1/*	$NetBSD: strfile.c,v 1.44 2022/08/07 11:06:18 andvar Exp $	*/
2
3/*-
4 * Copyright (c) 1989, 1993
5 *	The Regents of the University of California.  All rights reserved.
6 *
7 * This code is derived from software contributed to Berkeley by
8 * Ken Arnold.
9 *
10 * Redistribution and use in source and binary forms, with or without
11 * modification, are permitted provided that the following conditions
12 * are met:
13 * 1. Redistributions of source code must retain the above copyright
14 *    notice, this list of conditions and the following disclaimer.
15 * 2. Redistributions in binary form must reproduce the above copyright
16 *    notice, this list of conditions and the following disclaimer in the
17 *    documentation and/or other materials provided with the distribution.
18 * 3. Neither the name of the University nor the names of its contributors
19 *    may be used to endorse or promote products derived from this software
20 *    without specific prior written permission.
21 *
22 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
23 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
24 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
25 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
26 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
27 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
28 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
29 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
30 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
31 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32 * SUCH DAMAGE.
33 */
34
35#if HAVE_NBTOOL_CONFIG_H
36#include "nbtool_config.h"
37#endif
38
39#ifdef __NetBSD__
40#include <sys/cdefs.h>
41#ifndef lint
42__COPYRIGHT("@(#) Copyright (c) 1989, 1993\
43 The Regents of the University of California.  All rights reserved.");
44#endif /* not lint */
45
46#ifndef lint
47#if 0
48static char sccsid[] = "@(#)strfile.c	8.1 (Berkeley) 5/31/93";
49#else
50__RCSID("$NetBSD: strfile.c,v 1.44 2022/08/07 11:06:18 andvar Exp $");
51#endif
52#endif /* not lint */
53#endif /* __NetBSD__ */
54
55#include <sys/types.h>
56#include <sys/param.h>
57#include <ctype.h>
58#include <stdarg.h>
59#include <stdio.h>
60#include <stdlib.h>
61#include <string.h>
62#include <time.h>
63#include <unistd.h>
64#include <inttypes.h>
65#include <err.h>
66
67#include "strfile.h"
68
69#ifndef MAXPATHLEN
70#define	MAXPATHLEN	1024
71#endif	/* MAXPATHLEN */
72
73/*
74 *	This program takes a file composed of strings separated by
75 * lines starting with two consecutive delimiting character (default
76 * character is '%') and creates another file which consists of a table
77 * describing the file (structure from "strfile.h"), a table of seek
78 * pointers to the start of the strings, and the strings, each terminated
79 * by a null byte.  Usage:
80 *
81 *	% strfile [-iorsx] [ -cC ] sourcefile [ datafile ]
82 *
83 *	c - Change delimiting character from '%' to 'C'
84 *	s - Silent.  Give no summary of data processed at the end of
85 *	    the run.
86 *	o - order the strings in alphabetic order
87 *	i - if ordering, ignore case
88 *	r - randomize the order of the strings
89 *	x - set rotated bit
90 *
91 *		Ken Arnold	Sept. 7, 1978 --
92 *
93 *	Added ordering options.
