PM03792: NUM source fields used on the vglib.concatenateBytes function are showing incorrect value in the SIGN position

APAR status

• Closed as program error.

Error description

• Rational Business Developer V7.5.1.4:
  
  When running EGL generated COBOL programs and a NUM field is
  used as the source on the vglib.concatenateBytes function, an
  incorrect value is in the SIGN position.
  
  The problem is that when the following statement is generated:
  
    TSTWS.RESULT =
     VGLib.concatenateBytes(TSTWS.WHEREFLD,
       TSTWS.CNTRL-NO-ARRAY[TSTWS.CNTRL-NO-INDEX])
  
  It moves the numeric field into a temporary pic s9(13) field
  named EZELFV-ACE-69 and uses this as the source on the
  concatenation.  When this "compute" statement is done, COBOL
  will change the last byte from F0 to C0.
  
    COMPUTE EZELFV-ACE-69 = CNTRL-NO-ARRAY-8 IN CONTROLNO-WORK-7
      IN TSTWS-15 ( CNTRL-NO-INDEX-11 IN CONTROLNO-WORK-7 IN
        TSTWS-15 )
  
  The workaround is to use a statement like this:
  
   tstws.wherefld = clip(tstws.wherefld) +
     TSTWS.CNTRL-NO-ARRAY[TSTCNCTWS.CNTRL-NO-INDEX];
  
  This also provides better performance.
  

Local fix

• A testfix is available from support.
  

Problem summary

• Certain temporary variables were being skipped when deciding
  to set the sign bits for numerics.
  

Problem conclusion

• Software updated.
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

Fix information

• Fixed component name

  RBD EGL FOR COB

• Fixed component ID

  5724S5001

Applicable component levels

• R751 PSN

     UP