1 /* PSPP - a program for statistical analysis.
2 Copyright (C) 2006, 2010, 2011, 2012, 2014, 2016 Free Software Foundation, Inc.
4 This program is free software: you can redistribute it and/or modify
5 it under the terms of the GNU General Public License as published by
6 the Free Software Foundation, either version 3 of the License, or
7 (at your option) any later version.
9 This program is distributed in the hope that it will be useful,
10 but WITHOUT ANY WARRANTY; without even the implied warranty of
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 GNU General Public License for more details.
14 You should have received a copy of the GNU General Public License
15 along with this program. If not, see <http://www.gnu.org/licenses/>. */
20 #include "libpspp/compiler.h"
24 void i18n_done (void);
25 void i18n_init (void);
29 /* The encoding of literal strings in PSPP source code, as seen at execution
30 time. In fact this is likely to be some extended ASCII encoding, such as
31 UTF-8 or ISO-8859-1, but ASCII is adequate for our purposes. */
32 #define C_ENCODING "ASCII"
36 char recode_byte (const char *to, const char *from, char);
38 char *recode_string (const char *to, const char *from,
39 const char *text, int len);
40 char *recode_string_pool (const char *to, const char *from,
41 const char *text, int length, struct pool *);
42 struct substring recode_substring_pool (const char *to, const char *from,
43 struct substring text, struct pool *);
44 int recode_pedantically (const char *to, const char *from,
45 struct substring text, struct pool *,
46 struct substring *out);
48 size_t recode_string_len (const char *to, const char *from,
49 const char *text, int len);
51 char *utf8_encoding_trunc (const char *, const char *encoding,
53 size_t utf8_encoding_trunc_len (const char *, const char *encoding,
56 char *utf8_encoding_concat (const char *head, const char *tail,
57 const char *encoding, size_t max_len);
58 size_t utf8_encoding_concat_len (const char *head, const char *tail,
59 const char *encoding, size_t max_len);
61 size_t utf8_count_columns (const char *, size_t);
62 size_t utf8_columns_to_bytes (const char *, size_t, size_t n_columns);
64 char *utf8_to_filename (const char *filename);
65 char *filename_to_utf8 (const char *filename);
67 bool valid_encoding (const char *enc);
69 char get_system_decimal (void);
71 const char * get_default_encoding (void);
72 void set_default_encoding (const char *enc);
74 bool set_encoding_from_locale (const char *loc);
76 const char *uc_name (ucs4_t uc, char buffer[16]);
78 unsigned int utf8_hash_case_bytes (const char *, size_t n, unsigned int basis) WARN_UNUSED_RESULT;
79 unsigned int utf8_hash_case_string (const char *, unsigned int basis) WARN_UNUSED_RESULT;
80 int utf8_strcasecmp (const char *, const char *);
81 int utf8_strncasecmp (const char *, size_t, const char *, size_t);
82 int utf8_strverscasecmp (const char *, const char *);
83 char *utf8_to_upper (const char *);
84 char *utf8_to_lower (const char *);
85 char *utf8_to_title (const char *);
87 /* Information about character encodings. */
89 /* ISO C defines a set of characters that a C implementation must support at
90 runtime, called the C basic execution character set, which consists of the
93 A B C D E F G H I J K L M
94 N O P Q R S T U V W X Y Z
95 a b c d e f g h i j k l m
96 n o p q r s t u v w x y z
98 ! " # % & ' () * + , - . / :
99 ; < = > ? [ \ ] ^ _ { | } ~
100 space \a \b \r \n \t \v \f \0
102 The following is true of every member of the C basic execution character
103 set in all "reasonable" encodings:
105 1. Every member of the C basic character set is encoded.
107 2. Every member of the C basic character set has the same width in
108 bytes, called the "unit width". Most encodings have a unit width of
109 1 byte, but UCS-2 and UTF-16 have a unit width of 2 bytes and UCS-4
110 and UTF-32 have a unit width of 4 bytes.
112 3. In a stateful encoding, the encoding of members of the C basic
113 character set does not vary with shift state.
115 4. When a string is read unit-by-unit, a unit that has the encoded value
116 of a member of the C basic character set, EXCEPT FOR THE DECIMAL
117 DIGITS, always represents that member. That is, if the encoding has
118 multi-unit characters, the units that encode the C basic character
119 set are never part of a multi-unit character.
121 The exception for decimal digits is due to GB18030, which uses
122 decimal digits as part of multi-byte encodings.
124 All 8-bit and wider encodings that I have been able to find follow these
125 rules. 7-bit and narrower encodings (e.g. UTF-7) do not. I'm not too
126 concerned about that. */
130 /* Maximum width of a unit, in bytes. UTF-32 with 4-byte units is the widest
131 that I am aware of. */
134 /* Information about an encoding. */
137 /* Encoding name. IANA says character set names may be up to 40 US-ASCII
141 /* True if this encoding has a unit width of 1 byte, and every character
142 used in ASCII text files has the same value in this encoding. */
143 bool is_ascii_compatible;
145 /* True if this encoding has a unit width of 1 byte and appears to be
147 bool is_ebcdic_compatible;
149 /* Character information. */
150 int unit; /* Unit width, in bytes. */
151 char cr[MAX_UNIT]; /* \r in encoding, 'unit' bytes long. */
152 char lf[MAX_UNIT]; /* \n in encoding, 'unit' bytes long. */
153 char space[MAX_UNIT]; /* ' ' in encoding, 'unit' bytes long. */
156 bool get_encoding_info (struct encoding_info *, const char *name);
157 bool is_encoding_ascii_compatible (const char *encoding);
158 bool is_encoding_ebcdic_compatible (const char *encoding);
159 bool is_encoding_supported (const char *encoding);
161 bool is_encoding_utf8 (const char *encoding);
163 /* Database of encodings, by language or region. */
165 struct encoding_category
167 const char *category; /* e.g. "Arabic" or "Western European". */
168 const char **encodings; /* Encodings within the category. */
169 size_t n_encodings; /* Number of encodings in category. */
172 struct encoding_category *get_encoding_categories (void);
173 size_t get_n_encoding_categories (void);
175 /* Return the ISO two letter code for the current LC_MESSAGES
177 char *get_language (void);