docs/documentation/3.0/rspec-expectations/file.Changelog.html

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
  "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
<title>
  File: Changelog
    &mdash; Documentation by YARD 0.8.7.4
</title>
  <link rel="stylesheet" href="/stylesheets/docs/style.css" type="text/css" charset="utf-8" />
  <link rel="stylesheet" href="/stylesheets/docs/common.css" type="text/css" charset="utf-8" />
<script type="text/javascript" charset="utf-8">
  hasFrames = window.top.frames.main ? true : false;
  relpath = '';
  framesUrl = "frames.html#!file.Changelog.html";
</script>
  <script type="text/javascript" charset="utf-8" src="/javascripts/docs/jquery.js"></script>
  <script type="text/javascript" charset="utf-8" src="/javascripts/docs/app.js"></script>
  </head>
  <body>
    <div id="header">
      <div id="menu">
    <a href="_index.html">Index</a> &raquo; 
    <span class="title">File: Changelog</span>
  <div class="noframes"><span class="title">(</span><a href="." target="_top">no frames</a><span class="title">)</span></div>
</div>
      <div id="search">
    <a class="full_list_link" id="class_list_link"
        href="class_list.html">
      Class List
    </a>
    <a class="full_list_link" id="method_list_link"
        href="method_list.html">
      Method List
    </a>
    <a class="full_list_link" id="file_list_link"
        href="file_list.html">
      File List
    </a>
</div>
      <div class="clear"></div>
    </div>
    <iframe id="search_frame"></iframe>
    <div id="content"><div id='filecontents'><h3>3.0.4 / 2014-08-14</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v3.0.3...v3.0.4">Full Changelog</a></p>
<p>Bug Fixes:</p>
<ul>
<li>Fix <code>start_with</code> and <code>end_with</code> so that they work properly with
structs. (Myron Marston, #620)</li>
<li>Fix failure message generation so that structs are printed properly
in failures. Previously failure messages would represent them as
an array. (Myron Marston, #620)</li>
<li>Fix composable matcher support so that it does not wrongly treat
structs as arrays. (Myron Marston, #620)</li>
</ul>
<h3>3.0.3 / 2014-07-21</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v3.0.2...v3.0.3">Full Changelog</a></p>
<p>Bug Fixes:</p>
<ul>
<li>Fix issue with detection of generic operator matchers so they work
correctly when undefined. (Myron Marston, #597)</li>
<li>Don&#39;t inadvertently define <code>BasicObject</code> in 1.8.7. (Chris Griego, #603)</li>
<li>Fix <code>include</code> matcher so that it fails gracefully when matched against
an object that does not respond to <code>include?</code>. (Myron Marston, #607)</li>
</ul>
<h3>3.0.2 / 2014-06-19</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v3.0.1...v3.0.2">Full Changelog</a></p>
<p>Bug Fixes:</p>
<ul>
<li>Fix regression in <code>contain_exactly</code> (AKA <code>match_array</code>) that caused it
to wrongly pass when the expected array was empty. (Myron Marston, #581)</li>
<li>Provide a better error message when you use the <code>change(obj, :msg)</code>
form of the change matcher but forget the message argument. (Alex
Sunderland, #585)</li>
<li>Make the <code>contain_exactly</code> matcher work with arrays that contain hashes in
arbitrary ordering. (Sam Phippen, #578)</li>
</ul>
<h3>3.0.1 / 2014-06-12</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v3.0.0...v3.0.1">Full Changelog</a></p>
<p>Bug Fixes:</p>
<ul>
<li>Add a missing <code>require</code> that would cause the <code>respond_to</code> matcher to
fail when used in a project where the rest of RSpec (e.g. core and
expecatations) weren&#39;t being used. (Myron Marston, #566)</li>
<li>Structs are no longer treated as arrays when diffed. (Jon Rowe, #576)</li>
</ul>
<h3>3.0.0 / 2014-06-01</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v3.0.0.rc1...v3.0.0">Full Changelog</a></p>
<p>No code changes. Just taking it out of pre-release.</p>
<h3>3.0.0.rc1 / 2014-05-18</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v3.0.0.beta2...v3.0.0.rc1">Full Changelog</a></p>
<p>Breaking Changes for 3.0.0:</p>
<ul>
<li>Remove <code>matcher_execution_context</code> attribute from DSL-defined
custom matchers. (Myron Marston)</li>
<li>Remove <code>RSpec::Matchers::Pretty#_pretty_print</code>. (Myron Marston)</li>
<li>Remove <code>RSpec::Matchers::Pretty#expected_to_sentence</code>. (Myron Marston)</li>
<li>Rename <code>RSpec::Matchers::Configuration</code> constant to
<code>RSpec::Expectations::Configuration</code>. (Myron Marston)</li>
<li>Prevent <code>have_xyz</code> predicate matchers using private methods.
(Adrian Gonzalez)</li>
<li>Block matchers must now implement <code>supports_block_expectations?</code>.
(Myron Marston)</li>
<li>Stop supporting <code>require &#39;rspec-expectations&#39;</code>.
