-
The negative form of
allow_value
has been changed so that instead of asserting that any of the given values is an invalid value (allowing good values to pass through), assert that all values are invalid values (allowing good values not to pass through). This means that this test which formerly passed will now fail:expect(record).not_to allow_value('good value', *bad_values)
-
ensure_inclusion_of
,ensure_exclusion_of
, andensure_length_of
have been removed in favor of theirvalidate_*
counterparts. -
set_the_flash
andset_session
have been changed to more closely align with each other:set_the_flash
has been removed in favor ofset_flash
.set_session('foo')
is no longer valid syntax, please useset_session['foo']
instead.set_session['key'].to(nil)
will no longer pass when the key in question has not been set yet.
-
allow_value
may raise an error if the attribute in question contains custom logic to ignore certain values, resulting in a discrepancy between the value you provide and the value that the attribute is actually set to. Specifically, if the attribute cannot be changed from a non-nil value to a nil value, or vice versa, then you'll get a CouldNotSetAttributeError. The current behavior (which is to permit this) is misleading, as the test that you write by usingallow_value
is different from the test that actually ends up getting run.
-
So far the tests for the gem have been running against only SQLite. Now they run against PostgreSQL, too. As a result we were able to fix some Postgres-related bugs, specifically around
validate_uniqueness_of
:-
When scoped to a UUID column that ends in an "f", the matcher is able to generate a proper "next" value without erroring. (#402, #587, #662)
-
Support scopes that are PostgreSQL array columns. Please note that this is only supported for Rails 4.2 and greater, as versions before this cannot handle array columns correctly, particularly in conjunction with the uniqueness validator. (#554)
-
Fix so that when scoped to a text column and the scope is set to nil before running it through the matcher, the matcher does not fail. (#521, #607)
-
-
Fix
define_enum_for
so that it actually tests that the attribute is present in the list of defined enums, as you could fool it by merely defining a class method that was the pluralized version of the attribute name. In the same vein, passing a pluralized version of the attribute name todefine_enum_for
would erroneously pass, and now it fails. (#641) -
Fix
permit
so that it does not break the functionality of ActionController::Parameters#require. (#648, #675) -
Fix
validate_uniqueness_of
+scoped_to
so that it does not raise an error if a record exists where the scoped attribute is nil. (#677)
- Add
on
qualifier topermit
. This allows you to make an assertion that a restriction was placed on a slice of theparams
hash and not the entireparams
hash. Although we don't require you to use this qualifier, we do recommend it, as it's a more precise check. (#675)
-
ensure_length_of
has been renamed tovalidate_length_of
.ensure_length_of
is deprecated and will be removed in 3.0.0. -
set_the_flash
has been renamed toset_flash
.set_the_flash
is deprecated and will be removed in 3.0.0. -
set_session(:foo)
is deprecated in favor ofset_session[:foo]
.set_session(:foo)
will be invalid syntax in 3.0.0. -
Using
should set_session[:key].to(nil)
to assert that that a value has not been set is deprecated. Please useshould_not set_session[:key]
instead. In 3.0.0,should set_session[:key].to(nil)
will only pass if the value is truly nil.
