This reposirory will be known by a different brand Cicili
and archived. lcc moves to github.com/saman-pasha/cicili.
Lisp C Compiler aka. 'Cicili' programming language, which compiles Lisp-like syntax to C code and more extra features like method, lambda, defer.
- Install SBCL.
clang
required for compiling and linking.apt
orbrew
can be used to install Clang. Current used version:(clang-1700.0.13.3)
- lcc uses Libtool as default for perfoming better compiling and linking
C
code. Install it for your platform and put it in thePATH
environment variable. Compiler and linker could be set inconfig.lisp
file. Current used version:(GNU libtool) 2.5.4
- Download and copy lcc folder to
~/common-lisp
for enabling ASDF access to lcc package. - Write your own lcc code and save it in
.lcc
or.lisp
extension. - Copy
lcc.lisp
file from source folder into your project path. - Send your file as an argument to lcc.lisp.
sbcl --script lcc.lisp test.lisp
- If you are using EMACS editor, copy
mode.lisp
file content into.emacs
or.emacs.d/init.el
file for syntax highlighting.
- lcc now uses
IR
(Intermediate Representation) to handle more clauses and features. - modularizing lcc code makes clarity and easy to follow C code but makes debugging harder. refer to module test folder
module.lisp
sample. lambda
clause allows developer to write in-place function for sending as other function argument ordefer
destructure. refer to lambda test folderlambda.lisp
sample.defer
attribute. only available for variables defined bylet
expression. Allows developers to set a function how to destruct a variable or a pointer. refer to defer test folderdefer.lisp
sample.- auto deferral is a way let expressions will defined to automatically release dynamic memory allocated by
alloc
clause. refer to alloc test folderdefer.lisp
sample. method
clause will receive current instance or pointer asthis
parameter. Methods are defined outside a structure by access method operator->
placed between struct name and method name likeEmployee->Sign
. refer to method test foldermethod.lisp
sample.auto
variable type simplifies lambda and function pointer variables. alsotypeof
clause is added to use other variables type for define another variable.inline struct
can be defined in variable declaration, function parameters or outputs which permits to return multiple values from a function. refer to multimulti.lisp
file for complex samples.func
type allows developer to define a function pointer which wasn't available before.- refer to basic
basic.lisp
file for some struct definition samples. - refer to control
control.lisp
file for some control structures samples.
For basic variable definition refer to var var.lisp
file.
(var int amount)
(var double total)
(var double * total2)
int amount;
double total;
double * total2;
(var const int SIDE . 10)
(var const int * SIDE1 . #'(aof SIDE))
(var const int * const SIDE2 . #'(aof SIDE1))
const int SIDE = 10;
const int * SIDE1 = &SIDE;
const int * const SIDE2 = &SIDE1;
lcc Operator | C Operator |
---|---|
+ |
+ |
- |
- |
* |
* |
/ |
/ |
% |
% |
(set total (+ total amount))
(let ((int i . 3)
(int j . 7)
(int k))
(set k (+ i j)))
total = total + amount;
{
int i = 3, j = 7, k;
k = i + j;
}
lcc Operator | C Operator |
---|---|
++ |
prefix ++ |
-- |
prefix -- |
1+ |
postfix ++ |
1- |
postfix -- |
(source "main.c" ()
(include <stdio.h>)
(func main ()
(let ((int a . 5)
(int b . 5))
;; Print them and decrementing each time.
;; Use postfix mode for a and prefix mode for b.
(printf "\n%d %d" (1- a) (-- b))
(printf "\n%d %d" (1- a) (-- b))
(printf "\n%d %d" (1- a) (-- b))
(printf "\n%d %d" (1- a) (-- b))
(printf "\n%d %d" (1- a) (-- b)))))
#include <stdio.h>
int main()
{
{
int a = 5, b = 5;
//Print them and decrementing each time.
//Use postfix mode for a and prefix mode for b.
printf("\n%d %d", a--, --b);
printf("\n%d %d", a--, --b);
printf("\n%d %d", a--, --b);
printf("\n%d %d", a--, --b);
printf("\n%d %d", a--, --b);
}
}
lcc Operator | C Operator |
---|---|
== |
== |
!= |
!= |
> |
> |
< |
< |
>= |
>= |
<= |
<= |
lcc Operator | C Operator |
---|---|
and |
&& |
or |
|| |
not |
! |
lcc Operator | C Operator |
---|---|
<< |
<< |
>> |
>> |
~ |
~ |
bitand |
& |
bitor |
| |
xor |
^ |
^ |
^ |
lcc Operator | C Operator |
---|---|
set |
= |
+= |
+= |
-= |
-= |
*= |
*= |
/= |
/= |
%= |
%= |
<<= |
<<= |
>>= |
>>= |
lcc Operator | C Operator |
---|---|
? |
?: |
(set a (? (== b 2) 20 30))
a = (b == 2) ? 20 : 30;
lcc Operator | C Operator |
---|---|
sizeof |
sizeof() |
typeof |
typeof() |
aof |
& |
cof |
* |
ANSI C provides three types of data types:
- Primary(Built-in) Data Types: void, int, char, double and float.
