Mike,

The problem lies in using this column name in a view where the underlying items 
in the SELECT clause return a different data type.

CREATE VIEW viewname (columnName1...columnNameX) AS SELECT ...

  SELECT SYS_TABLE_ID,SYS_COLUMN_ID,SYS_COLUMN_NAME,SYS_NULLABLE,+
    SYS_DATA_TYPE,((IFLT(SYS_DATA_TYPE,0,(CHAR(61)), +
    NULL))+(IFEQ(SGET('00200000102222100',1, +
    ABS(SYS_DATA_TYPE)),'0',SYS_TYPE_NAME,(IFEQ(SGET('00200000102222100',1, +
    ABS(SYS_DATA_TYPE)),'1',(SYS_TYPE_NAME&CHAR(40)+CTXT(SYS_PRECISION)+ +
    CHAR(44)+CTXT(SYS_SCALE)+CHAR(41)),(IFGT(SYS_LENGTH,0,+
    (SYS_TYPE_NAME&CHAR(40)+CTXT(SYS_LENGTH)+ +
    CHAR(41)),'LONG'& SYS_TYPE_NAME))))))), +
    (IFLT(SYS_DATA_TYPE,0,NULL,SYS_COMMENT)), +
    (IFLT(SYS_DATA_TYPE,0,SYS_COMMENT,NULL)) FROM SYS_COLUMNS +
   WHERE SYS_COLUMN_NAME='BP1'

The above SELECT statement may take a long time to process. When you identify 
the SYS_TABLE_ID or id's 


SELECT SYS_TABLE_NAME FROM SYS_TABLES WHERE SYS_TABLE_ID IN (list of 
SYS_TABLE_ID's)



 
Jim Bentley,
American Celiac Society
1-504-737-3293


________________________________
 From: Michael J. Sinclair <[email protected]>
To: RBASE-L Mailing List <[email protected]> 
Sent: Monday, March 11, 2013 10:48 PM
Subject: [RBASE-L] - Problems with unload ALL
 

Hi,

I am trying to clean up my database (Rbase 9.5 32)
When I try to run this command...

UNLOAD ALL

I get this error message...

-ERROR- Column BP1 exists as both a computed and non-computed column. (195)

But when I try to find it by doing this...

LIST column BP1

I get this....

 Column Name        Table Name         Attributes
 ------------------ ------------------ ---------------------------------------
 BP1                VITALS0            Type   : TEXT 12
                                       Compute: 
((CTXT(SYSTOLIC1))+'/'+(CTXT(DIASTOLIC1)))
                    vitals1            Type   : TEXT 12
                                       Compute: 
((CTXT(SYSTOLIC1))+'/'+(CTXT(DIASTOLIC1)))

So how do I find the non-computed column??

Mike

Reply via email to