xref: /openbsd/usr.bin/sort/sort.h (revision 3d8817e4)
1 /*	$OpenBSD: sort.h,v 1.7 2007/08/21 20:29:25 millert Exp $	*/
2 
3 /*-
4  * Copyright (c) 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  * Peter McIlroy.
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  *	@(#)sort.h	8.1 (Berkeley) 6/6/93
35  */
36 
37 #include <sys/param.h>
38 
39 #include <db.h>
40 #include <err.h>
41 #include <errno.h>
42 #include <fcntl.h>
43 #include <limits.h>
44 #include <stdio.h>
45 #include <stdlib.h>
46 #include <string.h>
47 
48 #define NBINS 256
49 #define MAXMERGE 16
50 
51 /* values for masks, weights, and other flags. */
52 #define I 1		/* mask out non-printable characters */
53 #define D 2		/* sort alphanumeric characters only */
54 #define N 4		/* Field is a number */
55 #define F 8		/* weight lower and upper case the same */
56 #define R 16		/* Field is reversed with respect to the global weight */
57 #define BI 32		/* ignore blanks in icol */
58 #define BT 64		/* ignore blanks in tcol */
59 
60 /* masks for delimiters: blanks, fields, and termination. */
61 #define BLANK 1		/* ' ', '\t'; '\n' if -T is invoked */
62 #define FLD_D 2		/* ' ', '\t' default; from -t otherwise */
63 #define REC_D_F 4	/* '\n' default; from -T otherwise */
64 
65 #define min(a, b) ((a) < (b) ? (a) : (b))
66 #define max(a, b) ((a) > (b) ? (a) : (b))
67 
68 #define	FCLOSE(file) {							\
69 	if (EOF == fclose(file))					\
70 		err(2, "fclose");					\
71 }
72 
73 #define	EWRITE(ptr, size, n, f) {					\
74 	if (!fwrite(ptr, size, n, f))					\
75 		 err(2, "fwrite");					\
76 }
77 
78 /* length of record is currently limited to maximum string length (size_t) */
79 typedef size_t length_t;
80 
81 #define SALIGN(n) ((n+(sizeof(length_t)-1)) & ~(sizeof(length_t)-1))
82 
83 /* a record is a key/line pair starting at rec.data. It has a total length
84  * and an offset to the start of the line half of the pair.
85  */
86 typedef struct recheader {
87 	length_t length;
88 	length_t offset;
89 	u_char data[1];
90 } RECHEADER;
91 
92 typedef struct trecheader {
93 	length_t length;
94 	length_t offset;
95 } TRECHEADER;
96 
97 /* This is the column as seen by struct field.  It is used by enterfield.
98  * They are matched with corresponding coldescs during initialization.
99  */
100 struct column {
101 	struct coldesc *p;
102 	int num;
103 	int indent;
104 };
105 
106 /* a coldesc has a number and pointers to the beginning and end of the
107  * corresponding column in the current line.  This is determined in enterkey.
108  */
109 typedef struct coldesc {
110 	u_char *start;
111 	u_char *end;
112 	int num;
113 } COLDESC;
114 
115 /* A field has an initial and final column; an omitted final column
116  * implies the end of the line.  Flags regulate omission of blanks and
117  * numerical sorts; mask determines which characters are ignored (from -i, -d);
118  * weights determines the sort weights of a character (from -f, -r).
119  */
120 struct field {
121 	struct column icol;
122 	struct column tcol;
123 	u_int flags;
124 	u_char *mask;
125 	u_char *weights;
126 };
127 
128 union f_handle {
129 	int top;
130 	char **names;
131 };
132 extern int PANIC;	/* maximum depth of fsort before fmerge is called */
133 extern u_char ascii[NBINS], Rascii[NBINS], Ftable[NBINS], RFtable[NBINS];
134 extern u_char alltable[NBINS], dtable[NBINS], itable[NBINS];
135 extern u_char d_mask[NBINS];
136 extern int SINGL_FLD, SEP_FLAG, UNIQUE, STABLE;
137 extern int REC_D;
138 extern char *tmpdir;
139 extern int ND;		/* limit on number of -k options. */
140 
141 #include "extern.h"
142