ZF-5840: [2.0] Remove action-view-helper

Description

The action view helper is mostly used totally wrong (I never knew any actual use-case for it). Instead of using view helpers or models, people start to use actions as reusable units. If there's actually a use-case for the action view helper, it should be mentioned in the documentation, for what it actually is for, else I'd suggest to remove it (in 2.0, to keep BC before), and probably yet mark it as deprecated (see ZF-8170).

Comments

Though it fits perfectly in the use-at-will spirit of ZF, I do think that it costs way too much resources (as I recall right, it does an entire dispatch..), and asks for bad design. Therefore, let's get rid of it :D

I agree. We should still be able to use that helper if we want to. Anyway, regarding the performance problem, it is specified in the official doc, in the performance chapter, so anyone should know about it.

I decided to use it for a "tabbed menu". Each of the tabs was originally developed as its own action, but then I wrote another page that just calls the individual actions.

Uhm, do you know Zend_Navigation, Corey?

That still doesn't actually render the content of all the tabs to the HTML (which is what I intend to do - using jQuery to tabSwitch) I suppose a partial view helper would work as well, but some of these actions build up forms, etc. for editing - Perhaps the Action is not the place to build the form?

It probably isn't. Would you mind moving this discussion to a mailinglist, forum, or irc?

In my opinion, Dolf is right. Move this discussion to the mailinglist.

Personaly i avoid using the action view helper, because of the very big dispatch overhead. Better remove it and talk about a smaller and faster solution. Just my two cents.

There are a lot of discussions in Blogs, Forums and Chats with the topic "reusable content" or "widget problem". Every discussion leads to a "new" solution. I'am wondering a lot that the manual is missing this important topic.

Hopefully we will see a clear solution in 2.0. :-)

Is there any reason to keep this issue open?

I'll take that as a no ;)