- Derived Data Types: Array, References, and Pointers.
- User Defined Data Types: Structure, Union, and Enumeration.
lcc supports declaration and definition of all ANCI C data types.
lcc Data Type | C Data Type |
---|---|
nil |
NULL |
void |
void |
bool |
bool |
char |
char |
uchar |
unsigned char |
short |
short |
ushort |
unsigned short |
int |
int |
uint |
unsigned int |
long |
long |
ulong |
unsigned long |
llong |
long long |
ullong |
unsigned long long |
i8 |
int8_t |
u8 |
uint8_t |
i16 |
int16_t |
u16 |
uint16_t |
i32 |
int32_t |
u32 |
uint32_t |
i64 |
int64_t |
u64 |
uint64_t |
i128 |
__int128 |
u128 |
unsigned __int128 |
float |
float |
double |
double |
real |
long double |
auto |
__auto_type |
(source "main.c" ()
(func main ()
(let ((double price . 500.4) ; atom initialization
(double price_array [] . '{100.2 230.7 924.8}) ; list initialization
(double price_calc . #'(calculate_price)) ; initialization by output of a function
(auto identity . '(lambda ((int x)) (out int) (return x))))))) ; lambda initialization
int __lccLambda_main_178 (int x) {
return x ;
}
int main () {
{
double price = 500.4;
double price_array[] = {100.2, 230.7, 924.8};
double price_calc = calculate_price ();
__auto_type identity = __lccLambda_main_178 ;
}
}
A free variable can has some attributes or storage class. each attribute enclosed in braces or parentheses. free variables can only be defined as global variable for inside function variable declaration let
clause should be used.
- {auto}
- {register}
- {static}
- {extern}
{auto} (var int width)
{register} (var int height . 5)
(var char letter . #\A)
(var float age)
{extern} (var float area)
{static} (var double d)
;; actual initialization
(set width 10)
(set age 26.5)
auto int width;
register height = 5;
char letter = 'A';
float age;
extern float area;
static double d;
/* actual initialization */
width = 10;
age = 26.5;
A scoped variable can has some attributes or storage class. each attribute enclosed in braces or parentheses. let
clause allows to declare a defer
destructure for every variable as a lambda or a function which receives a pointer of variable type. useful for free struct pointers or any resource which stored inside a struct.
- {auto}
- {register}
- {static}
- {defer
'(lambda ((int * intPtr)) (printf "int gone out of scope\n"))
} variable destructor
(source "main.c" ()
(func main ()
(let ({static} (int width . 3)
{register} (int height . 4)
{defer '(lambda ((Employee ** empPtr))
(free (cof empPtr))
(printf "from defer, emp is freed\n"))}
(Employee * emp . #'(alloc (sizeof Employee))))
(printf "area: %d" (* width height)))))
void __lccLambda_main_178 (Employee ** empPtr) {
free ((*empPtr));
printf ("from defer, emp is freed\n");
}
int main () {
{
static int width = 3;
register int height = 4;
Employee * emp __attribute__((__cleanup__(__lccLambda_main_178))) = ((Employee *)malloc(sizeof(Employee)));
printf ("area: %d", (width * height));
}
}
(set width 60)
(set age 35)
(set width 65 age 40) ; multi assignment
width = 60;
age = 31;
width = 65;
age = 40;
(source "main.c" ()
(include <stdio.h>)
(func main ()
(let ((int age . 33))
(printf "I am %d years old.\n" age))))
#include <stdio.h>
int main()
{
{
int age = 33;
printf("I am %d years old.\n", age);
}
}
(source "main.c" ()
(include <stdio.h>)
(func main ()
(let ((float a))
(set a (cast float (/ 15 6)))
(printf "%f" a))))
#include <stdio.h>
main ()
{
{
float a;
a = (float) 15 / 6;
printf("%f", a);
}
}
lcc program involves one or many header or source forms calling targets. targets are translating its content forms to C code. header targets only compile its content without resolving, but source targets resolves attribue and method access of any struct variable. for example
(let ((Employee emp)
(Employee * empPtr))
($ emp id) ; do not needd resolve
($ empPtr id)) ; resolves pointer access
{
Employee emp;
Employee * empPtr;
emp.id;
empPtr->id;
}
each target must has a target c file, and a list of feature arguments.
