Metaprogramming
You are encouraged to solve this task according to the task description, using any language you may know.
Name and briefly demonstrate any support your language has for metaprogramming. Your demonstration may take the form of cross-references to other tasks on Rosetta Code. When possible, provide links to relevant documentation.
For the purposes of this task, "support for metaprogramming" means any way the user can effectively modify the language's syntax that's built into the language (like Lisp macros) or that's conventionally used with the language (like the C preprocessor). Such facilities need not be very powerful: even user-defined infix operators count. On the other hand, in general, neither operator overloading nor eval count. The task author acknowledges that what qualifies as metaprogramming is largely a judgment call.
ALGOL 68
ALGOL 68 allows the definition of new unary and binary operators, as well at the overloading of existing operators. This sample adds a COBOL-like INSPECT "statement" by defining suitable operators.
<lang algol68># This example uses ALGOL 68 user defined operators to add a COBOL-style #
- "INSPECT statement" to ALGOL 68 #
- #
- The (partial) syntax of the COBOL INSPECT is: #
- INSPECT string-variable REPLACING ALL string BY string #
- or INSPECT string-variable REPLACING LEADING string BY string #
- or INSPECT string-variable REPLACING FIRST string BY string #
- #
- Because "BY" is a reserved bold word in ALGOL 68, we use "WITH" instead #
- #
- We define unary operators INSPECT, ALL, LEADING and FIRST #
- and binary operators REPLACING and WITH #
- We choose the priorities of REPLACING and WITH so that parenthesis is not #
- needed to ensure the correct interpretation of the "statement" #
- #
- We also provide a unary DISPLAY operator for a partial COBOL DISPLAY #
- statement #
- INSPECTEE is returned by the INSPECT unary operator #
MODE INSPECTEE = STRUCT( REF STRING item, INT option );
- INSPECTTOREPLACE is returned by the binary REPLACING operator #
MODE INSPECTTOREPLACE
= STRUCT( REF STRING item, INT option, STRING to replace );
- REPLACEMENT is returned by the unary ALL, LEADING and FIRST operators #
MODE REPLACEMENT = STRUCT( INT option, STRING replace );
- REPLACING option codes, these are the option values for a REPLACEMENT #
INT replace all = 1; INT replace leading = 2; INT replace first = 3;
OP INSPECT = ( REF STRING s )INSPECTEE: ( s, 0 ); OP ALL = ( STRING replace )REPLACEMENT: ( replace all, replace ); OP LEADING = ( STRING replace )REPLACEMENT: ( replace leading, replace ); OP FIRST = ( STRING replace )REPLACEMENT: ( replace first, replace ); OP ALL = ( CHAR replace )REPLACEMENT: ( replace all, replace ); OP LEADING = ( CHAR replace )REPLACEMENT: ( replace leading, replace ); OP FIRST = ( CHAR replace )REPLACEMENT: ( replace first, replace );
OP REPLACING = ( INSPECTEE inspected, REPLACEMENT replace )INSPECTTOREPLACE:
( item OF inspected
, option OF replace
, replace OF replace
);
OP WITH = ( INSPECTTOREPLACE inspected, CHAR replace with )REF STRING:
BEGIN
STRING with := replace with;
inspected WITH with
END; # WITH #
OP WITH = ( INSPECTTOREPLACE inspected, STRING replace with )REF STRING:
BEGIN
STRING to replace = to replace OF inspected;
INT pos := 0;
STRING rest := item OF inspected;
STRING result := "";
IF option OF inspected = replace all
THEN
# replace all occurances of "to replace" with "replace with" #
WHILE string in string( to replace, pos, rest )
DO
result +:= rest[ 1 : pos - 1 ] + replace with;
rest := rest[ pos + UPB to replace : ]
OD
ELIF option OF inspected = replace leading
THEN
# replace leading occurances of "to replace" with "replace with" #
WHILE IF string in string( to replace, pos, rest )
THEN
pos = 1
ELSE
FALSE
FI
DO
result +:= replace with;
rest := rest[ 1 + UPB to replace : ]
OD
ELIF option OF inspected = replace first
THEN
# replace first occurance of "to replace" with "replace with" #
IF string in string( to replace, pos, rest )
THEN
result +:= rest[ 1 : pos - 1 ] + replace with;
rest := rest[ pos + UPB to replace : ]
FI
ELSE
# unsupported replace option #
write( ( newline, "*** unsupported INSPECT REPLACING...", newline ) );
stop
FI;
result +:= rest;
item OF inspected := result
END; # WITH #
OP DISPLAY = ( STRING s )VOID: write( ( s, newline ) );
PRIO REPLACING = 2, WITH = 1;
main: (
# test the INSPECT and DISPLAY "verbs" #
STRING text := "some text"; DISPLAY text;
INSPECT text REPLACING FIRST "e" WITH "bbc"; DISPLAY text;
INSPECT text REPLACING ALL "b" WITH "X"; DISPLAY text;
INSPECT text REPLACING ALL "text" WITH "some"; DISPLAY text;
INSPECT text REPLACING LEADING "som" WITH "k"; DISPLAY text
)</lang>
Output:
some text sombbc text somXXc text somXXc some kXXc some
C
C preprocessor can be used to extend language to some extent.
It's possible to create static assertions
<lang c> // http://stackoverflow.com/questions/3385515/static-assert-in-c
- define STATIC_ASSERT(COND,MSG) typedef char static_assertion_##MSG[(!!(COND))*2-1]
// token pasting madness:
- define COMPILE_TIME_ASSERT3(X,L) STATIC_ASSERT(X,static_assertion_at_line_##L)
- define COMPILE_TIME_ASSERT2(X,L) COMPILE_TIME_ASSERT3(X,L)
- define COMPILE_TIME_ASSERT(X) COMPILE_TIME_ASSERT2(X,__LINE__)
COMPILE_TIME_ASSERT(sizeof(long)==8); int main() {
COMPILE_TIME_ASSERT(sizeof(int)==4);
} </lang>
Another common usage is to create custom loops
<lang c> //requires C99
- define ITERATE_LIST(n, list) \
for(Node *n = (list)->head; n; n = n->next)
... ITERATE_LIST(n, list) {
printf("node value: %s\n", n->value);
} </lang>
For examples in real world, look FreeCiv, and OpenTTD macros(city_map_iterate for FreeCiv, FOR_ALL_ENGINES for OpenTTD).
Also, C does not support functions overloading, but becaus macro calls do not require type it's possible to emulate overloading to some extent
<lang c>
- define my_min(x, y) ((x) < (y) ? (x) : (y))
... printf("%f %d %ll\n", my_min(0.0f, 1.0f), my_min(1,2), my_min(1ll, 2ll)); </lang>
The Order programming language is implemented entirely using the C preprocessor, providing a portable, high-level, functional programming language that can be used to metaprogram any C99 project, in a fashion loosely similar to Lisp's macro system.
C#
Metaprogramming in C# can be achieved using the Text Template Transformation Toolkit. It is a textual preprocessor embedded in Visual Studio (it can also be executed from the command-line, e.g. in build scripts). It is language-agnostic, and therefore can generate code for C#, Visual Basic or other languages. This also means that it has no features which help manipulating the underlying language: it is purely textual, and does not include a C# parser to transform existing C# files (so you will need to roll your own or use Roslyn), and does not include utilities which would help with combining pieces of code.
Clojure
See Clojure Macros and Metaprogramming article.
Common Lisp
Built-In Fruits of Metaprogramming
Common Lisp is based on decades of metaprogramming, so programmers don't have to roll their own to benefit from it. For instance, the LOOP syntax is just a macro. Prior to becoming a standard language feature, it was just a library that users shared. The object system originated in the same way.
