Scrape out monkey patching #1245

pirj · 2020-12-18T19:35:35Z

Sibling PRs:

Prerequisite PRs:

Stop using globally exposed DSL rspec-rails#2420
Use globally unexposed DSL rspec-support#457
Add a check to support both RSpec 3 & 4 rspec-rails#2425
Remove temporary commit. CI is expected to fail on Core/Mocks/Support/Rails steps, as disable_monkey_patching! calls methods that were removed with no replacement. I'll use the maintenance-branch temporary commit hack to fraternize those sibling PRs.

http://rspec.info/blog/2013/07/the-plan-for-rspec-3/#zero_monkey_patching_mode

we do want to encourage people to switch to the new syntax, so we plan to make RSpec 3 print a warning on first usage of any the old syntax methods (should, should_not, should_receive, etc) unless the should syntax has been explicitly enabled. This should nudge folks towards the new syntax while keeping RSpec friendly to new users and will pave the way for the old syntax to be disabled by default in RSpec 4.

zero-monkey-patching mode for RSpec... We plan for these config options to become the defaults in RSpec 4.0, so that RSpec 4.0 will have zero monkey patching out of the box.

As for "disabled by default" vs "completely removed" and "default, out of the box" vs "impossible" I can only say that RSpec 4 was probably planned to be released earlier, as:

we'll probably be dropping support for 1.8.7 in RSpec 4

but we've also dropped 1.9, 2.0, 2.1 and 2.2

rspec/rspec-core#2301 (comment)

In RSpec 4, we plan to extract all monkey patching from RSpec and move it into a separate gem, so that monkey patching is opt-in instead of opt-out and users have to explicitly install and load a gem to get it.

rspec-should (or rspec-monkey as it's also about exposing example group DSL in the top-level/Module?) will be released later.

Those using the monkey-patched should syntax are not encouraged to update to RSpec 4 until this gem is extracted.

Those using the globally-exposed DSL are encouraged to use RSpec.describe/RSpec.shared_examples_for instead.

README.md

Changelog.md

features/built_in_matchers/README.md

maintenance-branch

lib/rspec/expectations/expectation_target.rb

spec/rspec/expectations/extensions/kernel_spec.rb

Changelog.md

pirj · 2020-12-25T14:17:43Z

      NoMethodError:
       undefined method `syntax=' for #<RSpec::Expectations::Configuration:0x000055ba304c1f18>
     # ./spec/rspec/core/configuration_spec.rb:2729:in `block in in_fully_monkey_patched_rspec_environment'

4-0-dev is ATM more like main in terms of monkey-patching. Are you aiming to make 4.0 somehow compatible with it?

I suggest following our practice:

Remove the commit from step 2. We will merge ignoring the failure.
Remove the commit from the other, check it passes with the other commit now on main.
Merge the other.
We will trigger builds for the ~~main~~ 4-0-dev branch of affected repositories to check if everything is in order.

pirj · 2020-12-26T18:58:02Z

Green 💚

pirj · 2020-12-26T19:05:23Z

Filed minitest/minitest#861

JonRowe · 2020-12-26T19:42:45Z

Filed minitest/minitest#861

I've already re-worked our build around it, its just the PR isn't merged yet

pirj · 2020-12-26T19:50:44Z

Me too minitest/minitest#862 😄

pirj · 2020-12-29T17:39:07Z

Anything left to be done here?

JonRowe · 2020-12-30T08:21:18Z

spec/rspec/expectations/extensions/kernel_spec.rb

-  it "accepts and interacts with a matcher" do
-    expect(@matcher).to receive(:matches?).with(@target).and_return(true)
-    expect(@target).to @matcher
-  end
-
-  it "asks for a failure_message when matches? returns false" do
-    expect(@matcher).to receive(:matches?).with(@target).and_return(false)
-    expect(@matcher).to receive(:failure_message).and_return("the failure message")
-    expect {
-      expect(@target).to @matcher
-    }.to fail_with("the failure message")
-  end


Did you find duplicates of these? I thought you were moving them (if so a comment saying "these are duplicates of file:line" etc is helpful when reviewing).

