>Hi everybody,
>
>Can you please suggest an efficient algorithm of re-numbering? I have a table with the sequence number field (per each unique Fk). The sequence_number is supposed to be correct 1,2,3, etc., but we have data problem and some numbers are repeated.
>
>How can I re-number my records?
>
>Thanks in advance.

Before, fix duplicate foreignKey manually ... then

SELECT thetable
COUNT TO maxId
SET ORDER TO tag_sequenceField DESC
LOCATE
IF sequenceField > m.maxId
     maxId = sequenceField
ENDIF
* BEGIN TRAN
SCAN
  IF m.maxId <> sequenceField
       sequenceField = sequenceField
       replace sequenceField with m.maxId
       * update ... SET foreignkey = thetable.sequenceField WHERE  foreignkey = m.sequenceField
  ENDIF
  maxId = m.maxId - 1
ENDSCAN
* END TRAN
* if you want start with 1, now you can reprocess 
* in ascending and replace sequenceField 
SET ORDER TO tag_sequenceField ASC
SCAN
  IF sequenceField <> RECNO()
       sequenceField = sequenceField
       replace sequenceField with RECNO()
       * update ... SET foreignkey = thetable.sequenceField WHERE  foreignkey = m.sequenceField
  ENDIF
ENDSCAN