1/*	$OpenBSD: strfile.c,v 1.30 2020/02/14 19:17:33 schwarze Exp $	*/
2/*	$NetBSD: strfile.c,v 1.4 1995/04/24 12:23:09 cgd Exp $	*/
3
4/*-
5 * Copyright (c) 1989, 1993
6 *	The Regents of the University of California.  All rights reserved.
7 *
8 * This code is derived from software contributed to Berkeley by
9 * Ken Arnold.
10 *
11 * Redistribution and use in source and binary forms, with or without
12 * modification, are permitted provided that the following conditions
13 * are met:
14 * 1. Redistributions of source code must retain the above copyright
15 *    notice, this list of conditions and the following disclaimer.
16 * 2. Redistributions in binary form must reproduce the above copyright
17 *    notice, this list of conditions and the following disclaimer in the
18 *    documentation and/or other materials provided with the distribution.
19 * 3. Neither the name of the University nor the names of its contributors
20 *    may be used to endorse or promote products derived from this software
21 *    without specific prior written permission.
22 *
23 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
24 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
25 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
26 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
27 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
28 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
29 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
30 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
31 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
32 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
33 * SUCH DAMAGE.
34 */
35
36#include <ctype.h>
37#include <err.h>
38#include <limits.h>
39#include <stdbool.h>
40#include <stdio.h>
41#include <stdlib.h>
42#include <string.h>
43#include <unistd.h>
44
45#include "strfile.h"
46
47/*
48 *	This program takes a file composed of strings separated by
49 * lines starting with two consecutive delimiting character (default
50 * character is '%') and creates another file which consists of a table
51 * describing the file (structure from "strfile.h"), a table of seek
52 * pointers to the start of the strings, and the strings, each terminated
53 * by a null byte.  Usage:
54 *
55 *	% strfile [-iorsx] [ -cC ] sourcefile [ datafile ]
56 *
57 *	c - Change delimiting character from '%' to 'C'
58 *	s - Silent.  Give no summary of data processed at the end of
59 *	    the run.
60 *	o - order the strings in alphabetic order
61 *	i - if ordering, ignore case
62 *	r - randomize the order of the strings
63 *	x - set rotated bit
64 *
65 *		Ken Arnold	Sept. 7, 1978 --
66 *
67 *	Added ordering options.
68 */
69
70#define	STORING_PTRS	(Oflag || Rflag)
71#define	CHUNKSIZE	512
72
73# define	ALLOC(ptr,sz)	do { \
74			if (ptr == NULL) \
75				ptr = calloc(CHUNKSIZE, sizeof *ptr); \
76			else if (((sz) + 1) % CHUNKSIZE == 0) \
77				ptr = reallocarray(ptr, \
78						   (sz) + CHUNKSIZE, \
79						   sizeof(*ptr)); \
80			if (ptr == NULL) \
81				err(1, NULL); \
82		} while (0)
83
84typedef struct {
85	char	first;
86	int32_t	pos;
87} STR;
88
89char	*Infile		= NULL,		/* input file name */
90	Outfile[PATH_MAX] = "",		/* output file name */
91	Delimch		= '%';		/* delimiting character */
92
93bool	Sflag		= false;	/* silent run flag */
94bool	Oflag		= false;	/* ordering flag */
95bool	Iflag		= false;	/* ignore case flag */
96bool	Rflag		= false;	/* randomize order flag */
97bool	Xflag		= false;	/* set rotated bit */
98long	Num_pts		= 0;		/* number of pointers/strings */
99
100int32_t	*Seekpts;
101
102FILE	*Sort_1, *Sort_2;		/* pointers for sorting */
103
104STRFILE	Tbl;				/* statistics table */
105
106STR	*Firstch;			/* first chars of each string */
107
108
109void add_offset(FILE *, int32_t);
110int cmp_str(const void *, const void *);
111void do_order(void);
112void getargs(int, char **);
113void randomize(void);
114char *unctrl(char);
115__dead void usage(void);
116
117/*
118 * main:
119 *	Drive the sucker.  There are two main modes -- either we store
120 *	the seek pointers, if the table is to be sorted or randomized,
121 *	or we write the pointer directly to the file, if we are to stay
122 *	in file order.  If the former, we allocate and re-allocate in
123 *	CHUNKSIZE blocks; if the latter, we just write each pointer,
124 *	and then seek back to the beginning to write in the table.
