When porting data from Turbo Image to Eloquence on Windows, does Eloquence
change the bit order to “Little Indian” to make Windows programs happy?

Or is it the Eloquence DBGET on Windows that flips’s the bits around?

The HP3000 (or PA-RISC) is referred to as “BIG Indian”. Bits are layout
right-to-left.

Windows (Or Intel) is referred to as “Little Indian”.
Bits are layout left-to-right.

Newer COBOL compilers have a COMP-5 data items that refers to the “Little
Indian” or “BIG Indian” depending on the machine where the program is
running, and COMP is the “BIG Indian bit order” only, correct me if I'm
wrong.

I moved an HP3000 COBOL Program to Windows; I changed all the COMP fields
to COMP-5 fields, compiled (ACU COBOL) and ran it.

The Image data that this program needs was migrated to Eloquence on
Windows and HPUX.

The program is using COMP-5 in the database buffers, why is it (while
running on Windows) correctly reading data from a Big Indian machine as
well as the Little Indian machine. When I change the COMP-5 fields to COMP
it reads computation data incorrectly from the database on both machines.
It must be the Eloquence DBGET that flips’s the bits on the fly.

I think the database is storing data in the big Indian format, on Windows
and HPUX, same as it was on MPE. But is there is a method to
change/control this craziness.

Mike Anderson
832-515.3868

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *