Hi,
please note: You can *not* fetch results that require earlier results
using the pg_* functions. Explaination: when you submit a query the php
script doesn't know about the content of all rows returned, instead the
interface API makes use of pointers to fetch only the requested rows
from the result.
That said, if you make a second call to pg_exec() consider saving all
the results somewhere else if you need the previous set of results.
Regards,
Nuno Silva
Luca Lazzeroni wrote:
> Thanks; in fact I've tryed the second solution and it works for me; but I
> can't understand why the script stop working during the upgrade of system.
> The most strange thing is that, after a deeper investigation on the code,
> I've found the PHP engine does something like closing the connection after
> the second pg_exec command.
> In fact, by using persistent connections I've partially corrected the
> problem and the codes doesn't complain about using an invalid (but not null)
> connection descriptor; but it stills be unable to get results from
> successive queries, also if result index are valid ! This sounds very
> strange for me.
>
> Thank you for your help !
>
> Luca Lazzeroni
>
> "Nuno Silva" <[EMAIL PROTECTED]> ha scritto nel messaggio
> [EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
>
>> Hello,
>>
>> if the second query loops inside a _for_ statement and needs values from
>> the first query you have two options:
>>
>> 1. save all the results from query one to some array and then loop
>> inside the array within the _for_ statement executing new sql;
>> 2. open two separate connections to the database and work with the two.
>> like this:
>> $conn1 = pg_connect(...);
>> $conn2 = pg_connect(...);
>> ...
>> qry1 = pg_exec($conn1, $sql1);
>> ...
>> for ($i=.....) {
>> ...
>> $qry2 = pg_exec($conn2, $sql2);
>> ...
>> }
>>
>> Hope this helps,
>> Nuno Silva
>>
>>
>> Luca Lazzeroni wrote:
>>
>>
>>> Hi,
>>> I've recently migrated a system with PHP 4.0.3pl1 and PostgreSQL 7.0.2 -
>>> RedHat Linux 5.1 (yes 5) to a new one with PHP 4.0.4pl1 - PostgreSQL
>>
> 7.0.2 -
>
>>> RedHat Linux 7.0.
>>>
>>> All my database applications work fine except one which does strange
>>
> things:
>
>>> this is a piece of code:
>>>
>>> ----8<------- CUT ----
>>>
>>> $conn = pg_connect("dbname=aaaa user=yyyyy password=xxxxxx");
>>>
>>> $sql = "select * from table_a where condition_a";
>>> $qry1 = pg_exec ($conn,$sql);
>>>
>>> $obj1 = pg_fetchobject($qry1,0);
>>>
>>> // Until here everything works fine
>>>
>>> $sql2 = "select * from table_b where condition_b=".$obj1->value_a;
>>> $qry2 = pg_exec ($conn,$sql2);
>>>
>>> ----8<------- CUT ----
>>>
>>> The second query is nested in a for-loop running for every object in the
>>> first query; for performance reason I need to maintain the two queries
>>> independent.
>>>
>>> After executing (correctly) the first one, the system starts doing
>>
> strange
>
>>> things with the second one:
>>>
>>> Warning: Unable to jump to row 0 on PostgreSQL result index 23 in
>>
> cerca.php3
>
>>> on line 467
>>>
>>> and it fails to read data.
>>> The strange thing is that the system does the same things for EVERY
>>
> database
>
>>> where I must use two consecutive queries !
>>>
>>> Is there anyone who can help me ?
>>>
>>> Thanks in advance
>>>
>>> Luca Lazzeroni
>>
>>
>> --
>> PHP General Mailing List (http://www.php.net/)
>> To unsubscribe, e-mail: [EMAIL PROTECTED]
>> For additional commands, e-mail: [EMAIL PROTECTED]
>> To contact the list administrators, e-mail: [EMAIL PROTECTED]
>>
--
PHP General Mailing List (http://www.php.net/)
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]
