 Once we are familiar with Scheme, we will show how the language can
49 be integrated in LilyPond files.
55 * Scheme simple data types::
56 * Scheme compound data types::
57 * Calculations in Scheme::
59 * Scheme conditionals::
63 @subsection Scheme sandbox
65 The LilyPond installation includes the Guile implementation of
66 Scheme. On most systems you can experiment in a Scheme sandbox by
67 opening a terminal window and typing @q{guile}. On some systems,
68 notably Windows, you may need to set the environment variable
69 @code{GUILE_LOAD_PATH} to the directory @code{../usr/share/guile/1.8}
70 in the LilyPond installation. For the full path to this directory
71 see @rlearning{Other sources of information}. Alternatively, Windows
72 users may simply choose @q{Run} from the Start menu and enter
75 However, a hands-on Scheme sandbox with all of Lilypond loaded is
76 available with this command line:
78 lilypond scheme-sandbox
82 Once the sandbox is running, you will receive a guile prompt:
88 You can enter Scheme expressions at this prompt to experiment with
89 Scheme. If you want to be able to use the GNU readline library for
90 nicer editing of the Scheme command line, check the file
91 @file{ly/scheme-sandbox.ly} for more information. If you already have
92 enabled the readline library for your interactive Guile sessions outside
93 of LilyPond, this should work in the sandbox as well.
95 @node Scheme variables
96 @subsection Scheme variables
98 Scheme variables can have any valid scheme value, including a Scheme
101 Scheme variables are created with @code{define}:
108 Scheme variables can be evaluated at the guile prompt simply by
109 typing the variable name:
117 Scheme variables can be printed on the display by using the display function:
125 Note that both the value @code{2} and the guile prompt @code{guile}
126 showed up on the same line. This can be avoided by calling the
127 newline procedure or displaying a newline character.
130 guile> (display a)(newline)
132 guile> (display a)(display "\n")
137 Once a variable has been created, its value can be changed with @code{set!}:
140 guile> (set! a 12345)
146 @node Scheme simple data types
147 @subsection Scheme simple data types
149 The most basic concept in a language is data typing: numbers, character
150 strings, lists, etc. Here is a list of simple Scheme data types that are
151 often used with LilyPond.
155 Boolean values are True or False. The Scheme for True is @code{#t}
156 and False is @code{#f}.
161 Numbers are entered in the standard fashion,
162 @code{1} is the (integer) number one, while @w{@code{-1.5}} is a
163 floating point number (a non-integer number).
166 Strings are enclosed in double quotes:
172 Strings may span several lines:
181 and the newline characters at the end of each line will be included
184 Newline characters can also be added by including @code{\n} in the
188 "this\nis a\nmultiline string"
192 Quotation marks and backslashes are added to strings
193 by preceding them with a backslash.
194 The string @code{\a said "b"} is entered as
202 There are additional Scheme data types that are not discussed here.
203 For a complete listing see the Guile reference guide,
204 @uref{http://www.gnu.org/software/guile/manual/html_node/Simple-Data-Types.html}.
206 @node Scheme compound data types
207 @subsection Scheme compound data types
209 There are also compound data types in Scheme. The types commonly used in
210 LilyPond programming include pairs, lists, alists, and hash tables.
215 * Association lists (alists)::
220 @unnumberedsubsubsec Pairs
222 The foundational compound data type of Scheme is the @code{pair}. As
223 might be expected from its name, a pair is two values glued together.
224 The operator used to form a pair is called @code{cons}.
232 Note that the pair is displayed as two items surrounded by
233 parentheses and separated by whitespace, a period (@code{.}), and
234 more whitespace. The period is @emph{not} a decimal point, but
235 rather an indicator of the pair.
237 Pairs can also be entered as literal values by preceding them with
238 a single quote character.
246 The two elements of a pair may be any valid Scheme value:
251 guile> '("blah-blah" . 3.1415926535)
252 ("blah-blah" . 3.1415926535)
256 The first and second elements of the pair can be accessed by the
257 Scheme procedures @code{car} and @code{cdr}, respectively.
260 guile> (define mypair (cons 123 "hello there")
271 Note: @code{cdr} is pronounced "could-er", according Sussman and
273 @uref{http://mitpress.mit.edu/sicp/full-text/book/book-Z-H-14.html#footnote_Temp_133}
276 @unnumberedsubsubsec Lists
278 A very common Scheme data structure is the @emph{list}. Formally,
279 a @q{proper} list is defined to be either the empty list with its
280 input form @code{'()} and length@tie{}0, or a pair whose
281 @code{cdr} in turn is a shorter list.
283 There are many ways of creating lists. Perhaps the most common is
284 with the @code{list} procedure:
287 guile> (list 1 2 3 "abc" 17.5)
291 As can be seen, a list is displayed in the form of individual elements
292 separated by whitespace and enclosed in parentheses. Unlike a pair,
293 there is no period between the elements.
