Macros

[Steven Picard]

Paul, I am just curious if you will eventually support Macros.

[Ionic Wind Support Team]

Maybe,  If I get the urge

[Steven Picard]

Good enough answer for me.     I know some programmers love them and some think they're evil but I personally believe they have their proper place.

[Zen]

Is a macro basicly like a subroutine but its run during the compilation and the result is placed into the code? I bit like a constant, where the compiler would then change all references to it with the value?

So if i had a macro ADD(x,y) x + y or however they are formatted

then in my code i could put ADD(1,2) and it would place the value 3 there. Rather than run the function at runtime? If this is how it works then i can see how they are usefull.

Lewis

[Steven Picard]

You're basically correct.

Macros eliminate functional call overhead but are not type safe, however, they have their definate strengths.ÂÃ,  Here's a basic C example:

Code Select Expand

#define MAX(A, B) ((A) > (B) ? (A) : (B))

int i = 8, j = 5;
cout << MAX(i++, j) << endl;ÂÃ,  ÂÃ,  // displays 9 (incorrect)

Here's another one:

Code Select Expand

#define NELEMS(A) (sizeof(A) / sizeof A[0])

int buf[] = { 1, 3, 5, 7, 9 };
cout << NELEMS(buf) << endl;ÂÃ,  ÂÃ,  ÂÃ,  ÂÃ,  ÂÃ, // displays 5

However, you can get pretty creative with macro's and it can be the root of all obfuscation.ÂÃ, 

[Ionic Wind Support Team]

Macros still generate code, just not the function.  It is not evaluated at compile time like an interpreter.  The preprocessor inserts the statement into your code just before it gets compiled into assembly. 

Code Select Expand


#define MAX(A, B) ((A) > (B) ? (A) : (B))

int i = 8, j = 5;
cout << MAX(i++, j) << endl;    // displays 9 (incorrect)


becomes

Code Select Expand


int i = 8, j = 5;
cout << ((i++) > (j) ? (i++) : (j)) << endl;    // displays 9 (incorrect)


Which is then compiled.  And the above is one of the inherent problems with macros .   'i' gets incremented twice instead of once since all the preprocessor does is a direct substitution without regard to context.  A MAX subroutine wouldn't exibit this behaviour.

[Steven Picard]

That's one of those gotchas.ÂÃ,  Macros can offer potential speed increase over function calling but can also increase the size of code tremendously if not used wisely.ÂÃ,  The pitfalls of macros:ÂÃ,  not type safe, bugs easily pop in if someone is not completely familar how they work, and when used excessively can make the compiled code much larger.

But most of all, what would obfuscation be without macros?

