Matt S Trout
dbix-class at trout.me.uk
Thu Jun 8 02:03:21 CEST 2006
Christopher H. Laco wrote:
> Dan Horne wrote:
>>> On Behalf Of Matt S Trout=
>>> Subject: Re: [Dbix-class] InflateColumn::DateTime
>>> Christopher H. Laco wrote:
>>>> Can someone give me a quick synopsis of how this works.
>>>> It looks like it's always loaded, and only intervenes when it sees a
>>>> column_info data_type of 'datetime'.
>>> It's not always loaded, you have to ask for it. Yeah, all it does is
>>> intervene in the case of a register_column call for a datetime.
>> Actually, this is something that I've been meaning to raise, but I've been
>> beaten to it. Oracle has no concept of a datetime data type, it just has
>> Now, I could just declare date columns as datetime columns in the table
>> class, but this seems to contravene the approach specified in the
>> DBIx::Class::ResultSource doco:
>> "Currently there is no standard set of values for the data_type. Use
>> whatever your database supports."
>> Now I guess that we could provide a method in
>> DBIx::Class::Storage::DBI::<DB> that returns the DB-specific date data type,
>> but will that start adding too much overhead?
> On top of that, I was just looking at DBIx::Class::DateTime::Epoch which
> I think is a more sane approach if you're start a schema from scratch.
No it isn't, because it stops you being able to do date math on your DB.
One of the intentions of this feature is that once it's fully supported
across all the databases we handle it'll make Epoch unnecessary. If the
::Epoch author's about on this list, shout because I'd appreciate your
More information about the Dbix-class