Calculate mean, and sample variance and sample standard deviation of some numbers:
<lang lisp> (loop for count from 1
for x in '(1 2 3 4 5)
summing x into sum
summing (* x x) into sum-of-squares
finally
(return
(let* ((mean (/ sum count))
(spl-var (- (* count sum-of-squares) (* sum sum)))
(spl-dev (sqrt (/ spl-var (1- count)))))
(values mean spl-var spl-dev)))) </lang>
=>
5/2 ; 105 ; 4.582576
Being a macro, if LOOP were removed from Lisp, it could be supplied by the application program. In fact, sometimes programs have included their own LOOP to work around bugs in some implementations.
Here is what CLISP makes of the above, by investigating the macro expansion using the ANSI standard macroexpand function:
<lang lisp> [5]>
(macroexpand'
(loop for count from 1
for x in '(1 2 3 4 5)
summing x into sum
summing (* x x) into sum-of-squares
finally
(return
(let* ((mean (/ sum count))
(spl-var (- (* count sum-of-squares) (* sum sum)))
(spl-dev (sqrt (/ spl-var (1- count)))))
(values mean spl-var spl-dev)))))
(MACROLET ((LOOP-FINISH NIL (SYSTEM::LOOP-FINISH-ERROR)))
(BLOCK NIL
(LET ((COUNT 1))
(LET ((#:LIST-3047 '(1 2 3 4 5)))
(PROGN
(LET ((X NIL))
(LET ((SUM-OF-SQUARES 0) (SUM 0))
(MACROLET ((LOOP-FINISH NIL '(GO SYSTEM::END-LOOP)))
(TAGBODY SYSTEM::BEGIN-LOOP (WHEN (ENDP #:LIST-3047) (LOOP-FINISH))
(SETQ X (CAR #:LIST-3047))
(PROGN (SETQ SUM (+ SUM X))
(SETQ SUM-OF-SQUARES (+ SUM-OF-SQUARES (* X X))))
(PSETQ COUNT (+ COUNT 1)) (PSETQ #:LIST-3047 (CDR #:LIST-3047))
(GO SYSTEM::BEGIN-LOOP) SYSTEM::END-LOOP
(MACROLET
((LOOP-FINISH NIL (SYSTEM::LOOP-FINISH-WARN) '(GO SYSTEM::END-LOOP)))
(PROGN
(RETURN
(LET*
((MEAN (/ SUM COUNT))
(SPL-VAR (- (* COUNT SUM-OF-SQUARES) (* SUM SUM)))
(SPL-DEV (SQRT (/ SPL-VAR (1- COUNT)))))
(VALUES MEAN SPL-VAR SPL-DEV)))))))))))))) ; T</lang>
Next, we can leave ANSI behind and call CLISP's internal code walker to expand the entire form, removing all traces of the definitions of local macros, leaving behind only pure code based on special forms and function calls:
<lang lisp>(system::expand-form
'(loop for count from 1
for x in '(1 2 3 4 5)
summing x into sum
summing (* x x) into sum-of-squares
finally
(return
(let* ((mean (/ sum count))
(spl-var (- (* count sum-of-squares) (* sum sum)))
(spl-dev (sqrt (/ spl-var (1- count)))))
(values mean spl-var spl-dev))))))
(BLOCK NIL
(LET ((COUNT 1))
(LET ((#:LIST-3230 '(1 2 3 4 5)))
(LET ((X NIL))
(LET ((SUM-OF-SQUARES 0) (SUM 0))
(TAGBODY SYSTEM::BEGIN-LOOP
(WHEN (ENDP #:LIST-3230) (GO SYSTEM::END-LOOP))
(SETQ X (CAR #:LIST-3230))
(PROGN (SETQ SUM (+ SUM X))
(SETQ SUM-OF-SQUARES (+ SUM-OF-SQUARES (* X X))))
(PSETQ COUNT (+ COUNT 1)) (PSETQ #:LIST-3230 (CDR #:LIST-3230))
(GO SYSTEM::BEGIN-LOOP) SYSTEM::END-LOOP
(RETURN-FROM NIL
(LET*
((MEAN (/ SUM COUNT))
(SPL-VAR (- (* COUNT SUM-OF-SQUARES) (* SUM SUM)))
(SPL-DEV (SQRT (/ SPL-VAR (1- COUNT)))))
(VALUES MEAN SPL-VAR SPL-DEV)))))))))</lang>
Implement monadic comprehensions
We can use Lisp macros, and other features, to add support to Lisp for monads, which come from functional languages. The following module of code provides a new macro form called COMPREHEND which works with monads. If we use the LIST monad, we get list comprehensions. For instance:
<lang lisp>
- The -> notation is not part of Lisp, it is used in examples indicate the output of a form.
(comprehend 'list-monad (cons x y) (x '(1 2 3)) (y '(A B C)))
-> ((1 . A) (1 . B) (1 . C)
(2 . A) (2 . B) (2 . C)
(3 . A) (3 . B) (3 . C))</lang>
As you can see, the comprehension processes all combinations of X and Y from both sets, and collects the application of (CONS X Y) to these elements.
In other words {∀x∀y:(cons x y) | x ∈ { 1, 2 ,3 } ∧ y ∈ { A, B, C }}
Other monads are possible: idenitity, state transfomer, etc. Some of these are provided in the code below.
Furthermore, a form called DEFINE-MONAD is provided to define new kinds of monads. It is used to define the basic monads. DEFINE-MONAD also optionally generates a (trivial) short-hand comprehension macro for your monad type. So instead of (comprehend 'list ...) it is possible to write is also (list-comp ...).
Note how the state transformer monad uses the identity monad comprehension in its definition.
Also, a monad is a class, and there is a way in the DEFINE-MONAD syntax to declare what the base classes are (multiple inheritance) as well as any additional custom slots.
Another example, using the identity monad. With the identity monad, the comprehension becomes a sequence of successive variable bindings, and a form evaluated in the scope of those bindings. It is basically like a "Lispified" Haskell DO syntax: <lang lisp>(identity-comp (list x y z) (x 1) (y (* 3 x)) (z (+ x y))) -> (1 3 4) </lang> I.e. combine the values X, Y and Z into a triplet list, were X is 1, Y is 3X, and Z is X + Y.
