Wow, I am amazed the last post on this thread is in 2006. Can anyone recall me? I was on this forum in 2003 last.
Printable View
Wow, I am amazed the last post on this thread is in 2006. Can anyone recall me? I was on this forum in 2003 last.
( a + ib ) * ( c + id ) = X + iYQuote:
Originally Posted by grady
X=a*c-b*d
let i=a*c ; j=b*d;
Y=b*c+a*d=(a+b)*(c+d)-a*c-b*d=(a+b)*(c+d)-i-j
three multiplications:
1 : (a+b)*(c+d)
2 : a*c
3 : b*d
i hacked something up. and i'm not using "hacked" lightly here, i'm sure it's pretty volatile. the only easy way i could think of is by making sure all the words in the possibly decrypted strings have vowels in them. words like "in, a, is, are" fail this pretty easily if they aren't shifted to the correct alphabet character, so it works fairly well. a dictionary lookup would be less volatile though. if it can't decrypt the line using that test, then it simply skips it and moves on to the next. what's the key is found all subsequent lines are decrypted in the same manner:Quote:
Originally Posted by tecknophreak
Code:#!/usr/bin/perl
my $shiftAmount = 0;
my $cryptFound = 0;
while(<>) {
my @chars = split //;
my $string;
if ($cryptFound == 1) {
@chars = shiftChars($shiftAmount, @chars);
print join ('', @chars);
next;
}
for(1..26) {
++$shiftAmount;
@chars = shiftChars(1, @chars);
$string = join('',@chars);
print $string;
if( isEnglish((split(/\s+/,$string))) == 1 ) {
$cryptFound = 1;
print "DECRYPTED: Shift +$shiftAmount\n";
print $string;
last;
}
}
}
sub isEnglish {
my (@words) = @_;
for(@words) {
if (not /[aeiou]/g) {
return 0;
}
}
return 1;
}
sub shiftChars {
my $shiftAmount = shift;
my @chars = @_;
for(1..$shiftAmount) {
for(@chars) {
next if(/\W/);
if($_ eq "z") {
$_ = a;
next;
}
if($_ eq "Z") {
$_ = A;
next;
}
++$_;
}
}
return @chars;
}
Code:fluxion@serenity:~/source$ cat caesar.in
ufyr'q qylry'q jgrrjc fcjncp bmgle rm rfyr bme? jmmiq jgic fc'q rpwgle rm hskn mtcp, zsr fc ayl'r osgrc kyic gr.
wms
ypc y
kmlicwdyac
fluxion@serenity:~/source$ ./caesarCipher.pl caesar.in
vgzs'r rzmsz'r khsskd gdkodq cnhmf sn sgzs cnf? knnjr khjd gd'r sqxhmf sn itlo nudq, ats gd bzm's pthsd lzjd hs.
what's santa's little helper doing to that dog? looks like he's trying to jump over, but he can't quite make it.
DECRYPTED: Shift +2
what's santa's little helper doing to that dog? looks like he's trying to jump over, but he can't quite make it.
you
are a
monkeyface
fluxion@serenity:~/source$
You've all had three years to strengthen your Kung Fu by pondering the reduction of the complex multiply from four multplications to three. Many have either moved up to management or lost their skill by writing reams of code that runs on a flatulent virtual machine. Is there one left among you whose Kung Fu is strong enough for this challenge:
A 2x2 matrix multiply requires 8 multiplications and 4 additions. Can you make a computer do a 2x2 matrix multiply by issuing only 3 arithmetic instructions?
You have awaken the beast...
Can you define what a "2x2 matrix" is, for those of us that may not know. I have a couple of ideas, but I'm not sure if they match what you're thinking.
The 2x2 matrix is the 2-row 2-column square matrix, and 8 multiplications and 4 additions are correct. Its the standard def of matrix. Check out wikipedia for it.
Well, the matrix multiplication algorithm looks like:
Now, I don't see any way to simplify this to three instructions -- at least not in general. Unless your particular CPU does vectored arithmetic (can do independent operations on many operands at the same time), in which case, you're cheating. ;) It still does all 8 multiplications and all 4 additions, it just does some of them at the same time as others.Code:[ a b ] [ c d ] = [ ac+bg ad+bh ]
[ e f ] [ g h ] = [ ec+fg ed+fh ]
It's OK to use vector instructions. The challenge is: can you do the multiply by "issuing only 3 arithmetic instructions" - as in machine instructions. It's very likely that the computer in front of you can do this.Quote:
Originally Posted by bwkaz
Another challenge:
Reverse this apropos quote ".esnesnon dna skcirt elpmis fo tol a lla s'tI"... on the GPU (your video card's processor).
