django-fluent-pages 1.0b3

Page tree customization

This module provides a page tree, where each node type can be a different model. This allows developers like yourself to structure your site tree as you see fit. For example:

• Build a tree structure of RST pages, by defining a RstPage type.

• Build a tree with widget-based pages, by integrating django-fluent-contents.

• Build a “product page”, which exposes all products as sub nodes.

• Build a tree of a homepage, subsection, and article node, each with custom fields like professional CMSes have.

Each node type can have it’s own custom fields, attributes and rendering.

In case you’re building a custom CMS, this module might just be suited for you, since it provides the tree for you, without bothering with anything else. The actual page contents is defined via page type plugins.

Configuration

Next, create a project which uses the CMS:

cd ..
django-admin.py startproject fluentdemo

To have a standard setup with django-fluent-contents integrated, use:

INSTALLED_APPS += (
    # The CMS apps
    'fluent_pages',

    # Required dependencies
    'mptt',
    'parler',
    'polymorphic',
    'polymorphic_tree',
    'slug_preview',

    # Optional widget pages via django-fluent-contents
    'fluent_pages.pagetypes.fluentpage',
    'fluent_contents',
    'fluent_contents.plugins.text',
    'django_wysiwyg',

    # Optional other CMS page types
    'fluent_pages.pagetypes.redirectnode',

    # enable the admin
    'django.contrib.admin',
)

DJANGO_WYSIWYG_FLAVOR = "yui_advanced"

Note each CMS application is optional. Only fluent_pages and mptt are required. The remaining apps add additional functionality to the system.

In urls.py:

urlpatterns += patterns('',
    url(r'', include('fluent_pages.urls'))
)

The database can be created afterwards:

./manage.py syncdb
./manage.py runserver

Custom page types

The key feature of this module is the support for custom node types. Take a look in the existing types at fluent_pages.pagetypes to see how it’s being done.

It boils down to creating a package with 2 files:

The models.py file should define the custom node type, and any fields it has:

from django.db import models
from django.utils.translation import ugettext_lazy as _
from fluent_pages.models import HtmlPage
from mysite.settings import RST_TEMPLATE_CHOICES


class RstPage(HtmlPage):
    """
    A page that renders RST code.
    """
    rst_content = models.TextField(_("RST contents"))
    template = models.CharField(_("Template"), max_length=200, choices=RST_TEMPLATE_CHOICES)

    class Meta:
        verbose_name = _("RST page")
        verbose_name_plural = _("RST pages")

A page_type_plugins.py file that defines the metadata, and rendering:

from fluent_pages.extensions import PageTypePlugin, page_type_pool
from .models import RstPage


@page_type_pool.register
class RstPagePlugin(PageTypePlugin):
    model = RstPage
    sort_priority = 10

    def get_render_template(self, request, rstpage, **kwargs):
        return rstpage.template

A template could look like:

{% extends "base.html" %}
{% load markup %}

{% block headtitle %}{{ page.title }}{% endblock %}

{% block main %}
  <h1>{{ page.title }}</h1>

  <div id="content">
    {{ page.rst_content|restructuredtext }}
  </div>
{% endblock %}

Et, voila: with very little code a custom CMS was just created.

Optionally, a model_admin can also be defined, to have custom field layouts or extra functionality in the edit or delete page.

Adding pages to the sitemap

Optionally, the pages can be included in the sitemap. Add the following in urls.py:

from fluent_pages.sitemaps import PageSitemap

sitemaps = {
    'pages': PageSitemap,
}

urlpatterns += patterns('',
    url(r'^sitemap.xml$', 'django.contrib.sitemaps.views.sitemap', {'sitemaps': sitemaps}),
)

Contributing

This module is designed to be generic. In case there is anything you didn’t like about it, or think it’s not flexible enough, please let us know. We’d love to improve it!

If you have any other valuable contribution, suggestion or idea, please let us know as well because we will look into it. Pull requests are welcome too. :-)