On Wednesday 10 March 2004 12:25 pm, Randall Skelton wrote:
> Is there a way to return an integer row number for a query? Note
> that there may be a large number of rows so I would rather not have
> joined selects...
Well...if your result has a unique column you can do something like
this:
ste
On Wed, Mar 10, 2004 at 17:25:49 -0500,
Randall Skelton <[EMAIL PROTECTED]> wrote:
>
> That would be ideal as it is theoretically possible for there to be
> missing rows due to sampling errors; nevertheless, a WHERE test doesn't
> seem obvious to me. Can you please post an example? The time
Randall Skelton <[EMAIL PROTECTED]> writes:
>> Couldn't you code this as a WHERE test on the timestamp?
> That would be ideal as it is theoretically possible for there to be
> missing rows due to sampling errors; nevertheless, a WHERE test doesn't
> seem obvious to me. Can you please post an ex
Maybe by using a sequence and a function.
The sequece to generate the row number.
The a function
1) to reset the sequence and
2) to perform a select with the first column nextval(seq) and the column the
timestamp
However, I am a newbie with PostgreSQL and I am not sure it this will work
cor
Randall Skelton <[EMAIL PROTECTED]> writes:
> Correction, I don't want to simply get the nth row, I want all rows
> that are divisible by n. Essentially, the timestamp is at a regular
> interval and I want a way of selecting rows at different sampling
> intervals.
Couldn't you code this as a W
> I to return a sequential row number beginning at 1?
>
> i.e.
>
> row| timestamp
> ---+
> 1 2004-02-01 23:15:00.824-05
> 2 2004-02-01 23:15:01.824-05
> 3 2004-02-01 23:15:02.824-05
> ...
>
> My reason for wanting row numbers is so I can use a 'MOD(row_
Is there a way to return an integer row number for a query? Note that
there may be a large number of rows so I would rather not have joined
selects...
For a rather simple query:
SELECT timestamp from test
WHERE timestamp > '2004-02-02 04:15:00.00 +0'
ANDtimestamp < '2004-02-02 04:15:10.00