All features could be omitted or if available accept #t
for default behaviour or #f
for do nothing.
- :std: writes standard libraries inclusion at top of target file.
(source "main.c"
(:std #t)
;; some forms
)
#include <stdio.h>
#include <stddef.h>
#include <stdint.h>
#include <stdlib.h>
#include <stdbool.h>
- :compile: used for compiling target file. Dafault behaviour is
-c target.c
. Could be a list of arguments that will send to compiler which has been set inconfig.lisp
. - :link: used for linking and builing target file as library or executable. It has not default behaviour. Could be a list of arguments that will send to linker which has been set in
config.lisp
.
;; MyMath library declaration
(header "mymath.h"
(:compile #f)
(guard __MYMATH_H__
{decl} (func obj1_does ((int) (int)) (out int))
{decl} (func obj2_does ((int) (int)) (out int))
{decl} (func obj3_does ((int) (int)) (out int))))
;; Default compilation
(source "obj1.c"
(:compile #t)
(include "mymath.h")
(func obj1_does ((int x) (int y)) (out int)
(return (+ x y))))
;; Custom compilation
(source "obj2.c"
(:compile "-c obj2.c -o objmul.lo")
(include "mymath.h")
(func obj2_does ((int x) (int y)) (out int)
(return (* x y))))
;; Library creation and linking
(source "obj3.c"
(:compile #t :link "-o libMyMath.la -L{$CWD} obj1.lo objmul.lo obj3.lo")
(include "mymath.h")
(func obj3_does ((int x) (int y)) (out int)
(return (obj1_does (obj2_does x y) (obj2_does x y)))))
;; Executable creation and linking
(source "main.c"
(:std #t :compile #t :link "-o CompileTest -L{$CWD} main.lo -lMyMath")
(include "mymath.h")
(func main ((int argc) (char * argv []))
(if (!= argc 3)
(block
(printf "two digits needed!")
(return EXIT_FAILURE)))
(let ((int x . #'(atoi (nth 1 argv)))
(int y . #'(atoi (nth 2 argv))))
(printf "MyMath lib outputs: %d\n" (obj3_does x y)))
(return EXIT_SUCCESS)))
lcc % sbcl --script lcc.lisp test/test.lisp
software type: "Darwin"
arg specified: test/mylib.lisp
lcc: specifying target mymath.h
lcc: resolving target mymath.h
lcc: specifying target obj1.c
lcc: resolving target obj1.c.run1.c
run out 1 > glibtool: compile: clang -g -O "" -c obj1.c.run1.c -fno-common -DPIC -o .libs/obj1.c.run1.o
run out 1 > glibtool: compile: clang -g -O "" -c obj1.c.run1.c -o obj1.c.run1.o >/dev/null 2>&1
lcc: resolving target obj1.c.run2.c
run out 2 > glibtool: compile: clang -g -O "" -c obj1.c.run2.c -fno-common -DPIC -o .libs/obj1.c.run2.o
run out 2 > glibtool: compile: clang -g -O "" -c obj1.c.run2.c -o obj1.c.run2.o >/dev/null 2>&1
lcc: resolving target obj1.c.run3.c
run out 3 > glibtool: compile: clang -g -O "" -c obj1.c.run3.c -fno-common -DPIC -o .libs/obj1.c.run3.o
run out 3 > glibtool: compile: clang -g -O "" -c obj1.c.run3.c -o obj1.c.run3.o >/dev/null 2>&1
lcc: compiling target obj1.c
glibtool: compile: clang -g -O "" -c obj1.c -fno-common -DPIC -o .libs/obj1.o
glibtool: compile: clang -g -O "" -c obj1.c -o obj1.o >/dev/null 2>&1
lcc: specifying target obj2.c
lcc: resolving target obj2.c.run1.c
run out 1 > glibtool: compile: clang -g -O "" -c obj2.c -fno-common -DPIC -o .libs/objmul.o
run out 1 > glibtool: compile: clang -g -O "" -c obj2.c -o objmul.o >/dev/null 2>&1
lcc: resolving target obj2.c.run2.c
run out 2 > glibtool: compile: clang -g -O "" -c obj2.c -fno-common -DPIC -o .libs/objmul.o
run out 2 > glibtool: compile: clang -g -O "" -c obj2.c -o objmul.o >/dev/null 2>&1
lcc: resolving target obj2.c.run3.c
