Filters

The ability for a column to be column filtered is determined by the following:

• The column was defined with createDataColumn or a valid accessorKey/accessorFn.
• column.enableColumnFilter is not set to false
• options.enableColumnFilters is not set to false
• options.enableFilters is not set to false

The ability for a column to be globally filtered is determined by the following:

• The column was defined with createDataColumn or a valid accessorKey/accessorFn.
• If provided, options.getColumnCanGlobalFilter returns true for the given column. If it is not provided, the column is assumed to be globally filterable.
• column.enableColumnFilter is not set to false
• options.enableColumnFilters is not set to false
• options.enableFilters is not set to false

Filter state is stored on the table instance using the following shape:

export type FiltersTableState = {
  columnFilters: ColumnFiltersState
  globalFilter: any
}

export type ColumnFiltersState = ColumnFilter[]

export type ColumnFilter = {
  id: string
  value: unknown
}

The following filter functions are built-in to the table core:

• includesString
  • Case-insensitive string inclusion
• includesStringSensitive
  • Case-sensitive string inclusion
• equalsString
  • Case-insensitive string equality
• equalsStringSensitive
  • Case-sensitive string equality
• arrIncludes
  • Item inclusion within an array
• arrIncludesAll
  • All items included in an array
• arrIncludesSome
  • Some items included in an array
• equals
  • Object/referential equality Object.is/===
• weakEquals
  • Weak object/referential equality ==
• inNumberRange
  • Number range inclusion

Every filter function receives:

• The row to filter
• The columnId to use to retrieve the row's value
• The filter value

and should return true if the row should be included in the filtered rows, and false if it should be removed.

This is the type signature for every filter function:

export type FilterFn<TGenerics extends TableGenerics> = {
  (
    row: Row<TGenerics>,
    columnId: string,
    filterValue: any,
    addMeta: (meta: TGenerics['FilterMeta']) => void
  ): boolean
  resolveFilterValue?: TransformFilterValueFn<TGenerics>
  autoRemove?: ColumnFilterAutoRemoveTestFn<TGenerics>
  addMeta?: (meta?: TGenerics['FilterMeta']) => void
}

export type TransformFilterValueFn<TGenerics extends TableGenerics> = (
  value: any,
  column?: Column<TGenerics>
) => unknown

export type ColumnFilterAutoRemoveTestFn<TGenerics extends TableGenerics> = (
  value: any,
  column?: Column<TGenerics>
) => boolean

export type CustomFilterFns<TGenerics extends TableGenerics> = Record<
  string,
  FilterFn<TGenerics>
>

This optional "hanging" method on any given filterFn allows the filter function to transform/sanitize/format the filter value before it is passed to the filter function.

This optional "hanging" method on any given filterFn is passed a filter value and expected to return true if the filter value should be removed from the filter state. eg. Some boolean-style filters may want to remove the filter value from the table state if the filter value is set to false.

Filter functions can be used/referenced/defined by passing the following to columnDefinition.filterFn or options.globalFilterFn:

• A string that references a built-in filter function
• A string that references a custom filter functions provided via the tableOptions.filterFns option
• A function directly provided to the columnDefinition.filterFn option

The final list of filter functions available for the columnDef.filterFn and ``tableOptions.globalFilterFn` options use the following type:

export type FilterFnOption<TGenerics extends TableGenerics> =
  | 'auto'
  | BuiltInFilterFn
  | keyof TGenerics['FilterFns']
  | FilterFn<TGenerics>

Filtering data can often expose additional information about the data that can be used to aid other future operations on the same data. A good exmaple of this concept is a ranking-system like that of match-sorter that simutaneously ranks, filters and sorts data. While utilities like match-sorter make a lot of sense for single-dimensional filter+sort tasks, the decoupled filtering/sorting architecture of building a table makes them very difficult and unperformant to use.

To make a ranking/filtering/sorting system work with tables, filterFns can optionally mark results with a filter meta value that can be used later to sort/group/etc the data to your liking. This is done by calling the addMeta function supplied to your custom filterFn.

