Menu
▾
▴
binarycloud-dev
|
From: Andreas A. <a.a...@th...> - 2001-08-17 11:20:19
|
I just startet to optimize the code to fit with r2. Mostly convention
changes and connectivity with $gDatasources.
I guess on monday its cleaned up and usable like this:
import('binarycloud.mgr.QueryManager');
$qm = new QueryManager('my_datasrc_hook');
$query = array('QUERY' => 'SELECT foo, bar FROM table....balh');
$values = $qm->DoCustomQuery($query, "", true);
I have to do minor changes to metabase drivers (GetColumnNames(), oci
already supports this). QM:_ProccessResult does need this. The alternative
is to implement the database specific functions that are required directly
into QM, but I don't like this.
Alex, anything specific I have to take care of when r2ifying the code?
Andi
|
|
From: Alex B. <en...@tu...> - 2001-08-17 18:54:24
|
> I just startet to optimize the code to fit with r2. Mostly convention
> changes and connectivity with $gDatasources.
>
> I guess on monday its cleaned up and usable like this:
>
> import('binarycloud.mgr.QueryManager');
> $qm = new QueryManager('my_datasrc_hook');
>
> $query = array('QUERY' => 'SELECT foo, bar FROM table....balh');
> $values = $qm->DoCustomQuery($query, "", true);
>
> I have to do minor changes to metabase drivers (GetColumnNames(), oci
> already supports this). QM:_ProccessResult does need this. The alternative
> is to implement the database specific functions that are required directly
> into QM, but I don't like this.
>
> Alex, anything specific I have to take care of when r2ifying the code?
hi Andreas,
to be honest I have only "cleaned" it - I have not studied the code enough
to give you a real opinion.
I can say that I would prefer that datasources not be a global variable -
because they will contain passwords, sids, etc.
If you look at user/conf/langs.php - you'll see that it sets a cariable in
the lang class: I'd like to have it happen the same way with QueryManager.
-------
Re changes to the metabase drivers:
-John D, JPC? Did you have to do this?
Andreas: we might be using a slightly old version of metabase... let me
check phpclasses and get back to you :)
-a
|
|
From: Andreas A. <a.a...@th...> - 2001-08-17 22:41:34
|
Hi,
> I can say that I would prefer that datasources not be a global variable -
> because they will contain passwords, sids, etc.
Agreed. We dont have to mess up global namespace with this sensitive stuff.
Does Init have to be extended? I thought of anohter setting
<database>datasources_hook</database>
to switch on/off database support. Excatly: to import the metabase stuff and
set up querymanager with the database config stored in datasources.
> If you look at user/conf/langs.php - you'll see that it sets a cariable in
> the lang class: I'd like to have it happen the same way with QueryManager.
No problem. Ok I'll have to change that. So QueryManager will become a
global like $Lang and not a aggregated-as-needed class.
> Andreas: we might be using a slightly old version of metabase... let me
> check phpclasses and get back to you :)
Hmm, I checked the current version at phpclasses.
We need this method in the drivers (e.g. for mysql):
Function GetColumnNames($result)
{
$result_value=intval($result);
if(!IsSet($this->columns[$result_value]))
{
$this->columns[$result_value]=array();
$columns=mysql_num_fields($result);
for($column=0;$column<$columns;$column++)
$this->columns[$result_value][mysql_field_name($result,$column)]=$column;
}
}
But It's missing in mSql and MySql drivers.
Andi
|
|
From: Alex B. <en...@tu...> - 2001-08-17 23:00:55
|
>> I can say that I would prefer that datasources not be a global variable -
>> because they will contain passwords, sids, etc.
> Agreed. We dont have to mess up global namespace with this sensitive stuff.
>
> Does Init have to be extended? I thought of anohter setting
> <database>datasources_hook</database>
> to switch on/off database support. Excatly: to import the metabase stuff and
> set up querymanager with the database config stored in datasources.
oh, yes, you're right.
I'll add a 'database' flag to the init array, and a method to init to import
QueryManager.
>> If you look at user/conf/langs.php - you'll see that it sets a cariable in
>> the lang class: I'd like to have it happen the same way with QueryManager.
> No problem. Ok I'll have to change that. So QueryManager will become a
> global like $Lang and not a aggregated-as-needed class.
Yes... though I am not married to that at the moment, I think it's probably
best.
>> Andreas: we might be using a slightly old version of metabase... let me
>> check phpclasses and get back to you :)
>
> Hmm, I checked the current version at phpclasses.
> We need this method in the drivers (e.g. for mysql):
>
> Function GetColumnNames($result)
> {
> $result_value=intval($result);
> if(!IsSet($this->columns[$result_value]))
> {
> $this->columns[$result_value]=array();
> $columns=mysql_num_fields($result);
> for($column=0;$column<$columns;$column++)
>
> $this->columns[$result_value][mysql_field_name($result,$column)]=$column;
> }
> }
>
> But It's missing in mSql and MySql drivers.
