m17nCharset (3m17n)

Charset objects and API for them.

The m17n library uses charset objects to represent a coded character sets (CCS). The m17n library supports many predefined coded character sets. r, application programs can add other charsets. A character can belong to multiple charsets.

The m17n library distinguishes the following three concepts:

• A code-point is a number assigned by the CCS to each character. Code-points may or may not be continuous. The type unsigned is used to represent a code-point. An invalid code-point is represented by the macro MCHAR_INVALID_CODE.

• A character index is the canonical index of a character in a CCS. The character that has the character index N occupies the Nth position when all the characters in the current CCS are sorted by their code-points. Character indices in a CCS are continuous and start with 0.

• A character code is the internal representation in the m17n library of a character. A character code is a signed integer of 21 bits or longer.

Each charset object defines how characters are converted between code-points and character codes. To encode means converting code-points to character codes and to decode means converting character codes to code-points.

#define MCHAR_INVALID_CODE

Invalid code-point. The macro MCHAR_INVALID_CODE gives the invalid code-point.

\fBMSymbol\fP \fBMcharset_ascii\fP

Symbol representing the charset ASCII. The symbol Mcharset_ascii has name 'ascii' and represents the charset ISO 646, USA Version X3.4-1968 (ISO-IR-6).

\fBMSymbol\fP \fBMcharset_iso_8859_1\fP

Symbol representing the charset ISO/IEC 8859/1. The symbol Mcharset_iso_8859_1 has name 'iso-8859-1' and represents the charset ISO/IEC 8859-1:1998.

\fBMSymbol\fP \fBMcharset_unicode\fP

Symbol representing the charset Unicode. The symbol Mcharset_unicode has name 'unicode' and represents the charset Unicode.

\fBMSymbol\fP \fBMcharset_m17n\fP

Symbol representing the largest charset. The symbol Mcharset_m17n has name 'm17n' and represents the charset that contains all characters supported by the m17n library.

\fBMSymbol\fP \fBMcharset_binary\fP

Symbol representing the charset for ill-decoded characters. The symbol Mcharset_binary has name 'binary' and represents the fake charset which the decoding functions put to an M-text as a text property when they encounter an invalid byte (sequence).

See Code Conversion for more details.

\fBMSymbol\fP \fBMmethod\fP

\fBMSymbol\fP \fBMdimension\fP

\fBMSymbol\fP \fBMmin_range\fP

\fBMSymbol\fP \fBMmax_range\fP

\fBMSymbol\fP \fBMmin_code\fP

\fBMSymbol\fP \fBMmax_code\fP

\fBMSymbol\fP \fBMascii_compatible\fP

\fBMSymbol\fP \fBMfinal_byte\fP

\fBMSymbol\fP \fBMrevision\fP

\fBMSymbol\fP \fBMmin_char\fP

\fBMSymbol\fP \fBMmapfile\fP

\fBMSymbol\fP \fBMparents\fP

\fBMSymbol\fP \fBMsubset_offset\fP

\fBMSymbol\fP \fBMdefine_coding\fP

\fBMSymbol\fP \fBMaliases\fP

\fBMSymbol\fP \fBMoffset\fP

Symbol for the offset type method of charset. The symbol Moffset has the name 'offset' and, when used as a value of Mmethod parameter of a charset, it means that the conversion of code-points and character codes of the charset is done by this calculation:

CHARACTER-CODE = CODE-POINT - MIN-CODE + MIN-CHAR

where, MIN-CODE is a value of Mmin_code parameter of the charset, and MIN-CHAR is a value of Mmin_char parameter.

\fBMSymbol\fP \fBMmap\fP

Symbol for the map type method of charset. The symbol Mmap has the name 'map' and, when used as a value of Mmethod parameter of a charset, it means that the conversion of code-points and character codes of the charset is done by map looking up. The map must be given by Mmapfile parameter.

\fBMSymbol\fP \fBMunify\fP

Symbol for the unify type method of charset. The symbol Munify has the name 'unify' and, when used as a value of Mmethod parameter of a charset, it means that the conversion of code-points and character codes of the charset is done by map looking up and offsetting. The map must be given by Mmapfile parameter. For this kind of charset, a unique continuous character code space for all characters is assigned.

If the map has an entry for a code-point, the conversion is done by looking up the map. Otherwise, the conversion is done by this calculation:

CHARACTER-CODE = CODE-POINT - MIN-CODE + LOWEST-CHAR-CODE

where, MIN-CODE is a value of Mmin_code parameter of the charset, and LOWEST-CHAR-CODE is the lowest character code of the assigned code space.

\fBMSymbol\fP \fBMsubset\fP

Symbol for the subset type method of charset. The symbol Msubset has the name 'subset' and, when used as a value of Mmethod parameter of a charset, it means that the charset is a subset of a parent charset. The parent charset must be given by Mparents parameter. The conversion of code-points and character codes of the charset is done conceptually by this calculation:

CHARACTER-CODE = PARENT-CODE (CODE-POINT) + SUBSET-OFFSET

where, PARENT-CODE is a pseudo function that returns a character code of CODE-POINT in the parent charset, and SUBSET-OFFSET is a value given by Msubset_offset parameter.

\fBMSymbol\fP \fBMsuperset\fP

Symbol for the superset type method of charset. The symbol Msuperset has the name 'superset' and, when used as a value of Mmethod parameter of a charset, it means that the charset is a superset of parent charsets. The parent charsets must be given by Mparents parameter.

\fBMSymbol\fP \fBMcharset\fP

The symbol Mcharset. Any decoded M-text has a text property whose key is the predefined symbol Mcharset. The name of Mcharset is 'charset'.

Generated automatically by Doxygen for The m17n Library from the source code.

Copyright (C) 2001 Information-technology Promotion Agency (IPA)

Copyright (C) 2001-2011 National Institute of Advanced Industrial Science and Technology (AIST)

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License <http://www.gnu.org/licenses/fdl.html>.