I'v enever seen this one before so any help would be greatly appreciated.

 SQL
 SELECT DISTINCT
 ;
 TABLE
 ON TABLE HOLD AS ELIG_DET1
 END
 -RUN
 DEFINES MAY BE SET TO OLD OR COMPILED
 DEFINES MAY BE SET TO OLD OR COMPILED
 1
 0 NUMBER OF RECORDS IN TABLE=       73  LINES=     73
 0
 1
 0 NUMBER OF RECORDS IN TABLE=       76  LINES=     76
 0
 1
 0 NUMBER OF RECORDS IN TABLE=       76  LINES=     76
 0
 1
 0 NUMBER OF RECORDS IN TABLE=       76  LINES=     76
 0
 1
 0 NUMBER OF RECORDS IN TABLE=      140  LINES=    140
 0
 1
 0 NUMBER OF RECORDS IN TABLE=      140  LINES=    140
 0
 (FOC010) THE NUMBER OF SORT FIELDS EXCEEDS THE MAXIMUM: SQLHLD06.E33
 (FOC009) INCOMPLETE REQUEST STATEMENT
 BYPASSING TO END OF COMMAND
 

Any ideas?

This message has been edited. Last edited by: rfbowley,

The E33 part is the FOCUS alias for the 33rd data column.

Processing of "DISTINCT" requires sorting. The limit (on the number of sort keys in one segment of the internal matrix) is 32.

Since your WHERE requires SUB_CNCL_DT, TYP_COV_FL, and IDE.CASE_STAT_CD to have specified values, it is pointless to include them in the list.


- - - - -

I suggest you identify a set of key fields that guarantee uniqueness of the remaining columns' values, and use a request of the form

JOIN ...

TABLE FILE ...
SUM
FST.X
FST.Y ...
BY A
BY B
...
WHERE ...


(where A,B,... are the keys [presumably numbering well below the limit of 32] and X,Y,... the dependent fields).

This message has been edited. Last edited by: j.gross,

At the Requst of Management, I have deleted the actual SQL code that contains 'identifiable' code from the original request.

The problem was, in a SQL report, there were a significant number of tables involved, with multi field joins, and a the SELECT DISTINCT cause the error. Changing to a simple SELECT solved the problem.

Many thanks to j.gross for the assit.