Use <code>require &#39;rspec/expectations&#39;</code> instead. (Myron Marston)</li>
</ul>
<p>Bug Fixes:</p>
<ul>
<li>Fix <code>NoMethodError</code> triggered by beta2 when <code>YARD</code> was loaded in
the test environment. (Myron Marston)</li>
<li>Fix <code>be_xyz</code> matcher to accept a <code>do...end</code> block. (Myron Marston)</li>
<li>Fix composable matcher failure message generation logic
so that it does not blow up when given <code>$stdout</code> or <code>$stderr</code>.
(Myron Marston)</li>
<li>Fix <code>change</code> matcher to work properly with <code>IO</code> objects.
(Myron Marston)</li>
<li>Fix <code>exist</code> matcher so that it can be used in composed matcher
expressions involving objects that do not implement <code>exist?</code> or
<code>exists?</code>. (Daniel Fone)</li>
<li>Fix composable matcher match logic so that it clones matchers
before using them in order to work properly with matchers
that use internal memoization based on a given <code>actual</code> value.
(Myron Marston)</li>
<li>Fix <code>be_xyz</code> and <code>has_xyz</code> predicate matchers so that they can
be used in composed matcher expressions involving objects that
do not implement the predicate method. (Daniel Fone)</li>
</ul>
<p>Enhancements:</p>
<ul>
<li>Document the remaining public APIs. rspec-expectations now has 100% of
the public API documented and will remain that way (as new undocumented
methods will fail the build). (Myron Marston)</li>
<li>Improve the formatting of BigDecimal objects in <code>eq</code> matcher failure
messages. (Daniel Fone)</li>
<li>Improve the failure message for <code>be_xyz</code> predicate matchers so
that it includes the <code>inspect</code> output of the receiver.
(Erik Michaels-Ober, Sam Phippen)</li>
<li>Add <code>all</code> matcher, to allow you to specify that a given matcher
matches all elements in a collection:
<code>expect([1, 3, 5]).to all( be_odd )</code>. (Adam Farhi)</li>
<li>Add boolean aliases (<code>&amp;</code>/<code>|</code>) for compound operators (<code>and</code>/<code>or</code>). (Adam Farhi)</li>
<li>Give users a clear error when they wrongly use a value matcher
in a block expectation expression (e.g. <code>expect { 3 }.to eq(3)</code>)
or vice versa.  (Myron Marston)</li>
</ul>
<h3>3.0.0.beta2 / 2014-02-17</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v3.0.0.beta1...v3.0.0.beta2">Full Changelog</a></p>
<p>Breaking Changes for 3.0.0:</p>
<ul>
<li>Remove deprecated support for accessing the <code>RSpec</code> constant using
<code>Rspec</code> or <code>Spec</code>. (Myron Marston)</li>
<li>Remove deprecated <code>RSpec::Expectations.differ=</code>. (Myron Marston)</li>
<li>Remove support for deprecated <code>expect(...).should</code>. (Myron Marston)</li>
<li>Explicitly disallow <code>expect { }.not_to change { }</code> with <code>by</code>,
<code>by_at_least</code>, <code>by_at_most</code> or <code>to</code>. These have never been supported
but did not raise explicit errors. (Myron Marston)</li>
<li>Provide <code>===</code> rather than <code>==</code> as an alias of <code>matches?</code> for
all matchers.  The semantics of <code>===</code> are closer to an RSpec
matcher than <code>==</code>. (Myron Marston)</li>
<li>Remove deprecated <code>RSpec::Matchers::OperatorMatcher</code> constant.
(Myron Marston)</li>
<li>Make <code>RSpec::Expectations::ExpectationNotMetError</code> subclass
<code>Exception</code> rather than <code>StandardError</code> so they can bypass
a bare <code>rescue</code> in end-user code (e.g. when an expectation is
set from within a rspec-mocks stub implementation). (Myron Marston)</li>
<li>Remove Test::Unit and Minitest 4.x integration. (Myron Marston)</li>
</ul>
<p>Enhancements:</p>
<ul>
<li>Simplify the failure message of the <code>be</code> matcher when matching against:
<code>true</code>, <code>false</code> and <code>nil</code>. (Sam Phippen)</li>
<li>Update matcher protocol and custom matcher DSL to better align
with the newer <code>expect</code> syntax. If you want your matchers to
maintain compatibility with multiple versions of RSpec, you can
alias the new names to the old. (Myron Marston)
<ul>
<li><code>failure_message_for_should</code> =&gt; <code>failure_message</code></li>
<li><code>failure_message_for_should_not</code> =&gt; <code>failure_message_when_negated</code></li>
<li><code>match_for_should</code> =&gt; <code>match</code></li>
<li><code>match_for_should_not</code> =&gt; <code>match_when_negated</code></li>
</ul></li>
<li>Improve generated descriptions from <code>change</code> matcher. (Myron Marston)</li>
<li>Add support for compound matcher expressions using <code>and</code> and <code>or</code>.
Simply chain them off of any existing matcher to create an expression
like <code>expect(alphabet).to start_with(&quot;a&quot;).and end_with(&quot;z&quot;)</code>.
(Eloy Espinaco)</li>
<li>Add <code>contain_exactly</code> as a less ambiguous version of <code>match_array</code>.