pirj · 2020-12-30T12:11:19Z

spec/rspec/expectations/extensions/kernel_spec.rb

-  it "accepts and interacts with a matcher" do
-    expect(@matcher).to receive(:matches?).with(@target).and_return(true)
-    expect(@target).to @matcher
-  end


This is covered by more specific specs:

rspec-expectations/spec/rspec/expectations/handler_spec.rb

Line 77 in bf9db76

expect(RSpec::Expectations::PositiveExpectationHandler.handle_matcher(actual, matcher)).to eq :this_value

rspec-expectations/spec/rspec/matchers/dsl_spec.rb

Line 825 in bf9db76

it "overrides the failure message for positive expectations" do

pirj · 2020-12-30T12:14:07Z

spec/rspec/expectations/extensions/kernel_spec.rb

-  it "asks for a failure_message when matches? returns false" do
-    expect(@matcher).to receive(:matches?).with(@target).and_return(false)
-    expect(@matcher).to receive(:failure_message).and_return("the failure message")
-    expect {
-      expect(@target).to @matcher
-    }.to fail_with("the failure message")
-  end


Covered in:

rspec-expectations/spec/rspec/expectations/handler_spec.rb

Line 99 in bf9db76

expect(::RSpec::Expectations).to receive(:fail_with).with("message", 1, 2)

rspec-expectations/spec/rspec/matchers/dsl_spec.rb

Line 825 in bf9db76

it "overrides the failure message for positive expectations" do

pirj · 2020-12-30T12:14:33Z

spec/rspec/expectations/extensions/kernel_spec.rb

-    }.to fail_with("the failure message")
-  end
-
-  context "on interpretters that have BasicObject" do


We do not support such interpreters.

pirj · 2020-12-30T12:15:25Z

spec/rspec/expectations/extensions/kernel_spec.rb

-    it 'does not break the deprecation check on BasicObject-subclassed proxy objects' do
-      begin
-        should_enabled = RSpec::Expectations::Syntax.should_enabled?
-        RSpec::Expectations::Syntax.enable_should unless should_enabled
-        proxy_class.new(BasicObject.new).should be_proxied
-      ensure
-        RSpec::Expectations::Syntax.disable_should if should_enabled
-      end


We are dropping support for monkey-patching should.

pirj · 2020-12-30T12:17:10Z

spec/rspec/expectations/extensions/kernel_spec.rb

-  it "accepts and interacts with a matcher" do
-    expect(@matcher).to receive(:matches?).with(@target).and_return(false)
-    allow(@matcher).to receive(:failure_message_when_negated)
-
-    expect(@target).not_to @matcher
-  end


Covered in:

rspec-expectations/spec/rspec/matchers/dsl_spec.rb

Line 455 in bf9db76

expect(matcher.failure_message_when_negated).to eq "expected 77 not to to be composed of 7 and 11"

rspec-expectations/spec/rspec/expectations/handler_spec.rb

Line 163 in bf9db76

matcher = double("matcher", :failure_message_when_negated => "Error!")

pirj · 2020-12-30T12:18:27Z

spec/rspec/expectations/extensions/kernel_spec.rb

-  it "asks for a failure_message_when_negated when matches? returns true" do
-    expect(@matcher).to receive(:matches?).with(@target).and_return(true)
-    expect(@matcher).to receive(:failure_message_when_negated).and_return("the failure message for should not")
-    expect {
-      expect(@target).not_to @matcher
-    }.to fail_with("the failure message for should not")
-  end


Covered in:

