Skip to content


Subversion checkout URL

You can clone with
Download ZIP
branch: master
Commits on Apr 10, 2010
  1. @jondkinney
Commits on Apr 9, 2010
  1. @jondkinney
Commits on Oct 5, 2009
  1. @manuelmeurer
Commits on Sep 25, 2009
  1. Adding test case to prove 1c873e0

  2. @htanata

    Fix "validates_length_of :is" for jQuery adapter.

    htanata authored committed
Commits on Sep 23, 2009
  1. @Nielsomat

    Ruby 1.9 compatibility: instance_eval => instance_exec

    Nielsomat authored
    When using Ruby 1.9, foo.instance_eval(&bar) doesn't implicitly
    pass foo as bar's first argument anymore. While instance_exec
    isn't available in Ruby < 1.8.7, this is a Rails plugin and
    ActiveSupport provides a backport of instance_exec.
Commits on Aug 24, 2009
Commits on Jul 25, 2009
  1. @alistairholt
Commits on Jul 3, 2009
  1. Using .underscore instead of .downcase on model class names. We want …

    …user_session_foo, not usersession_foo.
Commits on May 9, 2009
  1. Silencing the deprecation warning. Had to use ActiveSupport::Deprecat…

    …ion.silence, not Kernel.silence.
Commits on May 3, 2009
  1. Simplifying the form builder implementation to make it easier on the …

    …eye. Also making tag attribute altering adapters (such as validatious) work again, woot!
  2. Not running .callback_has_visible_attributes? on validation adapters …

    …that alters tag attributes. We need to know about all the validations before rendering the form for these types of adapters, because the form itself gets altered.
Commits on May 1, 2009
  1. Renaming LiveValidations::Adapter to LiveValidations::AdapterBase. It…

    … just makes more sense that way.
  2. Renaming and cleaning up a few things. No real changes.

    * Renaming the internal `perform_validations` method to `run_validations`.
    * Cleaning up the attribute hook code in the form builder.
    * Improved the error message when passing a invalid form builder object to <% form_for(the_object, :live_validations => true) %>
Commits on Apr 29, 2009
  1. @holman

    ValidationHook#message_for accepts optional options hash for I18N str…

    holman authored committed
    …ing {{interpolation}} purposes. Currently limited to adapter, specifically length validations.
    Signed-off-by: August Lilleaas <>
  2. @holman

    Adds option to only validate onBlur rather than as you type

    holman authored committed
    Signed-off-by: August Lilleaas <>
  3. @holman

    Removes old submodule for Rails

    holman authored committed
    Signed-off-by: August Lilleaas <>
Commits on Apr 28, 2009
Commits on Apr 21, 2009
  1. The concat call is wrapped in silence_warnings. This will silence the…

    … deprecation warning caused by passing block.binding to concat.
    It is deprecated in 2.3 (and possibly 2.2), but in order to preserve backwards compatibility, I'll leave it as it is until it breaks stuff.
  2. We should not add custom validations to the messages stack. The messa…

    …ge is already baked in to the custom validation. Closes #1
Commits on Mar 20, 2009
  1. Added some kind of odd support for i18n interpolation in the custom :…

    …message messages for the live validations. E.g. :message => "but srsly, {{model}}" will be translated and used as "but srsly, Post" in the live validations.
Commits on Mar 19, 2009
  1. Allowing passing symbols to form_fors with live_validations => true. …

    …It will look for an instance variable matching the symbol, or raise an error if no instance variable is found, or if the symbol isn't an active record instance.
  2. Testing giving the form builder an invalid object, and raising a more…

    … sensible error message when it occurs.
  3. The plugin doesn't blow up when the model has validations callback wi…

    …thout attributes (e.g. `validate` with a block).
Commits on Mar 18, 2009
  1. Changed the test testing attributes with validations but no form fiel…

    …ds to make more sense. Using an actual attribute in the validation, instead of a random one, etc.
  2. Making the error messages more sensible. Instead of rescuing all erro…

    …rs, the error that was raised is passed on. Also improved InlineTemplate so that syntax errors in the viewstring passed to 'render' provides a sensible error message.
  3. Improving the test suite by not using a vendored rails app in test/, …

    …but mocking it all out instead. This is done because it is convenient to be able to pass the view to render in the test as a string, compared to creating a bunch of views in the vendored rails application.
Something went wrong with that request. Please try again.