Note that it expects the expected array to be splatted as
individual args: <code>expect(array).to contain_exactly(1, 2)</code> is
the same as <code>expect(array).to match_array([1, 2])</code>. (Myron Marston)</li>
<li>Update <code>contain_exactly</code>/<code>match_array</code> so that it can match against
other non-array collections (such as a <code>Set</code>). (Myron Marston)</li>
<li>Update built-in matchers so that they can accept matchers as arguments
to allow you to compose matchers in arbitrary ways. (Myron Marston)</li>
<li>Add <code>RSpec::Matchers::Composable</code> mixin that can be used to make
a custom matcher composable as well. Note that custom matchers
defined via <code>RSpec::Matchers.define</code> already have this. (Myron
Marston)</li>
<li>Define noun-phrase aliases for built-in matchers, which can be
used when creating composed matcher expressions that read better
and provide better failure messages. (Myron Marston)</li>
<li>Add <code>RSpec::Machers.alias_matcher</code> so users can define their own
matcher aliases. The <code>description</code> of the matcher will reflect the
alternate matcher name. (Myron Marston)</li>
<li>Add explicit <code>be_between</code> matcher. <code>be_between</code> has worked for a
long time as a dynamic predicate matcher, but the failure message
was suboptimal. The new matcher provides a much better failure
message. (Erik Michaels-Ober)</li>
<li>Enhance the <code>be_between</code> matcher to allow for <code>inclusive</code> or <code>exclusive</code>
comparison (e.g. inclusive of min/max or exclusive of min/max).
(Pedro Gimenez)</li>
<li>Make failure message for <code>not_to be #{operator}</code> less confusing by
only saying it&#39;s confusing when comparison operators are used.
(Prathamesh Sonpatki)</li>
<li>Improve failure message of <code>eq</code> matcher when <code>Time</code> or <code>DateTime</code>
objects are used so that the full sub-second precision is included.
(Thomas Holmes, Jeff Wallace)</li>
<li>Add <code>output</code> matcher for expecting that a block outputs <code>to_stdout</code>
or <code>to_stderr</code>. (Luca Pette, Matthias Günther)</li>
<li>Forward a provided block on to the <code>has_xyz?</code> method call when
the <code>have_xyz</code> matcher is used. (Damian Galarza)</li>
<li>Provide integration with Minitest 5.x. Require
<code>rspec/expectations/minitest_integration</code> after loading minitest
to use rspec-expectations with minitest. (Myron Marston)</li>
</ul>
<p>Bug Fixes:</p>
<ul>
<li>Fix wrong matcher descriptions with falsey expected value (yujinakayama)</li>
<li>Fix <code>expect { }.not_to change { }.from(x)</code> so that the matcher only
passes if the starting value is <code>x</code>. (Tyler Rick, Myron Marston)</li>
<li>Fix hash diffing, so that it colorizes properly and doesn&#39;t consider trailing
commas when performing the diff. (Jared Norman)</li>
<li>Fix built-in matchers to fail normally rather than raising
<code>ArgumentError</code> when given an object of the wrong type to match
against, so that they work well in composite matcher expressions like
<code>expect([1.51, &quot;foo&quot;]).to include(a_string_matching(/foo/), a_value_within(0.1).of(1.5))</code>.
(Myron Marston)</li>
</ul>
<p>Deprecations:</p>
<ul>
<li>Retain support for RSpec 2 matcher protocol (e.g. for matchers
in 3rd party extension gems like <code>shoulda</code>), but it will print
a deprecation warning. (Myron Marston)</li>
</ul>
<h3>3.0.0.beta1 / 2013-11-07</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.99.2...v3.0.0.beta1">Full Changelog</a></p>
<p>Breaking Changes for 3.0.0:</p>
<ul>
<li>Remove explicit support for 1.8.6. (Jon Rowe)</li>
<li>Remove the deprecated <code>be_close</code> matcher, preferring <code>be_within</code> instead.
(Sam Phippen)</li>
<li>Remove the deprecated <code>have</code>, <code>have_at_least</code> and <code>have_at_most</code> matchers.
You can continue using those matchers through <a href="https://github.com/rspec/rspec-collection_matchers">https://github.com/rspec/rspec-collection_matchers</a>,
or you can rewrite your expectations with something like 
<code>expect(your_object.size).to eq(num)</code>. (Hugo Baraúna)</li>
<li>Rename <code>be_true</code> and <code>be_false</code> to <code>be_truthy</code> and <code>be_falsey</code>. (Sam Phippen)</li>
<li>Make <code>expect { }.to_not raise_error(SomeSpecificClass, message)</code>,
   <code>expect { }.to_not raise_error(SomeSpecificClass)</code> and
   <code>expect { }.to_not raise_error(message)</code> invalid, since they are prone
to hiding failures. Instead, use <code>expect { }.to_not raise_error</code> (with no
args). (Sam Phippen)</li>
<li>Within <code>RSpec::Matchers.define</code> blocks, helper methods made available
either via <code>def self.helper</code> or <code>extend HelperModule</code> are no longer
available to the <code>match</code> block (or any of the others). Instead
<code>include</code> your helper module and define the helper method as an
instance method. (Myron Marston)</li>
<li>Force upgrading Diff::LCS for encoding compatability with diffs. (Jon Rowe)</li>
</ul>
<p>Enhancements:</p>
<ul>
<li>Support <code>do..end</code> style block with <code>raise_error</code> matcher. (Yuji Nakayama)</li>
<li>Rewrote custom matcher DSL to simplify its implementation and solve a
few issues. (Myron Marston)</li>
<li>Allow early <code>return</code> from within custom matcher DSL blocks. (Myron
Marston)</li>
<li>The custom matcher DSL&#39;s <code>chain</code> can now accept a block. (Myron
Marston)</li>
<li>Support setting an expectation on a <code>raise_error</code> matcher via a chained
<code>with_message</code> method call. (Sam Phippen)</li>
</ul>
<p>Bug Fixes:</p>
<ul>
<li>Allow <code>include</code> and <code>match</code> matchers to be used from within a
DSL-defined custom matcher&#39;s <code>match</code> block. (Myron Marston)</li>
<li>Correct encoding error message on diff failure (Jon Rowe)</li>
</ul>
<p>Deprecations:</p>
<ul>
<li>Using the old <code>:should</code> syntax without explicitly configuring it is deprecated.
