Interesting. Comments... The COBOL standard does not assign those names you've given. Rather, the proposed followup standard does, except that it does not have those FLOAT-HEX-# usages. And in fact, I'm of the believe that the COBOL 2002 FP usages indicate the following:
1. FLOAT-SHORT COMP-1 2. FLOAT-LONG COMP-2 3. FLOAT-EXTENDED What led you to your belief? Open COBOL has implemented the above equivalences, as has Micro Focus which explicitly states "USAGE FLOAT-SHORT is equivalent to USAGE COMPUTATIONAL-1. USAGE FLOAT-LONG is equivalent to USAGE COMPUTATIONAL-2." 12) The usages float-short, float-long and float-extended define signed numeric data items that are held in a floating-point format suitable for the machine on which the runtime module is to run. The size and permitted range of values for these fields is defined by the implementor. Any value that may be held in a data item of usage float-short shall also be expressible in a data item of usage float-long. Any value that may be held in a data item of usage float-long shall also be expressible in a data item of usage float-extended. Do you also support the following? BINARY-SHORT = COMP PIC S9(4) BINARY-LONG = COMP PIC S9(9) BINARY-EXTENDED = COMP PIC S9(18) >________________________________ > From: Don Higgins <[email protected]> >To: [email protected] >Sent: Friday, July 5, 2013 5:08 AM >Subject: Re: Future of COBOL based on RDz policies was Re: RDz or >RDzEnterprise developers > > >The lack of full floating point support after all this time in IBM COBOL does >seem to indicate a decline in its support. For an example of what full >floating point support in mainframe COBOL looks like, check out the portable >open source zCOBOL which runs on Windows, Linux, and Apple OSX. The compiler >produces readable HLASM assembler source with data field labels which in turn >runs on z390 portable mainframe assembler and emulator written entirely in >open source J2SE java. zCOBOL supports HFP, BFP, and DFP using the 2002 COBOL >standard recommended naming conventions for the 9 different floating point >data types. Here is link to article with example floating point compute >statement using all 9 floating point types: > >http://www.z390.org/zcobol/demo/callcomp/zCOBOL_COMPUTE.pdf > >Don Higgins >[email protected] >www.don-higgins.net > >---------------------------------------------------------------------- >For IBM-MAIN subscribe / signoff / archive access instructions, >send email to [email protected] with the message: INFO IBM-MAIN > > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