run out 3 > glibtool: compile: clang -g -O "" -c obj2.c -fno-common -DPIC -o .libs/objmul.o
run out 3 > glibtool: compile: clang -g -O "" -c obj2.c -o objmul.o >/dev/null 2>&1
lcc: compiling target obj2.c
glibtool: compile: clang -g -O "" -c obj2.c -fno-common -DPIC -o .libs/objmul.o
glibtool: compile: clang -g -O "" -c obj2.c -o objmul.o >/dev/null 2>&1
lcc: specifying target obj3.c
lcc: resolving target obj3.c.run1.c
run out 1 > glibtool: compile: clang -g -O "" -c obj3.c.run1.c -fno-common -DPIC -o .libs/obj3.c.run1.o
run out 1 > glibtool: compile: clang -g -O "" -c obj3.c.run1.c -o obj3.c.run1.o >/dev/null 2>&1
run out 1 > glibtool: link: rm -fr .libs/libMyMath.a .libs/libMyMath.la
run out 1 > glibtool: link: ar cr .libs/libMyMath.a .libs/obj1.o .libs/objmul.o .libs/obj3.o
run out 1 > glibtool: link: ranlib .libs/libMyMath.a
run out 1 > glibtool: link: ( cd ".libs" && rm -f "libMyMath.la" && ln -s "../libMyMath.la" "libMyMath.la" )
lcc: resolving target obj3.c.run2.c
run out 2 > glibtool: compile: clang -g -O "" -c obj3.c.run2.c -fno-common -DPIC -o .libs/obj3.c.run2.o
run out 2 > glibtool: compile: clang -g -O "" -c obj3.c.run2.c -o obj3.c.run2.o >/dev/null 2>&1
run out 2 > glibtool: link: rm -fr .libs/libMyMath.a .libs/libMyMath.la
run out 2 > glibtool: link: ar cr .libs/libMyMath.a .libs/obj1.o .libs/objmul.o .libs/obj3.o
run out 2 > glibtool: link: ranlib .libs/libMyMath.a
run out 2 > glibtool: link: ( cd ".libs" && rm -f "libMyMath.la" && ln -s "../libMyMath.la" "libMyMath.la" )
lcc: resolving target obj3.c.run3.c
run out 3 > glibtool: compile: clang -g -O "" -c obj3.c.run3.c -fno-common -DPIC -o .libs/obj3.c.run3.o
run out 3 > glibtool: compile: clang -g -O "" -c obj3.c.run3.c -o obj3.c.run3.o >/dev/null 2>&1
run out 3 > glibtool: link: rm -fr .libs/libMyMath.a .libs/libMyMath.la
run out 3 > glibtool: link: ar cr .libs/libMyMath.a .libs/obj1.o .libs/objmul.o .libs/obj3.o
run out 3 > glibtool: link: ranlib .libs/libMyMath.a
run out 3 > glibtool: link: ( cd ".libs" && rm -f "libMyMath.la" && ln -s "../libMyMath.la" "libMyMath.la" )
lcc: compiling target obj3.c
glibtool: compile: clang -g -O "" -c obj3.c -fno-common -DPIC -o .libs/obj3.o
glibtool: compile: clang -g -O "" -c obj3.c -o obj3.o >/dev/null 2>&1
glibtool: link: rm -fr .libs/libMyMath.a .libs/libMyMath.la
glibtool: link: ar cr .libs/libMyMath.a .libs/obj1.o .libs/objmul.o .libs/obj3.o
glibtool: link: ranlib .libs/libMyMath.a
glibtool: link: ( cd ".libs" && rm -f "libMyMath.la" && ln -s "../libMyMath.la" "libMyMath.la" )
lcc: specifying target main.c
lcc: resolving target main.c.run1.c
run out 1 > glibtool: compile: clang -g -O "" -c main.c.run1.c -fno-common -DPIC -o .libs/main.c.run1.o
run out 1 > glibtool: compile: clang -g -O "" -c main.c.run1.c -o main.c.run1.o >/dev/null 2>&1
run out 1 > glibtool: link: clang -g -O "" -o CompileTest .libs/main.o -L/Users/a1/Projects/GitHub/lcc/test/ /Users/a1/Projects/GitHub/lcc/test/.libs/libMyMath.a
lcc: resolving target main.c.run2.c
run out 2 > glibtool: compile: clang -g -O "" -c main.c.run2.c -fno-common -DPIC -o .libs/main.c.run2.o
run out 2 > glibtool: compile: clang -g -O "" -c main.c.run2.c -o main.c.run2.o >/dev/null 2>&1
run out 2 > glibtool: link: clang -g -O "" -o CompileTest .libs/main.o -L/Users/a1/Projects/GitHub/lcc/test/ /Users/a1/Projects/GitHub/lcc/test/.libs/libMyMath.a
lcc: resolving target main.c.run3.c