It will continue to work but will emit a deprecation warning in RSpec 3 if
you do not explicitly enable it. (Sam Phippen)</li>
</ul>
<h3>2.99.2 / 2014-07-21</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.99.1...v2.99.2">Full Changelog</a></p>
<p>Bug Fixes:</p>
<ul>
<li>Fix regression in <code>Expectations#method_handle_for</code> where proxy objects
with method delegated would wrongly not return a method handle.
(Jon Rowe, #594)</li>
<li>Fix issue with detection of generic operator matchers so they work
correctly when undefined. (Myron Marston, #597)</li>
</ul>
<h3>2.99.1 / 2014-06-19</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.99.0...v2.99.1">Full Changelog</a></p>
<p>Bug Fixes:</p>
<ul>
<li>Fix typo in custom matcher <code>expected</code> deprecation warning -- it&#39;s
<code>expected_as_array</code>, not <code>expected_array</code>. (Frederick Cheung, #562)</li>
</ul>
<h3>2.99.0 / 2014-06-01</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.99.0.rc1...v2.99.0">Full Changelog</a></p>
<p>Enhancements:</p>
<ul>
<li>Special case deprecation message for <code>errors_on</code> with <code>rspec-rails</code> to be more useful.
(Aaron Kromer)</li>
</ul>
<h3>2.99.0.rc1 / 2014-05-18</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.99.0.beta2...2.99.0.rc1">Full Changelog</a></p>
<p>Deprecations:</p>
<ul>
<li>Deprecate <code>matcher_execution_context</code> attribute on DSL-defined
custom matchers. (Myron Marston)</li>
<li>Deprecate <code>RSpec::Matchers::Pretty#_pretty_print</code>. (Myron Marston)</li>
<li>Deprecate <code>RSpec::Matchers::Pretty#expected_to_sentence</code>. (Myron Marston)</li>
<li>Deprecate <code>RSpec::Matchers::Configuration</code> in favor of
<code>RSpec::Expectations::Configuration</code>. (Myron Marston)</li>
<li>Deprecate <code>be_xyz</code> predicate matcher on an object that doesn&#39;t respond to
<code>xyz?</code> or <code>xyzs?</code>. (Daniel Fone)</li>
<li>Deprecate <code>have_xyz</code> matcher on an object that doesn&#39;t respond to <code>has_xyz?</code>.
(Daniel Fone)</li>
<li>Deprecate <code>have_xyz</code> matcher on an object that has a private method <code>has_xyz?</code>.
(Jon Rowe)</li>
<li>Issue a deprecation warning when a block expectation expression is
used with a matcher that doesn&#39;t explicitly support block expectations
via <code>supports_block_expectations?</code>. (Myron Marston)</li>
<li>Deprecate <code>require &#39;rspec-expectations&#39;</code>. Use
<code>require &#39;rspec/expectations&#39;</code> instead. (Myron Marston)</li>
</ul>
<h3>2.99.0.beta2 / 2014-02-17</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.99.0.beta1...v2.99.0.beta2">Full Changelog</a></p>
<p>Deprecations:</p>
<ul>
<li>Deprecate chaining <code>by</code>, <code>by_at_least</code>, <code>by_at_most</code> or <code>to</code> off of
<code>expect { }.not_to change { }</code>. The docs have always said these are
not supported for the negative form but now they explicitly raise
errors in RSpec 3. (Myron Marston)</li>
<li>Change the semantics of <code>expect { }.not_to change { x }.from(y)</code>.
In RSpec 2.x, this expectation would only fail if <code>x</code> started with
the value of <code>y</code> and changed. If it started with a different value
and changed, it would pass. In RSpec 3, it will pass only if the
value starts at <code>y</code> and it does not change. (Myron Marston)</li>
<li>Deprecate <code>matcher == value</code> as an alias for <code>matcher.matches?(value)</code>,
in favor of <code>matcher === value</code>. (Myron Marston)</li>
<li>Deprecate <code>RSpec::Matchers::OperatorMatcher</code> in favor of
<code>RSpec::Matchers::BuiltIn::OperatorMatcher</code>. (Myron Marston)</li>
<li>Deprecate auto-integration with Test::Unit and minitest.
