I'm interested in doing complex joins with RDBO but
without having to map the results to specific Rose::DB::Object
classes.
I found a discussion along these lines from '05.
http://www.gossamer-threads.com/lists/catalyst/users/3095
excerpt:
"For funky SQL in the FROM clause or the column list,
John Siracusa wrote:
> On 6/14/07, Derek Watson <[EMAIL PROTECTED]> wrote:
>> 1. A fatal error is thrown when the object count is zero. I believe
>> this is happening around line 2749 of MakeMethods::Generic:
>>
>> $count =
>> $ft_manager->$ft_count_method(query => [EMAIL PROTECTED], db =>
>>
> "Lars" == Lars Haugseth <[EMAIL PROTECTED]> writes:
Lars> How to pick a separator character that is guaranteed not to appear in
Lars> the column values, though?
"StUfYFuNk". :)
10 points to those who remember where that's from. :)
--
Randal L. Schwartz - Stonehenge Consulting Services, I
On Monday 18 June 2007 17:32:38 Ovid wrote:
> - Original Message
> From: Derek Watson <[EMAIL PROTECTED]>
>
> > > It seems like a simple concat (pg operator ||) would work in this case:
> > >
> > > SELECT COUNT(t1.a_id || t1.b_id)
> >
> > Ugh, I forgot the DISTINCT again:
> >
> > SELECT CO
> "Ovid" == Ovid <[EMAIL PROTECTED]> writes:
Ovid> So what happens when you concatenate the IDs '1', '12' and '11', '2'? I
think you need a separator in there:
Ovid> SELECT COUNT(DISTINCT t1.a_id || '-' || t1.b_id)
We need join(). :)
--
Randal L. Schwartz - Stonehenge Consulting Ser
- Original Message
From: Derek Watson <[EMAIL PROTECTED]>
> > It seems like a simple concat (pg operator ||) would work in this case:
> >
> > SELECT COUNT(t1.a_id || t1.b_id)
>
> Ugh, I forgot the DISTINCT again:
>
> SELECT COUNT(DISTINCT t1.a_id || t1.b_id)
>
> But this is tested and wor
> It seems like a simple concat (pg operator ||) would work in this case:
>
> SELECT COUNT(t1.a_id || t1.b_id)
Ugh, I forgot the DISTINCT again:
SELECT COUNT(DISTINCT t1.a_id || t1.b_id)
But this is tested and works properly too.
-
> What you need is
>
> COUNT(DISTINCT some_function_that_joins_all_args(arg, arg))
>
> because you really want "4,5" to be counted distinctly from "4,6".
> So coalesce definitely doesn't do it.
Right,
It seems like a simple concat (pg operator ||) would work in this case:
SELECT COUNT(t1.a_id
> "John" == John Siracusa <[EMAIL PROTECTED]> writes:
John> You need the DISTINCT part if you join with any ...-to-many
John> relationship. I'm not sure in your example if the DISTINCT should
John> have been there, but I can imagine cases where it will need to be
John> there, so we need some
On 6/18/07, Derek Watson <[EMAIL PROTECTED]> wrote:
> Postgres defines the function separately from count(*):
>
> count(expression) number of input values for which the value of
> expression is not null
>
> In this case I would write
>
> SELECT COUNT(coalesce(t1.a_id, t1.b_id))
>
> Which would coun
> > Which dies because $count is assigned zero, making the statement
> > evaluate false.
>
> Whoops, careless copy/paste. That's fixed in SVN now.
Great, thanks very much!
>
> > 2. If I am counting objects related by a many-to-many where the
> > mapping table has a multi-valued primary key, the
On 6/14/07, Derek Watson <[EMAIL PROTECTED]> wrote:
> 1. A fatal error is thrown when the object count is zero. I believe
> this is happening around line 2749 of MakeMethods::Generic:
>
> $count =
> $ft_manager->$ft_count_method(query => [EMAIL PROTECTED], db =>
>
12 matches
Mail list logo