To see the original version of this code with lengthy comments, have a look in the Lisp Pastebin. http://paste.lisp.org/display/71196
<lang lisp>(defgeneric monadic-map (monad-class function))
(defgeneric monadic-join (monad-class container-of-containers &rest additional))
(defgeneric monadic-instance (monad-class-name))
(defmacro comprehend (monad-instance expr &rest clauses)
(let ((monad-var (gensym "CLASS-")))
(cond
((null clauses) `(multiple-value-call #'monadic-unit
,monad-instance ,expr))
((rest clauses) `(let ((,monad-var ,monad-instance))
(multiple-value-call #'monadic-join ,monad-var
(comprehend ,monad-var
(comprehend ,monad-var ,expr ,@(rest clauses))
,(first clauses)))))
(t (destructuring-bind (var &rest container-exprs) (first clauses)
(cond
((and var (symbolp var))
`(funcall (monadic-map ,monad-instance (lambda (,var) ,expr))
,(first container-exprs)))
((and (consp var) (every #'symbolp var))
`(multiple-value-call (monadic-map ,monad-instance
(lambda (,@var) ,expr))
,@container-exprs))
(t (error "COMPREHEND: bad variable specification: ~s" vars))))))))
(defmacro define-monad (class-name
&key comprehension
(monad-param (gensym "MONAD-"))
bases slots initargs
((:map ((map-param)
&body map-body)))
((:join ((join-param
&optional
(j-rest-kw '&rest)
(j-rest (gensym "JOIN-REST-")))
&body join-body)))
((:unit ((unit-param
&optional
(u-rest-kw '&rest)
(u-rest (gensym "UNIT-REST-")))
&body unit-body))))
`(progn
(defclass ,class-name ,bases ,slots)
(defmethod monadic-instance ((monad (eql ',class-name)))
(load-time-value (make-instance ',class-name ,@initargs)))
(defmethod monadic-map ((,monad-param ,class-name) ,map-param)
(declare (ignorable ,monad-param))
,@map-body)
(defmethod monadic-join ((,monad-param ,class-name)
,join-param &rest ,j-rest)
(declare (ignorable ,monad-param ,j-rest))
,@join-body)
(defmethod monadic-unit ((,monad-param ,class-name)
,unit-param &rest ,u-rest)
(declare (ignorable ,monad-param ,u-rest))
,@unit-body)
,@(if comprehension
`((defmacro ,comprehension (expr &rest clauses)
`(comprehend (monadic-instance ',',class-name)
,expr ,@clauses))))))
(defmethod monadic-map ((monad symbol) function)
(monadic-map (monadic-instance monad) function))
(defmethod monadic-join ((monad symbol) container-of-containers &rest rest)
(apply #'monadic-join (monadic-instance monad) container-of-containers rest))
(defmethod monadic-unit ((monad symbol) element &rest rest)
(apply #'monadic-unit (monadic-instance monad) element rest))
(define-monad list-monad
:comprehension list-comp :map ((function) (lambda (container) (mapcar function container))) :join ((list-of-lists) (reduce #'append list-of-lists)) :unit ((element) (list element)))
(define-monad identity-monad
:comprehension identity-comp :map ((f) f) :join ((x &rest rest) (apply #'values x rest)) :unit ((x &rest rest) (apply #'values x rest)))
(define-monad state-xform-monad
:comprehension state-xform-comp
:map ((f)
(lambda (xformer)
(lambda (s)
(identity-comp (values (funcall f x) new-state)
((x new-state) (funcall xformer s))))))
:join ((nested-xformer)
(lambda (s)
(identity-comp (values x new-state)
((embedded-xformer intermediate-state)
(funcall nested-xformer s))
((x new-state)
(funcall embedded-xformer intermediate-state)))))
:unit ((x) (lambda (s) (values x s))))</lang>
Python in Lisp
The CLPython project (http://common-lisp.net/project/clpython) provides a Python implementation embedded in Common Lisp. Python modules can be included in Lisp programs and interoperate with Lisp code. There is even a mixed-mode interactive loop ("REPL") where one can use a dialect which mixes Python and Lisp:
From the project documentation:
CLPython is able to turn a regular Lisp listener (REPL) into a "mixed-mode" listener that supports both Lisp and Python source as input:
clpython(213): (clpython:enter-mixed-lisp-python-syntax) ; The mixed Lisp/Python syntax mode is now enabled; ; Lispy *readtable* is now set. clpython(214): print 123 * 2 246 clpython(215): range(100)[98:2:-2] #(98 96 94 92 90 88 86 84 82 80 ...) clpython(216): (+ 1 2) 3It supports multi-line Python statements as long as the next lines are properly indented:
clpython(70): for i in range(4): print i, print i*2 0 0 1 2 2 4 3 6
Unfortunately, further metaprogramming within the Python is evidently discouraged (see Python section below).
D
Mixins enable string constants to be compiled as regular D code and inserted into the program. Combining this with compile time manipulation of strings enables the creation of domain-specific languages.
<lang d>enum GenStruct(string name, string fieldName) =
"struct " ~ name ~ "{ int " ~ fieldName ~ "; }";
// Equivalent to: struct Foo { int bar; } mixin(GenStruct!("Foo", "bar"));
void main() {
Foo f; f.bar = 10;
}</lang>
E
Forms of metaprogramming existant in E:
- Quasi-literals provide convenient notation for data structures and values for which there is not literal syntax, as discussed in String#E.
- E program fragments may be quoted, manipulated as an AST, and evaluated, similarly to Lisp; lexical environments are first-class objects (though static with respect to the evaluated code). Demonstrated in Runtime evaluation#E and Eval in environment#E.
- Control structures may be defined, as demonstrated in Extend your language#E.
Erlang
Metaprogramming, as understood for this task, is done with parse transformations in Erlang. This is what the documentation says: "Programmers are strongly advised not to engage in parse transformations".
Forth
Perhaps the most obvious use of Metaprogramming in Forth is Forth itself. The Forth virtual machine traditionally has a primitive operation called BRANCH, which as you could guess, does an unconditional branch to somewhere in the program. There is also ?BRANCH (sometimes called 0BRANCH) which branches only if the top of the Forth DATA stack is zero. There are typically also primitive "DO" and "LOOP" operators. All of these primitives cannot be used on their own and must be compiled into program code by the Forth compiler which needs to compute where to branch or loop to in the context of the code. The compiler itself is implemented in Forth and defines the syntax for conditional branches and loops. Understanding this code is not "needed" for application programming but exists inside the Forth system. The application programmer is free to use these same tools to create new branching and looping syntax if there is a need to do so. (FOR/NEXT, CASE ENDCASE etc.)
<LANG>\ BRANCH and LOOP COMPILERS
\ branch offset computation operators
- AHEAD ( -- addr) HERE 0 , ;
- BACK ( addr -- ) HERE - , ;
- RESOLVE ( addr -- ) HERE OVER - SWAP ! ;
\ LEAVE stack is called L0. It is initialized by QUIT.
- >L ( x -- ) ( L: -- x ) 2 LP +! LP @ ! ;
- L> ( -- x ) ( L: x -- ) LP @ @ -2 LP +! ;
\ finite loop compilers
- DO ( -- ) POSTPONE <DO> HERE 0 >L 3 ; IMMEDIATE
- ?DO ( -- ) POSTPONE <?DO> HERE 0 >L 3 ; IMMEDIATE
- LEAVE ( -- ) ( L: -- addr )
POSTPONE UNLOOP POSTPONE BRANCH AHEAD >L ; IMMEDIATE
- RAKE ( -- ) ( L: 0 a1 a2 .. aN -- )
BEGIN L> ?DUP WHILE RESOLVE REPEAT ; IMMEDIATE
- LOOP ( -- ) 3 ?PAIRS POSTPONE <LOOP> BACK RAKE ; IMMEDIATE
- +LOOP ( -- ) 3 ?PAIRS POSTPONE <+LOOP> BACK RAKE ; IMMEDIATE
\ conditional branches
- IF ( ? -- ) POSTPONE ?BRANCH AHEAD 2 ; IMMEDIATE
- THEN ( -- ) ?COMP 2 ?PAIRS RESOLVE ; IMMEDIATE
- ELSE ( -- ) 2 ?PAIRS POSTPONE BRANCH AHEAD SWAP 2
POSTPONE THEN 2 ; IMMEDIATE
\ infinite loop compilers
- BEGIN ( -- addr n) ?COMP HERE 1 ; IMMEDIATE
- AGAIN ( -- ) 1 ?PAIRS POSTPONE BRANCH BACK ; IMMEDIATE
- UNTIL ( ? -- ) 1 ?PAIRS POSTPONE ?BRANCH BACK ; IMMEDIATE
- WHILE ( ? -- ) POSTPONE IF 2+ ; IMMEDIATE
- REPEAT ( -- ) 2>R POSTPONE AGAIN 2R> 2- POSTPONE THEN ; IMMEDIATE</LANG>
Simple Usage Examples
<LANG> : CHARSET [CHAR] ~ [CHAR] ! DO I EMIT LOOP ;
- >DIGIT ( n -- c) DUP 9 > IF 7 + THEN [CHAR] 0 + ;
- -TRAILING ( adr len -- adr len') \ remove trailing blanks (spaces)
BEGIN
2DUP + 1- C@ BL = \ test if last char = blank
WHILE
1- \ dec. length
REPEAT ;</LANG>
FreeBASIC
Single line and multiple line macros can be used to modify or extend the language's syntax and are about as powerful as those found in the C language.
