Django Datatables

The django datatables library makes creating tables that make use of the datatables library simple, reusable, pythonic, djangoesque, and quite a bit fun.

Project Goals

• Allow creation of tables in a style similar to django forms.
• Remove tedious editing of datatables javascript config to match columns.
• Configure ajax URLs automagically.
• Simplify use of Django style URLs within the datatable

Installation

pip install classy-django-datatables

Quick Setup

Download the library and place it somewhere accessable in your PYTHONPATH. The following is a basic example to demonstrate the ease to get up and running.

settings.py

Add django_datables to the INSTALLED_APPS setting.

INSTALLED_APPS = [
    ...
    'django_datatables',
    ...
]

urls.py

Add the following line to urls.py.

    url(r'^__django_datatables__/', include('django_datatables.urls')),

views.py

    from django_datatables.datatable import *
    from django_datatables import column

    class StudyListDatatable(Datatable):
        code_name = column.TextColumn()

        class Meta:
            model = Study

    def study_list(request)
        datatable = StudyListDatatable()
        return render(request, 'datatables_demo.html',
            {"datatable": datatable}
        )

Template

    {{datatable.render}}

Building Up

Custom Title/Value

In the example shown, the code_name, as the variable name is automatically used to fetch the field and then used as the header for the column. There will often be cases where the variable name will not coincide to either of these and can be overritten with the following:

    created_date = column.TextColumn(title='Made on')
    scientist = column.TextColumn(value='scientist__scientist_name')

CSS Class

A css class to apply to each cell in the column.

    scientist = column.TextColumn(css_class='text-danger')

Joined tables

Fields in joined tables are accessed using the same syntax used in django.

    scientist = column.TextColumn(value='scientist__scientist_name')

Adding links

Links support django's URL dispatcher. Just add the URL name to the link attribute and the arguments that get passed to the link. You don't even need the column listed -- Django Datables will integentally fetch the needed field and populate the links accordingly.

    code_name = column.TextColumn(link='edit_study', link_args=['slug'])

Data from multiple fields

To pull data from multiple fields into one column, declare the column as a StringColumn. If needed, add the fields to be requested from the database in the Meta.extra_fields attribute. Finally, render the desired with the render_* method.

    name = column.StringColumn()
    class Meta:
        model = Employee
        extra_fields = ('first_name', 'last_name')
    def render_name(self, row):
        return "{} {}".format(row['first_name'], row['last_name']).strip()

Other Querysets

The initial queryset can be overridden if a more complex query is needed, or if a default filter needs to be in place.

   def get_initial_queryset(self, request):
        return Employee.objects.filter(manager=request.user)

Meta

model: the primary model to be displayed in the table

    model = Study

order_columns: a list of the columns that can be sorted

    order_columns = ['study_name', 'created_date', 'modified_date', 'scientist']

initial_order: the inital sort of the table

    initial_order = ['created_date', 'scientist']

searching: (default: false) Enable the search box

search_fields the fields that the search box will search for content. This can be more finely controlled in the filter_by_search() method.

    search_fields = ['study_name', 'code_name', 'scientist__scientist_name']

title: The title of the report. Only used for the filename and sheet name of the excel export.

    title = "Study List"

export_to_excel: If openpyxl is installed and set to true, will display a link to download an excel file containing all rows in the table.

    export_to_excel = True

Custom rendering

Any field can have it's render method extended using render_*

    def render_code_name(self, value):
        return value.lower()

    def render_created_date(self, value):
        return value.strftime("%m/%d/%Y")

Columns

Attributes

• title - Displayed in the header
• css_class - A CSS class to apply to the column
• value - The value in the database to use
• link - The django url name this column will link to
• link_args - the link arguments

The following column types are available in the django_datatables.column module.

TextColumn: A standard column that will display the contents of a single field.

ConstantTextColumn(text): Will display text independant of the database. Ex: Edit, or Delete

StringColumn: A column that will render text using multiple fields. Request the data with Meta.extra_fields and format the text with the render_* method.

CheckBoxColumn: Render a checkbox.

GlyphiconColumn(icon): Displayan icon from bootstrap's v3 glyphicon set.

FontAwesome4(icon): Display an icon from the Font Awesome 4 library. Ex: column.FontAwesome4Column('stop-circle fa-2x') (Must manually include bootstrap in source.)

DateColumn: Render a date in Y-m-d format.

Filters

Filter forms can be connected to the datatable by assigning a django form to Meta.filter_form. Naming the fields as django queryset keys (eg: name__icontains, count__gte) will auto filter the form as needed.

Filter example

class EmployeeFilterForm(forms.Form):
    last_name__icontains = forms.CharField(label="Last Name", required=False)


class StudyListDatatable(datatable.Datatable):

    name = column.StringColumn()
    class Meta:
        model = Employee
        filter_form = EmployeeFilterForm
        extra_fields = ('first_name', 'last_name')
    def render_name(self, row):
        return "{} {}".format(row['first_name'], row['last_name']).strip()

In the template, the form can be displayed with the following. There /must/ be a .datatable-form class attached to the form.

    <form class='datatable-form'>
        {{datatable.filter_form}}
        <button type="submit">Submit</button>
    </form>

Permissions

It's important to not just lock down the view, but also the ajax call that retrieves the data. Fortunately, authentication is easily handled with mixins. Django 1.9 ships with LoginRequiredMixin, UserPassesTestMixin, and PermissionRequiredMixin which handle most use cases. Ensure the permission-related mixins are stated first.

More information regarding mixins can be found at the official django authentication doc.

    from django.contrib.auth.mixins import LoginRequiredMixin

    class EmployeeListDatatable(LoginRequiredMixin, datatable.Datatable):
        ...

Testing

This project uses tox to test with a build matrix to ensure that the project works with multiple versions of Python and Django. Run tox to test. Run tox --skip-missing-interpreters to avoid failures if you don't have a particular Python version installed. See other options in the tox documentation.