Hi,
for some models I found that it’s easier to share the controller new and
update code just to share the view code. At the front of every such
action function I have to lookup if the object already exists, if so
“find” it from the database and update the “attributes=”. If there is no
such object, “new” it with the attributes I’ve got from the form.
After I’ve wrote such code the third time, I came up with this more
general solution:
class ActiveRecord::Base
def Base.find_for_create_modify(key, params)
old_one = find(key) unless key.nil?
if old_one
result = find key
result.attributes= params
else
result = new params
end
[result, old_one]
end
end
But I think, I can’t be the first one, facing this problem. So I wonder
if there might be a more ruby/rails-ish way to accomplish this.
best regards
Torsten
On 5/27/06, Torsten R. [email protected] wrote:
class ActiveRecord::Base
end
Rails mailing list
[email protected]
http://lists.rubyonrails.org/mailman/listinfo/rails
Hi Torsten
I don’t think the pattern you describe should be that common. Creating
a new record and updating an existing record are very different
operations. I don’t think it’s healthy to abstract that difference out
of your code like that.
More concretely, I believe #create and #update should be generally
distinct controller actions. It is possible that action #foo would
create or update, but then I’d do it with explicit code that informed
the surrounding logic whether create or update were performed (can be
easily done with a controller helper). Your method prevents that, and
I don’t think it’s a good idea. You can of course rewrite the Model
method to at least provide that info, but I still think the whole
thing should be dealt with at the Controller domain, not hidden inside
the Model.
Alder G. wrote:
Hi Torsten
Hi Alder,
thanks for replying.
I don’t think the pattern you describe should be that common. Creating
a new record and updating an existing record are very different
operations. I don’t think it’s healthy to abstract that difference out
of your code like that.
Where I see the most common in the implementation of both an update and
a create function is in the view code. So maybe I should try to use the
same view code for two different controller actions.
More concretely, I believe #create and #update should be generally
distinct controller actions. It is possible that action #foo would
create or update, but then I’d do it with explicit code that informed
the surrounding logic whether create or update were performed (can be
easily done with a controller helper). Your method prevents that, and
I don’t think it’s a good idea.
Actually the function returns both the new or updated object and if any,
the value of the old object.
You can of course rewrite the Model
method to at least provide that info, but I still think the whole
thing should be dealt with at the Controller domain, not hidden inside
the Model.
Point taken.
thanks and best regards
Torsten