Data validation

Registering a validator

When you create a validator, a good idea is to assign it as an alias that will refer to this particular validator function. Handsontable defines 5 aliases by default:

  • autocomplete for Handsontable.validators.AutocompleteValidator
  • date for Handsontable.validators.DateValidator
  • dropdown for Handsontable.validators.DropdownValidator
  • numeric for Handsontable.validators.NumericValidator
  • time for Handsontable.validators.TimeValidator

It gives users a convenient way for defining which validator should be used when table validation was triggered. User doesn't need to know which validator function is responsible for checking the cell value, he does not even need to know that there is any function at all. What is more, you can change the validator function associated with an alias without a need to change code that defines a table.

To register your own alias use Handsontable.validators.registerValidator() function. It takes two arguments:

  • validatorName - a string representing a validator function
  • validator - a validator function that will be represented by validatorName

If you'd like to register creditCardValidator under alias credit-card you have to call:

Handsontable.validators.registerValidator('credit-card', creditCardValidator);

Choose aliases wisely. If you register your validator under name that is already registered, the target function will be overwritten:

Handsontable.validators.registerValidator('date', creditCardValidator);

// Now 'date' alias points to `creditCardValidator` function, not Handsontable.validators.DateValidator

So, unless you intentionally want to overwrite an existing alias, try to choose a unique name. A good practice is prefixing your aliases with some custom name (for example your GitHub username) to minimize the possibility of name collisions. This is especially important if you want to publish your validator, because you never know aliases has been registered by the user who uses your validator.

Handsontable.validators.registerValidator('credit-card', creditCardValidator);

// Someone might already registered such alias

Handsontable.validators.registerValidator('', creditCardValidator); // That's better.

Using an alias

The final touch is to using the registered aliases, so that users can easily refer to it without the need to now the actual validator function is.

To sum up, a well prepared validator function should look like this:

  function customValidator(query, callback) {
    // ...your custom logic of the validator

    callback(/* Pass `true` or `false` based on your logic */);

  // Register an alias
  Handsontable.validators.registerValidator('my.custom', customValidator);


From now on, you can use customValidator like so:

var hot = new Handsontable(document.getElementById('container'), {
  data: someData,
  columns: [
      validator: 'my.custom'

Use the validator (see options page) method to easily validate synchronous or asynchronous changes to a cell. If you need more control, beforeValidate and afterValidate plugin hooks are available (see hooks page). In the below example, email_validator_fn is an async validator that resolves after 1000 ms.

Use the allowInvalid option (see options page) to define if the grid should accept input that does not validate. If you need to modify the input (e.g. censor bad words, uppercase first letter), use the plugin hook beforeChange (see hooks page).

By default all invalid cells are marked by htInvalid CSS class. If you want to change class to another you can basically add the invalidCellClassName option to Handsontable settings. For example:

// For whole table
invalidCellClassName: 'myInvalidClass',

// For specified columns
columns: [
  {data: 'firstName', invalidCellClassName: 'myInvalidClass'},
  {data: 'lastName', invalidCellClassName: 'myInvalidSecondClass'},
  {data: 'address'},

Callback console: [[row, col, oldValue, newValue], ...]

Edit the above grid to see callback

Note: Please keep in mind that changes in table are applied after running all validators (both synchronous and and asynchronous) from every changed cells.


Edit this page

Tutorial: Data validation