1 /* Line breaking of Unicode strings.
2 Copyright (C) 2001-2003, 2005-2021 Free Software Foundation, Inc.
3 Written by Bruno Haible <bruno@clisp.org>, 2001.
4
5 This file is free software.
6 It is dual-licensed under "the GNU LGPLv3+ or the GNU GPLv2+".
7 You can redistribute it and/or modify it under either
8 - the terms of the GNU Lesser General Public License as published
9 by the Free Software Foundation; either version 3, or (at your
10 option) any later version, or
11 - the terms of the GNU General Public License as published by the
12 Free Software Foundation; either version 2, or (at your option)
13 any later version, or
14 - the same dual license "the GNU LGPLv3+ or the GNU GPLv2+".
15
16 This file is distributed in the hope that it will be useful,
17 but WITHOUT ANY WARRANTY; without even the implied warranty of
18 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
19 Lesser General Public License and the GNU General Public License
20 for more details.
21
22 You should have received a copy of the GNU Lesser General Public
23 License and of the GNU General Public License along with this
24 program. If not, see <https://www.gnu.org/licenses/>. */
25
26 #ifndef _UNILBRK_H
27 #define _UNILBRK_H
28
29 /* Get size_t. */
30 #include <stddef.h>
31
32 #include "unitypes.h"
33
34 /* Get locale_charset() declaration. */
35 #include "localcharset.h"
36
37
38 #ifdef __cplusplus
39 extern "C" {
40 #endif
41
42
43 /* These functions are locale dependent. The encoding argument identifies
44 the encoding (e.g. "ISO-8859-2" for Polish). */
45
46
47 /* Line breaking. */
48
49 enum
50 {
51 UC_BREAK_UNDEFINED,
52 UC_BREAK_PROHIBITED,
53 UC_BREAK_POSSIBLE,
54 UC_BREAK_MANDATORY,
55 UC_BREAK_HYPHENATION
56 };
57
58 /* Determine the line break points in S, and store the result at p[0..n-1].
59 p[i] = UC_BREAK_MANDATORY means that s[i] is a line break character.
60 p[i] = UC_BREAK_POSSIBLE means that a line break may be inserted between
61 s[i-1] and s[i].
62 p[i] = UC_BREAK_HYPHENATION means that a hyphen and a line break may be
63 inserted between s[i-1] and s[i]. But beware of language dependent
64 hyphenation rules.
65 p[i] = UC_BREAK_PROHIBITED means that s[i-1] and s[i] must not be separated.
66 */
67 extern void
68 u8_possible_linebreaks (const uint8_t *s, size_t n,
69 const char *encoding, char *_UC_RESTRICT p);
70 extern void
71 u16_possible_linebreaks (const uint16_t *s, size_t n,
72 const char *encoding, char *_UC_RESTRICT p);
73 extern void
74 u32_possible_linebreaks (const uint32_t *s, size_t n,
75 const char *encoding, char *_UC_RESTRICT p);
76 extern void
77 ulc_possible_linebreaks (const char *s, size_t n,
78 const char *encoding, char *_UC_RESTRICT p);
79
80 /* Choose the best line breaks, assuming the uc_width function.
81 The string is s[0..n-1]. The maximum number of columns per line is given
82 as WIDTH. The starting column of the string is given as START_COLUMN.
83 If the algorithm shall keep room after the last piece, they can be given
84 as AT_END_COLUMNS.
85 o is an optional override; if o[i] != UC_BREAK_UNDEFINED, o[i] takes
86 precedence over p[i] as returned by the *_possible_linebreaks function.
87 The given ENCODING is used for disambiguating widths in uc_width.
88 Return the column after the end of the string, and store the result at
89 p[0..n-1].
90 */
91 extern int
92 u8_width_linebreaks (const uint8_t *s, size_t n, int width,
93 int start_column, int at_end_columns,
94 const char *o, const char *encoding,
95 char *_UC_RESTRICT p);
96 extern int
97 u16_width_linebreaks (const uint16_t *s, size_t n, int width,
98 int start_column, int at_end_columns,
99 const char *o, const char *encoding,
100 char *_UC_RESTRICT p);
101 extern int
102 u32_width_linebreaks (const uint32_t *s, size_t n, int width,
103 int start_column, int at_end_columns,
104 const char *o, const char *encoding,
105 char *_UC_RESTRICT p);
106 extern int
107 ulc_width_linebreaks (const char *s, size_t n, int width,
108 int start_column, int at_end_columns,
109 const char *o, const char *encoding,
110 char *_UC_RESTRICT p);
111
112
113 #ifdef __cplusplus
114 }
115 #endif
116
117
118 #endif /* _UNILBRK_H */