run out 3 > glibtool: compile: clang -g -O "" -c main.c.run3.c -fno-common -DPIC -o .libs/main.c.run3.o
run out 3 > glibtool: compile: clang -g -O "" -c main.c.run3.c -o main.c.run3.o >/dev/null 2>&1
run out 3 > glibtool: link: clang -g -O "" -o CompileTest .libs/main.o -L/Users/a1/Projects/GitHub/lcc/test/ /Users/a1/Projects/GitHub/lcc/test/.libs/libMyMath.a
lcc: compiling target main.c
glibtool: compile: clang -g -O "" -c main.c -fno-common -DPIC -o .libs/main.o
glibtool: compile: clang -g -O "" -c main.c -o main.o >/dev/null 2>&1
glibtool: link: clang -g -O "" -o CompileTest .libs/main.o -L/Users/a1/Projects/GitHub/lcc/test/ /Users/a1/Projects/GitHub/lcc/test/.libs/libMyMath.a
- Documentations: starts with semi-colon(s) ";"
;;; about a lisp file
;;;; author, licence and/or documentation about each target
(var long height) ; description of a form
(func sqr ((double a))
(out double)
;; some commented code or documentation inside code
(return (* a a)))
- Preprocessor Forms: a form which starts with at-sign "@" and accepts one argument. code form is used for writing C code inside lcc.
(@define (code "SHA1_ROTL(bits, word) (((word) << (bits)) | ((word) >> (32-(bits)))"))
(struct SHA512Context
(@ifdef USE_32BIT_ONLY)
(member uint32_t Intermediate_Hash[(/ SHA512HashSize 4)]) ; Message Digest
(member uint32_t Length[4]) ; Message length in bits
(@else) ; !USE_32BIT_ONLY
(member uint64_t Intermediate_Hash[(/ SHA512HashSize 8)]) ; Message Digest
(member uint64_t Length_High)
(member uint64_t Length_Low) ; Message length in bits
(@endif) ; USE_32BIT_ONLY
(member int_least16_t Message_Block_Index) ; Message_Block array index
(member uint8_t Message_Block[SHA512_Message_Block_Size]) ; 1024-bit message blocks
(member int Computed) ; Is the hash computed?
(member int Corrupted)) ; Cumulative corruption code
#define SHA1_ROTL(bits, word) (((word) << (bits)) | ((word) >> (32-(bits)))
typedef struct SHA512Context {
#ifdef USE_32BIT_ONLY
uint32_t Intermediate_Hash [SHA512HashSize / 4];
uint32_t Length [4];
#else
uint64_t Intermediate_Hash [SHA512HashSize / 8];
uint64_t Length_High;
uint64_t Length_Low;
#endif
int_least16_t Message_Block_Index;
uint8_t Message_Block [SHA512_Message_Block_Size];
int Computed;
int Corrupted;
} SHA512Context;
- Main Function: The main function is where program execution begins. Every lcc program must contain only one main function.
If form accepts 2 or 3 argument. condition, form for true evaluation of condition and form for false evaluation. third part(else) could be omitted. use block
form if you need more forms in each part.
(let ((int a . 5)
(int b . 6))
(if (> a b)
(printf "a is greater")
(printf "maybe b is greater")))
{
int a = 5;
int b = 6;
if (a > b)
printf("a is greater");
else
printf("maybe b is greater");
}
(let ((int a . 5)
(int b . 6))
(if (> a b)
(block
(printf "a is greater")
(set a (* a b)))
(block
(printf "maybe b is greater")
(set b (* b a)))))
{
int a = 5;
int b = 6;
if (a > b) {
printf("a is greater");
a = a * b;
} else {
printf("maybe b is greater");
b = b * a;
}
}
(let ((int a))
(printf "Please enter a number between 1 and 5: ")
(scanf "%d" (aof a))
(switch a
(case 1 (printf "You chose One") break)
(case 2 (printf "You chose Two") break)
(case 3 (printf "You chose Three") break)
(case 4 (printf "You chose Four") break)
(case 5 (printf "You chose Five") break)
(default (printf "Invalid Choice."))))