Instead, include <code>RSpec::Matchers</code> in the appropriate test case
base class yourself. (Myron Marston)</li>
<li>Deprecate treating <code>#expected</code> on a DSL-generated custom matcher
as an array when only 1 argument is passed to the matcher method.
In RSpec 3 it will be the single value in order to make diffs
work properly. (Jon Rowe)</li>
</ul>
<h3>2.99.0.beta1 / 2013-11-07</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.14.4...v2.99.0.beta1">Full Changelog</a></p>
<p>Deprecations</p>
<ul>
<li>Deprecate <code>have</code>, <code>have_at_least</code> and <code>have_at_most</code>. You can continue using those
matchers through <a href="https://github.com/rspec/rspec-collection_matchers">https://github.com/rspec/rspec-collection_matchers</a>, or
you can rewrite your expectations with something like
<code>expect(your_object.size).to eq(num)</code>. (Hugo Baraúna)</li>
<li>Deprecate <code>be_xyz</code> predicate matcher when <code>xyz?</code> is a private method.
(Jon Rowe)</li>
<li>Deprecate <code>be_true</code>/<code>be_false</code> in favour of <code>be_truthy</code>/<code>be_falsey</code>
(for Ruby&#39;s conditional semantics) or <code>be true</code>/<code>be false</code>
(for exact equality). (Sam Phippen)</li>
<li>Deprecate calling helper methods from a custom matcher with the wrong
scope. (Myron Marston)
<ul>
<li><code>def self.foo</code> / <code>extend Helper</code> can be used to add macro methods
(e.g. methods that call the custom matcher DSL methods), but should
not be used to define helper methods called from within the DSL
blocks.</li>
<li><code>def foo</code> / <code>include Helper</code> is the opposite: it&#39;s for helper methods
callable from within a DSL block, but not for defining macros.</li>
<li>RSpec 2.x allowed helper methods defined either way to be used for
either purpose, but RSpec 3.0 will not.</li>
</ul></li>
</ul>
<h3>2.14.5 / 2014-02-01</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.14.4...v2.14.5">Full Changelog</a></p>
<p>Bug fixes</p>
<ul>
<li>Fix wrong matcher descriptions with falsey expected value
(yujinakayama)</li>
</ul>
<h3>2.14.4 / 2013-11-06</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.14.3...v2.14.4">Full Changelog</a></p>
<p>Bug fixes</p>
<ul>
<li>Make the <code>match</code> matcher produce a diff output. (Jon Rowe, Ben Moss)</li>
<li>Choose encoding for diff&#39;s more intelligently, and when all else fails fall
back to default internal encoding with replacing characters. (Jon Rowe)</li>
</ul>
<h3>2.14.3 / 2013-09-22</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.14.2...v2.14.3">Full Changelog</a></p>
<p>Bug fixes</p>
<ul>
<li>Fix operator matchers (<code>should</code> syntax) when <code>method</code> is redefined on target.
(Brandon Turner)</li>
<li>Fix diffing of hashes with object based keys. (Jon Rowe)</li>
<li>Fix operator matchers (<code>should</code> syntax) when operator is defined via
<code>method_missing</code> (Jon Rowe)</li>
</ul>
<h3>2.14.2 / 2013-08-14</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.14.1...v2.14.2">Full Changelog</a></p>
<p>Bug fixes</p>
<ul>
<li>Fix <code>be_&lt;predicate&gt;</code> matcher to not support operator chaining like the
<code>be</code> matcher does (e.g. <code>be == 5</code>). This led to some odd behaviors
since <code>be_&lt;predicate&gt; == anything</code> returned a <code>BeComparedTo</code> matcher
and was thus always truthy. This was a consequence of the implementation
(e.g. subclassing the basic <code>Be</code> matcher) and was not intended behavior.
(Myron Marston).</li>
<li>Fix <code>change</code> matcher to compare using <code>==</code> in addition to <code>===</code>. This
is important for an expression like:
<code>expect {}.to change { a.class }.from(ClassA).to(ClassB)</code> because
<code>SomeClass === SomeClass</code> returns false. (Myron Marston)</li>
</ul>
<h3>2.14.1 / 2013-08-08</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.14.0...2.14.1">Full Changelog</a></p>
<p>Bug fixes</p>
<ul>
<li>Ensure diff output uses the same encoding as the encoding of
the string being diff&#39;d to prevent <code>Encoding::UndefinedConversionError</code>
errors (Jon Rowe).</li>
</ul>
<h3>2.14.0 / 2013-07-06</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.14.0.rc1...v2.14.0">Full Changelog</a></p>
<p>Bug fixes</p>
<ul>
<li>Values that are not matchers use <code>#inspect</code>, rather than <code>#description</code> for
documentation output (Andy Lindeman, Sam Phippen).</li>
<li>Make <code>expect(a).to be_within(x).percent_of(y)</code> work with negative y
(Katsuhiko Nishimra).</li>
<li>Make the <code>be_predicate</code> matcher work as expected used with <code>expect{...}.to
change...</code>  (Sam Phippen).</li>
</ul>
<h3>2.14.0.rc1 / 2013-05-27</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.13.0...v2.14.0.rc1">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li>Enhance <code>yield_control</code> so that you can specify an exact or relative
number of times: <code>expect { }.to yield_control.exactly(3).times</code>,
<code>expect { }.to yield_control.at_least(2).times</code>, etc (Bartek
Borkowski).</li>
<li>Make the differ that is used when an expectation fails better handle arrays
by splitting each element of the array onto its own line. (Sam Phippen)</li>
<li>Accept duck-typed strings that respond to <code>:to_str</code> as expectation messages.