Code Select Expand


#include <stdio.h>
#define k *(int*)
#define a if(
#define c ad()
#define i else
#define p while(
#define x *(char*)
#define b ==
#define V =calloc(1,99999)
#define f ()
#define J return
#define l ae(
#define n e)
#define u d!=
#define F int
#define y (j)
#define r m=
#define t +4
F d,z,C,h,P,K,ac,q,G,v,Q,R,D,L,W,M;
E(n{
x D++=e;
}
o f{
a L){
h=x L++;
a h b 2){
L=0;
h=W;
}
}
i h=fgetc(Q);
}
X f{
J isalnum(h)|h b 95;
}
Y f{
a h b 92){
o f;
a h b 110)h=10;
}
}
c{
F e,j,m;
p isspace(h)|h b 35){
a h b 35){
o f;
c;
a d b 536){
c;
E(32);
k d=1;
k(d t)=D;
}
p h!=10){
E(h);
o f;
}
E(h);
E(2);
}
o f;
}
C=0;
d=h;
a X f){
E(32);
M=D;
p X f){
E(h);
o f;
}
a isdigit(d)){
z=strtol(M,0,0);
d=2;
}
i{
x D=32;
d=strstr(R,M-1)-R;
x D=0;
d=d*8+256;
a d>536){
d=P+d;
a k d b 1){
L=k(d t);
W=h;
o f;
c;
}
}
}
}
i{
o f;
a d b 39){
d=2;
Y f;
z=h;
o f;
o f;
}
i a d b 47&h b 42){
o f;
p h){
p h!=42)o f;
o f;
a h b 47)h=0;
}
o f;
c;
}
i{
e="++#m--%am*@R<^1c/@%[_[H3c%@%[_[H3c+@.B#d-@%:_^BKd<<Z/03e>>`/03e<=0f>=/f<@.f>@1f==&g!='g&&k||#l&@.BCh^@.BSi|@.B+j~@/%Yd!@&d*@b";
p j=x e++){
r x e++;
z=0;
p(C=x e++-98)<0)z=z*64+C+64;
a j b d&(m b h|m b 64)){
a m b h){
o f;
d=1;
}
break;
}
}
}
}
}
l g){
p g&&g!=-1){
x q++=g;
g=g>>8;
}
}
A(n{
F g;
p n{
g=k e;
k e=q-e-4;
e=g;
}
}
s(g,n{
l g);
k q=e;
e=q;
q=q t;
J e;
}
H(n{
s(184,n;
}
B(n{
J s(233,n;
}
S(j,n{
l 1032325);
J s(132+j,n;
}
Z(n{
l 49465);
H(0);
l 15);
l e+144);
l 192);
}
N(j,n{
l j+131);
s((e<512)<<7|5,n;
}
T y{
F g,e,m,aa;
g=1;
a d b 34){
H(v);
p h!=34){
Y f;
x v++=h;
o f;
}
x v=0;
v=v t&-4;
o f;
c;
}
i{
aa=C;
r z;
e=d;
c;
a e b 2){
H(m);
}
i a aa b 2){
T(0);
s(185,0);
a e b 33)Z(m);
i l m);
}
i a e b 40){
w f;
c;
}
i a e b 42){
c;
e=d;
c;
c;
a d b 42){
c;
c;
c;
c;
e=0;
}
c;
T(0);
a d b 61){
c;
l 80);
w f;
l 89);
l 392+(e b 256));
}
i a n{
a e b 256)l 139);
i l 48655);
q++;
}
}
i a e b 38){
N(10,k d);
c;
}
i{
g=k e;
a!g)g=dlsym(0,M);
a d b 61&j){
c;
w f;
N(6,g);
}
i a u 40){
N(8,g);
a C b 11){
N(0,g);
l z);
c;
}
}
}
}
a d b 40){
a g b 1)l 80);
r s(60545,0);
c;
j=0;
p u 41){
w f;
s(2393225,j);
a d b 44)c;
j=j t;
}
k r j;
c;
a!g){
e=e t;
k e=s(232,k n;
}
i a g b 1){
s(2397439,j);
j=j t;
}
i{
s(232,g-q-5);
}
a j)s(50305,j);
}
}
O y{
F e,g,m;
a j--b 1)T(1);
i{
O y;
r 0;
p j b C){
g=d;
e=z;
c;
a j>8){
r S(e,m);
O y;
}
i{
l 80);
O y;
l 89);
a j b 4|j b 5){
Z(n;
}
i{
l n;
a g b 37)l 146);
}
}
}
a m&&j>8){
r S(e,m);
H(e^1);
B(5);
A(m);
H(n;
}
}
}
w f{
O(11);
}
U f{
w f;
J S(0,0);
}
I y{
F m,g,e;
a d b 288){
c;
c;
r U f;
c;
I y;
a d b 312){
c;
g=B(0);
A(m);
I y;
A(g);
}
i{
A(m);
}
}
i a d b 352|d b 504){
e=d;
c;
c;
a e b 352){
g=q;
r U f;
}
i{
a u 59)w f;
c;
g=q;
r 0;
a u 59)r U f;
c;
a u 41){
e=B(0);
w f;
B(g-q-5);
A(n;
g=e t;
}
}
c;
I(&m);
B(g-q-5);
A(m);
}
i a d b 123){
c;
ab(1);
p u 125)I y;
c;
}
i{
a d b 448){
c;
a u 59)w f;
K=B(K);
}
i a d b 400){
c;
k j=B(k j);
}
i a u 59)w f;
c;
}
}
ab y{
F m;
p d b 256|u-1&!j){
a d b 256){
c;
p u 59){
a j){
G=G t;
k d=-G;
}
i{
k d=v;
v=v t;
}
c;
a d b 44)c;
}
c;
}
i{
A(k(d t));
k d=q;
c;
c;
r 8;
p u 41){
k d=m;
r m t;
c;
a d b 44)c;
}
c;
K=G=0;
l 15042901);
r s(60545,0);
I(0);
A(K);
l 50121);
k r G;
}
}
}
main(g,n{
Q=stdin;
a g-->1){
e=e t;
Q=fopen(k e,"r");
}
D=strcpy(R V," int if else while break return for define main ")+48;
v V;
q=ac V;
P V;
o f;
c;
ab(0);
J(*(int(*)f)k(P+592))(g,n;
}


http://www.ioccc.org/years.html#2003

[Ionic Wind Support Team]

The speed increase argument really isn't valid anymore.  With advanced branch prediction of CPU's made in the last 6 years the CPU can branch to a subroutine in less time than the expanded macro in many cases.  Macro expansion may result in using more registers/stack to store temporary results then would have been required during the subroutine call, which can actually cause a speed decrease since there is a finite amount of registers.