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 * $FreeBSD: src/games/fortune/strfile/strfile.c,v 1.15.2.2 2001/03/05 11:52:37 kris Exp $
33 * $DragonFly: src/games/fortune/strfile/strfile.c,v 1.6 2008/07/10 18:29:51 swildner Exp $
35 * @(#) Copyright (c) 1989, 1993 The Regents of the University of California. All rights reserved.
36 * @(#)strfile.c 8.1 (Berkeley) 5/31/93
37 * $FreeBSD: src/games/fortune/strfile/strfile.c,v 1.15.2.2 2001/03/05 11:52:37 kris Exp $
40 #include <sys/param.h>
41 #include <netinet/in.h>
53 * This program takes a file composed of strings separated by
54 * lines starting with two consecutive delimiting character (default
55 * character is '%') and creates another file which consists of a table
56 * describing the file (structure from "strfile.h"), a table of seek
57 * pointers to the start of the strings, and the strings, each terminated
58 * by a null byte. Usage:
60 * % strfile [-iorsx] [ -cC ] sourcefile [ datafile ]
62 * C - Allow comments marked by a double delimiter at line's beginning
63 * c - Change delimiting character from '%' to 'C'
64 * s - Silent. Give no summary of data processed at the end of
66 * o - order the strings in alphabetic order
67 * i - if ordering, ignore case
68 * r - randomize the order of the strings
71 * Ken Arnold Sept. 7, 1978 --
73 * Added ordering options.
76 #define STORING_PTRS (Oflag || Rflag)
79 #define ALLOC(ptr,sz) do { \
81 ptr = malloc((unsigned int)(CHUNKSIZE * sizeof(*ptr))); \
82 else if (((sz) + 1) % CHUNKSIZE == 0) \
83 ptr = realloc(ptr, ((unsigned int)((sz) + CHUNKSIZE) * sizeof(*ptr))); \
85 fprintf(stderr, "out of space\n"); \
95 char *Infile
= NULL
, /* input file name */
96 Outfile
[MAXPATHLEN
] = "", /* output file name */
97 Delimch
= '%'; /* delimiting character */
99 int Cflag
= false; /* embedded comments */
100 int Sflag
= false; /* silent run flag */
101 int Oflag
= false; /* ordering flag */
102 int Iflag
= false; /* ignore case flag */
103 int Rflag
= false; /* randomize order flag */
104 int Xflag
= false; /* set rotated bit */
105 long Num_pts
= 0; /* number of pointers/strings */
109 FILE *Sort_1
, *Sort_2
; /* pointers for sorting */
111 STRFILE Tbl
; /* statistics table */
113 STR
*Firstch
; /* first chars of each string */
115 void add_offset(FILE *, long);
116 int cmp_str(const void *, const void *);
117 static int collate_range_cmp(int, int);
119 void getargs(int, char **);
120 void randomize(void);
125 * Drive the sucker. There are two main modes -- either we store
126 * the seek pointers, if the table is to be sorted or randomized,
127 * or we write the pointer directly to the file, if we are to stay
128 * in file order. If the former, we allocate and re-allocate in
129 * CHUNKSIZE blocks; if the latter, we just write each pointer,
130 * and then seek back to the beginning to write in the table.
133 main(int ac
, char *av
[])
137 long last_off
, length
, pos
, *p
;
140 static char string
[257];
142 setlocale(LC_ALL
, "");
144 getargs(ac
, av
); /* evalute arguments */
146 if ((inf
= fopen(Infile
, "r")) == NULL
) {
151 if ((outf
= fopen(Outfile
, "w")) == NULL
) {
156 fseek(outf
, sizeof(Tbl
), SEEK_SET
);
159 * Write the strings onto the file
163 Tbl
.str_shortlen
= ~((unsigned long)0);
165 Tbl
.str_version
= VERSION
;
167 add_offset(outf
, ftell(inf
));
170 sp
= fgets(string
, 256, inf
);
171 if (sp
== NULL
|| (sp
[0] == dc
&& sp
[1] == '\n')) {
173 length
= pos
- last_off
- (sp
? strlen(sp
) : 0);
177 add_offset(outf
, pos
);
178 if (Tbl
.str_longlen
< (unsigned long)length
)
179 Tbl
.str_longlen
= length
;
180 if (Tbl
.str_shortlen
> (unsigned long)length