295 A list can also be entered as a literal list by enclosing its
296 elements in parentheses, and adding a quote:
299 guile> '(17 23 "foo" "bar" "bazzle")
300 (17 23 "foo" "bar" "bazzle")
303 Lists are a central part of Scheme. In, fact, Scheme is considered
304 a dialect of lisp, where @q{lisp} is an abbreviation for
305 @q{List Processing}. Scheme expressions are all lists.
307 @node Association lists (alists)
308 @unnumberedsubsubsec Association lists (alists)
310 A special type of list is an @emph{association list} or @emph{alist}.
311 An alist is used to store data for easy retrieval.
313 Alists are lists whose elements are pairs. The @code{car} of each
314 element is called the @emph{key}, and the @code{cdr} of each element
315 is called the @emph{value}. The Scheme procedure @code{assoc} is
316 used to retrieve an entry from the alist, and @code{cdr} is used to
320 guile> (define my-alist '((1 . "A") (2 . "B") (3 . "C")))
322 ((1 . "A") (2 . "B") (3 . "C"))
323 guile> (assoc 2 my-alist)
325 guile> (cdr (assoc 2 my-alist))
330 Alists are widely used in LilyPond to store properties and other data.
333 @unnumberedsubsubsec Hash tables
335 A data structure that is used occasionally in LilyPond. A hash table
336 is similar to an array, but the indexes to the array can be any type
337 of Scheme value, not just integers.
339 Hash tables are more efficient than alists if there is a lot of data
340 to store and the data changes very infrequently.
342 The syntax to create hash tables is a bit complex, but you
343 can see examples of it in the LilyPond source.
346 guile> (define h (make-hash-table 10))
349 guile> (hashq-set! h 'key1 "val1")
351 guile> (hashq-set! h 'key2 "val2")
353 guile> (hashq-set! h 3 "val3")
357 Values are retrieved from hash tables with @code{hashq-ref}.
360 guile> (hashq-ref h 3)
362 guile> (hashq-ref h 'key2)
367 Keys and values are retrieved as a pair with @code{hashq-get-handle}.
368 This is a preferred way, because it will return @code{#f} if a key is
372 guile> (hashq-get-handle h 'key1)
374 guile> (hashq-get-handle h 'frob)
379 @node Calculations in Scheme
380 @subsection Calculations in Scheme
383 We have been using lists all along. A calculation, like @code{(+ 1 2)}
384 is also a list (containing the symbol @code{+} and the numbers 1
385 and@tie{}2). Normally lists are interpreted as calculations, and the
386 Scheme interpreter substitutes the outcome of the calculation. To enter a
387 list, we stop the evaluation. This is done by quoting the list with a
388 quote @code{'} symbol. So, for calculations do not use a quote.
390 Inside a quoted list or pair, there is no need to quote anymore. The
391 following is a pair of symbols, a list of symbols and a list of lists
396 #'(staff clef key-signature)
401 Scheme can be used to do calculations. It uses @emph{prefix}
402 syntax. Adding 1 and@tie{}2 is written as @code{(+ 1 2)} rather than the
403 traditional @math{1+2}.
410 Calculations may be nested; the result of a function may
411 be used for another calculation.
418 These calculations are examples of evaluations; an expression like
419 @code{(* 3 4)} is replaced by its value @code{12}.
421 Scheme calculations are sensitive to the differences between integers
422 and non-integers. Integer calculations are exact, while non-integers
423 are calculated to the appropriate limits of precision:
432 When the scheme interpreter encounters an expression that is a list,
433 the first element of the list is treated as a procedure to be
434 evaluated with the arguments of the remainder of the list. Therefore,
435 all operators in Scheme are prefix operators.
437 If the first element of a Scheme expression that is a list passed to
438 the interpreter is @emph{not} an operator or procedure, an error will
448 <unnamed port>:52:1: In expression (1 2 3):
449 <unnamed port>:52:1: Wrong type to apply: 1
454 Here you can see that the interpreter was trying to treat 1 as an
455 operator or procedure, and it couldn't. Hence the error is "Wrong
458 Therefore, to create a list we need to use the list operator, or to
459 quote the list so that the interpreter will not try to evaluate it.
469 This is an error that can appear as you are working with Scheme in LilyPond.
472 The same assignment can be done in completely in Scheme as well,
475 #(define twentyFour (* 2 twelve))
478 @c this next section is confusing -- need to rewrite
480 The @emph{name} of a variable is also an expression, similar to a
481 number or a string. It is entered as
488 @cindex quoting in Scheme
490 The quote mark @code{'} prevents the Scheme interpreter from substituting
491 @code{24} for the @code{twentyFour}. Instead, we get the name
496 @node Scheme procedures
497 @subsection Scheme procedures
499 Scheme procedures are executable scheme expressions that return a
500 value resulting from their execution. They can also manipulate
501 variables defined outside of the procedure.
