Currently, models can have any possible dataset. Almost all models use
dataset that selects from a single table, but there are models that use a
joined dataset. The most common case for model joined datasets is models
using the class_table_inheritance plugin.
In most cases, it should not be necessary to use a joined dataset.
Instead, the joined dataset can be wrapped in a subquery, and the main
query just selects from the subquery. For example, this model with a joined
dataset:
class Foo < Sequel::Model(DB[:foos].join(:bars,
:id=>:bar_id).select_all(:foos).select_append{bars[:name].as(:bar_name)})
SELECT foos.*, bars.name FROM foos INNER JOIN bars ON (bars.id =
foos.bar_id)
end
can be changed to:
class Foo < Sequel::Model(DB[:foos].join(:bars,
:id=>:bar_id).select_all(:foos).select_append{bars[:name].as(:bar_name)})
SELECT * FROM (SELECT foos.*, bars.name FROM foos INNER JOIN bars ON
(bars.id = foos.bar_id)) AS foos
end
One advantage of this is that you can now use unqualified values when
filtering:
Foo.first(:id=>1)
I recently added support to the class_table_inheritance plugin for an
:alias option that automatically uses a subquery in this manner.
Is there any user who uses joined datasets for models and would not be able
to deal with wrapping them in a subquery as described above? If so, please
describe your situation and why wrapping in a subquery will not work, so I
can decide whether that is something we want to support going forward.
Thanks,
Jeremy
--
You received this message because you are subscribed to the Google Groups
"sequel-talk" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/sequel-talk.
For more options, visit https://groups.google.com/d/optout.
