7
4
can we print 1 to 100 without using any if conditions and loops in c&c++?
Conditon: main point is you must not use recursion...and doesnt hardcode code in it for e.g
print(1 2 3..etc);
7
4
can we print 1 to 100 without using any if conditions and loops in c&c++?
Conditon: main point is you must not use recursion...and doesnt hardcode code in it for e.g
print(1 2 3..etc);
23
static int
x=1;a(){char b[8];printf("%d\n",x++);b[24]-=5*(1-x/101);}main(){a();return 0;}
The function a which prints the numbers does not call itself! I exploited a buffer overflow and changed the return address to make the program counter go over function a again as long as I need.
I don't know if this is considered to be a recursion, but I thought it would worth trying.
This code works on my 64-bit machines with gcc 4.6, for other platforms the last statement of function a, could be a little different.
Exp1: I allocated a dummy buffer on stack b, and then addressed a passed-by-end location, which is the location of return address. I anticipated the distance between start of buffer and return address location from disassembly of function a.
Exp2: Expression 5*(1-x/101), is 5 for all x<=100 and 0 for x=101. By looking at disassembly of main (in my case), if you decrease the return address by 5, you will set the PC to calling point of a again. In the updated codes, the return value of printf is used for checking loop condition.
Update: After applying ugoren's suggestions and some other changes:
x;a(){int b[2];b[3*(printf("%d\n",++x)&2)]-=5;}main(){a();}
Update2: After Removing function a:
x;main(){int b[2];b[6^printf("%d ",++x)&4]-=7;}
Update3:
x;main(b){(&b)[1|printf("%d ",++x)&2]-=7;}
Update4: Thanks to mbz :)
x;main(b){(&b)[3|printf("%d ",++x)]-=7;}
Perhaps the most interesting answer here! Could you explain a bit about how it's supposed to work? How did you come with b[24] for the location of the PC and how can I determine what I should use instead (32-bit cygwin, gcc 4.7.3)? What is the rationale behind the magical expression 5*(1-x/101)?
Also, is the static part of x's declaration needed? I'm not fully clear on how this works but since we aren't explicitly returning from a() wouldn't even an auto variable x remain in memory? And in this case it's even a global already.
Thanks again for enlivening this dull repetitive thread. – sundar - Reinstate Monica – 2013-08-17T09:50:43.637
1@sundar Thank you for pointing out static. You're right, It was necessary for my previous code and I forgot to remove it. I added some small explanations, please check them out in answer :) – saeedn – 2013-08-19T03:36:31.460
1>
int b[2] would be shorter. 2. return 0; isn't needed. 3. Changing the index on b instead of the subtracted value would be shorter. (of course, every such change could change machine code and break the solution, but I think it should be OK).1Another idea - maybe exploit the fact that printf returns a higher value at 100? – ugoren – 2013-08-19T05:02:54.343
@saeedn Thanks for the explanations, it's very clear now. I felt stupid after reading Exp2 though, I'd been trying to decode it as some overcomplicated int-to-float bit pattern thing like the famous "inverse square root" formula, missed this much simpler explanation. :) – sundar - Reinstate Monica – 2013-08-19T18:29:54.780
@ugoren Thanks for suggestions. I tried them. The printf return value idea was nice :) – saeedn – 2013-08-19T21:08:44.480
6^printf()&4 saves 2 chars. At 100, it changes b[2], which is a harmless overflow on my machine. P.S. Why don't you change the character count in the title? – ugoren – 2013-08-20T07:00:12.167
The net step is to get rid of a. main can do the work alone. – ugoren – 2013-08-20T07:27:45.457
@ugoren I worked on main and updated the answer ;) – saeedn – 2013-08-20T17:40:01.423
Nice. I used -=4, not -=7, but I guess both work. I was going to say it's done, but I saw your \n-> change, and decided not to give up. How about making b a parameter (and using &b[...])? – ugoren – 2013-08-20T18:10:25.343
@ugoren Yeah, that was good point. Unfortunately the parentheses around &b were necessary. I think it can't be shorter than this, can it? – saeedn – 2013-08-20T18:33:12.920
It can be shorter. Try %2d. I couldn't get rid of the parenthesis. I tried the x[y]->y[x] trick, but it stays the same length (but even weirder than now). – ugoren – 2013-08-21T07:08:14.213
@ugoren is there documentation somewhere for why that x[y] -> y[x] trick works? I remember that it works, but I don't remember why, and I can't find what to search for. (I know that it's equivalent to something which is the same as something else, which is the same as y[x], but idr what those 2 middle ones are.) – mbomb007 – 2017-02-14T22:57:55.807
@mbomb007, both are equivalent to x[y] = *(x+y) = *(y+x) = y[x]. – ugoren – 2017-02-15T09:57:38.240
14
C (gcc)
#define c printf("%d ",i++);
#define b c c c c c
#define a b b b b b
main(i){a a a a}
Assuming no command line arguments were passed.
4Using the preprocessor is almost like hard coding it. – Martin Thoma – 2013-01-01T16:52:09.847
agreed. This is hardwired. – Cong Hui – 2013-11-09T08:01:14.007
10
With templates.
#include<cstdio>
#define Z(A,B,C,D)template<A>struct P B{P(){C;printf("%d ",D);}};
Z(int N,,P<N-1>(),N)Z(,<1>,0,1)int main(){P<100>();}
2s/class/struct/;s/public://;s/static //;s/::/()./g saves 11 characters. – ceased to turn counterclockwis – 2012-12-30T19:49:36.417
7
Assuming the ? operator is allowed.