(Toby Ovod-Everett)</li>
</ul>
<p>Bug fixes</p>
<ul>
<li>Fix differ to not raise errors when dealing with differently-encoded
strings (Jon Rowe).</li>
<li>Fix <code>expect(something).to be_within(x).percent_of(y)</code> where x and y are both
integers (Sam Phippen).</li>
<li>Fix <code>have</code> matcher to handle the fact that on ruby 2.0,
<code>Enumerator#size</code> may return nil (Kenta Murata).</li>
<li>Fix <code>expect { raise s }.to raise_error(s)</code> where s is an error instance
on ruby 2.0 (Sam Phippen).</li>
<li>Fix <code>expect(object).to raise_error</code> passing. This now warns the user and
fails the spec (tomykaira).</li>
</ul>
<p>Deprecations</p>
<ul>
<li>Deprecate <code>expect { }.not_to raise_error(SpecificErrorClass)</code> or
<code>expect { }.not_to raise_error(&quot;some specific message&quot;)</code>. Using
these was prone to hiding failures as they would allow <em>any other
error</em> to pass. (Sam Phippen and David Chelimsky)</li>
</ul>
<h3>2.13.0 / 2013-02-23</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.12.1...v2.13.0">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li>Add support for percent deltas to <code>be_within</code> matcher:
<code>expect(value).to be_within(10).percent_of(expected)</code>
(Myron Marston).</li>
<li>Add support to <code>include</code> matcher to allow it to be given a list
of matchers as the expecteds to match against (Luke Redpath).</li>
</ul>
<p>Bug fixes</p>
<ul>
<li>Fix <code>change</code> matcher so that it dups strings in order to handle
mutated strings (Myron Marston).</li>
<li>Fix <code>should be =~ /some regex/</code> / <code>expect(...).to be =~ /some regex/</code>.
Previously, these either failed with a confusing <code>undefined method
matches?&#39; for false:FalseClass</code> error or were no-ops that didn&#39;t
actually verify anything (Myron Marston).</li>
<li>Add compatibility for diff-lcs 1.2 and relax the version
constraint (Peter Goldstein).</li>
<li>Fix DSL-generated matchers to allow multiple instances of the
same matcher in the same example to have different description
and failure messages based on the expected value (Myron Marston).</li>
<li>Prevent <code>undefined method #split for Array</code> error when dumping
the diff of an array of multiline strings (Myron Marston).</li>
<li>Don&#39;t blow up when comparing strings that are in an encoding
that is not ASCII compatible (Myron Marston).</li>
<li>Remove confusing &quot;Check the implementation of #==&quot; message
printed for empty diffs (Myron Marston).</li>
</ul>
<h3>2.12.1 / 2012-12-15</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.12.0...v2.12.1">Full Changelog</a></p>
<p>Bug fixes</p>
<ul>
<li>Improve the failure message for an expression like
<code>{}.should =~ {}</code>. (Myron Marston and Andy Lindeman)</li>
<li>Provide a <code>match_regex</code> alias so that custom matchers
built using the matcher DSL can use it (since <code>match</code>
is a different method in that context).
(Steven Harman)</li>
</ul>
<h3>2.12.0 / 2012-11-12</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.11.3...v2.12.0">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li>Colorize diffs if the <code>--color</code> option is configured. (Alex Coplan)</li>
<li>Include backtraces in unexpected errors handled by <code>raise_error</code>
matcher (Myron Marston)</li>
<li>Print a warning when users accidentally pass a non-string argument
as an expectation message (Sam Phippen)</li>
<li><code>=~</code> and <code>match_array</code> matchers output a more useful error message when
the actual value is not an array (or an object that responds to <code>#to_ary</code>)
(Sam Phippen)</li>
</ul>
<p>Bug fixes</p>
<ul>
<li>Fix <code>include</code> matcher so that <code>expect({}).to include(:a =&gt; nil)</code>
fails as it should (Sam Phippen).</li>
<li>Fix <code>be_an_instance_of</code> matcher so that <code>Class#to_s</code> is used in the
description rather than <code>Class#inspect</code>, since some classes (like
<code>ActiveRecord::Base</code>) define a long, verbose <code>#inspect</code>.
(Tom Stuart)</li>
</ul>
<h3>2.11.3 / 2012-09-04</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.11.2...v2.11.3">Full Changelog</a></p>
<p>Bug fixes</p>
<ul>
<li>Fix (and deprecate) <code>expect { }.should</code> syntax so that it works even
though it was never a documented or intended syntax. It worked as a
consequence of the implementation of <code>expect</code> in RSpec 2.10 and
earlier. (Myron Marston)</li>
<li><p>Ensure #== is defined on built in matchers so that they can be composed.
For example:</p>
<p>expect {
  user.emailed!