>
> Andi
I've cc'd manuel on this, see what he says :)
-alex
|
|
From: Andreas A. <a.a...@th...> - 2001-08-18 10:41:59
|
Hi Alex,
> I'll add a 'database' flag to the init array, and a method to
> init to import QueryManager.
If you did not already do this. Just wait a bit I've added this already just
bevore starting to lift the qm code ":)
<page>
<init>
<database>true</database>
</init>
<!-- extra
<source>hook</source> // hook in datasources
<mcrpyt>true</mrcypt> // import metabase mcrypt stuff
<manage>false</manage> // import metabase manager classes
</database>
</page>
> > So QueryManager will become a global like $Lang and not a
> > aggregated-as-needed class.
> Yes... though I am not married to that at the moment, I think
> it's probably best.
I also think thats best for now. Querymanager does have the capability to
load stored queries so there is no real need to have more than one instance
of it loaded, I guess. So you define, e.g. in page or entity manager, which
queries you need for this pagerender/operation/ety e.g. in page:
<database>
<queries>
<name_of_query>path.to.query</name_of_query)
</queries>
</database>
or something like that.
[metabase]
> I've cc'd manuel on this, see what he says :)
Cool.
Andi
|
|
From: John D. <jo...@we...> - 2001-08-18 05:37:36
|
On Fri, 17 Aug 2001, Alex Black wrote: > > > > I have to do minor changes to metabase drivers (GetColumnNames(), oci > > already supports this). QM:_ProccessResult does need this. The alternative > > is to implement the database specific functions that are required directly > > into QM, but I don't like this. > > > > Alex, anything specific I have to take care of when r2ifying the code? > > ------- > > Re changes to the metabase drivers: > > -John D, JPC? Did you have to do this? > Yes, we did have to hack metabase a bit. If you need the hacked versions, you know where you can get them (they also have LOB support which wasn't implemented quite nicely enough to be contributed back to Manuel; cleaning up this would also be a nice 2-3 hour project if someone is interested :)), but it'd be much preferable if we could run off the distributed metabase. -- John Donagher Application Engineer, Intacct Corp. Public key available off http://www.keyserver.net Key fingerprint = 4024 DF50 56EE 19A3 258A D628 22DE AD56 EEBE 8DDD |
|
From: Alex B. <en...@tu...> - 2001-08-20 19:32:38
|
can you give me an idea of what kind of hacks besides lob? _a > On Fri, 17 Aug 2001, Alex Black wrote: > >>> >>> I have to do minor changes to metabase drivers (GetColumnNames(), oci >>> already supports this). QM:_ProccessResult does need this. The alternative >>> is to implement the database specific functions that are required directly >>> into QM, but I don't like this. >>> >>> Alex, anything specific I have to take care of when r2ifying the code? >> >> ------- >> >> Re changes to the metabase drivers: >> >> -John D, JPC? Did you have to do this? >> > > Yes, we did have to hack metabase a bit. If you need the hacked versions, you > know where you can get them (they also have LOB support which wasn't > implemented quite nicely enough to be contributed back to Manuel; cleaning up > this would also be a nice 2-3 hour project if someone is interested :)), but > it'd be much preferable if we could run off the distributed metabase. |
|
From: John D. <jo...@we...> - 2001-08-21 00:37:55
|
Sounds like Andi got it working on Oracle without the hacks, so hopefully they aren't necessary.. John On Mon, 20 Aug 2001, Alex Black wrote: > can you give me an idea of what kind of hacks besides lob? > > _a > -- John Donagher Application Engineer, Intacct Corp. Public key available off http://www.keyserver.net Key fingerprint = 4024 DF50 56EE 19A3 258A D628 22DE AD56 EEBE 8DDD |
|
From: Andreas A. <a.a...@th...> - 2001-08-21 09:33:40
|
Hi John, > Sounds like Andi got it working on Oracle without the hacks, so > hopefully they aren't necessary.. Not really. I did not test on oracle. But the methods required by QueryManager are present in the oci driver file. So i guess it works. I've only testet it with MySQL. Andi |
|
From: alex b. <en...@tu...> - 2001-08-21 16:33:09
|
I will test with oracle tomorrow. _a ----- Original Message ----- From: "Andreas Aderhold" <a.a...@th...> To: <bin...@li...> Sent: Tuesday, August 21, 2001 2:12 AM Subject: RE: [binarycloud-dev] QueryManager > Hi John, > > > Sounds like Andi got it working on Oracle without the hacks, so > > hopefully they aren't necessary.. > Not really. I did not test on oracle. But the methods required by > QueryManager are present in the oci driver file. So i guess it works. I've > only testet it with MySQL. > > > Andi > > > _______________________________________________ > binarycloud-dev mailing list > bin...@li... > http://lists.sourceforge.net/lists/listinfo/binarycloud-dev > |
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X