-
Fix
delegate_method
so that it works again with shoulda-context. (#591) -
Fix
validate_uniqueness_of
when used withscoped_to
so that when one of the scope attributes is a polymorphic*_type
attribute and the model has another validation on the same attribute, the matcher does not fail with an error. (#592) -
Fix
has_many
used withthrough
so that when the association does not exist, and the matcher fails, it does not raise an error when producing the failure message. (#588) -
Fix
have_and_belong_to_many
used withjoin_table
so that it does not fail whenforeign_key
and/orassociation_foreign_key
was specified on the association as a symbol instead of a string. (#584) -
Fix
allow_value
when an i18n translation key is passed towith_message
and the:against
option is used to specify an alternate attribute. A bug here also happened to affectvalidate_confirmation_of
when an i18n translation key is passed towith_message
. (#593) -
Fix
class_name
qualifier for association matchers so that if the model being referenced is namespaced, the matcher will correctly resolve the class before checking it against the association'sclass_name
. (#537) -
Fix
validate_inclusion_of
used withwith_message
so that it fails if given a message that does not match the message on the validation. (#598) -
Fix
route
matcher so that when controller and action are specified in hash notation (e.g.posts#show
), route parameters such asid
do not need to be specified as a string but may be specified as a number as well. (#602) -
Fix
allow_value
,validate_numericality_of
andvalidate_inclusion_of
so that they handle RangeErrors emitted from ActiveRecord 4.2. These exceptions arise whenever we attempt to set an attribute using a value that lies outside the range of the column (assuming the column is an integer). RangeError is now treated specially, failing the test instead of bubbling up as an error. (#634, #637, #642)
-
Add ability to test
:primary_key
option on associations. (#597) -
Add
allow_blank
qualifier tovalidate_uniqueness_of
to complement theallow_blank
option. (#543) -
Change
set_session
so that #[] and #to qualifiers are optional, similar toset_flash
. That is, you can now sayshould set_session
to assert that any flash value has been set, orshould set_session.to('value')
to assert that any value in the session is 'value'. -
Change
set_session
so that its #to qualifier supports regexps, similar toset_flash
. -
Add
with_prefix
qualifier todelegate_method
to correspond to theprefix
option for Rails'sdelegate
macro. (#622) -
Add support for Rails 4.2, especially fixing
serialize
matcher to remove warning aboutserialized_attributes
being deprecated. (#627) -
Update
dependent
qualifier on association matchers to support:destroy
,:delete
,:nullify
,:restrict
,:restrict_with_exception
, and:restrict_with_error
. You can also passtrue
orfalse
to assert that the association has (or has not) been declared with any dependent option. (#631)
- Tweak
allow_value
failure message so that it reads a bit nicer when listing existing errors.
-
ensure_inclusion_of
has been renamed tovalidate_inclusion_of
.ensure_inclusion_of
is deprecated and will be removed in 3.0.0. -
ensure_exclusion_of
has been renamed tovalidate_exclusion_of
.ensure_exclusion_of
is deprecated and will be removed in 3.0.0.
-
Fix
delegate_method
so that it does not raise an error if the method that returns the delegate object is private. -
Warn when
ensure_inclusion_of
is chained with.in_array([false, true])
as well as with.in_array([true, false])
. -
Fix
set_session
so that theto
qualifier if given nil checks that the session variable in question was set to nil (previously this actually did nothing). -
Fix
filter_param
so that it works whenconfig.filter_parameters
contains regexes. -
Fix
delegate_method
so that it can be required independent of Active Support. -
Fix
validate_uniqueness_of
. When used against an unpersisted record whose model contained a non-nullable column other than the one being validated, the matcher would break. Even if the test set that column to a value beforehand, the record had to be persisted in order for the matcher to work. Now this is no longer the case and the record can remain unpersisted. -
Fix
validate_absence_of
: it required that a string be passed as the attribute name rather than a symbol (which is the usual and documented usage).
- Add new matcher
define_enum_for
to test usage of theenum
macro introduced in Rails 4.1.
-
have_and_belongs_to_many
now checks to make sure that the join table contains the correct columns for the left- and right-hand side of the association. -
Reword failure message for
delegate_method
so that it's a little more helpful.
-
If you have a Rails >= 4.1 project and you are running tests using Spring, matchers that depend on assertions within Rails' testing layer (e.g.
render_template
androute
) will no longer fail. -
Fix
permit
so that it can be used more than once in the same test. -
Revert change to
validate_uniqueness_of
made in 2.6.0 so that it no longer provides default values for non-primary, non-nullable columns. This approach was causing test failures because it makes the assumption that none of these columns allow only specific values, which is not true. If you get an error fromvalidate_uniqueness_of
, your best bet continues to be creating a record manually and callingvalidate_uniqueness_of
on that instead. -
The majority of warnings that the gem produced have been removed. The gem still produces warnings under Ruby 1.9.3; we will address this in a future release.