{
int a;
printf("Please enter a number between 1 and 5: ");
scanf("%d", &a);
switch (a) {
case 1:
printf("You chose One");
break;
case 2:
printf("You chose Two");
break;
case 3:
printf("You chose Three");
break;
case 4:
printf("You chose Four");
break;
case 5:
printf("You chose Five.");
break;
default:
printf("Invalid Choice");
break;
}
}
(let ((int n . 1)
(int times . 5))
(while (<= n times)
(printf "lcc while loops: %d\n" n)
(1+ n)))
{
int n = 1, times = 5;
while (n <= times) {
printf("C while loops: %d\n", n);
n++;
}
}
(let ((int n . 1)
(int times . 5))
(do
(printf "lcc do loops: %d\n" n)
(1+ n)
(<= n times))) ; last form of do clause checks the condition
{
int n = 1, times = 5;
do {
printf("C do loops: %d\n", n);
n++;
} while (n <= times)
}
(let ((int n)
(int times))
(for ((n . 1)
(times . 5)) ; initialize
(<= n times) ; test
((1+ n)) ; step
(printf "lcc for loop: %d\n" n)))
(for ((int n . 1)
(times . 2)) ; initialize
(<= n times) ; test
((1+ n)) ; step
(printf "another initialization for loop: %d\n" n))
{
int n;
int times;
for ( n = 1, times = 5; (n <= times); (n ++)) {
printf ("lcc for loop: %d\n", n);
}
}
for (int n = 1, times = 2; (n <= times); (n ++)) {
printf ("another initialization for loop: %d\n", n);
}
lcc has some points on functions:
- Use returns form for setting the return type. returns form must be first form of a function after arguments list. A fucntion without returns form will returns void instead of main which returns int.
- Function's attributes must set in declaration time. each attribute enclosed in braces or parentheses.
- {declare}
- {static}
- {inline}
- {extern}
- {resolve #f) means do not resolve this function
(source "main.c"
(:std #t :compile #t :link #t)
;; function declaration
{decl} (func addition ((int * a) (int * b)) (out int))
(func main ()
;; local variable definition
(let ((int answer)
(int num1 . 10)
(int num2 . 5)
(func aFuncPtr ((int * _) (int * _)) (out int) . addition)) ; function pointer
;; calling a function to get addition value
(set answer (addition (aof num1) (aof num2)))
(printf "The addition of two numbers is: %d\n" answer)
(set answer (aFuncPtr (aof num1) (aof num2)))
(printf "The addition of two numbers by function pointer is: %d\n" answer))
(return 0))
;; function returning the addition of two numbers
(func addition ((int * a) (int * b))
(out int)
(return (+ (cof a) (cof b)))))
#include <stdio.h>
#include <stddef.h>
#include <stdint.h>
#include <stdlib.h>
#include <stdbool.h>
int addition (int * a, int * b);
int main () {
{
int answer;
int num1 = 10;
int num2 = 5;
int (*aFuncPtr) (int * , int * ) = addition;
answer = addition ((&num1), (&num2));
printf ("The addition of two numbers is: %d\n", answer);
answer = aFuncPtr ((&num1), (&num2));
printf ("The addition of two numbers by function pointer is: %d\n", answer);
}
return 0;
}
int addition (int * a, int * b) {
return ((*a) + (*b));
}
- Functions can return multiple values by inline structs.