Below is an example using our own match-sorter-utils package (a utility fork of match-sorter) to rank, filter, and sort the data

import {
  Column,
  createTable,
  TableInstance,
  useTableInstance,
  ColumnFiltersState,
  getCoreRowModel,
  getFilteredRowModel,
  getFacetedRowModel,
  getFacetedUniqueValues,
  getFacetedMinMaxValues,
  getPaginationRowModel,
  sortingFns,
} from '@tanstack/react-table'

import {
  RankingInfo,
  rankItem,
  compareItems,
} from '@tanstack/match-sorter-utils'

let table = createTable()
  .setFilterMetaType<RankingInfo>()
  .setOptions({
    filterFns: {
      fuzzy: (row, columnId, value, addMeta) => {
        // Rank the item
        const itemRank = rankItem(row.getValue(columnId), value)

        // Store the ranking info
        addMeta(itemRank)

        // Return if the item should be filtered in/out
        return itemRank.passed
      },
    },
    sortingFns: {
      fuzzy: (rowA, rowB, columnId) => {
        let dir = 0

        // Only sort by rank if the column has ranking information
        if (rowA.columnFiltersMeta[columnId]) {
          dir = compareItems(
            rowA.columnFiltersMeta[columnId]!,
            rowB.columnFiltersMeta[columnId]!
          )
        }

        // Provide an alphanumeric fallback for when the item ranks are equal
        return dir === 0 ? sortingFns.alphanumeric(rowA, rowB, columnId) : dir
      },
    },
  })

filterFn?: FilterFn | keyof TGenerics['FilterFns'] | keyof BuiltInFilterFns

The filter function to use with this column.

Options:

• A string referencing a built-in filter function)
• A string referencing a custom filter function defined on the filterFns table option
• A custom filter function

enableColumnFilter?: boolean

Enables/disables the column filter for this column. For option priority, see Can-Filter Option Priority.

enableGlobalFilter?: boolean

Enables/disables the global filter for this column. For option priority, see Can-Filter Option Priority.

getCanFilter: () => boolean

Returns whether or not the column can be column filtered.

getCanGlobalFilter: () => boolean

Returns whether or not the column can be globally filtered.

getFilterIndex: () => number

Returns the index (including -1) of the column filter in the table's state.columnFilters array.

getIsFiltered: () => boolean

Returns whether or not the column is currently filtered.

getFilterValue: () => unknown

Returns the current filter value of the column.

setFilterValue: (updater: Updater<any>) => void

A function that sets the current filter value for the column. You can pass it a value or an updater function for immutability-safe operations on existing values.

getAutoFilterFn: (columnId: string) => FilterFn<TGenerics> | undefined

Returns an automatically calculated filter function for the column based off of the columns first known value.

getFilterFn: (columnId: string) => FilterFn<TGenerics> | undefined

Returns the filter function (either user-defined or automatic, depending on configuration) for the columnId specified.

type getFacetedRowModel = () => RowModel<TGenerics>

⚠️ Requires that you pass a valid getFacetedRowModel function to options.facetedRowModel. A default implementation is provided via the exported getFacetedRowModel function.

Returns the row model with all other column filters applied, excluding its own filter. Useful for displaying faceted result counts.

getFacetedUniqueValues: () => Map<any, number>

⚠️ Requires that you pass a valid getFacetedUniqueValues function to options.getFacetedUniqueValues. A default implementation is provided via the exported getFacetedUniqueValues function.

A function that computes and returns a Map of unique values and their occurences derived from column.getFacetedRowModel. Useful for displaying faceted result values.

getFacetedMinMaxValues: () => Map<any, number>

⚠️ Requires that you pass a valid getFacetedMinMaxValues function to options.getFacetedMinMaxValues. A default implementation is provided via the exported getFacetedMinMaxValues function.

A function that computes and returns a min/max tuple derived from column.getFacetedRowModel. Useful for displaying faceted result values.

columnFilters: Record<string, boolean>

The column filters map for the row. This object tracks whether a row is passing/failing specific filters by their column ID.

columnFiltersMeta: Record<string, TGenerics['FilterMeta']>

The column filters meta map for the row. This object tracks any filter meta for a row as optionally provided during the filtering process.

filterFromLeafRows?: boolean

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 this option to true will cause filtering to be done from leaf rows up (which means parent rows will be included so long as one of their child or grand-child rows is also included).

filterFns?: Record<string, FilterFn>

Normally set ahead of time when using the createTable() helper, this option allows you to define custom filter functions that can be referenced by their string key.