-
Revert changes to
validate_numericality_of
made in the last release, which made it so that comparison qualifiers specified on the validation are tested using a very small decimal number offset rather than a whole number by default, except if the matcher was qualified withonly_integer
. This means that prior to 2.6.0, if your validation specifiedonly_integer
and you did not, then after 2.6.0 that test would fail. This is now fixed. -
Fix regression in previous release where ActiveRecord matchers would not be included when ActiveRecord wasn't defined (i.e. if you were using ActiveModel only).
-
Revert the behavior of
allow_value
changed in 2.6.0 (it will no longer raise CouldNotClearAttribute). This was originally done as a part of a fix forvalidate_presence_of
when used in conjunction withhas_secure_password
. That fix has been updated so that it does not affectallow_value
. -
Fix callback matchers and correct test coverage.
-
Fix
permit
so that it does not interfere with different usages ofparams
in your controller action. Specifically, this will not raise an error:params.fetch(:foo, {}).permit(:bar, :baz)
(thepermit
will have no problems recognizing that :bar and :baz are permitted params). -
Fix
permit
on Rails 4.1 to use PATCH by default for #update instead of PUT. Previously you had to specify this manually. -
Fix
permit
so that it track multiple calls to #permit in your controller action. Previously only the last usage of #permit would be considered in determining whether the matcher matched. -
Fix
permit
so that if the route for your action requires params (such as id) then you can now specify those params:permit(:first_name, :last_name).for(:update, params: { id: 42 })
. -
Fix
delegate_method
so that it does not stub the target method forever, returning it to its original implementation after the match ends. -
Fix
validate_uniqueness_of
to work with Rails 4.1 enum columns.
- Teach
with_message
qualifier onallow_value
to accept a hash of i18n interpolation values:allow_value('foo').for(:attr).with_message(:greater_than, values: { count: 20 })
.
-
The boolean argument to
have_db_index
'sunique
option is now optional, for consistency with other matchers. -
Association matchers now test that the model being referred to (either implicitly or explicitly, using
:class_name
) actually exists. -
Add ability to test
:autosave
option on associations. -
Fix
validate_uniqueness_of(...).allow_nil
so that it can be used against an non-password attribute which is in a model thathas_secure_password
. Doing so previously would result in a "Password digest missing on new record" error. -
Fix description for
validate_numericality_of
so that if the matcher fails, the error message reported does not say the matcher accepts integer values if you didn't specify that. -
Fix
ensure_inclusion_of
so that you can use it against a boolean column (and pass boolean values toin_array
). There are two caveats:- You should not test that your attribute allows both true and false
(
.in_array([true, false]
); there's no way to test that it doesn't accept anything other than that. - You cannot test that your attribute allows nil (
.in_array([nil])
) if the column does not allow null values.
- You should not test that your attribute allows both true and false
(
-
Change
validate_uniqueness_of(...)
so that it provides default values for non-nullable attributes. -
Running
rake
now installs Appraisals before running the test suite. (Additionally, we now manage Appraisals using theappraisal
executable in Appraisal 1.0.0.) -
Add
allow_nil
option tovalidate_numericality_of
so that you can validate that numeric values are validated only if a value is supplied. -
Fix
validate_numericality_of
so that test fails when the value withgreater_than
,greater_than_or_equal_to
,less_than
,less_than_or_equal_ to
orequal_to
is not appropriate. -
Change
validate_presence_of
under Rails 4 so that if you are using it with a user whose modelhas_secure_password
and whose password is set to a value, you will be instructed to use a user whose password is blank instead. The reason for this change is due to the fact that Rails 4's version ofhas_secure_password
defines #password= such thatnil
will be ignored, which interferes with howvalidate_presence_of
works. -
Add ability to test
belongs_to
associations defined with:inverse_of
. -
Add back matchers that were removed in 2.0.0:
permit
, for testing strong parameters, anddelegate_method
, for testing delegation. -
Add new matchers for testing controller filters:
before_filter
,after_filter
, andaround_filter
(aliased tobefore_action
,after_action
andaround_action
for Rails 4). -
Fix
rescue_from
matcher so that it does not raise an error when testing a method handler which has been marked as protected or private. -
Fix compatibility issues with Rails 4.1:
set_the_flash
andhave_and_belongs_to_many
no longer raise errors- Minitest no longer prints warnings whenever shoulda-matchers is required
-
Fix Rails/Test::Unit integration to ensure that the test case classes we are re-opening actually exist.
