Is it possible to filter duplicates codes by filtering with diferent
dates???

I have a table from where i load into my tables, that table has not
got PK on cod1. When i try to load into my table1 found that there is
a lot of duplicates.

So i am trying to group then as:

if date1 is not null then return date1 as data1
elseif data2 is not null then return data2 as data2
elseif data3 is not null then return data3 as data3

select distinct cod1, max (data1)
from table
where date1 is not null
group by grouping sets (cod1 , date1 )

this query returns some blanks that i dont know how to manage.

Anyone has any idea ??

Thanks in advance ..
A.

COD1 DATE1

2006003183015 26/02/2007
23/02/2007
26/02/2007

COD1 DATE1

2006003183015 26/02/2007
23/02/2007
26/02/2007

COD1 DATE1

2006003183015 26/02/2007
23/02/2007
26/02/2007

On May 23, 7:16 pm, valigula <valig... (AT) gmail (DOT) com> wrote:
Just add the primary key and handle exceptions and be done with it.
Constraints are The Good Thing(tm), I don't really understand why
people try to avoid them.

Regards,
Vladimir M. Zakharychev
N-Networks, makers of Dynamic PSP(tm)
http://www.dynamicpsp.com

On May 23, 10:37 am, valigula <valig... (AT) gmail (DOT) com> wrote:
Is there some VALID reason you persist in inundating us with this same
message??? You've asked your question and provided this followup 3
times now. This is the surest way to get yourself, and your
questions, ignored.

It is painfully obvious that COD1 contains NULL values; since we don't
know what sort of data this is, or what it represents, it's your task
to determine how to replace these NULLs with usable values.

Stop posting your output to the group; we have three copies of it and
that is two copies too many.


David Fitzjarrell

On May 23, 10:37 am, valigula <valig... (AT) gmail (DOT) com> wrote:
Is there some VALID reason you persist in inundating us with this same
message??? You've asked your question and provided this followup 3
times now. This is the surest way to get yourself, and your
questions, ignored.

It is painfully obvious that COD1 contains NULL values; since we don't
know what sort of data this is, or what it represents, it's your task
to determine how to replace these NULLs with usable values.

Stop posting your output to the group; we have three copies of it and
that is two copies too many.


David Fitzjarrell

On May 23, 10:37 am, valigula <valig... (AT) gmail (DOT) com> wrote:
Is there some VALID reason you persist in inundating us with this same
message??? You've asked your question and provided this followup 3
times now. This is the surest way to get yourself, and your
questions, ignored.

It is painfully obvious that COD1 contains NULL values; since we don't
know what sort of data this is, or what it represents, it's your task
to determine how to replace these NULLs with usable values.

Stop posting your output to the group; we have three copies of it and
that is two copies too many.


David Fitzjarrell

On May 23, 10:37 am, valigula <valig... (AT) gmail (DOT) com> wrote:
Is there some VALID reason you persist in inundating us with this same
message??? You've asked your question and provided this followup 3
times now. This is the surest way to get yourself, and your
questions, ignored.

It is painfully obvious that COD1 contains NULL values; since we don't
know what sort of data this is, or what it represents, it's your task
to determine how to replace these NULLs with usable values.

Stop posting your output to the group; we have three copies of it and
that is two copies too many.


David Fitzjarrell

On May 23, 10:37 am, valigula <valig... (AT) gmail (DOT) com> wrote:
Is there some VALID reason you persist in inundating us with this same
message??? You've asked your question and provided this followup 3
times now. This is the surest way to get yourself, and your
questions, ignored.

It is painfully obvious that COD1 contains NULL values; since we don't
know what sort of data this is, or what it represents, it's your task
to determine how to replace these NULLs with usable values.

Stop posting your output to the group; we have three copies of it and
that is two copies too many.


David Fitzjarrell