For example, we can create a 'forall' loop to iterate through the characters of a string rather than use a traditional 'for' loop:
<lang freebasic>' FB 1.05.0 Win64
- Macro ForAll(C, S)
For _i as integer = 0 To Len(s)
- Define C (Chr(s[_i]))
- EndMacro
- Define In ,
Dim s As String = "Rosetta" ForAll(c in s)
Print c; " ";
Next
Print Sleep</lang>
- Output:
R o s e t t a
Haskell
Metaprogramming is implemented using Template Haskell.
J
J names have one of four different grammatic types: noun, verb, adverb, conjunction. Nouns and verbs are nothing special from a metaprogramming point of view. However, adverbs and conjunctions are evaluated at "parse time" and can be used to introduce expression variants. (The result of an adverb, or of a conjunction may be either a noun, a verb, an adverb or a conjunction.)
Additionally, J script blocks (a block of text terminated by a right parenthesis on a single line) can and are used with differing interpreters (which may be built into the language or user written).
The J implementation of the Y combinator could be considered an example of metaprogramming.
J's explicit definitions might also be considered an example of metaprogramming, since explicit definitions have data dependent syntactic types.
Julia
Julia's metaprogramming features are descibed in the online documentation at https://docs.julialang.org/en/v1/manual/metaprogramming/index.html
Here is an example of metaprogramming. Julia in base form does not have C's do { } while() statement.
Using metaprogramming, the do/while can be somewhat emulated: <lang julia>macro dowhile(condition, block)
quote
while true
$(esc(block))
if !$(esc(condition))
break
end
end
end
end
macro dountil(condition, block)
quote
while true
$(esc(block))
if $(esc(condition))
break
end
end
end
end
using Primes
arr = [7, 31]
@dowhile (!isprime(arr[1]) && !isprime(arr[2])) begin
println(arr) arr .+= 1
end println("Done.")
@dountil (isprime(arr[1]) || isprime(arr[2])) begin
println(arr) arr .+= 1
end println("Done.")
</lang>
- Output:
[7, 31] [8, 32] [9, 33] [10, 34] Done. [11, 35] [12, 36] Done.
Kotlin
Although Kotlin doesn't support 'true' meta-programming, it does have facilities which make it possible to create something which closely resembles a language extension (see Extend_your_language#Kotlin).
It is also possible to define infix functions which look like user defined operators:
<lang scala>// version 1.0.6
infix fun Double.pwr(exp: Double) = Math.pow(this, exp)
fun main(args: Array<String>) {
val d = 2.0 pwr 8.0 println(d)
}</lang>
- Output:
256.0
Lingo
Lingo allows to create (and pre-compile) arbitrary code at runtime. You can't really change the language's syntax, but you can overwrite (or extend) built-in commands. Here as example some code that overwrite's Lingo's halt command, which would normally exit the current program:
<lang lingo>r = RETURN str = "on halt"&r&"--do nothing"&r&"end" new(#script).scripttext = str</lang>
Lua
Due to the way Lua's syntactic sugar is designed, metatables can make some Lua code look like a Domain-Specific Language, despite technically being (mostly) just specialized operator overloading.
For example: <lang lua> class "foo" : inherits "bar" {
} </lang>
is perfectly valid syntax. (Lua does not having a built in concept of classes or inheritance.)
M2000 Interpreter
<lang M2000 Interpreter> Module Meta {
FunName$="Alfa"
Code1$=FunName$+"=lambda (X)->{"
Code2$={
=x**2
}
Code3$="}"
Inline code1$+code2$+code3$
Print Function(FunName$, 4)=16
} Meta </lang>
Mathematica
Mathematica can overload all symbols, though sometimes Unprotect has to be invoked. You can also introduce your own infix operators: <lang Mathematica>CircleTimes[x_, y_] := Mod[x, 10] Mod[y, 10] 14\[CircleTimes]13</lang>
- Output:
12
For more info see: Operators in Mathematica
Nemerle
Nemerle provides support for macros, which range from defining new infix operators (in fact many 'built-in' operators are macros) to new keywords or control structures.
See here, here, and here on the Nemerle wiki for more information.
Nim
Infix Operators
You can define your own infix operators: <lang nim>proc `^`*[T: SomeInteger](base, exp: T): T =
var (base, exp) = (base, exp)
result = 1
while exp != 0:
if (exp and 1) != 0:
result *= base
exp = exp shr 1
base *= base
echo 2 ^ 10 # 1024</lang>
Compile Time evaluation
when is a compile time if and can be used to prevent code from even being parsed at compile time, for example to write platform specific code:
<lang nim>when defined windows:
echo "Call some Windows specific functions here"
elif defined linux:
echo "Call some Linux specific functions here"
else:
echo "Code for the other platforms"</lang>
Normal code can be executed at compile time if it is in a static block:
<lang nim>static:
echo "Hello Compile time world: ", 2 ^ 10</lang>
As well as stored in compile time constants: <lang nim>const x = 2 ^ 10</lang>
Templates
The expensive procedure has to be evaluated even when debug is false:
<lang nim>import os
const debug = false
proc expensive: string =
sleep(milsecs = 100) result = "That was difficult"
proc log(msg: string) =
if debug: echo msg
for i in 1..10:
log expensive()
</lang> This can be prevented using templates, as template calls are replaced with the template body at compile time: <lang>template log(msg: string) =
if debug: echo msg
for i in 1..10:
log expensive()</lang>
Templates can use block syntax with statement parameters: <lang nim>template times(x, y: untyped): untyped =
for i in 1..x: y
10.times: # or times 10: or times(10):
echo "hi" echo "bye"</lang>
Term Rewriting Templates
Term Rewriting Templates can be used to write your own optimizations: <lang>template optLog1{a and a}(a): auto = a template optLog2{a and (b or (not b))}(a,b): auto = a template optLog3{a and not a}(a: int): auto = 0
var
x = 12 s = x and x # Hint: optLog1(x) --> ’x’ [Pattern]
r = (x and x) and ((s or s) or (not (s or s))) # Hint: optLog2(x and x, s or s) --> ’x and x’ [Pattern] # Hint: optLog1(x) --> ’x’ [Pattern]
q = (s and not x) and not (s and not x) # Hint: optLog3(s and not x) --> ’0’ [Pattern]</lang>
Macros
The most powerful metaprogramming capabilities are offered by macros. They can generate source code or even an AST directly.
dumpTree can be useful when creating an AST, as it show you the AST of any code:
<lang nim>import macros
dumpTree:
if x:
if y:
p0
else:
p1
else:
if y:
p2
else:
p3</lang>
This prints:
StmtList
IfStmt
ElifBranch
Ident !"x"
StmtList
IfStmt
ElifBranch
Ident !"y"
StmtList
Ident !"p0"
Else
StmtList
Ident !"p1"
Else
StmtList
IfStmt
ElifBranch
Ident !"y"
StmtList
Ident !"p2"
Else
StmtList
Ident !"p3"
Using this information we can create an if2 macro for two conditions, as is done in the "Extend your language" task.
