I am of course aware that the use of 'printable', engineering-notation
representations of floating-point values [and indeed of other data types
too] makes communication between Java and COBOL easy, at least for most
'reasonable', frequently encountered engineering-notation values.
What is involved here is one [or both] of the conversions
<internal HFP> ==> <engineering notation> ==> <internal BFP>
or
<internal BFP> ==> <engineering notation> ==> <internal HFP>
preceded and followed by the usual bullet-proofing in both cases.
Moreover, I have for my sins wrestled with ISV products that use such
schemes on behalf of clients, and I have (a) found their performance not
just unsatisfactory but radically so and (b) identified these conversions as
the villains of the piece.
I am reminded of a book called "Topology made easy", which accomplished its
objective by leaving the hard parts out.
John Gilmore
Ashland, MA 01721-1817
USA
_________________________________________________________________
FREE pop-up blocking with the new MSN Toolbar get it now!
http://toolbar.msn.click-url.com/go/onm00200415ave/direct/01/
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