Here's a new challenge. I was posed this problem earlier today, and it took me a lot longer to figure out than I originally thought it would. I have yet to put it into code myslef.
Create a program that takes a word, then outputs all possible scrambles of that word. A sample of its output might look similar to this:Do not use a recursive algorithm.Code:./scramble hack
hack
ahkc
kahc
akch
...
the following works only for 9 digit words or less
Code:#include <iostream>
#include <string>
using namespace std;
/* takes the j-th digit from
left to right */
const unsigned long P10[]=
{1,10,100,
1000,10000,100000,
1000000,10000000,
100000000};
typedef int* vec;
#define dig(i,j,len) (i/P10[len-j])%10
inline bool rep(unsigned long K,int l){
int rep_hash[10]={0,0,0,0,0,0,0,0,0,0};
bool okay=false;
int U;
for(U=1;U<=l;U++){
if( dig(K,U,l) > l || dig(K,U,l) ==0 ){
okay=true;break;
};
if(rep_hash[ dig(K,U,l) ] == 1){
okay=true;break;
};
rep_hash[ dig(K,U,l) ]=1;
};
return okay;
}
/* returns true if repeating digit
and returns false if no repeating
digit is found */
/* also needs to check for
digits that should not be larger
than l and the digit should be
different from 0*/
inline void print_perm(char* a,unsigned long T,int len){
for(int i=1;i<=len;i++)
cout<<a[ dig(T,i,len) - 1 ];
cout<<endl;
}
void all_scrambles(char* scramble){
int len=strlen(scramble);
unsigned long T=123456789;
unsigned long UPPERLIM=999999999;
T/=P10[9-len];
UPPERLIM/=P10[9-len];
while(++T<=UPPERLIM)
if(rep(T,len)==false)
print_perm(scramble,T,len);
}
int main(int argc,char** argv){
if(argc>1)
all_scrambles(argv[1]);
return 0;
}
i forgot to brag about it,actually that it does the job for 9 letters pretty fast
on a 1.6ghz, that is...
spx2@asfasfadgf ~
$ time ./blah.exe abcdefgh > blah.txt
real 0m27.601s
user 0m0.015s
sys 0m0.000s
the code needs trivial modifications to reach the general purpose of the challenge(for an arbitrary length word)
now some explanations on the algorithm used wichis also trivial.
one can understand the "ideea" behind the algorithm if he thinks at
the way a number is increased by on unit,if he can understand the
mecanism of transport for addition and all that concerns it then,then
it will be easy to understand the ++T line.
after that one must think the digits of T are actually the possitions
of the letters inside a new scramble of argv[1].
those positions have to meet some constraints described in the comments
wich are included in the source.
that is all
P.S. the typedef int* vec; i forgot to delete this
its useless,at first i wanted to use it but figgured that id be
better off without.
I thought we did this one a while ago. Anywho, it's fun:
I've got another algorithm out there somewhere which does much better, I'll see if I can find it.Code:#include <iostream>
#include <algorithm>
int main() {
char array[10] = "abcdefghi";
while (std::next_permutation(array, &array[9])) {
std::cout << array << std::endl;
}
}
There it is:
next_perm.h
test.cppCode:#ifndef NEXT_PERM_H
#define NEXT_PERM_H
template<class T>
inline void swap(T *lh, T *rh) {
T tmp(*lh);
*lh = *rh;
*rh = tmp;
}
template<class T>
inline void reverse(T *array, int size) {
int lh(0), rh(size - 1);
while (lh < rh) {
swap(&array[lh], &array[rh]);
++lh;
--rh;
}
}
template<class T>
bool next_perm(T *array, int length) {
int len(length - 1), replace(len), current;
while (--replace != -1) {
for (current = len; current != replace; --current) {
if (array[current] > array[replace]) {
swap(&array[current], &array[replace]);
reverse(&array[replace + 1], len - replace);
return true;
}
}
}
reverse(array, length);
return false;
}
#endif // NEXT_PERM_H
takes ~24 sec to do 9 on a 3.0 ghz machine, however, ABCDEFGH is only 8 letters, ABCDEFGHI is 9 letters, I don't know if that was a miss type.Code:#include <iostream>
#include "next_perm.h"
int main(int argc, char **argv) {
if (argc == 2) {
char *array(argv[1]);
int len(strlen(array));
std::cout << array << std::endl;
while (next_perm(array, len)) {
std::cout << array << std::endl;
}
}
}
somebody know how can i use shell variable inside sed?
i=string
sed 's/somestring/$i/'
??
