If you have a complicated query table for data entry, such as Select * from TableA where Field1 = :A......FieldN = :N order by Field1 and Field2. OK, you have a icon, report a summary report. Then you will find that you need to create another query for the report, such query may be Select * from TableA where Field1 = :A......FieldN = :N order by Field2 and Field3. Actually, the data of the report is actually the data of the data entry, and you may find the two queries are likely, but the order phase. Such design is OK but the maintainence is hard as we need to maintain two queries. If there any method to obtain data from a dataset for different purpose. One query for data entry, (ordered by keyA and keyB), same query for report (ordered by keyB and keyC).

你可以使用TBDECLientDataSet 如此, 可以隨時改變indexfieldnames

But Clientdataset only sort one column, same Select ..... order by keya while in application, I need Select ..... Order by Keyb, keyc, keyd **** It is because, I need group report.

But Clientdataset only sort one column, 可以多個column clientdataset1.indexfieldnames:='field1;field2'; 甚至若須要多column且有的Ascending, 有的Descending也可稍麻煩些須runTime Create index