Expanding Sub-Rows/Tree Data Guide

Shadcn Table has support for expanding sub-rows or tree data. This feature is useful for displaying hierarchical data. The sub-rows can be expanded and collapsed by clicking on the expand/collapse icon.

NOTE: This feature is for expanding rows of the same data type. If you want to add expansion of more data for the same row, check out the Detail Panel Feature Guide.

Relevant Table OptionsCopy Anchor Link

Relevant State OptionsCopy Anchor Link

Enable Expanding Sub-RowsCopy Anchor Link

To enable expanding sub-rows, you must first set the enableExpanding table option to true.

However, your data must also be formatted in a way to allow for expanding rows that are in some way related to each other. By default, Shadcn Table will look for a special subRows property on each row of your data and treat any array of rows that it finds as the sub-rows for that row. You can customize or override this behavior by passing a custom getSubRows table option.

const data = [
  {
    id: 1,
    name: 'John Doe',
    subRows: [
      {
        id: 2,
        name: 'Jane Doe',
      },
      //more sub rows...
    ],
  },
  //more rows...
];

const table = useShadcnTable({
  columns,
  data,
  enableExpanding: true,
  getSubRows: (originalRow) => originalRow.subRows, //default, can customize
});

return <ShadcnTable table={table} />;

Expand All Rows ButtonCopy Anchor Link

By default, Shadcn Table will show the expand all button in the expand column header. You can disable this by setting the enableExpandAll table option to false.

const table = useShadcnTable({
  columns,
  data,
  enableExpanding: true,
  enableExpandAll: false, //hide expand all button in header
});

Generate Sub Rows with getSubRowsCopy Anchor Link

If your data is not yet in a tree structure, but the data has relationships that can be parsed into a tree, you can use the getSubRows table option to let TanStack Table find the sub rows for each row.

There are a couple key things you have to do to make this work:

1. Only pass in root (top level) rows in your data prop.
2. Set the getSubRows table option to a function that scans all the rest of your data and returns the sub rows for a given row.

This can sometimes be useful in combination with lazy loading sub rows.

NOTE: Be conscious of the performance implications of the getSubRows function. It will be called for every row in your table, so it should be performant.

Expanded Rows Pagination BehaviorCopy Anchor Link

By default, Shadcn Table will treat expanded sub-rows the same as any other row when it comes to pagination. This means that some expanded rows may be on the next page. You can change this behavior by setting the paginateExpandedRows table option to false.

const table = useShadcnTable({
  columns,
  data,
  enableExpanding: true,
  paginateExpandedRows: false, //expanded rows will be on the same page as their parent row
});

Expanded Leaf Row Filtering BehaviorCopy Anchor Link

If you are using the filtering features alongside sub-row features, then there are a few behaviors and customizations you should be aware of.

Filter From Leaf RowsCopy Anchor Link

By default, filtering is done from parent rows down (so if a parent row is filtered out, all of its children will be filtered out as well). Setting the filterFromLeafRows table option to true will cause filtering to be done from leaf rows up (which means parent rows will be kept so long as one of their child, or grand-child, etc. rows pass the filtering).

const table = useShadcnTable({
  columns,
  data,
  enableExpanding: true,
  filterFromLeafRows: true, //search for child rows and preserve parent rows
});

Max Leaf Row Filter DepthCopy Anchor Link

By default, filtering is done for all rows (max depth of 100), no matter if they are root level parent rows or the child leaf rows of a parent row. Setting the maxLeafRowFilterDepth table option to 0 will cause filtering to only be applied to the root level parent rows, with all sub-rows remaining unfiltered. Similarly, setting this option to 1 will cause filtering to only be applied to child leaf rows 1 level deep, and so on.

This is useful for situations where you want a row's entire child hierarchy to be visible, regardless of the applied filter.

const table = useShadcnTable({
  columns,
  data,
  enableExpanding: true,
  maxLeafRowFilterDepth: 0, //When filtering root rows, keep all child rows of the passing parent rows
});

Expand All Rows By DefaultCopy Anchor Link

You can manage the initial state of the expanded rows with the expanded state option in either the initialState or state props.

For example, you may want all rows to be expanded by default. To do this, you can simply set the expanded state option to true.

const table = useShadcnTable({
  columns,
  data,
  enableExpanding: true,
  initialState: { expanded: true }, //all rows expanded by default
});

Expand Root Rows Only By DefaultCopy Anchor Link

Here is a slightly more complex initial expanded state example where all the root rows are expanded by default, but none of the sub rows themselves are expanded by default. We just need to find all of the root row ids and set their key in the expanded initialState option to true.

Customize Expand ColumnCopy Anchor Link

You can customize the expand column by using the displayColumnDefOptions table option.

const table = useShadcnTable({
  columns,
  data,
  enableGrouping: true,
  displayColumnDefOptions: {
    'st-row-expand': {
      enableResizing: true, //allow resizing
      size: 120, //make the expand column wider
    },
  },
});

Lazy Load Sub RowsCopy Anchor Link

If you have a ton of nested data that you want to display, but you don't want to fetch it all up front, you can set up Shadcn Table to only fetch the sub-rows data when the user expands the row.

There are quite a few ways in which you could implement fetching sub-rows lazily. This example is just one way to do it.

The main concept to understand from this example is that you can manage the expanded state option in your own scope, and fetch the data for your table based on that state.

How your data is structured from the server is up to you. It is usually easiest to have the server do the hard work and return the data in a nested tree structure, but you can also return the data in a flat structure and use the getSubRows table option to parse the data into a tree structure.