504 * Defining procedures::
509 @node Defining procedures
510 @unnumberedsubsubsec Defining procedures
512 Procedures are defined in Scheme with define
515 (define (function-name arg1 arg2 ... argn)
516 scheme-expression-that-gives-a-return-value)
519 For example, we could define a procedure to calculate the average:
522 guile> (define (average x y) (/ (+ x y) 2))
524 #<procedure average (x y)>
527 Once a procedure is defined, it is called by putting the procedure
528 name and the arguments in a list. For example, we can calculate
529 the average of 3 and 12:
532 guile> (average 3 12)
537 @unnumberedsubsubsec Predicates
539 Scheme procedures that return boolean values are often called
540 @emph{predicates}. By convention (but not necessity), predicate names
541 typically end in a question mark:
544 guile> (define (less-than-ten? x) (< x 10))
545 guile> (less-than-ten? 9)
547 guile> (less-than-ten? 15)
552 @unnumberedsubsubsec Return values
554 Scheme procedures always return a return value, which is the value
555 of the last expression executed in the procedure. The return
556 value can be any valid Scheme value, including a complex data
557 structure or a procedure.
559 Sometimes the user would like to have multiple Scheme expressions in
560 a procedure. There are two ways that multiple expressions can be
561 combined. The first is the @code{begin} procedure, which allows
562 multiple expressions to be evaluated, and returns the value of
566 guile> (begin (+ 1 2) (- 5 8) (* 2 2))
570 The second way to combine multiple expressions is in a @code{let} block.
571 In a let block, a series of bindings are created, and then a sequence
572 of expressions that can include those bindings is evaluated. The
573 return value of the let block is the return value of the last
574 statement in the let block:
577 guile> (let ((x 2) (y 3) (z 4)) (display (+ x y)) (display (- z 4))
578 @dots{} (+ (* x y) (/ z x)))
582 @node Scheme conditionals
583 @subsection Scheme conditionals
591 @unnumberedsubsubsec if
593 Scheme has an @code{if} procedure:
596 (if test-expression true-expression false-expression)
599 @var{test-expression} is an expression that returns a boolean
600 value. If @var{test-expression} returns @code{#t}, the if
601 procedure returns the value of @var{true-expression}, otherwise
602 it returns the value of @var{false-expression}.
607 guile> (if (> a b) "a is greater than b" "a is not greater than b")
608 "a is not greater than b"
612 @unnumberedsubsubsec cond
614 Another conditional procedure in scheme is @code{cond}:
617 (cond (test-expression-1 result-expression-sequence-1)
618 (test-expression-2 result-expression-sequence-2)
620 (test-expression-n result-expression-sequence-n))
628 guile> (cond ((< a b) "a is less than b")
629 ... ((= a b) "a equals b")
630 ... ((> a b) "a is greater than b"))
634 @node Scheme in LilyPond
635 @section Scheme in LilyPond
639 * LilyPond Scheme syntax::
640 * LilyPond variables::
641 * Input variables and Scheme::
642 * Importing Scheme in LilyPond::
643 * Object properties::
644 * LilyPond compound variables::
645 * Internal music representation::
648 @node LilyPond Scheme syntax
649 @subsection LilyPond Scheme syntax
653 The Guile interpreter is part of LilyPond, which means that
654 Scheme can be included in LilyPond input files. There are several
655 methods for including Scheme in LilyPond.
657 The simplest way is to use a hash mark@tie{}@code{#} before a Scheme
660 Now LilyPond's input is structured into tokens and expressions, much
661 like human language is structured into words and sentences. LilyPond
662 has a lexer that recognizes tokens (literal numbers, strings, Scheme
663 elements, pitches and so on), and a parser that understands the syntax,
664 @rcontrib{LilyPond grammar}. Once it knows that a particular syntax rule
665 applies, it executes actions associated with it.
667 The hash mark@tie{}@code{#} method of embedding Scheme is a natural fit
668 for this system. Once the lexer sees a hash mark, it calls the Scheme
669 reader to read one full Scheme expression (this can be an identifier, an
670 expression enclosed in parentheses, or several other things). After the
671 Scheme expression is read, it is stored away as the value for an
672 @code{SCM_TOKEN} in the grammar. Once the parser knows how to make use
673 of this token, it calls Guile for evaluating the Scheme expression.
674 Since the parser usually requires a bit of lookahead from the lexer to
675 make its parsing decisions, this separation of reading and evaluation
676 between lexer and parser is exactly what is needed to keep the execution
677 of LilyPond and Scheme expressions in sync. For this reason, you should
678 use the hash mark@tie{}@code{#} for calling Scheme whenever this is
681 Another way to call the Scheme interpreter from LilyPond is the use of
682 dollar@tie{}@code{$} instead of a hash mark for introducing Scheme
683 expressions. In this case, Lilypond evaluates the code right after the
684 lexer has read it. It checks the resulting type of the Scheme
685 expression and then picks a token type (one of several
686 @code{xxx_IDENTIFIER} in the syntax) for it. It creates a @emph{copy}
687 of the value and uses that for the value of the token. If the value of
688 the expression is void (Guile's value of @code{*unspecified*}), nothing
689 at all is passed to the parser.