94 */
95
96# define	STORING_PTRS	(Oflag || Rflag)
97# define	CHUNKSIZE	512
98
99# define	ALLOC(ptr,sz)	do { \
100			if (ptr == NULL) \
101				ptr = malloc(CHUNKSIZE * sizeof *ptr); \
102			else if (((sz) + 1) % CHUNKSIZE == 0) \
103				ptr = realloc(ptr, ((sz) + CHUNKSIZE) * sizeof *ptr); \
104			if (ptr == NULL) \
105				err(1, "out of space"); \
106		} while (0)
107
108typedef struct {
109	char	first;
110	off_t	pos;
111} STR;
112
113static char *Infile = NULL;		/* input file name */
114static char Outfile[MAXPATHLEN] = "";	/* output file name */
115static char Delimch = '%';		/* delimiting character */
116
117static int Sflag	= 0;		/* silent run flag */
118static int Oflag	= 0;		/* ordering flag */
119static int Iflag	= 0;		/* ignore case flag */
120static int Rflag	= 0;		/* randomize order flag */
121static int Xflag	= 0;		/* set rotated bit */
122static long Num_pts	= 0;		/* number of pointers/strings */
123
124static off_t *Seekpts;
125
126static FILE *Sort_1, *Sort_2;		/* pointers for sorting */
127
128static STRFILE Tbl;			/* statistics table */
129
130static STR *Firstch;			/* first chars of each string */
131
132
133static uint32_t h2nl(uint32_t h);
134static void getargs(int argc, char **argv);
135static void usage(const char *, ...) __dead __printflike(1, 2);
136static void add_offset(FILE *fp, off_t off);
137static void do_order(void);
138static int cmp_str(const void *vp1, const void *vp2);
139static void randomize(void);
140static void fwrite_be_offt(off_t off, FILE *f);
141
142
143/*
144 * main:
145 *	Drive the sucker.  There are two main modes -- either we store
146 *	the seek pointers, if the table is to be sorted or randomized,
147 *	or we write the pointer directly to the file, if we are to stay
148 *	in file order.  If the former, we allocate and re-allocate in
149 *	CHUNKSIZE blocks; if the latter, we just write each pointer,
150 *	and then seek back to the beginning to write in the table.
151 */
152int
153main(int ac, char **av)
154{
155	char		*sp, dc;
156	FILE		*inf, *outf;
157	off_t		last_off, length, pos;
158	int		first;
159	char		*nsp;
160	STR		*fp;
161	static char	string[257];
162	long		i;
163
164	/* sanity test */
165	if (sizeof(uint32_t) != 4)
166		errx(1, "sizeof(uint32_t) != 4");
167
168	getargs(ac, av);		/* evalute arguments */
169	dc = Delimch;
170	if ((inf = fopen(Infile, "r")) == NULL)
171		err(1, "open `%s'", Infile);
172
173	if ((outf = fopen(Outfile, "w")) == NULL)
174		err(1, "open `%s'", Outfile);
175	if (!STORING_PTRS)
176		(void) fseek(outf, sizeof Tbl, SEEK_SET);
177
178	/*
179	 * Write the strings onto the file
180	 */
181
182	Tbl.str_longlen = 0;
183	Tbl.str_shortlen = (unsigned int) 0x7fffffff;
184	Tbl.str_delim = dc;
185	Tbl.str_version = VERSION;
186	first = Oflag;
187	add_offset(outf, ftell(inf));
188	last_off = 0;
189	do {
190		sp = fgets(string, 256, inf);
191		if (sp == NULL || (sp[0] == dc && sp[1] == '\n')) {
192			pos = ftell(inf);
193			length = pos - last_off - (sp ? strlen(sp) : 0);
194			last_off = pos;
195			if (!length)
196				continue;
197			add_offset(outf, pos);
198			if ((off_t)Tbl.str_longlen < length)
199				Tbl.str_longlen = length;
200			if ((off_t)Tbl.str_shortlen > length)
201				Tbl.str_shortlen = length;
202			first = Oflag;
203		}
204		else if (first) {
205			for (nsp = sp; !isalnum((unsigned char)*nsp); nsp++)
206				continue;
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);
