> I need to collect, let's say, 5 fields from the table, but to return
> only unique values from 4 of them. I'll try to be more specific: I
> have fields F1, F2, F3, F4, F5, and F5 is always different (primary
> key). I need to return that field also as a result of a query, but to
> make the rest of the fields unique. Something like:

> SELECT DISTINCT F1,F2,F3,F4,F5 FROM table,

> just that I want F5 field to be excluded from duplicate checking.

> How can I do that ? I'm sure there is an easy
> answer , but it's so far from me at this moment.
> Best regards,
>        Zeljko

> We are born {*filter*}, wet, and hungry.
> Then things got worse.

> Remove "die.spammer" before replying

> Thanks for the reply. Unfortunately, this doesn't suit my needs,
> because I need to return primary key also, as a result of the query.
> Maybe my post was somewhat vague. I'll try to clear it up a bit. If I
> have this situation in the table:

> F1 F2 F3 F4 F5

> A B C D 1
> A B C D 2

> I need query to return the folowing:

> F1 F2 F3 F4 F5

> A B C D 1 <= "1" is the first occurence

> of combination "ABCD".

> Now, when I think of it a bit more, I'm not sure this is possible,
> but, "almost nothing" very well describes my SQL knowledge, so...

> >Hi Zeljko,

> >if you select a unique field in your query you cannot distinct duplicate
> >values in any field, because every record of the table will be in the
> >resulting recordset. If you want to distinct duplicate values you can
group
> >the records with the GROUP BY - command. But you have to include all
fields
> >from your select-statement, that aren't used with aggregate-functions
(SUM,
> >COUNT etc.). ie.

> >SELECT F1,F2,F3,F4, COUNT(F5) FROM table GROUP BY F1,F2,F3,F4;

> >With this query you will get the F1-F2-F3-F4 - combinations only once -
plus
> >the number of records with this combinations.

> >I hope this will help you.
> >--
> >CU  Ronald
> >-> Life would be easier if we had the source code.

> Best regards,
>        Zeljko

> We are born {*filter*}, wet, and hungry.
> Then things got worse.

> Remove "die.spammer" before replying

>SELECT F1, F2, F3, F4, MIN(F5)
>FROM TABLE
>GROUP BY F1, F2, F3, F4

>You will get one row for each F1-F2-F3-F4 combination, with the lowest F5
>value present for each combination. If any F5 values are NULL, they will not
>appear in the result.

>> Hi, Ronald.

>> Thanks for the reply. Unfortunately, this doesn't suit my needs,
>> because I need to return primary key also, as a result of the query.
>> Maybe my post was somewhat vague. I'll try to clear it up a bit. If I
>> have this situation in the table:

>> F1 F2 F3 F4 F5

>> A B C D 1
>> A B C D 2

>> I need query to return the folowing:

>> F1 F2 F3 F4 F5

>> A B C D 1 <= "1" is the first occurence

>> of combination "ABCD".

>> Now, when I think of it a bit more, I'm not sure this is possible,
>> but, "almost nothing" very well describes my SQL knowledge, so...

>> >Hi Zeljko,

>> >if you select a unique field in your query you cannot distinct duplicate
>> >values in any field, because every record of the table will be in the
>> >resulting recordset. If you want to distinct duplicate values you can
>group
>> >the records with the GROUP BY - command. But you have to include all
>fields
>> >from your select-statement, that aren't used with aggregate-functions
>(SUM,
>> >COUNT etc.). ie.

>> >SELECT F1,F2,F3,F4, COUNT(F5) FROM table GROUP BY F1,F2,F3,F4;

>> >With this query you will get the F1-F2-F3-F4 - combinations only once -
>plus
>> >the number of records with this combinations.

>> >I hope this will help you.
>> >--
>> >CU  Ronald
>> >-> Life would be easier if we had the source code.

>> Best regards,
>>        Zeljko

>> We are born {*filter*}, wet, and hungry.
>> Then things got worse.

>> Remove "die.spammer" before replying