Request handling
Incoming requests to display a page are handled by the filter chain which is the place to start if you want to add your own business logic to Magnolia CMs.
A typical use case is to integrate a new framework with Magnolia. Integration of a new framework is generally accomplished by implementing the following steps:
- Read the request and decide if the framework should handle it. Implement a filter to do this task and place the filter in the proper order in the filter chain. You can save the resulting model to be rendered as an attribute of the request's MgnlContext object as follows:
MgnlContext.setAttribute("xyModel", model)
- Access the the model in your rendering system using JSTL like:
${xyModel.title}
The following diagram shows a typical request processing in Magnolia. The RenderingFilter determines which template to use to render the page. Each template is associated with a template type.
Extension points
These points are where Magnolia request handling can be extended with additional custom business logic:| Area | Object | Interface / Parent Class | Definition | Remarks |
|---|---|---|---|---|
| Filter chain | Filter | AbstractMgnlFilter | config repository at /server/filters | The filters in the filter chain are executed in the configured order in the repository. Objects you store as attributes of the request's MgnlContext object can be used as JSTL variables later. |
| Template | TemplateRenderer | TemplateRenderer | Define a template renderer with a new type in a module's template-renderers section. | Use this new type in the template's type attribute |
| Paragraph | Paragraph Renderer | ParagraphRenderer | Define a paragraph renderer with a new type in a module's paragraphs-renderers section. | Use this new type in the paragraph's type attribute |