2 * Copyright (c) 1989, 1993
3 * The Regents of the University of California. All rights reserved.
5 * This code is derived from software contributed to Berkeley by
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
11 * 1. Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in the
15 * documentation and/or other materials provided with the distribution.
16 * 3. Neither the name of the University nor the names of its contributors
17 * may be used to endorse or promote products derived from this software
18 * without specific prior written permission.
20 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
21 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
24 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
26 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
27 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
29 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32 * @(#) Copyright (c) 1989, 1993 The Regents of the University of California. All rights reserved.
33 * @(#)strfile.c 8.1 (Berkeley) 5/31/93
34 * $FreeBSD: src/games/fortune/strfile/strfile.c,v 1.15.2.2 2001/03/05 11:52:37 kris Exp $
37 #include <sys/param.h>
38 #include <netinet/in.h>
51 * This program takes a file composed of strings separated by
52 * lines starting with two consecutive delimiting character (default
53 * character is '%') and creates another file which consists of a table
54 * describing the file (structure from "strfile.h"), a table of seek
55 * pointers to the start of the strings, and the strings, each terminated
56 * by a null byte. Usage:
58 * % strfile [-iorsx] [ -cC ] sourcefile [ datafile ]
60 * C - Allow comments marked by a double delimiter at line's beginning
61 * c - Change delimiting character from '%' to 'C'
62 * s - Silent. Give no summary of data processed at the end of
64 * o - order the strings in alphabetic order
65 * i - if ordering, ignore case
66 * r - randomize the order of the strings
69 * Ken Arnold Sept. 7, 1978 --
71 * Added ordering options.
74 #define STORING_PTRS (Oflag || Rflag)
77 #define ALLOC(ptr, sz) do { \
79 ptr = malloc((unsigned int)(CHUNKSIZE * sizeof(*ptr))); \
80 else if (((sz) + 1) % CHUNKSIZE == 0) \
81 ptr = realloc(ptr, ((unsigned int)((sz) + CHUNKSIZE) * sizeof(*ptr))); \
83 fprintf(stderr, "out of space\n"); \
93 static char *Infile = NULL; /* input file name */
94 static char Outfile[MAXPATHLEN] = ""; /* output file name */
95 static char Delimch = '%'; /* delimiting character */
97 static int Cflag = false; /* embedded comments */
98 static int Sflag = false; /* silent run flag */
99 static int Oflag = false; /* ordering flag */
100 static int Iflag = false; /* ignore case flag */
101 static int Rflag = false; /* randomize order flag */
102 static int Xflag = false; /* set rotated bit */
103 static long Num_pts = 0; /* number of pointers/strings */
105 static long *Seekpts;
107 static FILE *Sort_1, *Sort_2; /* pointers for sorting */
109 static STRFILE Tbl; /* statistics table */
111 static STR *Firstch; /* first chars of each string */
113 static void add_offset(FILE *, long);
114 static int cmp_str(const void *, const void *);
115 static int collate_range_cmp(int, int);
116 static void do_order(void);
117 static void getargs(int, char **);
118 static void randomize(void);
119 static void usage(void);
123 * Drive the sucker. There are two main modes -- either we store
124 * the seek pointers, if the table is to be sorted or randomized,
125 * or we write the pointer directly to the file, if we are to stay
126 * in file order. If the former, we allocate and re-allocate in
127 * CHUNKSIZE blocks; if the latter, we just write each pointer,
128 * and then seek back to the beginning to write in the table.
