src/libpspp/i18n.h

   1 /* PSPP - a program for statistical analysis.
   2    Copyright (C) 2006, 2010, 2011, 2012, 2014, 2016 Free Software Foundation, Inc.
   3
   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.
   8
   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.
  13
  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/>. */
  16
  17 #ifndef I18N_H
  18 #define I18N_H
  19
  20 #include <stdbool.h>
  21 #include <unistr.h>
  22
  23 void  i18n_done (void);
  24 void  i18n_init (void);
  25
  26 #define UTF8 "UTF-8"
  27
  28 /* The encoding of literal strings in PSPP source code, as seen at execution
  29    time.  In fact this is likely to be some extended ASCII encoding, such as
  30    UTF-8 or ISO-8859-1, but ASCII is adequate for our purposes. */
  31 #define C_ENCODING "ASCII"
  32
  33 struct pool;
  34
  35 char recode_byte (const char *to, const char *from, char);
  36
  37 char *recode_string (const char *to, const char *from,
  38                      const char *text, int len);
  39 char *recode_string_pool (const char *to, const char *from,
  40                           const char *text, int length, struct pool *);
  41 struct substring recode_substring_pool (const char *to, const char *from,
  42                                         struct substring text, struct pool *);
  43 int recode_pedantically (const char *to, const char *from,
  44                          struct substring text, struct pool *,
  45                          struct substring *out);
  46
  47 size_t recode_string_len (const char *to, const char *from,
  48                           const char *text, int len);
  49
  50 char *utf8_encoding_trunc (const char *, const char *encoding,
  51                            size_t max_len);
  52 size_t utf8_encoding_trunc_len (const char *, const char *encoding,
  53                                 size_t max_len);
  54
  55 char *utf8_encoding_concat (const char *head, const char *tail,
  56                             const char *encoding, size_t max_len);
  57 size_t utf8_encoding_concat_len (const char *head, const char *tail,
  58                                  const char *encoding, size_t max_len);
  59
  60 char *utf8_to_filename (const char *filename);
  61 char *filename_to_utf8 (const char *filename);
  62
  63 bool valid_encoding (const char *enc);
  64
  65 char get_system_decimal (void);
  66
  67 const char * get_default_encoding (void);
  68 void set_default_encoding (const char *enc);
  69
  70 bool set_encoding_from_locale (const char *loc);
  71
  72 const char *uc_name (ucs4_t uc, char buffer[16]);
  73
  74 unsigned int utf8_hash_case_bytes (const char *, size_t n, unsigned int basis);
  75 unsigned int utf8_hash_case_string (const char *, unsigned int basis);
  76 int utf8_strcasecmp (const char *, const char *);
  77 int utf8_strncasecmp (const char *, size_t, const char *, size_t);
  78 int utf8_strverscasecmp (const char *, const char *);
  79 char *utf8_to_upper (const char *);
  80 char *utf8_to_lower (const char *);
  81 \f
  82 /* Information about character encodings. */
  83
  84 /* ISO C defines a set of characters that a C implementation must support at
  85    runtime, called the C basic execution character set, which consists of the
  86    following characters:
  87
  88        A B C D E F G H I J K L M
  89        N O P Q R S T U V W X Y Z
  90        a b c d e f g h i j k l m
  91        n o p q r s t u v w x y z
  92        0 1 2 3 4 5 6 7 8 9
  93        ! " # % & ' ( ) * + , - . / :
  94        ; < = > ? [ \ ] ^ _ { | } ~
  95        space \a \b \r \n \t \v \f \0
  96
  97    The following is true of every member of the C basic execution character
  98    set in all "reasonable" encodings:
  99
 100        1. Every member of the C basic character set is encoded.
 101
 102        2. Every member of the C basic character set has the same width in
 103           bytes, called the "unit width".  Most encodings have a unit width of
 104           1 byte, but UCS-2 and UTF-16 have a unit width of 2 bytes and UCS-4
 105           and UTF-32 have a unit width of 4 bytes.
 106
 107        3. In a stateful encoding, the encoding of members of the C basic
 108           character set does not vary with shift state.
 109
 110        4. When a string is read unit-by-unit, a unit that has the encoded value
 111           of a member of the C basic character set, EXCEPT FOR THE DECIMAL
 112           DIGITS, always represents that member.  That is, if the encoding has
 113           multi-unit characters, the units that encode the C basic character
 114           set are never part of a multi-unit character.
 115
 116           The exception for decimal digits is due to GB18030, which uses
 117           decimal digits as part of multi-byte encodings.
 118
 119    All 8-bit and wider encodings that I have been able to find follow these
 120    rules.  7-bit and narrower encodings (e.g. UTF-7) do not.  I'm not too
 121    concerned about that. */
 122
 123 #include <stdbool.h>
 124
 125 /* Maximum width of a unit, in bytes.  UTF-32 with 4-byte units is the widest
 126    that I am aware of. */
 127 #define MAX_UNIT 4
 128
 129 /* Information about an encoding. */
 130 struct encoding_info
 131   {
 132     /* Encoding name.  IANA says character set names may be up to 40 US-ASCII
 133        characters. */
 134     char name[41];
 135
 136     /* True if this encoding has a unit width of 1 byte, and every character
 137        used in ASCII text files has the same value in this encoding. */
 138     bool is_ascii_compatible;
 139
 140     /* True if this encoding has a unit width of 1 byte and appears to be
 141        EBCDIC-based.  */
 142     bool is_ebcdic_compatible;
 143
 144     /* Character information. */
 145     int unit;                   /* Unit width, in bytes. */
 146     char cr[MAX_UNIT];          /* \r in encoding, 'unit' bytes long. */
 147     char lf[MAX_UNIT];          /* \n in encoding, 'unit' bytes long. */
 148     char space[MAX_UNIT];       /* ' ' in encoding, 'unit' bytes long. */
 149   };
 150
 151 bool get_encoding_info (struct encoding_info *, const char *name);
 152 bool is_encoding_ascii_compatible (const char *encoding);
 153 bool is_encoding_ebcdic_compatible (const char *encoding);
 154 bool is_encoding_supported (const char *encoding);
 155
 156 bool is_encoding_utf8 (const char *encoding);
 157 \f
 158 /* Database of encodings, by language or region. */
 159
 160 struct encoding_category
 161   {
 162     const char *category;       /* e.g. "Arabic" or "Western European". */
 163     const char **encodings;     /* Encodings within the category. */
 164     size_t n_encodings;         /* Number of encodings in category. */
 165   };
 166
 167 struct encoding_category *get_encoding_categories (void);
 168 size_t get_n_encoding_categories (void);
 169
 170 #endif /* i18n.h */