Re: COBOL Bit Manipulation

[Jon Paris <jon.paris@xxxxxxxxxxxxxx>]

Please note - THIS IS UNTESTED - but syntactically valid <grin>

It may not be quite right but you should be able to see the basic principal in action. You start by dividing by a factor of 2 to shift the bit you want to the least significant position. Division by 2 then pushes that value into the remainder. 

computes could probably be used to simplify the process some but I wanted you to see the mechanics of what was happening.

The alternative is to use the same approach but to repetitively divide by 2 to push the bits one by one into the remainder and then store the bit values in an array.

       Identification Division. Program-Id. BITSET.
       Data Division.
       Working-Storage Section.
       01  x.
           05 bin-value       pic 9(4) comp-4 value zero.
           05 filler redefines bin-value.
              10 filler       pic x.
              10 char-value   pic x.

       77  bit-number         pic 9.
       77  actual-bit         pic 9.
       77  top-bits           pic 9(4).
       77  shift-value        pic 9(4).
       77  left-bits          pic 9(4).
       77  right-bits         pic 9(4).

      * this contains the byte that is bit sensitive
       77  char-bits          pic x.

       Procedure Division.
       Bit-Play.
           move char-bits to char-value.

      * Isolate a specific bit
           If bit-number < 8
              compute shift-value = 2 ** (8 - bit-number)
              divide bin-value by shift-value
                  giving left-bits
                  remainder right-bits
           End-If.       
           divide left-bits by 2
                  giving left-bits
                  remainder actual-bit.

      * To set the bit
           move 1 to actual-bit.
           multiply left-bits by 2 giving bin-value.
           add actual-bit to bin-value.
           
           If bit-number < 8
              multiply bin-value by shift-value
              add right-bits to bin-value
           End-If.
 
Hope this helps - sorry I haven't got time to be sure it works.


Jon Paris

www.Partner400.com
www.SystemiDeveloper.com



On Jan 12, 2011, at 3:32 PM, Jason Abreu wrote:

> I realize that the PIC 1 clause is for display usage only.  It was just 
> the first thing that had come to mind.
>
> I can get this:
>
> 01        WORK.
>   05      CHAR        PIC  X(01)  VALUE X'00'.
>   05      BIN         REDEFINES   CHAR
>                       PIC  9(01)  COMP-3.
>
> But the question is still how to do bit manipulation to set a specific 
> bit to "on" or "off."
>
> I do not know any RPG and, due to in-house coding standards, am looking 
> for a COBOL solution.  Ideally, I would be able to redefine a binary 
> data type into bits and use 88-level items to turn them on or off.
>
> Thanks,
>
> Jason Abreu
> Abreu Innovations, Inc.
> jason.abreu@xxxxxxxxxxxxxxxxxxxx
> http://www.abreuinnovations.com/
>
> On 1/12/2011 3:19 PM, Jon Paris wrote:
> > Well Pic 1 means a single byte indicator so that won't help.
> >
> > It is years since I did this but in the pre-ILE days I did it with a simple divide/multiply routine. Don't have a sample of the code handy but it is not hard to do. Depending on what you want, you can either divide by the appropriate power of 2 to shift the bit in question to the least significant position, then divide by 2 to isolate the bit in question (in the remainder) change that value as you wish and reverse the process.
> >
> > If you want to test/set any arbitrary bit then it is easier to just use a divide by 2 loop and store the remainders in an array. Test/change as desired and rebuild by a repetitive multiply loop.
> >
> > These days I would code it in RPGIV and use the bit oriented functions in RPG IV to do the work. Then call the RPGIV routine form the COBOL code - probably much easier.
> >
> >
> > Jon Paris
> >
> > www.Partner400.com
> > www.SystemiDeveloper.com
> >
> >
> >
> > On Jan 12, 2011, at 2:59 PM, Jason Abreu wrote:
> >
> > > Does anybody have experience with bit manipulation in COBOL?
> > >
> > > I would like to be able to switch individual bits on/off in a single byte.
> > >
> > > I imagined something like:
> > >
> > > 01        WORK.
> > >   05      CHAR-BYTE      PIC X(01)    VALUE X'40'.
> > >   05      BINARY-REP     REDEFINES    CHAR-BYTE.
> > >     10    BIT-1          PIC 1.
> > >       88  BIT-1-ON                    VALUE B'1'.
> > >       88  BIT-1-OFF                   VALUE B'1'.
> > >     ... repeat for all 8 bits.
> > >
> > >
> > > So usage could be something like
> > >
> > > SET BIT-1-ON TO TRUE.
> > > MOVE CHAR-BYTE TO FILE-FIELD.
> > >
> > > Any ideas on this?
> > > Thanks, ahead of time!
> > > --
> > > Jason Abreu
> > > Abreu Innovations, Inc.
> > > jason.abreu@xxxxxxxxxxxxxxxxxxxx
> > > http://www.abreuinnovations.com/
> > > --
> > > This is the COBOL Programming on the iSeries/AS400 (COBOL400-L) mailing list
> > > To post a message email: COBOL400-L@xxxxxxxxxxxx
> > > To subscribe, unsubscribe, or change list options,
> > > visit: http://lists.midrange.com/mailman/listinfo/cobol400-l
> > > or email: COBOL400-L-request@xxxxxxxxxxxx
> > > Before posting, please take a moment to review the archives
> > > at http://archive.midrange.com/cobol400-l.
> -- 
> This is the COBOL Programming on the iSeries/AS400 (COBOL400-L) mailing list
> To post a message email: COBOL400-L@xxxxxxxxxxxx
> To subscribe, unsubscribe, or change list options,
> visit: http://lists.midrange.com/mailman/listinfo/cobol400-l
> or email: COBOL400-L-request@xxxxxxxxxxxx
> Before posting, please take a moment to review the archives
> at http://archive.midrange.com/cobol400-l.