PARI/GP
The primary means of metaprogramming for GP is to extend it with PARI. As an example, defining an infix @ operator could work like
In src/functions/symbolic_operators/min0:
Function: _@_ Help: x@y: compute the lesser of x and y, or 0, whichever is larger. Section: symbolic_operators C-Name: gmin0 Prototype: GG Description: (small, small):small smin0ss($1, $2) (mp, mp):mp gmin0($1, $2) (gen, gen):gen gmin0($1, $2)
In (e.g.) basemath/somefile.c: <lang C>long smin0ss(long a, long b) {
long c = a < b ? a : b; return c > 0 ? c : 0;
}
GEN
gmin0(GEN a, GEN b)
{
GEN c = gcmp(a, b) < 1 ? a : b; /* copy pointer */ return signe(c) > 0 ? gcopy(c) : gen_0;
}</lang>
Perl
You can textually transform code with a source filter, a module that when used modifies the following lines of source. Filter::Util::Call provides a general means of writing source filters. Filter::Simple is an interface to Filter::Util::Call that lets you elide a lot of boilerplate code. More important, Filter::Simple can hide the contents of quoting constructs from your filter, obviating the biggest dangers of textual metaprogramming. For example, given the following module:
<lang perl>package UnicodeEllipsis;
use Filter::Simple;
FILTER_ONLY code => sub { s/…/../g };</lang>
this program:
<lang perl>use UnicodeEllipsis;
print join(' … ', 1 … 5), "\n";</lang>
prints:
1 … 2 … 3 … 4 … 5
Devel::Declare lets you define a new keyword by setting up a hook to be run whenever the parser encounters a bareword of your choice. Devel::Declare is powerful, but it has a reputation for being difficult to understand. See this blog post for a relatively simple usage example.
Perl 6
Perl 6 makes it very easy to do metaprogramming. It is a basic goal of the language.
It is trivial to add a new operator. Most Perl 6 operators are written as normal multiple-dispatch functions in the setting (known as a "prelude" in other languages, but in Perl 6 the setting is a lexical scope notionally surrounding your compilation unit).
There is no a built in factorial operator Perl 6. It was purposely left out to use as a demonstration of how easy it is to add it. :-)
<lang perl6>sub postfix:<!> { [*] 1..$^n } say 5!; # prints 120</lang>
You may augment a base class with a new method, as long as you declare that you are going to cheat.
Here we add a new method to do natural sorting to the base class Any. (List and Array are both subclasses of Any)
<lang perl6>use MONKEY-TYPING; # Needed to do runtime augmentation of a base class.
augment class List {
method nsort { self.list.sort: {$^a.lc.subst(/(\d+)/, -> $/ {0 ~ $0.chars.chr ~ $0 }, :g) ~ "\x0" ~ $^a} }
};
say ~<a201st a32nd a3rd a144th a17th a2 a95>.nsort; say ~<a201st a32nd a3rd a144th a17th a2 a95>.sort;</lang>
Prints
a2 a3rd a17th a32nd a95 a144th a201st a144th a17th a2 a201st a32nd a3rd a95
Grammar mixins work in Perl 6 because grammar rules are just methods in grammar classes, and Perl 6 automatically writes a JIT lexer for you whenever you derive a new language. This functionality already works internally in the standard parser—what is not yet implemented is the augment slang hook to allow user code to do this mixin. Perl 6 itself is already parsed using such grammar mixins to provide extensible quoting and regex syntax. For example, every time you pick your own quote characters, you're actually deriving a new Perl 6 dialect that supports those start and stop characters. Likewise any switches to impose single or double-quote semantics, or heredoc semantics, is merely a grammar mixin on the basic Q language. <lang perl6>say "Foo = $foo\n"; # normal double quotes say Q:qq 【Foo = $foo\n】; # a more explicit derivation, new quotes</lang>
PicoLisp
As in any Lisp, metaprogramming is an essential aspect of PicoLisp. In most cases normal functions are used to extend the language (see Extend your language#PicoLisp), read-macros operate on the source level, and also runtime macros are used occasionally.
PostScript
PostScript allows the reification of stack, scoping (dynamic scoping is default, but lexical scoping can be implemented using immediate loading), bindings using dictionaries, and even control flow. Here is an example of implementation of if statement
<lang postscript>
/ift { 4 dict begin
[/.if /.then] let*
count array astore /.stack exch def
/_restore {clear .stack aload pop}.
.stack aload pop .if {
_restore .then
} {
_restore
} ifelse
end}. </lang> The standard if expression in PostScript does not take a predicate. Instead it acts on the boolean value on top of the stack. This newly created word allows us to do <lang postscript> >| 2 {1 gt} {=} ift 2 </lang> Instead of
<lang postscript> >| 2 dup 1 gt {=} ift 2 </lang>
Note that even the let expression was implemented using meta programming <lang postscript> /let* {reverse {exch def} forall}. </lang>
Python
Metaprogramming is frowned on in Python and considered un-pythonic. The only widely known example of metaprogramming in Python was an implementation of a goto (and comefrom) keyword done as an April-fools joke.
Another more recent library that shows it can be done in Python is MacroPy.
This is example is taken from MacroPy's GitHub page. It creates a macro called expand that, when invoked, generates the AST for a function in place of the original expression.
<lang python> from macropy.core.macros import * from macropy.core.quotes import macros, q, ast, u
macros = Macros()
@macros.expr def expand(tree, **kw):
addition = 10 return q[lambda x: x * ast[tree] + u[addition]]
</lang>
It is then invoked like this: <lang python> func = expand[1 + 2] print func(5) </lang>
Racket
Racket has an extremely rich set of metaprogramming tools, which scale from simple pattern-based macros to implementing entire new languages with their own syntax, such as Datalog and Algol 60. Many parts of Racket itself, including its class-based object system, are implemented as macros that expand to a much smaller set of core forms.
As a descendent of the Scheme tradition, Racket provides hygienic pattern-based macros, allows the use of the full Racket language (including programmer-defined extensions) in implementing macros, and supports locally-defined macros.
Racket adds many extensions to this tradition, such as syntax-parse, which simplifies writing robust macros with good error reporting. For more information on Racket's metaprogramming features, see the relevant chapters of The Racket Guide and The Racket Reference.
For a simple example, this is the definition and a use of the macro list-when: <lang racket>#lang racket
(define-syntax-rule (list-when test body)
(if test
body
'()))
(let ([not-a-string 42])
(list-when (string? not-a-string) (string->list not-a-string)))</lang>
Unlike a plain function, which would eagerly evaluate its arguments, list-when only evaluates its body expression when its test expression passes: otherwise, it evaluates to the empty list. Therefore, the example above does not produce an error.
Alternatively, list-when could be defined using syntax-parse, which provides better error messages for syntax errors:
<lang racket>(require (for-syntax syntax/parse))
(define-syntax list-when
(syntax-parser
[(_ test:expr body:expr)
#'(if test
body
null)]))</lang>
Rascal
Rascal has been developed for metaprogramming. Rascal modules already have functionality to analyse Java code (see documentation).
Syntax Definition
In Rascal, grammars can be easily defined. The example below shows the syntax definition for the easy languages C and E1. ViewParseTree visualises the parse tree and lets the user interactively check whether sentences belong to the grammar. The greater than symbol in language E1 means that multiplication has a higher priority than addition.
