Events assigned to the table can be exceptionally useful for user interaction, however you must be aware that DataTables will add and remove rows from the DOM as they are needed (i.e. when paging only the visible elements are actually available in the DOM). As such, this can lead to the odd hiccup when working with events.

Each column has an optional rendering control called columns.render which can be used to process the content of each cell before the data is used. columns.render has a wide array of options available to it for rendering different types of data orthogonally (ordering, searching, display etc), but it can be used very simply to manipulate the content of a cell, as shown here.

As is described by the basic DOM positioning example you can use the dom initialization parameter to move DataTables features around the table to where you want them.

Complex headers (using colspan / rowspan) can be used to group columns of similar information in DataTables, creating a very powerful visual effect.
In addition to the basic behaviour, DataTables can also take colspan and rowspans into account when working with hidden columns. The colspan and rowspan attributes for each cell are automatically calculated and rendered on the page for you. This allows the columns.visible option and column().visible() method to take into account rowspan / colspan cells, drawing the header correctly.

As well as being able to pass language information to DataTables through the language initialization option, you can also store the language information in a file, which DataTables can load by Ajax using the language.url option.

When working with DataTables over multiple pages it is often useful to set the initialization defaults to common values (for example you might want to set dom to a common value so all tables get the same layout). This can be done using the $.fn.dataTable.defaults object. This object will take all of the same parameters as the DataTables initialization object, but in this case you are setting the default for all future initializations of DataTables.

Although DataTables doesn't have row grouping built-in (picking one of the many methods available would overly limit the DataTables core), it is most certainly possible to give the look and feel of row grouping.

Through the use of the header and footer callback manipulation functions provided by DataTables (headerCallback and footerCallback), it is possible to perform some powerful and useful data manipulation functions, such as summarising data in the table.

DataTables inserts DOM elements around the table to control DataTables features, and you can make use of this mechanism as well to insert your own custom elements. In this example a div with a class of '-string toolbar' is created using dom, and jQuery then used to insert HTML into that element to create the toolbar. You could put whatever HTML you want into the toolbar!

The following example shows how a callback function can be used to format a particular row at draw time. For each row that is generated for display, the createdRow function is called once and once only. It is passed the create row node which can then be modified.

Exporting data from a table can often be a key part of a complex application. The Buttons extension for DataTables provides three plug-ins that provide overlapping functionality for data export.