131 main(int argc, char *argv[])
135 long last_off, length, pos, *p;
138 static char string[257];
140 setlocale(LC_ALL, "");
142 getargs(argc, argv); /* evalute arguments */
144 if ((inf = fopen(Infile, "r")) == NULL) {
149 if ((outf = fopen(Outfile, "w")) == NULL) {
154 fseek(outf, sizeof(Tbl), SEEK_SET);
157 * Write the strings onto the file
161 Tbl.str_shortlen = ~((unsigned long)0);
163 Tbl.str_version = VERSION;
165 add_offset(outf, ftell(inf));
168 sp = fgets(string, 256, inf);
169 if (sp == NULL || (sp[0] == dc && sp[1] == '\n')) {
171 length = pos - last_off - (sp ? strlen(sp) : 0);
175 add_offset(outf, pos);
176 if (Tbl.str_longlen < (unsigned long)length)
177 Tbl.str_longlen = length;
178 if (Tbl.str_shortlen > (unsigned long)length)
179 Tbl.str_shortlen = length;
183 for (nsp = sp; !isalnum((unsigned char)*nsp); nsp++)
185 ALLOC(Firstch, Num_pts);
186 fp = &Firstch[Num_pts - 1];
187 if (Iflag && isupper((unsigned char)*nsp))
188 fp->first = tolower((unsigned char)*nsp);
191 fp->pos = Seekpts[Num_pts - 1];
194 } while (sp != NULL);
197 * write the tables in
201 Tbl.str_numstr = Num_pts - 1;
204 Tbl.str_flags |= STR_COMMENTS;
212 Tbl.str_flags |= STR_ROTATED;
215 printf("\"%s\" created\n", Outfile);
217 puts("There was 1 string");
219 printf("There were %ld strings\n", Num_pts - 1);
220 printf("Longest string: %lu byte%s\n", Tbl.str_longlen,
221 Tbl.str_longlen == 1 ? "" : "s");
222 printf("Shortest string: %lu byte%s\n", Tbl.str_shortlen,
223 Tbl.str_shortlen == 1 ? "" : "s");
227 Tbl.str_version = htonl(Tbl.str_version);
228 Tbl.str_numstr = htonl(Tbl.str_numstr);
229 Tbl.str_longlen = htonl(Tbl.str_longlen);
230 Tbl.str_shortlen = htonl(Tbl.str_shortlen);
231 Tbl.str_flags = htonl(Tbl.str_flags);
232 fwrite((char *)&Tbl, sizeof(Tbl), 1, outf);
234 for (p = Seekpts, cnt = Num_pts; cnt--; ++p)
236 fwrite((char *)Seekpts, sizeof(*Seekpts), (int)Num_pts, outf);
243 * This routine evaluates arguments from the command line
246 getargs(int argc, char **argv)
250 while ((ch = getopt(argc, argv, "Cc:iorsx")) != -1)
252 case 'C': /* embedded comments */
255 case 'c': /* new delimiting char */
257 if (!isascii(Delimch)) {
258 printf("bad delimiting character: '\\%o\n'",
259 (unsigned char)Delimch);
262 case 'i': /* ignore case in ordering */
265 case 'o': /* order strings */
268 case 'r': /* randomize pointers */
271 case 's': /* silent */
274 case 'x': /* set the rotated bit */
286 strcpy(Outfile, *argv);
289 puts("No input file name");
292 if (*Outfile == '\0') {
293 strcpy(Outfile, Infile);
294 strcat(Outfile, ".dat");
302 "strfile [-Ciorsx] [-c char] source_file [output_file]\n");
308 * Add an offset to the list, or write it out, as appropriate.
311 add_offset(FILE *fp, long off)
317 fwrite(&net, 1, sizeof(net), fp);
319 ALLOC(Seekpts, Num_pts + 1);
320 Seekpts[Num_pts] = off;
327 * Order the strings alphabetically (possibly ignoring case).
336 Sort_1 = fopen(Infile, "r");
337 Sort_2 = fopen(Infile, "r");
338 qsort((char *)Firstch, (int)Tbl.str_numstr, sizeof(*Firstch), cmp_str);
346 Tbl.str_flags |= STR_ORDERED;
350 collate_range_cmp (int c1, int c2)
352 static char s1[2], s2[2];
357 if ((ret = strcoll(s1, s2)) != 0)
364 * Compare two strings in the file
367 cmp_str(const void *s1, const void *s2)
370 int c1, c2, n1, n2, r;
372 #define SET_N(nf,ch) (nf = (ch == '\n'))
373 #define IS_END(ch,nf) (ch == EOF || (ch == (unsigned char)Delimch && nf))
375 p1 = (const STR *)s1;
376 p2 = (const STR *)s2;
378 c1 = (unsigned char)p1->first;
379 c2 = (unsigned char)p2->first;
380 if ((r = collate_range_cmp(c1, c2)) != 0)
383 fseek(Sort_1, p1->pos, SEEK_SET);
384 fseek(Sort_2, p2->pos, SEEK_SET);
388 while (!isalnum(c1 = getc(Sort_1)) && c1 != '\0' && c1 != EOF)
390 while (!isalnum(c2 = getc(Sort_2)) && c2 != '\0' && c2 != EOF)
393 while (!IS_END(c1, n1) && !IS_END(c2, n2)) {
400 if ((r = collate_range_cmp(c1, c2)) != 0)
412 return (collate_range_cmp(c1, c2));
417 * Randomize the order of the string table. We must be careful
418 * not to randomize across delimiter boundaries. All
419 * randomization is done within each block.
428 Tbl.str_flags |= STR_RANDOM;
429 cnt = Tbl.str_numstr;
432 * move things around randomly
435 for (sp = Seekpts; cnt > 0; cnt--, sp++) {
436 i = arc4random_uniform(cnt);