(source "main.c" (:std #t :compile #t :link #t)
(func aMultiReturnFunc ((int x) (int y)) (out '{(int a) (int b)})
(return '{ x y }))
(func aMultiReturnFuncS ((int x) (int y)) (out '{(int a) (int b)})
(let (((typeof (aMultiReturnFuncS x y)) s . '{ x y }))
(return s)))
(func main ()
(let ((int n . 3)
(int t . 4)
((typeof (aMultiReturnFunc 1 1)) mr)
((typeof (aMultiReturnFuncS 1 1)) mrt))
(set mr (aMultiReturnFunc n t))
(printf "a: %d, b: %d\n" ($ mr a) ($ mr b))
(set mrt (aMultiReturnFuncS (++ n) (++ t)))
(printf "a: %d, b: %d\n" ($ mrt a) ($ mrt b)))))
typedef struct __lccStruct_aMultiReturnFunc_177 {
int a;
int b;
} __lccStruct_aMultiReturnFunc_177;
__lccStruct_aMultiReturnFunc_177 aMultiReturnFunc (int x, int y) {
return ((__lccStruct_aMultiReturnFunc_177){x , y});
}
typedef struct __lccStruct_aMultiReturnFuncS_178 {
int a;
int b;
} __lccStruct_aMultiReturnFuncS_178;
__lccStruct_aMultiReturnFuncS_178 aMultiReturnFuncS (int x, int y) {
{
typeof(aMultiReturnFuncS (x , y)) s = {x , y};
return ((__lccStruct_aMultiReturnFuncS_178)s);
}
}
int main () {
{
int n = 3;
int t = 4;
typeof(aMultiReturnFunc (1, 1)) mr;
typeof(aMultiReturnFuncS (1, 1)) mrt;
mr = aMultiReturnFunc (n , t);
printf ("a: %d, b: %d\n", (mr . a), (mr . b));
mrt = aMultiReturnFuncS ((++n ), (++t ));
printf ("a: %d, b: %d\n", (mrt . a), (mrt . b));
}
}
(var double amount [5])
(var int digits [] . '{ 1 2 3 4 5 })
(var char hw [][5] . '{ "Hello" "World" })
int digits[] = {1, 2, 3, 4, 5};
char hw[][5] = { "Hello" "World" };
(var int myArray [5])
;; Initializing elements of array seperately
(for ((int n . 0))
(< n (/ (sizeof myArray) (sizeof int)))
((1+ n))
(set (nth n myArray) n))
int myArray[5];
// Initializing elements of array seperately
for(int n = 0; n < sizeof(myArray) / sizeof(int); n++)
{
myArray[n] = n;
}
(var char name [6] . '{#\C #\l #\o #\u #\d #\Null})
(var char name [] . "Cloud")
(var char * name . "Cloud")
char name[6] = {'C', 'l', 'o', 'u', 'd', '\0'};
char name[] = "Cloud";
char * name = "Cloud";
#\Null
#\Space
#\Newline
#\Tab
#\Page
#\Rubout
#\Linefeed
#\Return
#\Backspace
(var int * width)
(var int * letter)
int *width;
char *letter;
(source "main.c" ()
(include <stdio.h>)
(func main ((int argc) (char * argv []))
(let ((int n . 20)
(int * pntr)) ; actual and pointer variable declaration
(set pntr (aof n)) ; store address of n in pointer variable
(printf "Address of n variable: %x\n" (aof n))
;; address stored in pointer variable
(printf "Address stored in pntr variable: %x\n" pntr)
;; access the value using the pointer
(printf "Value of *pntr variable: %d\n" (cof pntr)))
(return 0))
#include<stdio.h>
int main (int argc, char *argv[])
{
{
int n = 20, *pntr; /* actual and pointer variable declaration */
pntr = &n; /* store address of n in pointer variable */
printf("Address of n variable: %x\n", &n);
/* address stored in pointer variable */
printf("Address stored in pntr variable: %x\n", pntr);
/* access the value using the pointer */
printf("Value of *pntr variable: %d\n", *pntr);
}
return 0;
}
C dynamic memory allocation functions malloc()
, calloc()
, realloc()
, free()
are available. Other keyword alloc
that works in let
initialization part which automatically defers and freeing allocated memory when the variable gone out of let scope. Auto deferral could be replaced by {defer (aFuncPonter | lambda)} which takes a pointer to pointer variable.
(let ((char * mem_alloc . #'(malloc (* 15 (sizeof char))))) ; memory allocated dynamically
(if (== mem_alloc nil) (printf "Couldn't able to allocate requested memory\n"))
(free mem_alloc))
{
char * mem_alloc = malloc(15 * sizeof(char)); /* memory allocated dynamically */
if (mem_alloc == NULL) {
printf("Couldn't able to allocate requested memory\n");
}
free(mem_alloc);
}
Allocation with alloc
and equivalent code in C:
(func main ()
(let ({defer '(lambda ((int * xPtr)) (printf "x was %d\n" (cof xPtr)))}
(int x . 6)
(int * ax . #'(alloc 5 (sizeof int))))
(printf "x is %d\n" x))))
void __lccLambda_main_178 (int * xPtr) {
printf("x was %d\n", (*xPtr));
}
void __lccLambda_main_179 (int ** ax) {
free((*ax));
}
int main () {
{
int x __attribute__((__cleanup__(__lccLambda_main_178))) = 6;
int * ax __attribute__((__cleanup__(__lccLambda_main_179))) = ((int *)calloc(5, sizeof(int)));
printf("x is %d\n", x);
}
}
(let ((int n_rows . 4)
(int n_columns . 5)
(int ** matrix . #'(alloc (* (* n_rows n_columns) (sizeof int)))))
(printf "Matrix allocated\n"))
void __lccLambda_main_178 (int *** matrix) {
free((*matrix));
}
int main () {
{
int n_rows = 4;
int n_columns = 5;
int ** matrix __attribute__((__cleanup__(__lccLambda_main_178))) = ((int **)malloc(((n_rows * n_columns) * sizeof(int))));
printf ("Matrix allocated\n");
}
}
Allocation by alloc
and equivalent calloc
:
(let ((char * safe_alloc . #'(alloc 15 (sizeof char))))
(printf "Memory allocated safely\n"))
void __lccLambda_main_178 (char ** safe_alloc) {
free((*safe_alloc));
}
int main () {
{
char * safe_alloc __attribute__((__cleanup__(__lccLambda_main_178))) = ((char *)calloc(15, sizeof(char)));
printf ("Memory allocated safely\n");
}
}
declare
clause is for declaring one or more variable(s) at the end of nested struct declaration just for anonymous structures.