}.to change { user.last_emailed_at }.to be_within(1.second).of(Time.zone.now)</p></li>
</ul>
<h3>2.11.2 / 2012-07-25</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.11.1...v2.11.2">Full Changelog</a></p>
<p>Bug fixes</p>
<ul>
<li>Define <code>should</code> and <code>should_not</code> on <code>Object</code> rather than <code>BasicObject</code>
on MacRuby. On MacRuby, <code>BasicObject</code> is defined but is not the root
of the object hierarchy. (Gabriel Gilder)</li>
</ul>
<h3>2.11.1 / 2012-07-08</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.11.0...v2.11.1">Full Changelog</a></p>
<p>Bug fixes</p>
<ul>
<li>Constrain <code>actual</code> in <code>be_within</code> matcher to values that respond to <code>-</code> instead
of requiring a specific type.
<ul>
<li><code>Time</code>, for example, is a legit alternative.</li>
</ul></li>
</ul>
<h3>2.11.0 / 2012-07-07</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.10.0...v2.11.0">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li>Expand <code>expect</code> syntax so that it supports expections on bare values
in addition to blocks (Myron Marston).</li>
<li>Add configuration options to control available expectation syntaxes
(Myron Marston):
<ul>
<li><code>RSpec.configuration.expect_with(:rspec) { |c| c.syntax = :expect }</code></li>
<li><code>RSpec.configuration.expect_with(:rspec) { |c| c.syntax = :should }</code></li>
<li><code>RSpec.configuration.expect_with(:rspec) { |c| c.syntax = [:should, :expect] }</code></li>
<li><code>RSpec.configuration.add_should_and_should_not_to Delegator</code></li>
</ul></li>
</ul>
<p>Bug fixes</p>
<ul>
<li>Allow only <code>Numeric</code> values to be the &quot;actual&quot; in the <code>be_within</code> matcher.
This prevents confusing error messages. (Su Zhang @zhangsu)</li>
<li>Define <code>should</code> and <code>should_not</code> on <code>BasicObject</code> rather than <code>Kernel</code>
on 1.9. This makes <code>should</code> and <code>should_not</code> work properly with
<code>BasicObject</code>-subclassed proxy objects like <code>Delegator</code>. (Myron
Marston)</li>
</ul>
<h3>2.10.0 / 2012-05-03</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.9.1...v2.10.0">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li>Add new <code>start_with</code> and <code>end_with</code> matchers (Jeremy Wadsack)</li>
<li>Add new matchers for specifying yields (Myron Marston):
<ul>
<li><code>expect {...}.to yield_control</code></li>
<li><code>expect {...}.to yield_with_args(1, 2, 3)</code></li>
<li><code>expect {...}.to yield_with_no_args</code></li>
<li><code>expect {...}.to yield_successive_args(1, 2, 3)</code></li>
</ul></li>
<li><code>match_unless_raises</code> takes multiple exception args</li>
</ul>
<p>Bug fixes</p>
<ul>
<li>Fix <code>be_within</code> matcher to be inclusive of delta.</li>
<li>Fix message-specific specs to pass on Rubinius (John Firebaugh)</li>
</ul>
<h3>2.9.1 / 2012-04-03</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.9.0...v2.9.1">Full Changelog</a></p>
<p>Bug fixes</p>
<ul>
<li>Provide a helpful message if the diff between two objects is empty.</li>
<li>Fix bug diffing single strings with multiline strings.</li>
<li>Fix for error with using custom matchers inside other custom matchers
(mirasrael)</li>
<li>Fix using execution context methods in nested DSL matchers (mirasrael)</li>
</ul>
<h3>2.9.0 / 2012-03-17</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.8.0...v2.9.0">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li>Move built-in matcher classes to RSpec::Matchers::BuiltIn to reduce pollution
of RSpec::Matchers (which is included in every example).</li>
<li>Autoload files with matcher classes to improve load time.</li>
</ul>
<p>Bug fixes</p>
<ul>
<li>Align <code>respond_to?</code> and <code>method_missing</code> in DSL-defined matchers.</li>
<li>Clear out user-defined instance variables between invocations of DSL-defined
matchers.</li>
<li>Dup the instance of a DSL generated matcher so its state is not changed by
subsequent invocations.</li>
<li>Treat expected args consistently across positive and negative expectations
(thanks to Ralf Kistner for the heads up)</li>
</ul>
<h3>2.8.0 / 2012-01-04</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.8.0.rc2...v2.8.0">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li>Better diff output for Hash (Philippe Creux)</li>
<li>Eliminate Ruby warnings (Olek Janiszewski)</li>
</ul>
<h3>2.8.0.rc2 / 2011-12-19</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.8.0.rc1...v2.8.0.rc2">Full Changelog</a></p>
<p>No changes for this release. Just releasing with the other rspec gems.</p>
<h3>2.8.0.rc1 / 2011-11-06</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.7.0...v2.8.0.rc1">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li>Use classes for the built-in matchers (they&#39;re faster).</li>
<li>Eliminate Ruby warnings (Matijs van Zuijlen)</li>
</ul>
