Am 08.02.2016 um 21:17 schrieb Vitaly Burovoy:
> On 2/8/16, Johannes <jotpe@posteo.de> wrote:
>> Am 08.02.2016 um 20:15 schrieb David G. Johnston:
>>> On Mon, Feb 8, 2016 at 12:05 PM, Johannes <jotpe@posteo.de> wrote:
>>>
>>>> Hi,
>>>>
>>>> is there a best practice to share data between two select statements?
>>>>
>>>> Imaging following situation: I want to receive two result sets from two
>>>> tables, referring to a specific id from table t0 AND I try not to query
>>>> for that specific id a second time.
>>>>
>>>> Table t0 returns 1 row and table t1 returns multiple rows.
>>>>
>>>> begin;
>>>> select id, col1, col2, ... from t0 where id = (select max(id) from t0
>>>> where col1 = value1 and col2 = value2 and ...);
>>>> select col1 from t1 where t0_id = (select max(id) from t0 where col1 =
>>>> value1 and col2 = value2 and ...);
>>>> commit;
>>>
>>>
>>> Please confirm:
>>>
>>> You want the result of "SELECT max(id) FROM t0" to be used in the
>>> second
>>> query without having to recompute it?
>>
>> Yes.
>>
>>> What client are you using to execute these statements?
>>
>> JDBC. I execute both statements at once and iterate through the resultsets.
>>
>> Johannes
>
> Hmm. Could you clarify why you don't want to pass id from the first
> query to the second one:
>
> select col1 from t1 where t0_id = value_id_from_the_first_query
>
Of course I could do that, but in that case I would not ask.
I thougt there could be a better solution to execute all statements at
once. Saving roundtrips, increase speed, a more sophistacted solution,
learn something new...
Johannes