211			else
212				fp->first = *nsp;
213			fp->pos = Seekpts[Num_pts - 1];
214			first = 0;
215		}
216	} while (sp != NULL);
217
218	/*
219	 * write the tables in
220	 */
221
222	(void) fclose(inf);
223
224	if (Oflag)
225		do_order();
226	else if (Rflag)
227		randomize();
228
229	if (Xflag)
230		Tbl.str_flags |= STR_ROTATED;
231
232	if (!Sflag) {
233		printf("\"%s\" created\n", Outfile);
234		if (Num_pts == 2)
235			puts("There was 1 string");
236		else
237			printf("There were %d strings\n", (int)(Num_pts - 1));
238		printf("Longest string: %lu byte%s\n", (unsigned long)Tbl.str_longlen,
239		       Tbl.str_longlen == 1 ? "" : "s");
240		printf("Shortest string: %lu byte%s\n", (unsigned long)Tbl.str_shortlen,
241		       Tbl.str_shortlen == 1 ? "" : "s");
242	}
243
244	(void) fseek(outf, (off_t) 0, SEEK_SET);
245	Tbl.str_version = h2nl(Tbl.str_version);
246	Tbl.str_numstr = h2nl(Num_pts - 1);
247	Tbl.str_longlen = h2nl(Tbl.str_longlen);
248	Tbl.str_shortlen = h2nl(Tbl.str_shortlen);
249	Tbl.str_flags = h2nl(Tbl.str_flags);
250	(void) fwrite((char *) &Tbl, sizeof Tbl, 1, outf);
251	if (STORING_PTRS) {
252		for (i = 0; i < Num_pts; i++)
253			fwrite_be_offt(Seekpts[i], outf);
254	}
255	fflush(outf);
256	if (ferror(outf))
257		err(1, "fwrite %s", Outfile);
258	(void) fclose(outf);
259	exit(0);
260}
261
262/*
263 *	This routine evaluates arguments from the command line
264 */
265static void
266getargs(int argc, char **argv)
267{
268	int	ch;
269	extern	int optind;
270	extern	char *optarg;
271	size_t	len;
272
273	while ((ch = getopt(argc, argv, "c:iorsx")) != -1)
274		switch(ch) {
275		case 'c':			/* new delimiting char */
276			Delimch = *optarg;
277			if (!isascii(Delimch)) {
278				printf("bad delimiting character: '\\%o\n'",
279				       Delimch);
280			}
281			break;
282		case 'i':			/* ignore case in ordering */
283			Iflag++;
284			break;
285		case 'o':			/* order strings */
286			Oflag++;
287			break;
288		case 'r':			/* randomize pointers */
289			Rflag++;
290			break;
291		case 's':			/* silent */
292			Sflag++;
293			break;
294		case 'x':			/* set the rotated bit */
295			Xflag++;
296			break;
297		case '?':
298		default:
299			usage(NULL);
300		}
301	argv += optind;
302
303	if (*argv) {
304		Infile = *argv;
305		if (*++argv) {
306			len = strlcpy(Outfile, *argv, sizeof(Outfile));
307			if (len >= sizeof(Outfile)) {
308				usage("Too long output filename");
309			}
310		}
311	}
312	if (!Infile) {
313		usage("No input file name");
314	}
315	if (*Outfile == '\0') {
316		len = snprintf(Outfile, sizeof(Outfile), "%s.dat", Infile);
317		if (len >= sizeof(Outfile)) {
318			usage("Too long input filename");
319		}
320	}
321}
322
323static void
324usage(const char *fmt, ...)
325{
326	if (fmt) {
327		va_list ap;
328		va_start(ap, fmt);
329		vwarnx(fmt, ap);
330		va_end(ap);
331	}
332	(void) fprintf(stderr,
333	    "Usage: %s [-iorsx] [-c char] sourcefile [datafile]\n",
334	    getprogname());
335	exit(1);
336}
337
338/*
339 * add_offset:
340 *	Add an offset to the list, or write it out, as appropriate.
341 */
342static void
343add_offset(FILE *fp, off_t off)
344{
345
346	if (!STORING_PTRS) {
347		fwrite_be_offt(off, fp);
348	} else {
349		ALLOC(Seekpts, Num_pts + 1);
350		Seekpts[Num_pts] = off;
351	}
352	Num_pts++;
353}
354
355/*
356 * do_order:
357 *	Order the strings alphabetically (possibly ignoring case).
