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. All advertising materials mentioning features or use of this software
17 * must display the following acknowledgement:
18 * This product includes software developed by the University of
19 * California, Berkeley and its contributors.
20 * 4. Neither the name of the University nor the names of its contributors
21 * may be used to endorse or promote products derived from this software
22 * without specific prior written permission.
24 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
25 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
26 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
27 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
28 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
29 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
30 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
31 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
32 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
33 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
39 static const char copyright[] =
40 "@(#) Copyright (c) 1989, 1993\n\
41 The Regents of the University of California. All rights reserved.\n";
45 static const char sccsid[] = "@(#)strfile.c 8.1 (Berkeley) 5/31/93";
48 #include <sys/cdefs.h>
49 __FBSDID("$FreeBSD$");
51 # include <sys/param.h>
52 # include <sys/endian.h>
63 * This program takes a file composed of strings separated by
64 * lines starting with two consecutive delimiting character (default
65 * character is '%') and creates another file which consists of a table
66 * describing the file (structure from "strfile.h"), a table of seek
67 * pointers to the start of the strings, and the strings, each terminated
68 * by a null byte. Usage:
70 * % strfile [-iorsx] [ -cC ] sourcefile [ datafile ]
72 * C - Allow comments marked by a double delimiter at line's beginning
73 * c - Change delimiting character from '%' to 'C'
74 * s - Silent. Give no summary of data processed at the end of
76 * o - order the strings in alphabetic order
77 * i - if ordering, ignore case
78 * r - randomize the order of the strings
81 * Ken Arnold Sept. 7, 1978 --
83 * Added ordering options.
89 # define STORING_PTRS (Oflag || Rflag)
90 # define CHUNKSIZE 512
92 # define ALLOC(ptr,sz) { \
94 ptr = malloc(CHUNKSIZE * sizeof *ptr); \
95 else if (((sz) + 1) % CHUNKSIZE == 0) \
96 ptr = realloc(ptr, ((sz) + CHUNKSIZE) * sizeof *ptr); \
98 fprintf(stderr, "out of space\n"); \
112 static char *Infile = NULL, /* input file name */
113 Outfile[MAXPATHLEN] = "", /* output file name */
114 Delimch = '%'; /* delimiting character */
116 static int Cflag = FALSE; /* embedded comments */
117 static int Sflag = FALSE; /* silent run flag */
118 static int Oflag = FALSE; /* ordering flag */
119 static int Iflag = FALSE; /* ignore case flag */
120 static int Rflag = FALSE; /* randomize order flag */
121 static int Xflag = FALSE; /* set rotated bit */
122 static uint32_t Num_pts = 0; /* number of pointers/strings */
124 static off_t *Seekpts;
126 static FILE *Sort_1, *Sort_2; /* pointers for sorting */
128 static STRFILE Tbl; /* statistics table */
130 static STR *Firstch; /* first chars of each string */
132 static void add_offset(FILE *, off_t);
133 static int cmp_str(const void *, const void *);
134 static int stable_collate_range_cmp(int, int);
135 static void do_order(void);
136 static void getargs(int, char **);
137 static void randomize(void);
138 static void usage(void);
142 * Drive the sucker. There are two main modes -- either we store
143 * the seek pointers, if the table is to be sorted or randomized,
144 * or we write the pointer directly to the file, if we are to stay
145 * in file order. If the former, we allocate and re-allocate in
146 * CHUNKSIZE blocks; if the latter, we just write each pointer,
147 * and then seek back to the beginning to write in the table.
