/* find mattress words: words consisting of two halves in which the first
* part can be transformed by a one-to-one mapping of letters to become
* the second half. e.g. m -> r, a -> e, t -> s.
*
* filters for "interesting" words, where there is at least one
* repeated letter in each half of the word.
*
* Some of the more interesting ones: Zeffirelli, Weisenheimer, needlessly,
* Shakespearizes, donnybrook, Appomattox, anastomose. It finds 720 words
* in a 173528-word word list I have. */
#include <stdio.h>
#include <string.h>
#include <ctype.h>
#define wordsiz 256
char word[wordsiz], modword[wordsiz];
/* Canonicalizes in such a way that cryptograms get the same string.
* Returns whether the canonicalization was "interesting" ---
* i.e. repeated characters were found. */
int canonicalize(char *input, char *output, int len) {
int ii, interesting = 0, distinct = 0;
char corresp[256] = { 0 };
for (ii = 0; ii != len; ii++) {
int c = tolower(input[ii]);
if (!corresp[c]) corresp[c] = ++distinct;
else interesting = 1;
output[ii] = corresp[c];
}
return interesting;
}
int main() {
int len;
while (fgets(word, wordsiz, stdin)) {
if (!strchr(word, '\n')) {
fprintf(stderr, "Word too long: %s...\n", word);
return 1;
}
len = strlen(word) - 1; /* -1 to remove \n */
if (len % 2) continue; /* can't split word in half */
len /= 2;
/* remove this conditional to print uninteresting words like 'adherent'
* too. */
if (!canonicalize(word, modword, len)) continue;
canonicalize(word + len, modword + len, len);
if (!memcmp(modword, modword + len, len)) fputs(word, stdout);
}
return 0;
}
--
<[EMAIL PROTECTED]> Kragen Sitaker <http://www.pobox.com/~kragen/>
The Internet stock bubble didn't burst on 1999-11-08. Hurrah!
<URL:http://www.pobox.com/~kragen/bubble.html>
The power didn't go out on 2000-01-01 either. :)
