Are you sure about that, Sergey? I thought that HAVING was applied after almost all other criteria in the SELECT clause; that is, applied to the preliminary result set and that was the difference between HAVING and WHERE.

>No, only ORDER BY clause is applied to the whole result set. The HAVING clause is applied to each query in UNION. You can use derived table to apply additional conditions
>

>SELECT * FROM (
>	SELECT ;
>	   File.* ;
>	   FROM File ;
>	   WHERE <unique conditions ;
>	UNION SELECT ;
>	   File.* ;
>	   FROM File ;
>	   WHERE <more unique conditions ;
>	UNION SELECT ;
>	   File.* ;
>	   FROM Files ;
>	   WHERE <more unique conditions ;
>	   ) dt1
>WHERE ;
>   < 3 conditions common to the above)
>INTO CURSOR x
>

>
>>
>>Usually I use the "HAVING" clause in a SELECT statement to limit result rows in a situation where I am doing a GROUP BY at the same time (like a count using a GROUP BY).
>>
>>Is it legit - i.e. will I get correct results - to use filter conditions in the HAVING clause to avoid repeating a filter condition in a UNION? i.e. I have a 3 way UNION each of which has a couple of unique conditions but then all 3 of them have 3 more conditions. Instead of repeating these last 3 conditions 3 times, I thought maybe I could put them in as a "HAVING" as I think I remember reading that that works like a WHERE clause and is processed after the UNION is done. BTW, the unique conditions are complex enough I don't want to try to code a complicated set of conditions joined by two "OR"s. By doing it as a UNION, it is much more readable.
>>
>>Just want to confirm. Here is the example:
>>
>>

>>
>>SELECT ;
>>   File.* ;
>>   FROM File ;
>>   WHERE <unique conditions> ;
>>UNION SELECT ;
>>   File.* ;
>>   FROM File ;
>>   WHERE <more unique conditions> ;
>>UNION SELECT ;
>>   File.* ;
>>   FROM Files ;
>>   WHERE <more unique conditions> ;
>>HAVING ;
>>   < 3 conditions common to the above)
>>INTO CURSOR x
>>
>>

>>
>>Thanks,
>>Albert