125 */
126int
127main(int ac, char *av[])
128{
129	bool		first;
130	char		*sp, dc;
131	FILE		*inf, *outf;
132	int32_t		last_off, length, pos;
133	int32_t		*p;
134	int		cnt;
135	char		*nsp;
136	STR		*fp;
137	static char	string[257];
138
139	if (pledge("stdio rpath wpath cpath", NULL) == -1)
140		err(1, "pledge");
141
142	getargs(ac, av);		/* evalute arguments */
143	dc = Delimch;
144	if ((inf = fopen(Infile, "r")) == NULL)
145		err(1, "%s", Infile);
146
147	if ((outf = fopen(Outfile, "w")) == NULL)
148		err(1, "%s", Outfile);
149
150	if (pledge("stdio", NULL) == -1)
151		err(1, "pledge");
152
153	if (!STORING_PTRS)
154		(void) fseek(outf, sizeof Tbl, SEEK_SET);
155
156	/*
157	 * Write the strings onto the file
158	 */
159
160	Tbl.str_longlen = 0;
161	Tbl.str_shortlen = (unsigned int) 0xffffffff;
162	Tbl.str_delim = dc;
163	Tbl.str_version = VERSION;
164	first = Oflag;
165	add_offset(outf, ftell(inf));
166	last_off = 0;
167	do {
168		sp = fgets(string, sizeof(string), inf);
169		if (sp == NULL || (sp[0] == dc && sp[1] == '\n')) {
170			pos = ftell(inf);
171			length = pos - last_off - (sp ? strlen(sp) : 0);
172			last_off = pos;
173			if (!length)
174				continue;
175			add_offset(outf, pos);
176			if (Tbl.str_longlen < (u_int32_t)length)
177				Tbl.str_longlen = length;
178			if (Tbl.str_shortlen > (u_int32_t)length)
179				Tbl.str_shortlen = length;
180			first = Oflag;
181		} else if (first) {
182			for (nsp = sp; !isalnum((unsigned char)*nsp); nsp++)
183				continue;
184			ALLOC(Firstch, Num_pts);
185			fp = &Firstch[Num_pts - 1];
186			if (Iflag && isupper((unsigned char)*nsp))
187				fp->first = tolower((unsigned char)*nsp);
188			else
189				fp->first = *nsp;
190			fp->pos = Seekpts[Num_pts - 1];
191			first = false;
192		}
193	} while (sp != NULL);
194
195	/*
196	 * write the tables in
197	 */
198
199	(void) fclose(inf);
200	Tbl.str_numstr = Num_pts - 1;
201	if (Tbl.str_numstr == 0)
202		Tbl.str_shortlen = 0;
203
204	if (Oflag)
205		do_order();
206	else if (Rflag)
207		randomize();
208
209	if (Xflag)
210		Tbl.str_flags |= STR_ROTATED;
211
212	if (!Sflag) {
213		printf("\"%s\" created\n", Outfile);
214		if (Tbl.str_numstr == 1)
215			puts("There was 1 string");
216		else
217			printf("There were %u strings\n", Tbl.str_numstr);
218		printf("Longest string: %lu byte%s\n",
219			  (unsigned long) Tbl.str_longlen,
220		       Tbl.str_longlen == 1 ? "" : "s");
221		printf("Shortest string: %lu byte%s\n",
222			  (unsigned long) Tbl.str_shortlen,
223		       Tbl.str_shortlen == 1 ? "" : "s");
224	}
225
226	(void) fseek(outf, 0, SEEK_SET);
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	(void) fwrite(&Tbl.str_version,  sizeof(Tbl.str_version),  1, outf);
233	(void) fwrite(&Tbl.str_numstr,   sizeof(Tbl.str_numstr),   1, outf);
234	(void) fwrite(&Tbl.str_longlen,  sizeof(Tbl.str_longlen),  1, outf);
235	(void) fwrite(&Tbl.str_shortlen, sizeof(Tbl.str_shortlen), 1, outf);
236	(void) fwrite(&Tbl.str_flags,    sizeof(Tbl.str_flags),    1, outf);
237	(void) fwrite( Tbl.stuff,	 sizeof(Tbl.stuff),	   1, outf);
238	if (STORING_PTRS) {
239		for (p = Seekpts, cnt = Num_pts; cnt--; ++p) {
240			*p = htonl(*p);
241			(void) fwrite(p, sizeof(*p), 1, outf);
242		}
243	}
244	if (fclose(outf))
245		err(1, "fclose `%s'", Outfile);
246	return 0;
247}
248
249/*
250 *	This routine evaluates arguments from the command line
251 */
252void
253getargs(int argc, char *argv[])
254{
255	int	ch;
256
257	while ((ch = getopt(argc, argv, "c:hiorsx")) != -1) {
258		switch(ch) {
259		case 'c':			/* new delimiting char */
260			Delimch = *optarg;
261			if (!isascii((unsigned char)Delimch)) {
262				printf("bad delimiting character: '\\%o\n'",
263				       Delimch);
264			}
265			break;
266		case 'i':			/* ignore case in ordering */
267			Iflag = true;
268			break;
269		case 'o':			/* order strings */
270			Oflag = true;
271			break;
272		case 'r':			/* randomize pointers */
273			Rflag = true;
274			break;
275		case 's':			/* silent */
276			Sflag = true;
277			break;
278		case 'x':			/* set the rotated bit */
279			Xflag = true;
280			break;
281		case 'h':
282		default:
283			usage();
284		}
285	}
286	argv += optind;
287
288	if (*argv) {
289		Infile = *argv;
290		if (*++argv)
291			(void) strlcpy(Outfile, *argv, sizeof Outfile);
292	}
293	if (!Infile) {
294		puts("No input file name");
295		usage();
296	}
297	if (*Outfile == '\0') {
298		(void) strlcpy(Outfile, Infile, sizeof(Outfile));
299		if (strlcat(Outfile, ".dat", sizeof(Outfile)) >= sizeof(Outfile))
300			errx(1, "`%s': name too long", Infile);
301	}
302}
303
304void
305usage(void)
306{
307	(void) fprintf(stderr,
308	    "%s [-iorsx] [-c char] sourcefile [datafile]\n", getprogname());
309	exit(1);
310}
311
312/*
313 * add_offset:
314 *	Add an offset to the list, or write it out, as appropriate.