Use $
form for struct's member access and ->
form for member access of pointer of struct. Both $
and ->
have other utility if a function defined in source targets and doesn't have {resolve #f}
attribute. $
operator resolves attribute access for both instance and pointer variables, also ->
operator won't work for pointer access instead it resolves to method access for structures. inline
functions or methods in header files also don't have resolving process. Methods can be defined with method
clause and naming convention Struct->Method
. All methods have this
parametr automatically and will receive calling instance or pointer.
(header "course.h" ()
(struct Course
(member char WebSite [50])
(member char Subject [50])
(member int Price))
{decl} (method Course->Print ()))
(source "course.c" (:std #t :compile #t :link #t)
(include "course.h")
(var Course c1 . '{"domain.com" "Compilers" 100})
(var Course * pc1 . #'(aof c1))
(method Course->Print ()
(printf "Course: %s in %s for %d$\n"
($ this Subject)
($ this WebSite)
($ this Price)))
(func main ()
(-> c1 Print)
(-> pc1 Print)))
// course.h
typedef struct Course {
char WebSite[50];
char Subject[50];
int Price;
} Course;
void Course_Print (Course * this);
// course.c
Course c1 = {"domain.com", "Compilers", 100};
Course * pc1 = (&c1);
void Course_Print (Course * this) {
printf ("Course: %s in %s for %d$\n", (this ->Subject), (this ->WebSite), (this ->Price));
}
int main () {
Course_Print(&c1);
Course_Print(pc1);
}
declare
form is for declaring one or more variable(s) at the end of nested union declaration just for anonymous unions.
Use $
form for union's member access and ->
form for member access of pointer of union. Union supports resolver like Struct.
(struct USHAContext
(member int whichSha) ; which SHA is being used
(union
(member SHA1Context sha1Context)
(member SHA224Context sha224Context)
(member SHA256Context sha256Context)
(member SHA384Context sha384Context)
(member SHA512Context sha512Context)
(declare ctx)))
typedef struct USHAContext {
int whichSha;
union {
SHA1Context sha1Context;
SHA224Context sha224Context;
SHA256Context sha256Context;
SHA384Context sha384Context;
SHA512Context sha512Context;
} ctx;
} USHAContext;
(enum
(shaSuccess . 0)
(shaNull) ; Null pointer parameter
(shaInputTooLong) ; input data too long
(shaStateError) ; called Input after FinalBits or Result
(shaBadParam)) ; passed a bad parameter
enum {
shaSuccess = 0,
shaNull,
shaInputTooLong,
shaStateError,
shaBadParam
};
(guard __STUDENT_H__
(struct Student
(member char name [50])
(member char family [50])
(member int class_no)))
#ifndef __STUDENT_H__
#define __STUDENT_H__
typedef struct Student {
char name [50];
char family [50];
int class_no;
} Student;
#endif /* __STUDENT_H__ */
(typedef int * intptr_t)
typedef int * intptr_t;
sbcl --script /path/to/lcc.lisp /path/to/some-lcc-file.lisp {args}
Available arguments:
- --debug : will prints too many details about specifying, resolving and compiling.
- --verbose : adds
-v
option toclang
andlibtool
commands to print more details about compiling and linking. usefull when linking many complex libraries.
{$CWD}
placeholder is available inside :compile
and :link
command for every targets.
C++ compiler could be used instead of C compiler then some features availables:
&
modifier in function argument for pass by reference.- Default value for members of structs.
func
form for defining a member function inside of structs. Call these methods by$
member access operator(($ emp Sign) aDoc)
.