Example:

const table = createTable().setOptions({
  filterFns: {
    myCustomFilterFn: (rows, columnIds, filterValue) => {
      // return the filtered rows
    },
  },
})

const column = table.createDataColumn('key', {
  filterFn: 'myCustomFilterFn',
})

enableFilters?: boolean

Enables/disables all filters for the table. For option priority, see Can-Filter Option Priority.

manualFiltering?: boolean

Disables the getFilteredRowModel from being used to filter data. This may be useful if your table needs to dynamically support both client-side and server-side filtering.

onColumnFiltersChange?: OnChangeFn<ColumnFiltersState>

If provided, this function will be called with an updaterFn when state.columnFilters changes. This overrides the default internal state management, so you will need to persist the state change either fully or partially outside of the table.

enableColumnFilters?: boolean

Enables/disables all column filters for the table. For option priority, see Can-Filter Option Priority.

getFilteredRowModel?: (
  instance: TableInstance<TGenerics>
) => () => RowModel<TGenerics>

If provided, this function is called once per table instance and should return a new function which will calculate and return the row model for the table when it's filtered.

• For server-side filtering, this function is unnecessary and can be ignored since the server should already return the filtered row model.
• For client-side filtering, this function is required. A default implementation is provided via any table adapter's { getFilteredRowModel } export.

Example:

import { getFilteredRowModel } from '@tanstack/[adapter]-table'

useTableInstance(table, {
  getFilteredRowModel: getFilteredRowModel(),
})

getColumnFacetedRowModel: (columnId: string) => RowModel<TGenerics>

Returns the faceted row model for a given columnId.

globalFilterFn?: FilterFn | keyof TGenerics['FilterFns'] | keyof BuiltInFilterFns

The filter function to use for global filtering.

Options:

• A string referencing a built-in filter function)
• A string referencing a custom filter function defined on the filterFns table option
• A custom filter function

onGlobalFilterChange?: OnChangeFn<GlobalFilterState>

If provided, this function will be called with an updaterFn when state.globalFilter changes. This overrides the default internal state management, so you will need to persist the state change either fully or partially outside of the table.

enableGlobalFilter?: boolean

Enables/disables the global filter for the table. For option priority, see Can-Filter Option Priority.

getColumnCanGlobalFilter?: (column: Column<TGenerics>) => boolean

If provided, this function will be called with the column and should return true or false to indicate whether this column should be used for global filtering.

setColumnFilters: (updater: Updater<ColumnFiltersState>) => void

Sets or updates the state.columnFilters state.

resetColumnFilters: (defaultState?: boolean) => void

Resets the columnFilters state to initialState.columnFilters, or true can be passed to force a default blank state reset to [].

getPreFilteredRowModel: () => RowModel<TGenerics>

Returns the row model for the table before any column filtering has been applied.

getFilteredRowModel: () => RowModel<TGenerics>

Returns the row model for the table after column filtering has been applied.

setGlobalFilter: (updater: Updater<any>) => void

Sets or updates the state.globalFilter state.

resetGlobalFilter: (defaultState?: boolean) => void

Resets the globalFilter state to initialState.globalFilter, or true can be passed to force a default blank state reset to undefined.

getGlobalAutoFilterFn: (columnId: string) => FilterFn<TGenerics> | undefined

Currently, this function returns the built-in includesString filter function. In future releases, it may return more dynamic filter functions based on the nature of the data provided.

getGlobalFilterFn: (columnId: string) => FilterFn<TGenerics> | undefined

Returns the global filter function (either user-defined or automatic, depending on configuration) for the table.

getGlobalFacetedRowModel: () => RowModel<TGenerics>

Returns the faceted row model for the global filter.

getGlobalFacetedUniqueValues: () => Map<any, number>

Returns the faceted unique values for the global filter.

getGlobalFacetedMinMaxValues: () => [number, number]

Returns the faceted min and max values for the global filter.