358 */
359static void
360do_order(void)
361{
362	int	i;
363	off_t	*lp;
364	STR	*fp;
365
366	Sort_1 = fopen(Infile, "r");
367	Sort_2 = fopen(Infile, "r");
368	qsort((char *) Firstch, (int) Tbl.str_numstr, sizeof *Firstch, cmp_str);
369	i = Tbl.str_numstr;
370	lp = Seekpts;
371	fp = Firstch;
372	while (i--)
373		*lp++ = fp++->pos;
374	(void) fclose(Sort_1);
375	(void) fclose(Sort_2);
376	Tbl.str_flags |= STR_ORDERED;
377}
378
379static int
380cmp_str(const void *vp1, const void *vp2)
381{
382	const STR	*p1, *p2;
383	int	c1, c2;
384	int	n1, n2;
385
386	p1 = (const STR *)vp1;
387	p2 = (const STR *)vp2;
388
389# define	SET_N(nf,ch)	(nf = (ch == '\n'))
390# define	IS_END(ch,nf)	(ch == Delimch && nf)
391
392	c1 = p1->first;
393	c2 = p2->first;
394	if (c1 != c2)
395		return c1 - c2;
396
397	(void) fseek(Sort_1, p1->pos, SEEK_SET);
398	(void) fseek(Sort_2, p2->pos, SEEK_SET);
399
400	n1 = 0;
401	n2 = 0;
402	while (!isalnum(c1 = getc(Sort_1)) && c1 != '\0')
403		SET_N(n1, c1);
404	while (!isalnum(c2 = getc(Sort_2)) && c2 != '\0')
405		SET_N(n2, c2);
406
407	while (!IS_END(c1, n1) && !IS_END(c2, n2)) {
408		if (Iflag) {
409			if (isupper(c1))
410				c1 = tolower(c1);
411			if (isupper(c2))
412				c2 = tolower(c2);
413		}
414		if (c1 != c2)
415			return c1 - c2;
416		SET_N(n1, c1);
417		SET_N(n2, c2);
418		c1 = getc(Sort_1);
419		c2 = getc(Sort_2);
420	}
421	if (IS_END(c1, n1))
422		c1 = 0;
423	if (IS_END(c2, n2))
424		c2 = 0;
425	return c1 - c2;
426}
427
428/*
429 * randomize:
430 *	Randomize the order of the string table.  We must be careful
431 *	not to randomize across delimiter boundaries.  All
432 *	randomization is done within each block.
433 */
434static void
435randomize(void)
436{
437	int	cnt, i;
438	off_t	tmp;
439	off_t	*sp;
440
441	srandom((int)(time(NULL) + getpid()));
442
443	Tbl.str_flags |= STR_RANDOM;
444	cnt = Tbl.str_numstr;
445
446	/*
447	 * move things around randomly
448	 */
449
450	for (sp = Seekpts; cnt > 0; cnt--, sp++) {
451		i = random() % cnt;
452		tmp = sp[0];
453		sp[0] = sp[i];
454		sp[i] = tmp;
455	}
456}
457
458/*
459 * fwrite_be_offt:
460 *	Write out the off parameter as a 64 bit big endian number
461 */
462
463static void
464fwrite_be_offt(off_t off, FILE *f)
465{
466	int		i;
467	unsigned char	c[8];
468
469	for (i = 7; i >= 0; i--) {
470		c[i] = off & 0xff;
471		off >>= 8;
472	}
473	fwrite(c, sizeof(c), 1, f);
474}
475
476static uint32_t
477h2nl(uint32_t h)
478{
479        unsigned char c[4];
480        uint32_t rv;
481
482        c[0] = (h >> 24) & 0xff;
483        c[1] = (h >> 16) & 0xff;
484        c[2] = (h >>  8) & 0xff;
485        c[3] = (h >>  0) & 0xff;
486        memcpy(&rv, c, sizeof rv);
487
488        return (rv);
489}
490