/React Native


Provides efficient data processing and access to the ListView component. A ListViewDataSource is created with functions for extracting data from the input blob, and comparing elements (with default implementations for convenience). The input blob can be as simple as an array of strings, or an object with rows nested inside section objects.

To update the data in the datasource, use cloneWithRows (or cloneWithRowsAndSections if you care about sections). The data in the data source is immutable, so you can't modify it directly. The clone methods suck in the new data and compute a diff for each row so ListView knows whether to re-render it or not.

In this example, a component receives data in chunks, handled by _onDataArrived, which concats the new data onto the old data and updates the data source. We use concat to create a new array - mutating this._data, e.g. with this._data.push(newRowData), would be an error. _rowHasChanged understands the shape of the row data and knows how to efficiently compare it.

getInitialState: function() {
  var ds = new ListViewDataSource({rowHasChanged: this._rowHasChanged});
  return {ds};
_onDataArrived(newData) {
  this._data = this._data.concat(newData);
    ds: this.state.ds.cloneWithRows(this._data)



You can provide custom extraction and hasChanged functions for section headers and rows. If absent, data will be extracted with the defaultGetRowData and defaultGetSectionHeaderData functions.

The default extractor expects data of one of the following forms:

 { sectionID_1: { rowID_1: <rowData1>, ... }, ... }


 { sectionID_1: [ <rowData1>, <rowData2>, ... ], ... }


 [ [ <rowData1>, <rowData2>, ... ], ... ]

The constructor takes in a params argument that can contain any of the following:

  • getRowData(dataBlob, sectionID, rowID);
  • getSectionHeaderData(dataBlob, sectionID);
  • rowHasChanged(prevRowData, nextRowData);
  • sectionHeaderHasChanged(prevSectionData, nextSectionData);

cloneWithRows(dataBlob, rowIdentities)

Clones this ListViewDataSource with the specified dataBlob and rowIdentities. The dataBlob is just an arbitrary blob of data. At construction an extractor to get the interesting information was defined (or the default was used).

The rowIdentities is a 2D array of identifiers for rows. ie. [['a1', 'a2'], ['b1', 'b2', 'b3'], ...]. If not provided, it's assumed that the keys of the section data are the row identities.

Note: This function does NOT clone the data in this data source. It simply passes the functions defined at construction to a new data source with the data specified. If you wish to maintain the existing data you must handle merging of old and new data separately and then pass that into this function as the dataBlob.

cloneWithRowsAndSections(dataBlob, sectionIdentities, rowIdentities)

This performs the same function as the cloneWithRows function but here you also specify what your sectionIdentities are. If you don't care about sections you should safely be able to use cloneWithRows.

sectionIdentities is an array of identifiers for sections. ie. ['s1', 's2', ...]. If not provided, it's assumed that the keys of dataBlob are the section identities.

Note: this returns a new object!



rowShouldUpdate(sectionIndex, rowIndex)

Returns if the row is dirtied and needs to be rerendered

getRowData(sectionIndex, rowIndex)

Gets the data required to render the row.


Gets the rowID at index provided if the dataSource arrays were flattened, or null of out of range indexes.


Gets the sectionID at index provided if the dataSource arrays were flattened, or null for out of range indexes.


Returns an array containing the number of rows in each section


Returns if the section header is dirtied and needs to be rerendered


Gets the data required to render the section header

© 2015–2017 Facebook Inc.
Licensed under the Creative Commons Attribution 4.0 International Public License.