<lang rascal>extend ViewParseTree;
layout Whitespace = [\ \t\n]*; syntax A = "a"; syntax B = "b"; start syntax C = "c" | A C B;
layout Whitespace = [\ \t\n]*; lexical Integer = [0-9]+; start syntax E1 = Integer
| E "*" E
> E "+" E | "(" E ")"
;</lang>
An example of the parse viewer for E1
Syntax Tree Traversal
Furthermore, Rascal has built-in functions to traverse trees. This can be used to visit all the nodes in the abstract syntax trees that are automatically generated by Rascal. This provides a powerful tool to analyse code. The following example counts for each operator how many of these the programme contains. <lang rascal>map[str, int] operatorUsage(PROGRAM P) {
m = ();
visit(P){
case add(_,_): m["add"] ? 0 += 1;
case sub(_,_): m["sub"] ? 0 += 1;
case conc(_,_): m["conc"] ? 0 += 1;
}
return m;
}</lang> Where the abstract syntax is defined as follows <lang rascal>public data TYPE = natural() | string();
public alias PicoId = str;
public data PROGRAM =
program(list[DECL] decls, list[STATEMENT] stats);
public data DECL =
decl(PicoId name, TYPE tp);
public data EXP =
id(PicoId name) | natCon(int iVal) | strCon(str sVal) | add(EXP left, EXP right) | sub(EXP left, EXP right) | conc(EXP left, EXP right) ;
public data STATEMENT =
asgStat(PicoId name, EXP exp)
| ifElseStat(EXP exp, list[STATEMENT] thenpart, list[STATEMENT] elsepart) | ifThenStat(EXP exp, list[STATEMENT] thenpart) | whileStat(EXP exp, list[STATEMENT] body) | doUntilStat(EXP exp, list[STATEMENT] body) | unlessStat(EXP exp, list[STATEMENT] body)
- </lang>
Pico in Rascal
This is part of the Pico syntax expressed in Rascal.
<lang rascal>module lang::pico::Syntax
import Prelude;
lexical Id = [a-z][a-z0-9]* !>> [a-z0-9]; lexical Natural = [0-9]+ ; lexical String = "\"" ![\"]* "\"";
layout Layout = WhitespaceAndComment* !>> [\ \t\n\r%];
lexical WhitespaceAndComment
= [\ \t\n\r] | @category="Comment" "%" ![%]+ "%" | @category="Comment" "%%" ![\n]* $ ;
start syntax Program
= program: "begin" Declarations decls {Statement ";"}* body "end" ;
syntax Declarations
= "declare" {Declaration ","}* decls ";" ;
syntax Declaration = decl: Id id ":" Type tp;
syntax Type
= natural:"natural" | string :"string" ;
syntax Statement
= asgStat: Id var ":=" Expression val
| ifElseStat: "if" Expression cond "then" {Statement ";"}* thenPart "else" {Statement ";"}* elsePart "fi"
| ifThenStat: "if" Expression cond "then" {Statement ";"}* thenPart "fi"
| whileStat: "while" Expression cond "do" {Statement ";"}* body "od"
| doUntilStat: "do" {Statement ";"}* body "until" Expression cond "od"
| unlessStat: Statement "unless" Expression cond
;
syntax Expression
= id: Id name
| strCon: String string
| natCon: Natural natcon
| bracket "(" Expression e ")"
> left conc: Expression lhs "||" Expression rhs
> left ( add: Expression lhs "+" Expression rhs
| sub: Expression lhs "-" Expression rhs
)
;
public start[Program] program(str s) {
return parse(#start[Program], s);
}
public start[Program] program(str s, loc l) {
return parse(#start[Program], s, l);
} </lang>
REXX
<lang rexx>/*┌───────────────────────────────────────────────────────────────────┐
│ The REXX language doesn't allow for the changing or overriding of │ │ syntax per se, but any of the built-in-functions (BIFs) can be │ │ overridden by just specifying your own. │ │ │ │ To use the REXX's version of a built-in-function, you simply just │ │ enclose the BIF in quotation marks (and uppercase the name). │ │ │ │ The intent is two-fold: the REXX language doesn't have any │ │ reserved words, nor reserved BIFs (Built-In-Functions). │ │ │ │ So, if you don't know that VERIFY is a BIF, you can just code │ │ a subroutine (or function) with that name (or any name), and not │ │ worry about your subroutine being pre-empted. │ │ │ │ Second: if you're not satisfied how a BIF works, you can code │ │ your own. This also allows you to front-end a BIF for debugging │ │ or modifying the BIF's behavior. │ └───────────────────────────────────────────────────────────────────┘ */
yyy='123456789abcdefghi'
rrr = substr(yyy,5) /*returns efghi */ mmm = 'SUBSTR'(yyy,5) /*returns 56789abcdefgji */ sss = "SUBSTR"(yyy,5) /* (same as above) */ exit /*stick a fork in it, we're done.*/
/*──────────────────────────────────SUBSTR subroutine───────────────────*/ substr: return right(arg(1),arg(2))
/*┌───────────────────────────────────────────────────────────────────┐
│ Also, some REXX interpreters treat whitespace(s) as blanks when │ │ performing comparisons. Some of the whitespace characters are: │ │ │ │ NL (newLine) │ │ FF (formFeed) │ │ VT (vertical tab) │ │ HT (horizontal tab or TAB) │ │ LF (lineFeed) │ │ CR (carriage return) │ │ EOF (end-of-file) │ │ and/or others. │ │ │ │ Note that some of the above are ASCII or EBCDIC specific. │ │ │ │ Some REXX interpreters use the OPTIONS statement to force │ │ REXX to only treat blanks as spaces. │ │ │ │ (Both the verb and option may be in lower/upper/mixed case.) │ │ │ │ REXX interpreters which don't recognize any option won't treat │ │ the (below) statement as an error. │ └───────────────────────────────────────────────────────────────────┘ */
options strict_white_space_comparisons /*can be in lower/upper/mixed.*/</lang>
Ring
The next program add new method to the object class during the runtime <lang ring> o1 = new point { x=10 y=20 z=30 } addmethod(o1,"print", func { see x + nl + y + nl + z + nl } ) o1.print() Class point
x y z
</lang>
The next example presents how to create a class that defines two instructions The first instruction is : I want window The second instruction is : Window title = Expression Also keywords that can be ignored like the ‘the’ keyword
<lang ring> New App {
I want window
The window title = "hello world"
}
Class App
func geti
if nIwantwindow = 0
nIwantwindow++
ok
func getwant
if nIwantwindow = 1
nIwantwindow++
ok
func getwindow
if nIwantwindow = 2
nIwantwindow= 0
see "Instruction : I want window" + nl
ok
if nWindowTitle = 0
nWindowTitle++
ok
func settitle cValue
if nWindowTitle = 1
nWindowTitle=0
see "Instruction : Window Title = " + cValue + nl
ok
private
# Attributes for the instruction I want window
i want window
nIwantwindow = 0
# Attributes for the instruction Window title
# Here we don't define the window attribute again
title
nWindowTitle = 0
# Keywords to ignore, just give them any value
the=0
</lang>
Ruby
An rudimentary example of metaprogramming is presented in this simple identification system template: <lang ruby>class IDVictim
# Create elements of this man, woman, or child's identification. attr_accessor :name, :birthday, :gender, :hometown # Allows you to put in a space for anything which is not covered by the # preexisting elements. def self.new_element(element) attr_accessor element end
end</lang>
The "self.new_element" class method allows one to (later) specify a new attribute to be added to the defaults of "name", "birthday", "gender", and "hometown".
Run BASIC
(This is not really metaprogramming, at least not under any useful meaning...)
<lang runbasic>' --------------------------------------------------- ' create a file to be run ' RB can run the entire program ' or execute a function withing the RUNNED program ' --------------------------------------------------- open "runned.bas" for output as #f ' open runned.bas as output
print #f, "text$ = ""I'm rinning the complete program. ' print this program to the output Or you can run a function. The program or function within the RUNNED program can execute all Run BASIC commands."""
print #f, " x = displayText(text$)"
print #f, " ' besides RUNNING the entireprogram Function displayText(text$) ' we will execute this function only print text$ ' end function"
' ---------------------------------------- ' Execute the entire RUNNED program ' ---------------------------------------- RUN "runned.bas",#handle ' setup run command to execute runned.bas and give it a handle render #handle ' render the handle will execute the program
' ---------------------------------------- ' Execute a function in the RUNNED program ' ---------------------------------------- RUN "runned.bas",#handle ' setup run command to execute runned.bas and give it a handle
- handle displayText("Welcome!") ' only execute the function withing the runned program
render #handle ' render the handle will execute the program</lang>
Rust
Rust supports extensive metaprogramming via macros. Note that rust macros differ from, say, C preprocessor macros in that they are not mere text substitution (so operator precedence is preserved and name shadowing is not an issue). Here is an example from rustbyexample.com that implements and tests the +=, -=, and *= operators for Vectors.
