In essence, I agree with what you're saying...
in my base class i have:
relationships (
privacy_override__subordinates=> {
class=> 'Overrides',
key_columns=> { id=> 'id__viewed_by' },
type=> 'one to one',
},
privacy_override_
On 7/29/06 7:17 PM, Kurt Hansen wrote:
> $b1 = Basket->new(...);
> $p1 = Person->new(...);
> $b1->persons($p1);
> $b1->save(); # "I'd enter one entry"
>
> $b2 = Basket->new(...);
> $p2 = Person->new(...);
> $b2->persons($p2);
> $b2->save(); # "I entered another entry"
>
>
John Siracusa wrote:
>
>On 7/29/06 3:53 PM, Kurt Hansen wrote:
>
>
>>package Basket;
>>[...]
>>relationships => [
>> persons => {
>> type => 'one to many',
>> class => 'Person',
>> column_map => { Item => 'Item' },
>>},
>>[...]
>>I'd ent
>> I found a bug a couple weeks back where I had a DB structure that had
>> this odd table hanging around where the only field was the primary
>> key, which was also a foreign key. I found at the time that you could
>> create a new() on that object, which would save() fine. But if you
>> ever wante
On 7/29/06 4:21 PM, Jonathan Vanasco wrote:
> what we should have asked is for the planner to conditionally join
> the table -- so we get the null values if the condition isn't met
Ah, gotcha.
>> If a were to add the ability to augment the join conditions in a Manager
>> query, what would the sy
On Jul 29, 2006, at 10:05 AM, John Siracusa wrote:
> Join conditions are determined by the relationship or fk metadata,
> and they
> do not currently support nested boolean logic. But it shouldn't
> matter
> whether those conditions are in the ON (...) clause or in the WHERE
> clause,
> sho
On 7/29/06 4:19 PM, John Siracusa wrote:
> For more information on the methods created for each relationship, look at the
> "method map" section of the relationship docs and follow the links to the
> descriptions of the actual methods:
Oops, hit send by accident. Here's the link:
http://search.c
On 7/29/06 3:53 PM, Kurt Hansen wrote:
> package Basket;
> [...]
> relationships => [
>persons => {
> type => 'one to many',
> class => 'Person',
> column_map => { Item => 'Item' },
> },
> [...]
> I'd enter one entry with a transactio
Hello,
I just solved a problem that was most likely due to my lack of
understanding of RDBO. I just wanted to pass it along in case other
newbies run into the same issue.
Breaking the problem down to its simplest form, I have two tables in a
one-to many relationship. The problem was that subse
On Jul 29, 2006, at 9:44 AM, John Siracusa wrote:
> A what now? Data structure? These are objects! No poking around
> as if
> they were mere hashes! :) Methods are the approved way to modify
> the tree
> of objects. Fiddling around with the undocumented internals will
> only lead
> to
On 7/28/06 9:02 AM, Ritesh Errno wrote:
> I am following the canoical method of auto init like this -
>
> __PACKAGE__->meta->table("footable");
> __PACKAGE__->meta->auto_initialize;
>
> But it does not pick up the foreign keys, since the table -> class mapping
> does not follow the convention.
>
On 7/28/06 8:43 PM, Jonathan Vanasco wrote:
> My code below creates SQL that is essentially:
> SELECT t1.* , t2.* FROM useraccount_link t1 LEFT OUTER JOIN
> useraccount_link_privacysetting_override t2 ON(t1.id =
> t2.useraccount_link_id)
> WHERE
> t1.is_deleted = false AND
> t1.useraccount_id = 12
First Rose DB rocks!
I am following the canoical method of auto init like this -
__PACKAGE__->meta->table("footable");
__PACKAGE__->meta->auto_initialize;
But it does not pick up the foreign keys, since the table -> class mapping does not follow the convention. I tried adding:
__PACKAGE__
On 7/28/06 5:36 PM, Jonathan Vanasco wrote:
> the issue is because:
>
> privacy_override__subordinates == the object name, on the table
> user_link_privacy_override
>
> with_objects takes the arguments of the relationship object
> query builders take the arguments of the table name
The query bui
On 7/28/06 2:29 PM, Randal L. Schwartz wrote:
> So I have $client who has abstract table B which is implemented as B1, B2, B3,
> and B4, all inheriting B. There are no B rows. However, there's table A
> which asks for a "B" as a foreign key reference.
If there are no B rows, how does that work i
On 7/28/06 1:55 PM, Jonathan Vanasco wrote:
> if i load something with a with_objects, and get a structure like this:
A what now? Data structure? These are objects! No poking around as if
they were mere hashes! :) Methods are the approved way to modify the tree
of objects. Fiddling around wit
On 7/28/06 9:59 AM, Danial Pearce wrote:
> I found a bug a couple weeks back where I had a DB structure that had
> this odd table hanging around where the only field was the primary
> key, which was also a foreign key. I found at the time that you could
> create a new() on that object, which would
On 7/28/06 5:34 AM, Danial Pearce wrote:
> If you use the "with" support in the load() method, you get no SQL
> when you turn $Rose::DB::Object::Debug to 1. I tried to write a patch
> for you, but i got lost :-)
That's because it's really the Manager doing the loading for you in that
case. Try tu
On 7/26/06 2:16 PM, Wiggins d'Anconia wrote:
> And I can call the get_objects* methods directly but was wondering why
> the 'object_class' argument must be passed each time, if I have already
> set a 'sub object_class' with that data in the Manager class that I am
> using.
Yeah, that sounds reason
This is killing me..
My code below creates SQL that is essentially:
SELECT t1.* , t2.* FROM useraccount_link t1 LEFT OUTER JOIN
useraccount_link_privacysetting_override t2 ON(t1.id =
t2.useraccount_link_id)
WHERE
t1.is_deleted = false AND
t1.use
If you use the "with" support in the load() method, you get no SQL
when you turn $Rose::DB::Object::Debug to 1. I tried to write a patch
for you, but i got lost :-)
Also, is there any sort of "cascade" support for save()? I notice
there is support in delete(). If there isn't, would the best
replying to myself:
the issue is because:
privacy_override__subordinates == the object name, on the table
user_link_privacy_override
with_objects takes the arguments of the relationship object
query builders take the arguments of the table name
or=> [
New to Rose:: so please be gentle.
I have run into the same problem with needing to be able to call the
manager methods without knowing their names ahead of time. See recent
post here:
http://sourceforge.net/mailarchive/forum.php?thread_id=11198586&forum_id=46239
And I can call the get_objects*
I found a bug a couple weeks back where I had a DB structure that had
this odd table hanging around where the only field was the primary
key, which was also a foreign key. I found at the time that you could
create a new() on that object, which would save() fine. But if you
ever wanted to ca
24 matches
Mail list logo
