Character Translation


     GoldED  implements  several   different  proposals  for   character
     translation in imported and exported messages:

        FSC-0050  "Charset Identifier" by Thomas Sundblom.
        FSC-0051  "I51" by Thomas Gradin.
        FSC-0054  "CHARSET proposal" by Duncan McNutt.
        No FSC    "Composed Characters" by Andre van de Wijdeven.

     FSC-0050 is  currently known  to be  implemented in  the OPMED 3.xx
     message editor, and  in Opus 1.7x.  It uses the  same identifier as
     FSC-0054  (a  ^aCHARSET  kludge),  but  is  a  lot simpler (but not
     necessarily better).

     The  "I51"  and  "CHARSET"  proposals  are  in the process of being
     merged  to  one  proposal,  which  should combine the advantages of
     both. They are both based on  using the LATIN-1 (also known as  ISO
     8859-1) character set  for extended ASCII.  The LATIN-1 set  is the
     same  set  used  by  Windows  3.xx,  Amiga  and  many  other non-PC
     computers. In addition to LATIN-1,  I51 defines a set of  so-called
     escape sequences for characters not found in the LATIN-1 set.

     "Composed  Characters"  became  quite  popular  in Holland, but the
     author decided  to drop  his proposal  because it  relied on escape
     sequences  using  the  so  called  "soft-cr" (141d, 8Dh) character.
     GoldED  will  continue  to  support  Composed  as  long as it seems
     necessary.

     If you want to know more about the details, I suggest you read  the
     proposals or contact the authors.

     GoldED currently supports two  types of translation tables,  the *.
     ESC files and the *.CHS files.


                        The ESC translation tables


     The  *.ESC  files  are  used  for  import translation of the escape
     sequences defined in I51 and Composed Characters.

     In the ESC files, the  semicolon is used for comments.  The *first*
     non-comment line defines the charset the escape code are mapped TO.
     This  is  normally  IBMPC,  and  should  not  be changed. Any other
     non-comment line is treated  as an escape sequence  definition with
     this syntax:

          <esc1><esc2><space><map chars>[; comment/description]

     Leading spaces are  *not* allowed in  ESC files. <esc1>  and <esc2>
     are the two characters that define the escape sequence. <space>  is
     ignored and can be used to make the table look better. <map  chars>
     defines  the  local  representation  of  the escape sequence, up to
     three characters. Normally you would only map to one extended ascii
     character. The map chars  can be either the  characters themselves,
     or  decimal  or  hexadecimal  numbers  of  the  form  "\d<dec>"  or
     "\x<hex>" (like in the C programming language).


                          The CHS translation tables


     The *.CHS files are used  for import and export translation  of the
     CHARSET type character sets, and export of I51 and Composed  escape
     sequences.

     The CHS files uses the format of the raw text files provided in the
     CHARSET3.ZIP example implementation of FSC-0054. Study some of  the
     files provided if you want to know how to define them.

     The  two  keywords  XLATESCSET  and  XLATCHARSET are used to define
     which files belong  to what import  and export set.  You can define
     more than one import and export set for each file.

     The keyword XLATIMPORT defines which  charset you have on your  own
     machine  -  this  would  normally  be  "IBMPC". It can be useful to
     change  this  (using  the  Random  System)  in  areas where another
     character set than IBMPC is the dominant (like Amiga or  MacIntosh,
     whatever).

     The keyword XLATEXPORT defines the charset your messages should  be
     exported to, if any.

     Confused?  Yeah,  I  know  -  this  is  a confusing subject, and my
     implementation and documentation is not perfect. Normally you  will
     not have to  worry about it.  Turn it off  completely if you  don't
     understand it.