691 This is, in fact, exactly the same mechanism that Lilypond employs when
692 you call any variable or music function by name, as @code{\name}, with
693 the only difference that the name is determined by the Lilypond lexer
694 without consulting the Scheme reader, and thus only variable names
695 consistent with the current Lilypond mode are accepted.
697 The immediate action of @code{$} can lead to surprises, @ref{Input
698 variables and Scheme}. Using @code{#} where the parser supports it
699 is usually preferable. Inside of music expressions, expressions
700 created using @code{#} @emph{are} interpreted as music. However,
701 they are @emph{not} copied before use. If they are part of some
702 structure that might still get used, you may need to use
703 @code{ly:music-deep-copy} explicitly.
707 There are also @q{list splicing} operators @code{$@@} and @code{#@@}
708 that insert all elements of a list in the surrounding context.
710 Now let's take a look at some actual Scheme code. Scheme procedures can
711 be defined in LilyPond input files:
714 #(define (average a b c) (/ (+ a b c) 3))
717 Note that LilyPond comments (@code{%} and @code{%@{ %@}}) cannot
718 be used within Scheme code, even in a LilyPond input file, because
719 the Guile interpreter, not the LilyPond lexer, is reading
720 the Scheme expression. Comments in Guile Scheme are entered
724 ; this is a single-line comment
727 This a (non-nestable) Guile-style block comment
728 But these are rarely used by Schemers and never in
733 For the rest of this section, we will assume that the data is entered
734 in a music file, so we add@tie{}@code{#}s at the beginning of each Scheme
737 All of the top-level Scheme expressions in a LilyPond input file can
738 be combined into a single Scheme expression by the use of the
739 @code{begin} statement:
748 @node LilyPond variables
749 @subsection LilyPond variables
751 LilyPond variables are stored internally in the form of Scheme
765 This means that LilyPond variables are available
766 for use in Scheme expressions. For example, we could use
769 twentyFour = #(* 2 twelve)
773 which would result in the number 24 being stored in the
774 LilyPond (and Scheme) variable @code{twentyFour}.
776 The usual way to refer to Lilypond variables, @ref{LilyPond Scheme
777 syntax}, is to call them using a backslash, i.e., @code{\twentyFour}.
778 Since this creates a copy of the value for most of LilyPond's internal
779 types, in particular music expressions, music functions don't usually
780 create copies of material they change. For this reason, music
781 expressions given with @code{#} should usually not contain material that
782 is not either created from scratch or explicitly copied rather than
785 @node Input variables and Scheme
786 @subsection Input variables and Scheme
788 The input format supports the notion of variables: in the following
789 example, a music expression is assigned to a variable with the name
793 traLaLa = @{ c'4 d'4 @}
798 There is also a form of scoping: in the following example, the
799 @code{\layout} block also contains a @code{traLaLa} variable, which is
800 independent of the outer @code{\traLaLa}.
803 traLaLa = @{ c'4 d'4 @}
804 \layout @{ traLaLa = 1.0 @}
808 In effect, each input file is a scope, and all @code{\header},
809 @code{\midi}, and @code{\layout} blocks are scopes nested inside that
812 Both variables and scoping are implemented in the GUILE module system.
813 An anonymous Scheme module is attached to each scope. An assignment of
817 traLaLa = @{ c'4 d'4 @}
821 is internally converted to a Scheme definition:
824 (define traLaLa @var{Scheme value of `@code{@dots{}}'})
827 This means that LilyPond variables and Scheme variables may be freely
828 mixed. In the following example, a music fragment is stored in the
829 variable @code{traLaLa}, and duplicated using Scheme. The result is
830 imported in a @code{\score} block by means of a second variable
834 traLaLa = { c'4 d'4 }
836 #(define newLa (map ly:music-deep-copy
837 (list traLaLa traLaLa)))
839 (make-sequential-music newLa))
844 @c Due to parser lookahead
846 This is actually a rather interesting example. The assignment will only
847 take place after the parser has ascertained that nothing akin to
848 @code{\addlyrics} follows, so it needs to check what comes next. It
849 reads @code{#} and the following Scheme expression @emph{without}
850 evaluating it, so it can go ahead with the assignment, and
851 @emph{afterwards} execute the Scheme code without problem.
853 @node Importing Scheme in LilyPond
854 @subsection Importing Scheme in LilyPond
858 The above example shows how to @q{export} music expressions from the
859 input to the Scheme interpreter. The opposite is also possible. By
860 placing it after @code{$}, a Scheme
861 value is interpreted as if it were entered in LilyPond syntax.
862 Instead of defining @code{\twice}, the example above could also have
867 $(make-sequential-music newLa)
870 You can use @code{$} with a Scheme expression anywhere you could use
871 @code{\@var{name}} after having assigned the Scheme expression to a
872 variable @var{name}. This replacement happens in the @q{lexer}, so
873 Lilypond is not even aware of the difference.
875 One drawback, however, is that of timing. If we had been using @code{$}
876 instead of @code{#} for defining @code{newLa} in the above example, the
877 following Scheme definition would have failed because @code{traLaLa}
878 would not yet have been defined. For an explanation of this timing
879 problem, @ref{LilyPond Scheme syntax}.
