cfad47cfa3/t3compiler/tads3/include/charset.h

Show revision log
 
4b825dc642cb6eb9a060e54bf8d69288fbee4904cfad47cfa334b206c65f22086bcc5d63e6f70944
1
#charset "us-ascii"



2



3
/*



4
 *   Copyright (c) 2001, 2006 Michael J. Roberts



5
 *   



6
 *   This file is part of TADS 3.



7
 *   



8
 *   This header defines the CharacterSet intrinsic class.  



9
 */



10



11
#ifndef _CHARSET_H_



12
#define _CHARSET_H_



13



14
/* include our base class definition */



15
#include "systype.h"



16



17
/*



18
 *   The CharacterSet intrinsic class provides information on character set



19
 *   translations and can be used to translate between the Unicode character



20
 *   set that the T3 VM uses internally for string values and the local



21
 *   character set or sets used for display, keyboard input, and file I/O.  



22
 */



23
intrinsic class CharacterSet 'character-set/030001': Object



24
{



25
    /*



26
     *   Constructor:



27
     *   



28
     *   new CharacterSet(charsetName) - creates an object to represent the



29
     *   named local character set.  Certain character set names are



30
     *   pre-defined:



31
     *   



32
     *   us-ascii - the plain 7-bit ASCII character set



33
     *.  utf-8 - Unicode UTF-8 (a multi-byte unicode encoding)



34
     *.  utf-16le - little-endian 16-bit Unicode



35
     *.  utf-16be - big-endian 16-bit Unicode



36
     *   



37
     *   In addition, any character set for which the VM has an external



38
     *   mapping file can be used.  Check your platform-specific T3



39
     *   installation notes for infomration on how character set mapping



40
     *   files are implemented on your version of T3.



41
     *   



42
     *   A CharacterSet can be created for a non-existent mapping, but the



43
     *   object cannot be used to perform any mappings; an



44
     *   UnknownCharacterSetException will be thrown if any mapping is



45
     *   attempted with a CharacterSet object that has non-existent local



46
     *   mappings.  You can determine if the local mapping exists with the



47
     *   isMappingKnown method.  



48
     */



49



50
    /*



51
     *   Get the name of the character set.  This simply returns the name



52
     *   that was given to construct the character set. 



53
     */



54
    getName();



55



56
    /*



57
     *   Determine if the mapping is known.  This returns true if the



58
     *   character set has a known local mapping, nil if not.  Note that it



59
     *   doesn't matter whether or not the character set is actually in use



60
     *   on the local platform; all that matters is that a T3 mapping file



61
     *   is available on this machine. 



62
     */



63
    isMappingKnown();



64



65
    /*



66
     *   Determine if a character or string of characters is mappable to this



67
     *   character set.  If the input is an integer, it represents the



68
     *   Unicode character code for a single character; if the input is a



69
     *   string, each character in the string is checked.  This returns true



70
     *   if every character given has a valid mapping in the local character



71
     *   set, nil if not.  Note that if a string is given, and even one



72
     *   character is not mappable, this returns nil.  



73
     */



74
    isMappable(val);



75



76
    /*



77
     *   Determine if a character or string of characters is "round-trip"



78
     *   mappable to this character set.  If the input is an integer, it



79
     *   represents a Unicode character code to be tested; if the input is a



80
     *   string, each character in the string is tested.  Returns true if



81
     *   every character given has a valid round-trip mapping, nil if not.



82
     *   



83
     *   A character has a round-trip mapping if it can be mapped to this



84
     *   local character set and then back to Unicode to yield the original



85
     *   character.  If a character has a round-trip mapping, then in general



86
     *   the character has an exact representation in the local character set



87
     *   (as opposed to an approximation: if 'a-umlaut' maps to a simple



88
     *   unaccented 'a', or to 'ae', then it has only an approximated



89
     *   representation).  



90
     */



91
    isRoundTripMappable(val);



92
}



93



94
#endif /* _CHARSET_H_ */