Skip to content

Latest commit

 

History

History
31 lines (22 loc) · 1.52 KB

02_page_controller.md

File metadata and controls

31 lines (22 loc) · 1.52 KB

Page controller

A page controller handles a specific request on the server, by combining model and view and send back to the client the rendered view. Normally, it is bound to a specific URL.

It normally handles one logical page, or one specific action. This makes it extremely simple.

Once it receives the request, it extracts data from this request (e.g. HTTP headers, query parameters, cookies, form contents), performs operations on the model, then select the proper view to perform rendering. A template mechanism is generally used to render model data while separating rendering/presentation logic. It can render a whole page, a subset of it, or data that are parsed by the client-side javascript.

Testing the page controller can be complex due to the need to simulate an HTTP request. Ideally, one would partition non-HTTP related functionality and make them testable without an HTTP context, otherwise the web framework environment must be present and set up.

The page controller often is implemented as a reimplementation of a base class, to provide reusable functionality that is common to all page controllers.

In general, functionalities like session management, authentication, and similar low-level operations are not handled at the page controller level. A different technique is used: Filters (aka Middleware)

FIXME: common functionality between different page controllers may require inheritance. This can force hierarchies that not necessarily are appropriate, and force refactorings as new pages are added.