883 A further convenience can be the @q{list splicing} operators @code{$@@}
884 and @code{#@@} for inserting the elements of a list in the surrounding
885 context. Using those, the last part of the example could have been
893 Here, every element of the list stored in @code{newLa} is taken in
894 sequence and inserted into the list, as if we had written
897 @{ #(first newLa) #(second newLa) @}
900 Now in all of these forms, the Scheme code is evaluated while the
901 input is still being consumed, either in the lexer or in the parser.
902 If you need it to be executed at a later point of time, check out
903 @ref{Void scheme functions}, or store it in a procedure:
907 (ly:set-option 'point-and-click #f))
916 Mixing Scheme and LilyPond variables is not possible with the
917 @option{--safe} option.
920 @node Object properties
921 @subsection Object properties
923 Object properties are stored in LilyPond in the form of alist-chains,
924 which are lists of alists. Properties are set by adding values at
925 the beginning of the property list. Properties are read by retrieving
926 values from the alists.
928 Setting a new value for a property requires assigning a value to
929 the alist with both a key and a value. The LilyPond syntax for doing
933 \override Stem.thickness = #2.6
936 This instruction adjusts the appearance of stems. An alist entry
937 @code{'(thickness . 2.6)} is added to the property list of the
939 object. @code{thickness} is measured relative to the thickness of
940 staff lines, so these stem lines will be @code{2.6} times the
941 width of staff lines. This makes stems almost twice as thick as their
942 normal size. To distinguish between variables defined in input files (like
943 @code{twentyFour} in the example above) and variables of internal
944 objects, we will call the latter @q{properties} and the former
945 @q{variables.} So, the stem object has a @code{thickness} property,
946 while @code{twentyFour} is a variable.
948 @cindex properties vs. variables
949 @cindex variables vs. properties
951 @c todo -- here we're getting interesting. We're now introducing
952 @c LilyPond variable types. I think this deserves a section all
955 @node LilyPond compound variables
956 @subsection LilyPond compound variables
967 @unnumberedsubsubsec Offsets
969 Two-dimensional offsets (X and Y coordinates) are stored as @emph{pairs}.
970 The @code{car} of the offset is the X coordinate, and the @code{cdr} is
974 \override TextScript.extra-offset = #'(1 . 2)
977 This assigns the pair @code{(1 . 2)} to the @code{extra-offset}
979 TextScript object. These numbers are measured in staff-spaces, so
980 this command moves the object 1 staff space to the right, and 2 spaces up.
982 Procedures for working with offsets are found in @file{scm/lily-library.scm}.
985 @unnumberedsubsubsec Fractions
987 Fractions as used by LilyPond are again stored as @emph{pairs}, this
988 time of unsigned integers. While Scheme can represent rational numbers
989 as a native type, musically @samp{2/4} and @samp{1/2} are not the same,
990 and we need to be able to distinguish between them. Similarly there are
991 no negative @q{fractions} in LilyPond's mind. So @code{2/4} in LilyPond
992 means @code{(2 . 4)} in Scheme, and @code{#2/4} in LilyPond means
993 @code{1/2} in Scheme.
996 @unnumberedsubsubsec Extents
998 Pairs are also used to store intervals, which represent a range of numbers
999 from the minimum (the @code{car}) to the maximum (the @code{cdr}).
1000 Intervals are used to store the X- and Y- extents of printable objects.
1001 For X extents, the @code{car} is the left hand X coordinate, and the
1002 @code{cdr} is the right hand X coordinate. For Y extents, the @code{car}
1003 is the bottom coordinate, and the @code{cdr} is the top coordinate.
1005 Procedures for working with intervals are found in
1006 @file{scm/lily-library.scm}. These procedures should be used when possible
1007 to ensure consistency of code.
1009 @node Property alists
1010 @unnumberedsubsubsec Property alists
1012 A property alist is a LilyPond data structure that is an alist whose
1013 keys are properties and whose values are Scheme expressions that give
1014 the desired value for the property.
1016 LilyPond properties are Scheme symbols, such as @code{'thickness}.
1019 @unnumberedsubsubsec Alist chains
1021 An alist chain is a list containing property alists.
1023 The set of all properties that will apply to a grob is typically
1024 stored as an alist chain. In order to find the value for a particular
1025 property that a grob should have, each alist in the chain is searched in
1026 order, looking for an entry containing the property key. The first alist
1027 entry found is returned, and the value is the property value.
1029 The Scheme procedure @code{chain-assoc-get} is normally used to get
1030 grob property values.
1032 @node Internal music representation
1033 @subsection Internal music representation
1035 Internally, music is represented as a Scheme list. The list contains
1036 various elements that affect the printed output. Parsing is the process
1037 of converting music from the LilyPond input representation to the
1038 internal Scheme representation.
