Defining Czech character set

[greggvdw]

How do I define the Czech (or any odd alphabet) in the special names section. This is in order for it to respond to both Ascii and Ebcdic collating sequences. We have implemented our COBOL based information management system at a Czech company along with our GUI translator. Now the only problem that remains is that whenever a sort is performed, the data being sorted does not correspond with the values assigned by the ASCII and EBCDIC functions.
The mainframe is Unix and the terminals are NT.

 

[StephenJSpiro]

First, The Language must be supported by the compiler. This may require an add-on from the vendor.
Second, The "names" of most alphabet names are also implementor-defined.
Third, Czech ASCII and Czech EBCDIC will, of course, be different alphabets.
Fourth, assuming you have the required alphabets supported on your system, you can specify a different alphabet (from your SPECIAL NAMES) in the COLLATING SEQUENCE phrase of the SORT statement.
Fifth, if worse comes to worst, you can define the alphabet and sort-sequence in SPECIAL NAMES, but you have to do it character by character, and include any odd characters which might show up in the sort fields.

Stephen J Spiro
Member, J4 COBOL Standards Committee
check it out at

http://people.ne.mediaone.net/doncobol/

stephenjspiro at hotmail.com