#780 Map codepages automatically

[Jan-Erik Lärka]

ooRexx use the codepage currently active in Windows commandline that it run under while the Windows GUI run under a different codepage. English speaking countries use characters a-z while other contries often have a need for extended national alphabets.

As ooRexx call OS functions such as SysFileTree or connect to applications with ooSQLite, OLE, Java, OpenOffice etc. things get tricky as the ooRexx script developer has to convert string back and forth.

The Windows version would benefit if conversion can be done transparently internally for supporting functions or provide a tokenizer translation mechanism for scripts written in one codepage to run in another codepage as defined in the national setting for the active user.

[Erich]

Hi Jan-Erik,
ooRexx has no concept of codepages - it uses code points without assigning any meaning to them.

To make an ooRexx script run under codepage 1252, issue a CHCP 1252 command before starting the script.

scripts written in one codepage to run in another codepage
Rexx scripts are plain text files which don't know what codepage they are written in

To convert a string between Windows codepages, you can use something like below (where cpIn/cpOut are the codepages to convert from/to)

call SysToUnicode string, cpIn, , out.
call SysFromUnicode out.!text, cpOut, , , out.
return out.!text