149 int main(int ac, char *av[])
153 off_t last_off, pos, *p;
159 static char string[257];
161 (void) setlocale(LC_ALL, "");
163 getargs(ac, av); /* evalute arguments */
165 if ((inf = fopen(Infile, "r")) == NULL) {
170 if ((outf = fopen(Outfile, "w")) == NULL) {
175 (void) fseek(outf, (long) sizeof Tbl, 0);
178 * Write the strings onto the file
182 Tbl.str_shortlen = 0xffffffff;
184 Tbl.str_version = VERSION;
186 add_offset(outf, ftello(inf));
189 sp = fgets(string, 256, inf);
190 if (sp == NULL || (sp[0] == dc && sp[1] == '\n')) {
192 length = (size_t)(pos - last_off) -
193 (sp != NULL ? strlen(sp) : 0);
197 add_offset(outf, pos);
198 if ((size_t)Tbl.str_longlen < length)
199 Tbl.str_longlen = length;
200 if ((size_t)Tbl.str_shortlen > length)
201 Tbl.str_shortlen = length;
205 for (nsp = sp; !isalnum((unsigned char)*nsp); nsp++)
207 ALLOC(Firstch, Num_pts);
208 fp = &Firstch[Num_pts - 1];
209 if (Iflag && isupper((unsigned char)*nsp))
210 fp->first = tolower((unsigned char)*nsp);
213 fp->pos = Seekpts[Num_pts - 1];
216 } while (sp != NULL);
219 * write the tables in
223 Tbl.str_numstr = Num_pts - 1;
226 Tbl.str_flags |= STR_COMMENTS;
234 Tbl.str_flags |= STR_ROTATED;
237 printf("\"%s\" created\n", Outfile);
239 puts("There was 1 string");
241 printf("There were %u strings\n", Num_pts - 1);
242 printf("Longest string: %u byte%s\n", Tbl.str_longlen,
243 Tbl.str_longlen == 1 ? "" : "s");
244 printf("Shortest string: %u byte%s\n", Tbl.str_shortlen,
245 Tbl.str_shortlen == 1 ? "" : "s");
249 Tbl.str_version = htobe32(Tbl.str_version);
250 Tbl.str_numstr = htobe32(Tbl.str_numstr);
251 Tbl.str_longlen = htobe32(Tbl.str_longlen);
252 Tbl.str_shortlen = htobe32(Tbl.str_shortlen);
253 Tbl.str_flags = htobe32(Tbl.str_flags);
254 (void) fwrite((char *) &Tbl, sizeof Tbl, 1, outf);
256 for (p = Seekpts, cnt = Num_pts; cnt--; ++p)
258 (void) fwrite(Seekpts, sizeof *Seekpts, (size_t) Num_pts, outf);
265 * This routine evaluates arguments from the command line
267 void getargs(argc, argv)
273 while ((ch = getopt(argc, argv, "Cc:iorsx")) != EOF)
275 case 'C': /* embedded comments */
278 case 'c': /* new delimiting char */
280 if (!isascii(Delimch)) {
281 printf("bad delimiting character: '\\%o\n'",
282 (unsigned char)Delimch);
285 case 'i': /* ignore case in ordering */
288 case 'o': /* order strings */
291 case 'r': /* randomize pointers */
294 case 's': /* silent */
297 case 'x': /* set the rotated bit */
309 (void) strcpy(Outfile, *argv);
312 puts("No input file name");
315 if (*Outfile == '\0') {
316 (void) strcpy(Outfile, Infile);
317 (void) strcat(Outfile, ".dat");
323 (void) fprintf(stderr,
324 "strfile [-Ciorsx] [-c char] source_file [output_file]\n");
330 * Add an offset to the list, or write it out, as appropriate.
332 void add_offset(fp, off)
339 beoff = htobe64(off);
340 fwrite(&beoff, 1, sizeof beoff, fp);
342 ALLOC(Seekpts, Num_pts + 1);
343 Seekpts[Num_pts] = off;
350 * Order the strings alphabetically (possibly ignoring case).
358 Sort_1 = fopen(Infile, "r");
359 Sort_2 = fopen(Infile, "r");
360 qsort(Firstch, (size_t) Tbl.str_numstr, sizeof *Firstch, cmp_str);
366 (void) fclose(Sort_1);
367 (void) fclose(Sort_2);
368 Tbl.str_flags |= STR_ORDERED;
371 static int stable_collate_range_cmp(c1, c2)
374 static char s1[2], s2[2];
379 if ((ret = strcoll(s1, s2)) != 0)
386 * Compare two strings in the file
396 # define SET_N(nf,ch) (nf = (ch == '\n'))
397 # define IS_END(ch,nf) (ch == EOF || (ch == (unsigned char) Delimch && nf))
399 p1 = (const STR *) s1;
400 p2 = (const STR *) s2;
402 c1 = (unsigned char) p1->first;
403 c2 = (unsigned char) p2->first;
404 if ((r = stable_collate_range_cmp(c1, c2)) != 0)
407 (void) fseeko(Sort_1, p1->pos, 0);
408 (void) fseeko(Sort_2, p2->pos, 0);
412 while (!isalnum(c1 = getc(Sort_1)) && c1 != '\0' && c1 != EOF)
414 while (!isalnum(c2 = getc(Sort_2)) && c2 != '\0' && c2 != EOF)
417 while (!IS_END(c1, n1) && !IS_END(c2, n2)) {
424 if ((r = stable_collate_range_cmp(c1, c2)) != 0)
435 return (stable_collate_range_cmp(c1, c2));
440 * Randomize the order of the string table. We must be careful
441 * not to randomize across delimiter boundaries. All
442 * randomization is done within each block.
452 Tbl.str_flags |= STR_RANDOM;
453 cnt = Tbl.str_numstr;
456 * move things around randomly
459 for (sp = Seekpts; cnt > 0; cnt--, sp++) {