Hi Ed,

CONTINUE after an optimized LOCATE uses the work done in the locate. It does not retrieve newly changed records that would be included if the LOCATE was done at the time of the Continue. That's the same as the SCAN. It won't find records that meet the scans criteria but that were changed after the scan started, hence the SCAN NOOPTIMIZE. They both, however, do check the record previously determined to have matched, to see if it still matches the LOCATE or SCAN's conditions.

Charlie

>SCAN FOR processes records in the current sort order, identifying eligible records based on the FOR condition. LOCATE positions to the first record meeting the specified FOR condition. If the FOR condition is Rushmore optimizable, the eligible records can be established on either; if not, the records are examined from the logical BOF a record at a time in the current sort order. So they both perform in a similar fashion - the biggest difference is I'm not sure if CONTINUE has to repeat the check for eligible records.