xref: /freebsd/usr.bin/sort/coll.h (revision 2008043f386721d58158e37e0d7e50df8095942d)
1 
2 /*-
3  * SPDX-License-Identifier: BSD-2-Clause
4  *
5  * Copyright (C) 2009 Gabor Kovesdan <gabor@FreeBSD.org>
6  * Copyright (C) 2012 Oleg Moskalenko <mom040267@gmail.com>
7  * All rights reserved.
8  *
9  * Redistribution and use in source and binary forms, with or without
10  * modification, are permitted provided that the following conditions
11  * are met:
12  * 1. Redistributions of source code must retain the above copyright
13  *    notice, this list of conditions and the following disclaimer.
14  * 2. Redistributions in binary form must reproduce the above copyright
15  *    notice, this list of conditions and the following disclaimer in the
16  *    documentation and/or other materials provided with the distribution.
17  *
18  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
19  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
20  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
21  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
22  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
23  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
24  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
25  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
26  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
27  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
28  * SUCH DAMAGE.
29  */
30 
31 #if !defined(__COLL_H__)
32 #define	__COLL_H__
33 
34 #include "bwstring.h"
35 #include "sort.h"
36 
37 /*
38  * Sort hint data for -n
39  */
40 struct n_hint
41 {
42 	unsigned long long	 n1;
43 	unsigned char		 si;
44 	bool			 empty;
45 	bool			 neg;
46 };
47 
48 /*
49  * Sort hint data for -g
50  */
51 struct g_hint
52 {
53 	double			 d;
54 	bool			 nan;
55 	bool			 notnum;
56 };
57 
58 /*
59  * Sort hint data for -M
60  */
61 struct M_hint
62 {
63 	int			 m;
64 };
65 
66 /*
67  * Sort hint data for -R
68  *
69  * This stores the first 12 bytes of the digest rather than the full output to
70  * avoid increasing the size of the 'key_hint' object via the 'v' union.
71  */
72 struct R_hint
73 {
74 	unsigned char		 cached[12];
75 };
76 
77 /*
78  * Status of a sort hint object
79  */
80 typedef enum
81 {
82 	HS_ERROR = -1, HS_UNINITIALIZED = 0, HS_INITIALIZED = 1
83 } hint_status;
84 
85 /*
86  * Sort hint object
87  */
88 struct key_hint
89 {
90 	hint_status		status;
91 	union
92 	{
93 		struct n_hint		nh;
94 		struct g_hint		gh;
95 		struct M_hint		Mh;
96 		struct R_hint		Rh;
97 	}			v;
98 };
99 
100 /*
101  * Key value
102  */
103 struct key_value
104 {
105 	struct bwstring		*k; /* key string */
106 	struct key_hint		 hint[0]; /* key sort hint */
107 } __packed;
108 
109 /*
110  * Set of keys container object.
111  */
112 struct keys_array
113 {
114 	struct key_value	 key[0];
115 };
116 
117 /*
118  * Parsed -k option data
119  */
120 struct key_specs
121 {
122 	struct sort_mods	 sm;
123 	size_t			 c1;
124 	size_t			 c2;
125 	size_t			 f1;
126 	size_t			 f2;
127 	bool			 pos1b;
128 	bool			 pos2b;
129 };
130 
131 /*
132  * Single entry in sort list.
133  */
134 struct sort_list_item
135 {
136 	struct bwstring		*str;
137 	struct keys_array	 ka;
138 };
139 
140 /*
141  * Function type, used to compare two list objects
142  */
143 typedef int (*listcoll_t)(struct sort_list_item **ss1, struct sort_list_item **ss2);
144 
145 extern struct key_specs *keys;
146 extern size_t keys_num;
147 
148 /*
149  * Main localised symbols. These must be wint_t as they may hold WEOF.
150  */
151 extern wint_t symbol_decimal_point;
152 extern wint_t symbol_thousands_sep;
153 extern wint_t symbol_negative_sign;
154 extern wint_t symbol_positive_sign;
155 
156 /* funcs */
157 
158 cmpcoll_t get_sort_func(struct sort_mods *sm);
159 
160 struct keys_array *keys_array_alloc(void);
161 size_t keys_array_size(void);
162 struct key_value *get_key_from_keys_array(struct keys_array *ka, size_t ind);
163 void set_key_on_keys_array(struct keys_array *ka, struct bwstring *s, size_t ind);
164 void clean_keys_array(const struct bwstring *s, struct keys_array *ka);
165 
166 struct sort_list_item *sort_list_item_alloc(void);
167 void sort_list_item_set(struct sort_list_item *si, struct bwstring *str);
168 void sort_list_item_clean(struct sort_list_item *si);
169 size_t sort_list_item_size(struct sort_list_item *si);
170 
171 int preproc(struct bwstring *s, struct keys_array *ka);
172 int top_level_str_coll(const struct bwstring *, const struct bwstring *);
173 int key_coll(struct keys_array *ks1, struct keys_array *ks2, size_t offset);
174 int str_list_coll(struct bwstring *str1, struct sort_list_item **ss2);
175 int list_coll_by_str_only(struct sort_list_item **ss1, struct sort_list_item **ss2);
176 int list_coll(struct sort_list_item **ss1, struct sort_list_item **ss2);
177 int list_coll_offset(struct sort_list_item **ss1, struct sort_list_item **ss2, size_t offset);
178 
179 listcoll_t get_list_call_func(size_t offset);
180 
181 #endif /* __COLL_H__ */
182