#define f(a)a a a a a
int main(i){f(f(f(printf(i<102?"%d ":0,i++);)))}
Edit: ""->0
If ? is too similar to an if statement, then use this instead (78)
#define f(a)a a a a
#define g(a)f(a)a
int main(i){f(g(g(printf("%d ",i++);)))}
? is a conditional is it not? – Claudiu – 2013-01-10T14:40:58.350
vignesh4303 disallowed "if conditions", not all conditionals, which is why I included both answers. – cardboard_box – 2013-01-11T02:24:58.410
3
1..100
Equally doable, though no shorter, with 1..1e2. – Iszi – 2013-11-16T16:21:38.300
3
this is the best I can think of, assuming using the preprocessor is fine.
#include <stdio.h>
#define a(i)i,i+1,i+2,i+3
#define b(i)a(i),a(i+4),a(i+8),a(i+12)
#define c(i)b(i),b(i+16)
#define e c(1),c(33)
#define f %d %d %d %d
#define g f f f f f f f f
#define r(m) #m
#define s(m) r(m)
int main(){printf(s(g g g f),e,c(65),a(97));return 0;}
3
#include <cstdio>
template<int i>void p(){printf("%d ",i);p<i+1>();}
template<>void p<101>(){}
int main(){p<1>();}
Isn't this recursion? – Bryan Chen – 2013-09-18T05:47:28.830
3
1:100
7Doesn't the question want a code in C? – saeedn – 2013-08-15T21:40:17.360
2
>>> range(1,101)
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100]
2
print(list(range(1,101)))
it is a inbuilt function that uses recursion! – Pranit Bauva – 2013-01-05T14:21:16.677
5I guess it depends on the interpreter if range uses recursion. I don't use recursion in my solution. I don't care if the inbuilt functions use recursion. If you want to take a look at every possible interpreter / compiler, you can't answer this question with anything else than assembly. – Martin Thoma – 2013-01-05T14:46:42.687
9The garbage collector uses recursion! So does the kernel of your OS. So, computers are not allowed. – boothby – 2013-08-15T17:20:52.637
1
p *(1..100)
(Thanks to histocrat)
Previous 14-character solution:
p *1.upto(100)
This is a non-competitive answer (not C/C++ as requested)
2p *(1..100) saves 3 characters. – histocrat – 2013-01-05T00:05:37.923
@histocrat Thanks, I updated my answer. – knut – 2013-01-05T00:08:07.117
Can't you remove the brackets too? – dingledooper – 2019-10-01T04:17:33.867
0
p=lambda d:print d+1,;map(p,range(100))
p=lambda d:print(d+1,end=' ');set(map(p,range(100)))
or print numbers on separate lines (28 char)
set(map(print,range(1,101)))
or (tested in Python 3.3.2, set items printing in ascending order - 24 chars)
print(set(range(1,101)))
0
echo {1..100}
Some more text so it posts.
0
alert([...Array(101).keys()].slice(1))
Creates an array of length 101, uses the spread operator to put the array keys (0 to 100) into an array, removes the first element (0) then alerts the numbers within the array.
0
Scala (22)
1 to 100 foreach print
@downvoter care to comment ? – lhk – 2015-12-22T10:51:20.233
0
print ^101
If one wants to be able to read the numbers and therefore spaces between the numbers would be nice, the following will do the trick.
say ~ ^101
0
Here's a non-trivial Perl approach (not like say for 1 .. 100). The program uses perl's regular expression engine to count its own characters (two times), and that's the reason why I couldn't golfify the content of $k. It tries to find a (minimal, non-greedy) group of arbitrary single characters until (*FAIL).
$k='$k =~ /^(.+?)(?{print length($1) . "\n"})(*FAIL)/#'x2;eval$k;
The output is a list of all integers from 0 to 100 with newlines.
Clever solution on many levels, nice. I don't understand the *FAIL though, is it some special string to the regex engine? If not, won't the * have its usual metacharacter meaning here and mess things up? – sundar - Reinstate Monica – 2013-08-20T18:49:37.783
1
@sundar see the "Special Backtracking Control Verbs" section in perlre :)
– memowe – 2013-08-22T00:39:03.2070
alert|[1 100].fill
-1
IntStream.range(1,101).forEach(System.out::println);
Welcome to the site. I am not familiar with java but forEach looks an awful lot like a loop which seems to be forbidden by the challenge. – Post Rock Garf Hunter – 2017-02-14T21:42:05.703
It's a stream function. It performs an operation on each element of the stream. It's not a loop. – aglassman – 2017-02-15T21:22:01.317
I would disagree that a stream function is distinct from a loop but I don't think this argument is worthwhile. This question has been closed as unclear for the very reason that "loop" is not very well defined. – Post Rock Garf Hunter – 2017-02-15T21:25:44.577
This challenge restricts to C/C++, so Java is probably unacceptable here. – Erik the Outgolfer – 2017-05-15T16:59:57.030
-1
Looping? No this is list comprehension.
d=[0,1,2,3,4,5,6,7,8,9];print[x*10+y+1 for x in d for y in d]
That's stretching the "no hardcoding" pretty hard. – ugoren – 2013-08-16T07:02:30.827
print "1 to 100"orprint 1 2 3 4 5 6 (etc)? – beary605 – 2012-12-27T07:19:17.440@beary605 print 1 2 3..100 – BlueBerry - Vignesh4303 – 2012-12-27T07:21:12.047
4Does
gotocount? – Ry- – 2012-12-31T00:50:11.423Do preprocessor conditions count? – ApproachingDarknessFish – 2014-01-16T21:50:03.377