-
Fix
ensure_length_of
so that it uses the right message to validate whenis_equal_to
is specified in conjunction with a custom message. -
The
route
matcher now accepts specifying a controller/action pair as a string instead of only a hash (e.g.route(...).to('posts#index')
instead ofroute(...).to(controller: 'posts', action: 'index')
). -
The
ensure_inclusion_of
matcher now works with a decimal column. -
Under Rails 3, if you had an association matcher chained with the the
order
submatcher -- e.g.should have_many(:foos).order(:bar)
-- and your association had an:include
on it, using the matcher would raise an error. This has been fixed. -
Fix
validate_uniqueness_of
so it doesn't fail if the attribute under test has a limit of fewer than 16 characters. -
You can now test that your
has_many :through
orhas_one :through
associations are defined with a:source
option. -
Add new matcher
validates_absence_of
. -
Update matchers so that they use
failure_message
andfailure_message_when_negated
to define error messages. These are new methods in the upcoming RSpec 3 release which replacefailure_message_for_should
andfailure_message_for_should_not
. We've kept backward compatibility so all of your existing tests should still work -- this is just to make sure when RSpec 3 is released you don't get a bunch of warnings.
-
Fix a bug with the
validate_numericality_of
matcher that would not allow thewith_message
option on certain submatchers. -
Fix a regression with context-dependent validations in ActiveResource
-
shoulda-matchers is now fully compatible with Rails 4.
-
When not using RSpec, shoulda-matchers is now auto-included into ActiveSupport::TestCase instead of Test::Unit::TestCase (in Rails 4 the former no longer inherits from the latter).
-
Fix a bug in
ensure_inclusion_of
that would cause issues with usingin_array
with an integer value. -
Add support for PostgreSQL UUID columns to
validates_uniqueness_of
(#334). -
Fix
validates_numericality_of
so thatis_equal_to
submatcher works correctly (#326). -
Fix context support for validation matchers and disallowed values (#313).
-
Add a
counter_cache
submatcher forbelongs_to
associations (#311). -
Add a
rescue_from
matcher for Rails controllers which checks that the correct ActiveSupport call has been made and that the handlers exist without actually throwing an exception (#287). -
Changed the scope of AssociationMatcher methods from protected to private.
-
Extracted
#order
,#through
, and#dependent
from AssociationMatcher as their own submatchers.
-
Fix
have_and_belong_to_many
matcher issue for Rails 4. -
Fix
validate_uniqueness_of.scoped_to
issue when the scoped field is already taken (#207). -
Add comparison submatchers to
validate_numericality_of
to correspond to the comparison options you can give tovalidates_numericality_of
(#244).
-
Add missing
failure_message_for_should_not
implementations tovalidate_numericality_of
and its submatchers -
Support validation contexts for testing validations
on: :create
and when using custom contexts likemodel.valid?(:my_context)
. -
Fix a bug in validations with autosaved models.
-
Fix maximum value detection for the
ensure_inclusion_of
andensure_exclusion_of
matchers. -
Add
:odd
and:even
options to thevalidate_numericality_of
matcher. -
Add
:touch
option to AssociationMatcher. -
Ruby 2.0.0 is now officially supported.
-
Fix the issue where using
%{attribute}
or%{model}
in I18n translations raised exceptions. -
Support datetime columns in
validate_uniqueness_of.scoped_to
. -
Add
allow_nil
option to thevalidate_uniqueness_of
matcher.
-
Remove the following matchers:
assign_to
respond_with_content_type
query_the_database
validate_format_of
have_sent_email
permit
(strong parameters matcher)delegate_method
-
For more information about 2.0 changes, see: http://robots.thoughtbot.com/post/47031676783/shoulda-matchers-2-0.
- Revert previous change in AllowValueMatcher that added a check for a properly-set attribute.
-
AllowValueMatcher checks that the right value is used for attempts at setting the attribute with it.
- Please note that previously-passing tests might now fail. It is likely that it's not a bug, but please make sure that the code you're testing is written properly before submitting an issue.