<h3>2.7.0 / 2011-10-16</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.6.0...v2.7.0">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li><code>HaveMatcher</code> converts argument using <code>to_i</code> (Alex Bepple &amp; Pat Maddox)</li>
<li>Improved failure message for the <code>have_xxx</code> matcher (Myron Marston)</li>
<li><code>HaveMatcher</code> supports <code>count</code> (Matthew Bellantoni)</li>
<li>Change matcher dups <code>Enumerable</code> before the action, supporting custom
<code>Enumerable</code> types like <code>CollectionProxy</code> in Rails (David Chelimsky)</li>
</ul>
<p>Bug fixes</p>
<ul>
<li>Fix typo in <code>have(n).xyz</code> documentation (Jean Boussier)</li>
<li>fix <code>safe_sort</code> for ruby 1.9.2 (<code>Kernel</code> now defines <code>&lt;=&gt;</code> for Object) (Peter
van Hardenberg)</li>
</ul>
<h3>2.6.0 / 2011-05-12</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.5.0...v2.6.0">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li><code>change</code> matcher accepts regexps (Robert Davis)</li>
<li>better descriptions for <code>have_xxx</code> matchers (Magnus Bergmark)</li>
<li><code>range.should cover(*values)</code> (Anders Furseth)</li>
</ul>
<p>Bug fixes</p>
<ul>
<li>Removed non-ascii characters that were choking rcov (Geoffrey Byers)</li>
<li>change matcher dups arrays and hashes so their before/after states can be
compared correctly.</li>
<li>Fix the order of inclusion of RSpec::Matchers in Test::Unit::TestCase and
MiniTest::Unit::TestCase to prevent a SystemStackError (Myron Marston)</li>
</ul>
<h3>2.5.0 / 2011-02-05</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.4.0...v2.5.0">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li><code>should exist</code> works with <code>exist?</code> or <code>exists?</code> (Myron Marston)</li>
<li><code>expect { ... }.not_to do_something</code> (in addition to <code>to_not</code>)</li>
</ul>
<p>Documentation</p>
<ul>
<li>improved docs for raise_error matcher (James Almond)</li>
</ul>
<h3>2.4.0 / 2011-01-02</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.3.0...v2.4.0">Full Changelog</a></p>
<p>No functional changes in this release, which was made to align with the
rspec-core-2.4.0 release.</p>
<p>Enhancements</p>
<ul>
<li>improved RDoc for change matcher (Jo Liss)</li>
</ul>
<h3>2.3.0 / 2010-12-12</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.2.1...v2.3.0">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li>diff strings when include matcher fails (Mike Sassak)</li>
</ul>
<h3>2.2.0 / 2010-11-28</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.1.0...v2.2.0">Full Changelog</a></p>
<h3>2.1.0 / 2010-11-07</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.0.1...v2.1.0">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li><code>be_within(delta).of(expected)</code> matcher (Myron Marston)</li>
<li>Lots of new Cucumber features (Myron Marston)</li>
<li>Raise error if you try <code>should != expected</code> on Ruby-1.9 (Myron Marston)</li>
<li>Improved failure messages from <code>throw_symbol</code> (Myron Marston)</li>
</ul>
<p>Bug fixes</p>
<ul>
<li>Eliminate hard dependency on <code>RSpec::Core</code> (Myron Marston)</li>
<li><code>have_matcher</code> - use pluralize only when ActiveSupport inflections are indeed
defined (Josep M Bach)</li>
<li>throw_symbol matcher no longer swallows exceptions (Myron Marston)</li>
<li>fix matcher chaining to avoid name collisions (Myron Marston)</li>
</ul>
<h3>2.0.0 / 2010-10-10</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.0.0.rc...v2.0.0">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li>Add match_for_should_not method to matcher DSL (Myron Marston)</li>
</ul>
<p>Bug fixes</p>
<ul>
<li><code>respond_to</code> matcher works correctly with <code>should_not</code> with multiple methods
(Myron Marston)</li>
<li><code>include</code> matcher works correctly with <code>should_not</code> with multiple values
(Myron Marston)</li>
</ul>
<h3>2.0.0.rc / 2010-10-05</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.0.0.beta.22...v2.0.0.rc">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li><code>require &#39;rspec/expectations&#39;</code> in a T::U or MiniUnit suite (Josep M. Bach)</li>
</ul>
<p>Bug fixes</p>
<ul>
<li>change by 0 passes/fails correctly (Len Smith)</li>
<li>Add description to satisfy matcher</li>
</ul>
<h3>2.0.0.beta.22 / 2010-09-12</h3>
<p><a href="http://github.com/rspec/rspec-expectations/compare/v2.0.0.beta.20...v2.0.0.beta.22">Full Changelog</a></p>
<p>Enhancements</p>
<ul>
<li>diffing improvements
<ul>
<li>diff multiline strings</li>
<li>don&#39;t diff single line strings</li>
<li>don&#39;t diff numbers (silly)</li>
<li>diff regexp + multiline string</li>
</ul></li>
</ul>
<p>Bug fixes
    * <code>should[_not]</code> change now handles boolean values correctly</p>
</div></div>
    <div id="footer">
  Generated on Thu Jan 29 13:56:05 2015 by
  <a href="http://yardoc.org" title="Yay! A Ruby Documentation Tool" target="_parent">yard</a>
  0.8.7.4 (ruby-2.1.5).
</div>
  </body>
</html>