1040 When a music expression is parsed, it is converted into a set of
1041 Scheme music objects. The defining property of a music object is that
1042 it takes up time. The time it takes up is called its @emph{duration}.
1043 Durations are expressed as a rational number that measures the length
1044 of the music object in whole notes.
1046 A music object has three kinds of types:
1049 music name: Each music expression has a name. For example, a note
1050 leads to a @rinternals{NoteEvent}, and @code{\simultaneous} leads to
1051 a @rinternals{SimultaneousMusic}. A list of all expressions
1052 available is in the Internals Reference manual, under
1053 @rinternals{Music expressions}.
1056 @q{type} or interface: Each music name has several @q{types} or
1057 interfaces, for example, a note is an @code{event}, but it is also a
1058 @code{note-event}, a @code{rhythmic-event}, and a
1059 @code{melodic-event}. All classes of music are listed in the
1060 Internals Reference, under
1061 @rinternals{Music classes}.
1064 C++ object: Each music object is represented by an object of the C++
1068 The actual information of a music expression is stored in properties.
1069 For example, a @rinternals{NoteEvent} has @code{pitch} and
1070 @code{duration} properties that store the pitch and duration of that
1071 note. A list of all properties available can be found in the
1072 Internals Reference, under @rinternals{Music properties}.
1074 A compound music expression is a music object that contains other
1075 music objects in its properties. A list of objects can be stored in
1076 the @code{elements} property of a music object, or a single @q{child}
1077 music object in the @code{element} property. For example,
1078 @rinternals{SequentialMusic} has its children in @code{elements},
1079 and @rinternals{GraceMusic} has its single argument in
1080 @code{element}. The body of a repeat is stored in the @code{element}
1081 property of @rinternals{RepeatedMusic}, and the alternatives in
1084 @node Building complicated functions
1085 @section Building complicated functions
1087 This section explains how to gather the information necessary
1088 to create complicated music functions.
1091 * Displaying music expressions::
1092 * Music properties::
1093 * Doubling a note with slurs (example)::
1094 * Adding articulation to notes (example)::
1097 @node Displaying music expressions
1098 @subsection Displaying music expressions
1100 @cindex internal storage
1101 @cindex displaying music expressions
1102 @cindex internal representation, displaying
1103 @cindex displayMusic
1104 @funindex \displayMusic
1106 When writing a music function it is often instructive to inspect how
1107 a music expression is stored internally. This can be done with the
1108 music function @code{\displayMusic}
1112 \displayMusic @{ c'4\f @}
1127 'AbsoluteDynamicEvent
1131 (ly:make-duration 2 0 1/1)
1133 (ly:make-pitch 0 0 0))))
1136 By default, LilyPond will print these messages to the console along
1137 with all the other messages. To split up these messages and save
1138 the results of @code{\display@{STUFF@}}, you can specify an optional
1143 \displayMusic #(open-output-file "display.txt") @{ c'4\f @}
1147 This will overwrite a previous output file whenever it is called; if you
1148 need to write more than one expression, you would use a variable for
1149 your port and reuse it:
1152 port = #(open-output-file "display.txt")
1153 \displayMusic \port @{ c'4\f @}
1154 \displayMusic \port @{ d'4 @}
1155 #(close-output-port port)
1159 Guile's manual describes ports in detail. Closing the port is actually
1160 only necessary if you need to read the file before Lilypond finishes; in
1161 the first example, we did not bother to do so.
1163 A bit of reformatting makes the above information easier to read:
1166 (make-music 'SequentialMusic
1168 (make-music 'NoteEvent
1169 'articulations (list
1170 (make-music 'AbsoluteDynamicEvent
1173 'duration (ly:make-duration 2 0 1/1)
1174 'pitch (ly:make-pitch 0 0 0))))
1177 A @code{@{ @dots{} @}} music sequence has the name
1178 @code{SequentialMusic}, and its inner expressions are stored as a list
1179 in its @code{'elements} property. A note is represented as a
1180 @code{NoteEvent} object (storing the duration and pitch properties) with
1181 attached information (in this case, an @code{AbsoluteDynamicEvent} with
1182 a @code{"f"} text property) stored in its @code{articulations} property.
1185 @code{\displayMusic} returns the music it displays, so it will get
1186 interpreted as well as displayed. To avoid interpretation, write
1187 @code{\void} before @code{\displayMusic}.
1189 @node Music properties
1190 @subsection Music properties
1192 TODO -- make sure we delineate between @emph{music} properties,
1193 @emph{context} properties, and @emph{layout} properties. These
1194 are potentially confusing.
1196 Let's look at an example:
1200 \displayMusic \someNote
1205 (ly:make-duration 2 0 1/1)
1207 (ly:make-pitch 0 0 0))
1210 The @code{NoteEvent} object is the representation of @code{someNote}.
1211 Straightforward. How about putting c' in a chord?
1215 \displayMusic \someNote
1223 (ly:make-duration 2 0 1/1)
1225 (ly:make-pitch 0 0 0))))
1228 Now the @code{NoteEvent} object is the first object of the
1229 @code{'elements} property of @code{someNote}.