-
Use DisallowValueMatcher for
disallows_value_of
method. -
Assert
class_name
value on real class name for AssociationMatcher. -
Correct the variable used for
validate_confirmation_of
matcher description.
- Properly-released version of 1.5.3.
- Alleviate the need to add
rspec
gem to your app.
-
Bump version dependency of Bourne to allow for Mocha upgrade.
-
Should fix incompatibility with MiniTest.
-
Deprecate the following matchers:
assign_to
respond_with_content_type
query_the_database
validate_format_of
have_sent_email
permit
(strong parameters matcher)delegate_method
-
Use RSpec's native
configure.include
syntax for including matchers into RSpec (#204). -
Do not force MiniTest loading when test-unit is available (this was fixed before 1.3.0 then reverted in 1.3.0).
- Add a new
delegate_method
matcher.
-
Fix an issue when used with Test::Unit on the allow value matcher.
-
Fix an issue with using
ensure_inclusion_of(:attr)
given an array of true or false values.
-
Add
strict
option to validation matchers. -
Verify that arguments to
set_the_flash
matcher are valid. -
Fix issue in ValidateUniquenessMatcher that could cause an error on postgres.
-
You can now pass an array to
ensure_exclusion_of
usingin_array
. -
Allow testing of
:foreign_key
option forhas_one
relationships using the association matcher. -
Fix bug where
ensure_length_of
would pass if the given string was too long. -
allow_blank
will now allow values such as: ' ', '\n', and '\r'. -
Test outside values for
ensure_inclusion_of
when given an array. -
Fix the output of the
set_the_flash
matcher.
-
validate_format_of
will acceptallow_blank(bool)
andallow_nil(bool)
. -
Prefer Test::Unit to MiniTest when loading integrations so that RubyMine is happy (#88).
-
validates_uniqueness_of
will now create a record if one does not exist. Previously, users were required to create a record in the database before using this matcher. -
Fix an edge case when where the matchers weren't loaded into Test::Unit when mixing RSpec and Test::Unit tests and also loading both the 'rspec-rails' gem and 'shoulda-matchers' gem from the same Gemfile group, namely [:test, :development].
-
controller.should_not render_partial
now correctly matchesrender partial: "partial"
.
-
ensure_inclusion_of
now has anin_array
parameter:ensure_inclusion_of(:attr).in_array(['foo', 'bar'])
. It cannot be used with the.in_range
option. (vpereira) -
ensure_in_inclusion_of
within_array
will acceptallow_blank(bool)
andallow_nil(false)
-
Test against Rails 3.2.
-
Fix
ensure_length_of
to use all possible I18n error messages. -
have_db_index.unique(nil)
used to function exactly the same ashave_db_index
with no unique option. It now functions the same ashave_db_index.unique(false)
. -
In 1.1.0,
have_sent_email
checked all emails to ensure they matched. It now checks that only one email matches, which restores 1.0.0 behavior.
-
Add
only_integer
option tovalidate_numericality_of
:should validate_numericality_of(:attribute).only_integer
-
Add a
query_the_database
matcher:it { should query_the_database(4.times).when_calling(:complicated_method) }
it { should query_the_database(4.times).or_less.when_calling(:complicated_method) }
it { should_not query_the_database.when_calling(:complicated_method) }
-
Database columns are now correctly checked for primality. E.G., this works now:
it { should have_db_column(:id).with_options(:primary => true) }
-
The flash matcher can check specific flash keys using [], like so:
it { should set_the_flash[:alert].to("Password doesn't match") }
-
The
have_sent_email
matcher can checkreply_to
:it { should have_sent_email.reply_to([user, other]) }
-
Add
validates_confirmation_of
matcher:it { should validate_confirmation_of(:password) }
-
Add
serialize
matcher:it { should serialize(:details).as(Hash).as_instance_of(Hash) }
-
shoulda-matchers checks for all possible I18n keys, instead of just e.g.
activerecord.errors.messages.blank
-
Add
accept_nested_attributes
matcher -
Our very first dependency: ActiveSupport >= 3.0.0