This shows that you can simply useCode:# echo "abc" | sed s/abc/`echo "123"`/
123
EDIT:: be careful of the ` ' and " ! They are all different and you need to terminate each opening with a closing of the same type! ' and " can be used interchangeably, while ` is the one that really causes things to be interpreted by the shell. It (the executing `) is usually above the tab button.Code:i=string
sed s/somestring/`echo $i`/
Hey, look, another Useless Use Of Backticks! :p Forget the backticks (especially when all they do is call echo), just use the value directly. ;)
In this case, the problem seems to be that when you put sed's argument in single quotes, which is the "standard" way to run sed, the shell doesn't interpret any special characters. (This is why it's the "standard" way to run sed; some of the shell special characters need to be left alone by the shell, so that sed can interpret them.) That includes the $ character, which starts a variable reference; if you use $abc inside single quotes, you'll get $abc out, not the value of the shell variable abc.
So a better fix would be to run sed like this:
i=string
sed -e "s/some string/$i/"
(I left the double quotes in so that you could use whitespace in either the match or the replacement. Either way, you'll have to replace any $ characters in both the match and the replacement if you don't want the shell to replace them with a variable's value, but without the double quotes, you'd have to escape spaces, tabs, and newlines also.)
Can someone do a simple client-server game for Linux using FIFO? It's about being simple when it comes to code...I mean C++ code.
Who can do a primitive Linux shell? This will have to get a command from the command prompt,solve it than after it finishes,comes back and waits for an other command.
When i compile this code with gcc i get the error message "Segmentation fault" . Can anyone correct the code please?
hartson: I'm not sure those are exactly the kind of programming challenges that were intended. Most of them so far have been relatively small with only a few lines of code. Neither of yours are going to fall under that category.;)
Incidentally I've done both in my senior project from school that I'm still occasionally adding to. It's a client-server game with a console that acts like a Linux shell (albeit an extremely simple one). I would post the code, but it's neither complete nor fit for third party consumption, and it's about 4500 lines as it stands right now.;)
trying to learn C++ it totally sucks...... i need to write a blackjack program and well im having to many problems with stupid syntax ........ I like web programming (php , )
cybertron is right. A good challenge is probably about 100-20 lines of code or less.
Here is a common problem:
Write a program that takes an unsigned integer, and outputs it in binary form.
What format is the decimal being taken in as? Decimal, hex?
Guess it doesn't really matter. C++ attached.
c++ suxs in my opinion...... way to fussy and not for me.... ill stick to web programming php///// using mysql / etc ..........
tecknophreak saves the day again. :)
What happens when you need to make something that is impossible with php?Quote:
Originally Posted by crow2icedearth
If C++ isn't for you, then try python, java etc etc... there are just too many languages to swear off programming without trying them.
right now im *attempting* to create some sort of mud client in python, what i have is a main function that gives a desciption and asks for an input, and in it there's an if/elif/else statement that calls other functions based on the input.
At the end of each function, the main descript one will be called.
i just dont like c++ much so far. i might if i get good at it........ who knows
C++ is not easy, but it gives you a lot of power once you get to know it (sometimes too much;)).
boxxertrumps: Are you saying that the main function is going to call individual functions that will then call the main function again? If so, eventually you're going to exhaust your stack space and your app will crash. Just let the functions return normally and add a loop to the main function. Of course if that's not what you're planning to do then never mind.:)
i just need the descript function called after its finished excecuting.
Fair enough. Just wanted to make sure you weren't a new programmer doing something silly.:)
i am a new programmer, i just thought making a text game would be a good place to start.
(a+ib)*(c+id) = (a+ib)*(c+id)*(c-id)/(c-id)
= (a+ib)*(c2-d2)/(c-id)
or (c+id)*(a2-b2)/(a-ib)
So here we have to do only 3 multiplication.
Write a program that can multiply two really big positive integers (i.e. 200 digits or so).
Can't help but wonder, or ask,Quote:
Originally Posted by o0zi
Is that about Viet Nam? the burning THACHED-ROOFED COTTAGES thing?
P.S. this was @ Burninator
just wondering...
Has anyone figured out Levels Medium and Hard. I've been trying to figure it out and haven't gotten anywhere with medium. I did happen to figure out easy though.
Quote:
Originally Posted by Sepero
I need this in Php