1231 The @code{display-scheme-music} function is the function used by
1232 @code{\displayMusic} to display the Scheme representation of a music
1236 #(display-scheme-music (first (ly:music-property someNote 'elements)))
1241 (ly:make-duration 2 0 1/1)
1243 (ly:make-pitch 0 0 0))
1246 Then the note pitch is accessed through the @code{'pitch} property
1247 of the @code{NoteEvent} object,
1250 #(display-scheme-music
1251 (ly:music-property (first (ly:music-property someNote 'elements))
1254 (ly:make-pitch 0 0 0)
1257 The note pitch can be changed by setting this @code{'pitch} property,
1259 @funindex \displayLilyMusic
1262 #(set! (ly:music-property (first (ly:music-property someNote 'elements))
1264 (ly:make-pitch 0 1 0)) ;; set the pitch to d'.
1265 \displayLilyMusic \someNote
1271 @node Doubling a note with slurs (example)
1272 @subsection Doubling a note with slurs (example)
1274 Suppose we want to create a function that translates input like
1275 @code{a} into @code{@{ a( a) @}}. We begin by examining the internal
1276 representation of the desired result.
1279 \displayMusic@{ a'( a') @}
1292 (ly:make-duration 2 0 1/1)
1294 (ly:make-pitch 0 5 0))
1303 (ly:make-duration 2 0 1/1)
1305 (ly:make-pitch 0 5 0))))
1308 The bad news is that the @code{SlurEvent} expressions
1309 must be added @q{inside} the note (in its @code{articulations}
1312 Now we examine the input,
1320 (ly:make-duration 2 0 1/1)
1322 (ly:make-pitch 0 5 0))))
1325 So in our function, we need to clone this expression (so that we have
1326 two notes to build the sequence), add a @code{SlurEvent} to the
1327 @code{'articulations} property of each one, and finally make a
1328 @code{SequentialMusic} with the two @code{NoteEvent} elements. For adding to a
1329 property, it is useful to know that an unset property is read out as
1330 @code{'()}, the empty list, so no special checks are required before we
1331 put another element at the front of the @code{articulations} property.
1334 doubleSlur = #(define-music-function (note) (ly:music?)
1335 "Return: @{ note ( note ) @}.
1336 `note' is supposed to be a single note."
1337 (let ((note2 (ly:music-deep-copy note)))
1338 (set! (ly:music-property note 'articulations)
1339 (cons (make-music 'SlurEvent 'span-direction -1)
1340 (ly:music-property note 'articulations)))
1341 (set! (ly:music-property note2 'articulations)
1342 (cons (make-music 'SlurEvent 'span-direction 1)
1343 (ly:music-property note2 'articulations)))
1344 (make-music 'SequentialMusic 'elements (list note note2))))
1348 @node Adding articulation to notes (example)
1349 @subsection Adding articulation to notes (example)
1351 The easy way to add articulation to notes is to merge two music
1352 expressions into one context.
1353 However, suppose that we want to write a music function that does this.
1354 This will have the additional advantage that we can use that music
1355 function to add an articulation (like a fingering instruction) to a
1356 single note inside of a chord which is not possible if we just merge
1359 A @code{$variable} inside the @code{#@{@dots{}#@}} notation is like
1360 a regular @code{\variable} in classical LilyPond notation. We
1368 will not work in LilyPond. We could avoid this problem by attaching
1369 the articulation to an empty chord,
1372 @{ << \music <> -. -> >> @}
1376 but for the sake of this example, we will learn how to do this in
1377 Scheme. We begin by examining our input and desired output,
1386 (ly:make-duration 2 0 1/1)
1388 (ly:make-pitch -1 0 0))))
1401 (ly:make-duration 2 0 1/1)
1403 (ly:make-pitch -1 0 0))
1406 We see that a note (@code{c4}) is represented as an @code{NoteEvent}
1407 expression. To add an accent articulation, an @code{ArticulationEvent}
1408 expression must be added to the @code{articulations} property of the
1409 @code{NoteEvent} expression.
1411 To build this function, we begin with
1414 (define (add-accent note-event)
1415 "Add an accent ArticulationEvent to the articulations of `note-event',
1416 which is supposed to be a NoteEvent expression."
1417 (set! (ly:music-property note-event 'articulations)
1418 (cons (make-music 'ArticulationEvent
1419 'articulation-type "accent")
1420 (ly:music-property note-event 'articulations)))
1424 The first line is the way to define a function in Scheme: the function
1425 name is @code{add-accent}, and has one variable called
1426 @code{note-event}. In Scheme, the type of variable is often clear
1427 from its name. (this is good practice in other programming languages,
1431 "Add an accent@dots{}"
1435 is a description of what the function does. This is not strictly
1436 necessary, but just like clear variable names, it is good practice.