<lang rust>// dry.rs use std::ops::{Add, Mul, Sub};
macro_rules! assert_equal_len {
// The `tt` (token tree) designator is used for
// operators and tokens.
($a:ident, $b: ident, $func:ident, $op:tt) => (
assert!($a.len() == $b.len(),
"{:?}: dimension mismatch: {:?} {:?} {:?}",
stringify!($func),
($a.len(),),
stringify!($op),
($b.len(),));
)
}
macro_rules! op {
($func:ident, $bound:ident, $op:tt, $method:ident) => (
fn $func<T: $bound<T, Output=T> + Copy>(xs: &mut Vec<T>, ys: &Vec<T>) {
assert_equal_len!(xs, ys, $func, $op);
for (x, y) in xs.iter_mut().zip(ys.iter()) {
*x = $bound::$method(*x, *y);
// *x = x.$method(*y);
}
}
)
}
// Implement `add_assign`, `mul_assign`, and `sub_assign` functions. op!(add_assign, Add, +=, add); op!(mul_assign, Mul, *=, mul); op!(sub_assign, Sub, -=, sub);
mod test {
use std::iter;
macro_rules! test {
($func: ident, $x:expr, $y:expr, $z:expr) => {
#[test]
fn $func() {
for size in 0usize..10 {
let mut x: Vec<_> = iter::repeat($x).take(size).collect();
let y: Vec<_> = iter::repeat($y).take(size).collect();
let z: Vec<_> = iter::repeat($z).take(size).collect();
super::$func(&mut x, &y);
assert_eq!(x, z);
}
}
}
}
// Test `add_assign`, `mul_assign` and `sub_assign` test!(add_assign, 1u32, 2u32, 3u32); test!(mul_assign, 2u32, 3u32, 6u32); test!(sub_assign, 3u32, 2u32, 1u32);
}</lang>
- Output:
$ rustc --test dry.rs && ./dry running 3 tests test test::mul_assign ... ok test test::add_assign ... ok test test::sub_assign ... ok test result: ok. 3 passed; 0 failed; 0 ignored; 0 measured
Scala
<lang Scala>import scala.language.experimental.macros import scala.reflect.macros.Context
object Macros {
def impl(c: Context) = {
import c.universe._
c.Expr[Unit](q"""println("Hello World")""")
}
def hello: Unit = macro impl
}</lang>
SNOBOL4
There are several features of SNOBOL4 which could be considered meta-programming. The first of these is the ability to define synonyms for existing operators or functions, a feature which can help in creating DSLs of sorts in SNOBOL4 programs. For example the following code will alias the built-in function IDENT to SAME and the unary operator * to $:
<lang snobol4>
OPSYN('SAME','IDENT')
OPSYN('$','*',1)
</lang>
This is a simplistic use of OPSYN, however. More interesting is the aliasing of a function to an operator:
<lang snobol4>
OPSYN('F','*',1)
</lang>
In this setup, calling F(X) is the same as using the sequence *X which, in more complicated expressions, could result in better readability.
Other metaprogramming features supported would include the use of unevaluated expressions. If, in code, E is an expression it has a value as soon as it is defined and/or assigned to. *E, on the other hand, has a value only when it is evaluated either in the context of a pattern or in the context of an EVAL. The following example shows the motivation for unevaluated expressions in pattern matching contexts:
<lang snobol4>
&ANCHOR = 0 ; &TRIM = 1
WORD = BREAK(' .,') . W SPAN(' .,')
STRING1 = INPUT :F(ERROR)
STRING2 = INPUT :F(ERROR)
LOOP STRING1 WORD = :F(OUTPUT)
STRING2 ' ' W ANY(' .,') :F(LOOP)
LIST = LIST W ', ' :(LOOP)
OUTPUT OUTPUT = LIST END </lang>
In this code, two strings are input and a list of words appearing in both strings is generated. The problem with this code is that the pattern structure ' ' W ANY(' .,') is built on each iteration. Since pattern building is expensive, putting it in a loop like this is bad form. It cannot be moved outside of the loop, however, since the value of W changes for each iteration. The solution to this is to defer the evaluation of the variable W until it is needed while keeping the rest of the pattern intact:
<lang snobol4>
&ANCHOR = 0 ; &TRIM = 1
WORD = BREAK(' .,') . W SPAN(' .,')
FINDW = ' ' *W ANY(' .,')
STRING1 = INPUT :F(ERROR)
STRING2 = INPUT :F(ERROR)
LOOP STRING1 WORD = :F(OUTPUT)
STRING2 FINDW :F(LOOP)
LIST = LIST W ', ' :(LOOP)
OUTPUT OUTPUT = LIST
END
</lang>
In this code, the pattern is constructed only once in the line FINDW = ' ' *W ANY(' .,'). The value of the variable W, however, is only provided when FINDW is used in a pattern match. In this case it is given its value from the line before when STRING1 is matched against the pattern WORD. In this way the expense of building the pattern is paid only once, but the flexibility of matching a sequence of values is retained.
The final example of metaprogramming that's available lies in the idiosyncratic way that user-defined functions work in SNOBOL4. The fact that the DEFINE function can be recalled at any point to redefine any function is a powerful feature that can lead to very efficient code. (It can also lead to very unreadable code, of course, if not properly used.)
Consider this hand-waving example for the motivation: <lang snobol4>
- This example provides a bizarrely-expensive addition operation.
- It assumes the existence of an expensive procedure—say a database
- lookup—to extract the value to be added. This version uses the
- typical initialize-on-definition approach to implementation.
DEFINE('XADD(X)','XADD')
ADDVALUE = CALL_SOME_EXPENSIVE_OPERATION() :(XADD.END)
XADD XADD = X + ADDVALUE :(RETURN) XADD.END </lang>
In normal operation the interpreter will execute the DEFINE function and then execute the ADDVALUE = ... line, branching *past* the actual body of the function to the label XADD.END. If, however, there are many such functions, and especially if there's the possibility that these functions are never actually called, this could render program startup very slow. For purposes of amortizing initialization time, or for purposes of saving unnecessary initialization, the following code is better:
<lang snobol4>
DEFINE('XADD(X)','XADD.INIT') :(XADD.END)
XADD.INIT ADDVALUE = CALL_SOME_EXPENSIVE_OPERATION()
DEFINE('XADD(X)','XADD')
XADD XADD = X + ADDVALUE :(RETURN) XADD.END </lang>
The code now defines the XADD function and immediately, without doing initialization, jumps to the XADD.END label, bypassing both initialization and the function body. The trick here is that it defines the entry point of the function to be the XADD.INIT label. Now the first time XADD is called, control is transferred to XADD.INIT, the expensive initialization is performed, then the XADD function is *redefined* to point to the XADD label as the entry point. From this point onward all calls to XADD only perform the calculation, not the expensive initialization while the expensive initialization isn't paid at all unless the function is used at least once.
There are, of course, many other uses for function redefinition in this style which are suitable for metaprogramming efforts. Indeed such features are used prominently in the debugging subsystems of SNOBOL4 implementations.