)
181 Tbl
.str_shortlen
= length
;
185 for (nsp
= sp
; !isalnum((unsigned char)*nsp
); nsp
++)
187 ALLOC(Firstch
, Num_pts
);
188 fp
= &Firstch
[Num_pts
- 1];
189 if (Iflag
&& isupper((unsigned char)*nsp
))
190 fp
->first
= tolower((unsigned char)*nsp
);
193 fp
->pos
= Seekpts
[Num_pts
- 1];
196 } while (sp
!= NULL
);
199 * write the tables in
203 Tbl
.str_numstr
= Num_pts
- 1;
206 Tbl
.str_flags
|= STR_COMMENTS
;
214 Tbl
.str_flags
|= STR_ROTATED
;
217 printf("\"%s\" created\n", Outfile
);
219 puts("There was 1 string");
221 printf("There were %ld strings\n", Num_pts
- 1);
222 printf("Longest string: %lu byte%s\n", Tbl
.str_longlen
,
223 Tbl
.str_longlen
== 1 ? "" : "s");
224 printf("Shortest string: %lu byte%s\n", Tbl
.str_shortlen
,
225 Tbl
.str_shortlen
== 1 ? "" : "s");
229 Tbl
.str_version
= htonl(Tbl
.str_version
);
230 Tbl
.str_numstr
= htonl(Tbl
.str_numstr
);
231 Tbl
.str_longlen
= htonl(Tbl
.str_longlen
);
232 Tbl
.str_shortlen
= htonl(Tbl
.str_shortlen
);
233 Tbl
.str_flags
= htonl(Tbl
.str_flags
);
234 fwrite((char *)&Tbl
, sizeof(Tbl
), 1, outf
);
236 for (p
= Seekpts
, cnt
= Num_pts
; cnt
--; ++p
)
238 fwrite((char *)Seekpts
, sizeof(*Seekpts
), (int)Num_pts
, outf
);
245 * This routine evaluates arguments from the command line
248 getargs(int argc
, char **argv
)
252 while ((ch
= getopt(argc
, argv
, "Cc:iorsx")) != -1)
254 case 'C': /* embedded comments */
257 case 'c': /* new delimiting char */
259 if (!isascii(Delimch
)) {
260 printf("bad delimiting character: '\\%o\n'",
261 (unsigned char)Delimch
);
264 case 'i': /* ignore case in ordering */
267 case 'o': /* order strings */
270 case 'r': /* randomize pointers */
273 case 's': /* silent */
276 case 'x': /* set the rotated bit */
288 strcpy(Outfile
, *argv
);
291 puts("No input file name");
294 if (*Outfile
== '\0') {
295 strcpy(Outfile
, Infile
);
296 strcat(Outfile
, ".dat");
304 "strfile [-Ciorsx] [-c char] sourcefile [datafile]\n");
310 * Add an offset to the list, or write it out, as appropriate.
313 add_offset(FILE *fp
, long off
)
319 fwrite(&net
, 1, sizeof(net
), fp
);
321 ALLOC(Seekpts
, Num_pts
+ 1);
322 Seekpts
[Num_pts
] = off
;
329 * Order the strings alphabetically (possibly ignoring case).
338 Sort_1
= fopen(Infile
, "r");
339 Sort_2
= fopen(Infile
, "r");
340 qsort((char *)Firstch
, (int)Tbl
.str_numstr
, sizeof(*Firstch
), cmp_str
);
348 Tbl
.str_flags
|= STR_ORDERED
;
352 collate_range_cmp (int c1
, int c2
)
354 static char s1
[2], s2
[2];
363 if ((ret
= strcoll(s1
, s2
)) != 0)
370 * Compare two strings in the file
373 cmp_str(const void *s1
, const void *s2
)
376 int c1
, c2
, n1
, n2
, r
;
378 #define SET_N(nf,ch) (nf = (ch == '\n'))
379 #define IS_END(ch,nf) (ch == EOF || (ch == (unsigned char)Delimch && nf))
381 p1
= (const STR
*)s1
;
382 p2
= (const STR
*)s2
;
384 c1
= (unsigned char)p1
->first
;
385 c2
= (unsigned char)p2
->first
;
386 if ((r
= collate_range_cmp(c1
, c2
)) != 0)
389 fseek(Sort_1
, p1
->pos
, SEEK_SET
);
390 fseek(Sort_2
, p2
->pos
, SEEK_SET
);
394 while (!isalnum(c1
= getc(Sort_1
)) && c1
!= '\0' && c1
!= EOF
)
396 while (!isalnum(c2
= getc(Sort_2
)) && c2
!= '\0' && c2
!= EOF
)
399 while (!IS_END(c1
, n1
) && !IS_END(c2
, n2
)) {
406 if ((r
= collate_range_cmp(c1
, c2
)) != 0)
418 return (collate_range_cmp(c1
, c2
));
423 * Randomize the order of the string table. We must be careful
424 * not to randomize across delimiter boundaries. All
425 * randomization is done within each block.
436 Tbl
.str_flags
|= STR_RANDOM
;
437 cnt
= Tbl
.str_numstr
;
440 * move things around randomly
443 for (sp
= Seekpts
; cnt
> 0; cnt
--, sp
++) {