1438 You may wonder why we modify the note event directly instead of working
1439 on a copy (@code{ly:music-deep-copy} can be used for that). The reason
1440 is a silent contract: music functions are allowed to modify their
1441 arguments: they are either generated from scratch (like user input) or
1442 are already copied (referencing a music variable with @samp{\name} or
1443 music from immediate Scheme expressions @samp{$(@dots{})} provides a
1444 copy). Since it would be inefficient to create unnecessary copies, the
1445 return value from a music function is @emph{not} copied. So to heed
1446 that contract, you must not use any arguments more than once, and
1447 returning it counts as one use.
1449 In an earlier example, we constructed music by repeating a given music
1450 argument. In that case, at least one repetition had to be a copy of its
1451 own. If it weren't, strange things may happen. For example, if you use
1452 @code{\relative} or @code{\transpose} on the resulting music containing
1453 the same elements multiple times, those will be subjected to
1454 relativation or transposition multiple times. If you assign them to a
1455 music variable, the curse is broken since referencing @samp{\name} will
1456 again create a copy which does not retain the identity of the repeated
1459 Now while the above function is not a music function, it will normally
1460 be used within music functions. So it makes sense to heed the same
1461 contract we use for music functions: the input may be modified for
1462 producing the output, and the caller is responsible for creating copies
1463 if it still needs the unchanged argument itself. If you take a look at
1464 LilyPond's own functions like @code{music-map}, you'll find that they
1465 stick with the same principles.
1467 Where were we? Now we have a @code{note-event} we may modify, not
1468 because of using @code{ly:music-deep-copy} but because of a long-winded
1469 explanation. We add the accent to its @code{'articulations} list
1473 (set! place new-value)
1476 Here, what we want to set (the @q{place}) is the @code{'articulations}
1477 property of @code{note-event} expression.
1480 (ly:music-property note-event 'articulations)
1483 @code{ly:music-property} is the function used to access music properties
1484 (the @code{'articulations}, @code{'duration}, @code{'pitch}, etc, that we
1485 see in the @code{\displayMusic} output above). The new value is the
1486 former @code{'articulations} property, with an extra item: the
1487 @code{ArticulationEvent} expression, which we copy from the
1488 @code{\displayMusic} output,
1491 (cons (make-music 'ArticulationEvent
1492 'articulation-type "accent")
1493 (ly:music-property result-event-chord 'articulations))
1496 @code{cons} is used to add an element to the front of a list without
1497 modifying the original list. This is what we want: the same list as
1498 before, plus the new @code{ArticulationEvent} expression. The order
1499 inside the @code{'articulations} property is not important here.
1501 Finally, once we have added the accent articulation to its
1502 @code{articulations} property, we can return @code{note-event}, hence
1503 the last line of the function.
1505 Now we transform the @code{add-accent} function into a music
1506 function (a matter of some syntactic sugar and a declaration of the type
1507 of its sole @q{real} argument).
1510 addAccent = #(define-music-function (note-event)
1512 "Add an accent ArticulationEvent to the articulations of `note-event',
1513 which is supposed to be a NoteEvent expression."
1514 (set! (ly:music-property note-event 'articulations)
1515 (cons (make-music 'ArticulationEvent
1516 'articulation-type "accent")
1517 (ly:music-property note-event 'articulations)))
1521 We may verify that this music function works correctly,
1524 \displayMusic \addAccent c4
1533 * Tweaking with Scheme::
1536 @c @nod e Tweaking with Scheme
1537 @c @sectio n Tweaking with Scheme
1539 We have seen how LilyPond output can be heavily modified using
1541 @code{\override TextScript.extra-offset = ( 1 . -1)}. But
1542 we have even more power if we use Scheme. For a full explanation
1543 of this, see the @ref{Scheme tutorial}, and
1544 @ref{Interfaces for programmers}.
1546 We can use Scheme to simply @code{\override} commands,
1548 TODO Find a simple example
1549 @c This isn't a valid example with skylining
1550 @c It works fine without padText -td
1554 @lilypond[quote,verbatim,ragged-right]
1555 padText = #(define-music-function (padding) (number?)
1557 \once \override TextScript.padding = #padding
1561 c'''4^"piu mosso" b a b
1563 c4^"piu mosso" d e f
1565 c4^"piu mosso" fis a g
1571 We can use it to create new commands:
1573 @c Check this is a valid example with skylining
1574 @c It is - 'padding still works
1577 @lilypond[quote,verbatim,ragged-right]
1578 tempoPadded = #(define-music-function (padding tempotext)
1581 \once \override Score.MetronomeMark.padding = #padding
1582 \tempo \markup { \bold #tempotext }
1586 \tempo \markup { "Low tempo" }
1588 \tempoPadded #4.0 "High tempo"
1594 Even music expressions can be passed in:
1596 @lilypond[quote,verbatim,ragged-right]
1597 pattern = #(define-music-function (x y) (ly:music? ly:music?)
1604 \pattern {d16 dis} { ais16-> b\p }