315 */
316void
317add_offset(FILE *fp, int32_t off)
318{
319	int32_t net;
320
321	if (!STORING_PTRS) {
322		net = htonl(off);
323		fwrite(&net, 1, sizeof net, fp);
324	} else {
325		ALLOC(Seekpts, Num_pts + 1);
326		Seekpts[Num_pts] = off;
327	}
328	Num_pts++;
329}
330
331/*
332 * do_order:
333 *	Order the strings alphabetically (possibly ignoring case).
334 */
335void
336do_order(void)
337{
338	int	i;
339	int32_t	*lp;
340	STR	*fp;
341
342	Sort_1 = fopen(Infile, "r");
343	Sort_2 = fopen(Infile, "r");
344	qsort((char *) Firstch, (int) Tbl.str_numstr, sizeof *Firstch, cmp_str);
345	i = Tbl.str_numstr;
346	lp = Seekpts;
347	fp = Firstch;
348	while (i--)
349		*lp++ = fp++->pos;
350	(void) fclose(Sort_1);
351	(void) fclose(Sort_2);
352	Tbl.str_flags |= STR_ORDERED;
353}
354
355/*
356 * cmp_str:
357 *	Compare two strings in the file
358 */
359char *
360unctrl(char c)
361{
362	static char	buf[3];
363
364	if (isprint((unsigned char)c)) {
365		buf[0] = c;
366		buf[1] = '\0';
367	} else if (c == 0177) {
368		buf[0] = '^';
369		buf[1] = '?';
370	} else {
371		buf[0] = '^';
372		buf[1] = c + 'A' - 1;
373	}
374	return buf;
375}
376
377int
378cmp_str(const void *p1, const void *p2)
379{
380	bool	n1, n2;
381	int	c1, c2;
382
383# define	SET_N(nf,ch)	(nf = (ch == '\n'))
384# define	IS_END(ch,nf)	(ch == Delimch && nf)
385
386	c1 = ((STR *)p1)->first;
387	c2 = ((STR *)p2)->first;
388	if (c1 != c2)
389		return c1 - c2;
390
391	(void) fseek(Sort_1, ((STR *)p1)->pos, SEEK_SET);
392	(void) fseek(Sort_2, ((STR *)p2)->pos, SEEK_SET);
393
394	n1 = false;
395	n2 = false;
396	while (!isalnum(c1 = getc(Sort_1)) && c1 != '\0')
397		SET_N(n1, c1);
398	while (!isalnum(c2 = getc(Sort_2)) && c2 != '\0')
399		SET_N(n2, c2);
400
401	while (!IS_END(c1, n1) && !IS_END(c2, n2)) {
402		if (Iflag) {
403			if (isupper(c1))
404				c1 = tolower(c1);
405			if (isupper(c2))
406				c2 = tolower(c2);
407		}
408		if (c1 != c2)
409			return c1 - c2;
410		SET_N(n1, c1);
411		SET_N(n2, c2);
412		c1 = getc(Sort_1);
413		c2 = getc(Sort_2);
414	}
415	if (IS_END(c1, n1))
416		c1 = 0;
417	if (IS_END(c2, n2))
418		c2 = 0;
419	return c1 - c2;
420}
421
422/*
423 * randomize:
424 *	Randomize the order of the string table.  We must be careful
425 *	not to randomize across delimiter boundaries.  All
426 *	randomization is done within each block.
427 */
428void
429randomize(void)
430{
431	int	cnt, i;
432	int32_t	tmp;
433	int32_t	*sp;
434
435	Tbl.str_flags |= STR_RANDOM;
436	cnt = Tbl.str_numstr;
437
438	/*
439	 * move things around randomly
440	 */
441
442	for (sp = Seekpts; cnt > 0; cnt--, sp++) {
443		i = arc4random_uniform(cnt);
444		tmp = sp[0];
445		sp[0] = sp[i];
446		sp[i] = tmp;
447	}
448}
449