Shen
Being a Lisp, metaprogramming is easily achievable in Shen through macros. However, defining macros is only possible when the typechecker is off. <lang shen>(define make-list
[A|D] -> [cons (make-list A) (make-list D)] X -> X)
(defmacro info-macro
[info Exp] -> [output "~A: ~A~%" (make-list Exp) Exp])
(info (* 5 6)) \\ outputs [* 5 6]: 30</lang> Like most macro systems, defmacro looks like a function that takes a sexp and returns one. However, Shen's defmacro is special in that it allows arbitrary activation of sexps. <lang shen>(0-) (defmacro +-macro
[A + B] -> [+ A B])
macro +-macro
(1-) (1 + (* 2 3)) 7</lang> It's important to be careful when using macros like this; this example would be bad because + is sometimes used as an argument to a function (e.g. (fold-left + 0) would compile to (+ fold-left 0)). However, the fact that a symbol can at once match a macro and denote a function can give the illusion of optional arguments or polyadicity. This is how many mathematical operators and functions like append work while retaining their type signature: <lang shen>(2-) (tc +) true
(3+) (+ 1 2 3) 6 : number
(4+) + + : (number --> (number --> number))
(5-) (tc -) false
(6-) (macroexpand [+ 1 2 3]) [+ 1 [+ 2 3]]</lang>
Sidef
Sidef recognizes all mathematical operators in Unicode and allows the user to define methods that behave like infix operators, even for built-in types. <lang ruby>class Number {
method ⊕(arg) {
self + arg
}
}
say (21 ⊕ 42)</lang>
Another example of metaprogramming, is the definition of methods at run-time:
<lang ruby>var colors = Hash(
'black' => "000",
'red' => "f00",
'green' => "0f0",
'yellow' => "ff0",
'blue' => "00f",
'magenta' => "f0f",
'cyan' => "0ff",
'white' => "fff",
)
for color,code in colors {
String.def_method("in_#{color}", func (self) {
'' + self + ''
})
}
say "blue".in_blue say "red".in_red say "white".in_white</lang>
- Output:
<span style="color: #00f">blue</span> <span style="color: #f00">red</span> <span style="color: #fff">white</span>
Tcl
Metaprogramming is considered to be normal in Tcl; the whole language was designed to support new operations that work with a similar level of integration to existing commands (and indeed, the standard commands are not syntactically special in any way), and the upvar and uplevel commands are specifically for this sort of use. Moreover, there are no language keywords that need to be worked around; words/tokens can be used to mean anything necessary. For example:
<lang tcl>proc loopVar {var from lower to upper script} {
if {$from ne "from" || $to ne "to"} {error "syntax error"}
upvar 1 $var v
if {$lower <= $upper} {
for {set v $lower} {$v <= $upper} {incr v} {
uplevel 1 $script
}
} else {
# $upper and $lower really the other way round here
for {set v $lower} {$v >= $upper} {incr v -1} {
uplevel 1 $script
}
}
}</lang>
The above creates a new loopVar command that might be used like this:
<lang tcl>loopVar x from 1 to 4 {
loopVar y from $x to 6 {
puts "pair is ($x,$y)"
if {$y >= 4} break
}
}</lang> Which prints this:
pair is (1,1) pair is (1,2) pair is (1,3) pair is (1,4) pair is (2,2) pair is (2,3) pair is (2,4) pair is (3,3) pair is (3,4) pair is (4,4)
As you can see, the new looping command is wholly integrated with the rest of the Tcl language.
Code generation is also possible. The easiest approach is to use the list command to generate substitution-free command, leaving all substitutions to places that are written by the programmer directly.
Finally, the total lack of keywords is exemplified by this classic fragment of Tcl:
<lang tcl>set set set</lang>
In this, the first set is a command (that updates a variable), the second is the name of a variable in the current namespace, and the third is a string that will be placed in a variable.
TXR
TXR has a built-in Lisp dialect called TXR Lisp, which supports meta-programming, some of which is patterned after ANSI Common Lisp. TXR provides:
- run-time access to its parser for Lisp expressions:
(read "(+ a b c)"); - a parser for regular exprssions:
(regex-parse "a.*b")which produces abstract syntax; - a run-time compiler for converting regular expression abstract syntax to compiled regular expression object;
- a
evalfunction which expands and evaluates Lisp abstract syntax; - global as well as lexically scoped macros, for both compound forms (with automatically destructured parameter lists) and symbols (symbol macros): the operators
defmacro,defsymacro,macroletandsymacrolet; - structural quasiquote for convenient macro writing.
Example define a while loop which supports break and continue. Two forms of break are supported break which causes the loop to terminate with the return value nil and (break <form>) which returns the specified value.
<lang txrlisp>(defmacro whil ((condition : result) . body)
(let ((cblk (gensym "cnt-blk-"))
(bblk (gensym "brk-blk-")))
^(macrolet ((break (value) ^(return-from ,',bblk ,value)))
(symacrolet ((break (return-from ,bblk))
(continue (return-from ,cblk)))
(block ,bblk
(for () (,condition ,result) ()
(block ,cblk ,*body)))))))
(let ((i 0))
(whil ((< i 100))
(if (< (inc i) 20)
continue)
(if (> i 30)
break)
(prinl i)))
(prinl
(sys:expand
'(whil ((< i 100))
(if (< (inc i) 20)
continue)
(if (> i 30)
break)
(prinl i))))</lang>
- Output:
20
21
22
23
24
25
26
27
28
29
30
(block #:brk-blk-0062
(for () ((< i 100) ())
() (block #:cnt-blk-0061
(if (< (sys:setq i (succ i))
20) (return-from
#:cnt-blk-0061))
(if (> i 30)
(return-from
#:brk-blk-0062))
(prinl i))))
zkl
Meta programming is quite limited unless you hack the compiler (which is written in zkl) or you write a DSL. Otherwise, there is a "kinda like" C pre-processor support. <lang zkl>#define name [0|1]
- if [0|1|name]
- else, #endif</lang>
<lang zkl>//Full zkl functionality but limited access to the parse stream; only #defines
- ifdef name
- fcn name {code}</lang>
<lang zkl>// Shove text into the parse stream
- text name text
- tokenize name, #tokenize f, #tokenize f(a)</lang>
<lang zkl>#<<<# text, any text, inside #<<<# pairs is ignored
- <<<#</lang>
<lang zkl>string:=
- <<<
"here docs: all text in #<<< pairs is collected into one [long] line and passed verbatim to the tokenizer. Illustrated here as quoted (\") strings can not span lines.";
- <<<
println(string); // contains newlines</lang> In addition, there is a concept of "parse space/time" - it is after parsing and before compiling where the full power of the language can be used to "so stuff". For example, enums can be implemented like so: <lang zkl>const{ var _n=-1; var [proxy] N=fcn{ _n+=1 } } const X=N; // → 0 println(_n); // → 2 code time is after const time const Y=N,Z=N; // → 1,2</lang>
- Programming Tasks
- Solutions by Programming Task
- BBC BASIC/Omit
- ALGOL 68
- C
- C sharp
- Clojure
- Common Lisp
- D
- E
- Erlang
- Forth
- FreeBASIC
- Haskell
- J
- Julia
- Kotlin
- Lingo
- Lua
- M2000 Interpreter
- Mathematica
- Nemerle
- Nim
- PARI/GP
- Perl
- Perl 6
- PicoLisp
- PostScript
- Initlib
- Python
- Racket
- Rascal
- REXX
- Ring
- Ruby
- Run BASIC
- Rust
- Scala
- SNOBOL4
- Shen
- Sidef
- Tcl
- TXR
- Zkl
- Ada/Omit
- AWK/Omit
- Bc/Omit
- Go/Omit
- JavaScript/Omit
- Sed/Omit