rspec-expectations/spec/rspec/expectations/handler_spec.rb

Line 182 in bf9db76

it "raises an expectation failure when `matches?` returns true" do

References to duplicated examples for a removed spec added

http://rspec.info/blog/2013/07/the-plan-for-rspec-3/#zero_monkey_patching_mode > we do want to encourage people to switch to the new syntax, so we plan to make RSpec 3 print a warning on first usage of any the old syntax methods (should, should_not, should_receive, etc) unless the should syntax has been explicitly enabled. This should nudge folks towards the new syntax while keeping RSpec friendly to new users and will pave the way for the old syntax to be disabled by default in RSpec 4. > zero-monkey-patching mode for RSpec... We plan for these config options to become the defaults in RSpec 4.0, so that RSpec 4.0 will have zero monkey patching out of the box. As for "disabled by default" vs "completely removed" and "default, out of the box" vs "impossible" I can only say that RSpec 4 was probably planned to be released earlier, as: > we'll probably be dropping support for 1.8.7 in RSpec 4 but we've also dropped 1.9, 2.0, 2.1 and 2.2 rspec/rspec-core#2301 (comment) > In RSpec 4, we plan to extract all monkey patching from RSpec and move it into a separate gem, so that monkey patching is opt-in instead of opt-out and users have to explicitly install and load a gem to get it. `rspec-should` (or `rspec-monkey` as it's also about exposing example group DSL in the top-level/Module?) will be released later. Those using the monkey-patched `should` syntax are not encouraged to update to RSpec 4 until this gem is extracted. Those using the globally-exposed DSL are encouraged to use `RSpec.describe`/`RSpec.shared_examples_for` instead.

This commit was imported from rspec/rspec-expectations@566311c.

…rspec/remove-monkey-patching Scrape out monkey patching --- This commit was imported from rspec/rspec-expectations@ebab7e2.

pirj self-assigned this Dec 18, 2020

pirj force-pushed the remove-monkey-patching branch 2 times, most recently from d22a6f9 to fbd4401 Compare December 18, 2020 21:39

pirj mentioned this pull request Dec 18, 2020

Scrape out monkey patching rspec/rspec-mocks#1365

Merged

1 task

pirj force-pushed the remove-monkey-patching branch from fbd4401 to 133e5a8 Compare December 18, 2020 22:58

pirj mentioned this pull request Dec 19, 2020

Scrape out monkey patching rspec/rspec-core#2803

Merged

1 task

pirj marked this pull request as ready for review December 19, 2020 00:52

pirj requested review from JonRowe and benoittgt December 19, 2020 00:52

pirj mentioned this pull request Dec 19, 2020

Stop using globally exposed DSL rspec/rspec-rails#2420

Merged

pirj force-pushed the remove-monkey-patching branch 2 times, most recently from eb8c832 to 1071be7 Compare December 21, 2020 23:15

pirj mentioned this pull request Dec 21, 2020

Use globally unexposed DSL rspec/rspec-support#457

Merged

pirj force-pushed the remove-monkey-patching branch from 056c92f to 1071be7 Compare December 22, 2020 09:15

JonRowe reviewed Dec 22, 2020

View reviewed changes

README.md Outdated Show resolved Hide resolved

pirj force-pushed the remove-monkey-patching branch from 61c0bfc to 8f3c3da Compare December 23, 2020 16:50

pirj mentioned this pull request Dec 23, 2020

Add a check to support both RSpec 3 & 4 rspec/rspec-rails#2425

Merged

pirj requested a review from JonRowe December 23, 2020 17:36

pirj force-pushed the remove-monkey-patching branch from 4504b25 to 8f3c3da Compare December 23, 2020 20:49

JonRowe reviewed Dec 23, 2020

View reviewed changes

Changelog.md Outdated Show resolved Hide resolved

features/built_in_matchers/README.md Outdated Show resolved Hide resolved

JonRowe reviewed Dec 23, 2020

View reviewed changes

maintenance-branch Outdated Show resolved Hide resolved

JonRowe reviewed Dec 23, 2020

View reviewed changes

lib/rspec/expectations/expectation_target.rb Outdated Show resolved Hide resolved

pirj force-pushed the remove-monkey-patching branch 2 times, most recently from 3daf73c to 73c60f5 Compare December 23, 2020 23:26

pirj commented Dec 23, 2020

View reviewed changes

spec/rspec/expectations/extensions/kernel_spec.rb Outdated Show resolved Hide resolved

pirj commented Dec 23, 2020

View reviewed changes

Changelog.md Outdated Show resolved Hide resolved

pirj mentioned this pull request Dec 24, 2020

Remove support for legacy RSpec matchers (pre 3) #1253

Merged

pirj added the rspec 4 label Dec 24, 2020

pirj force-pushed the remove-monkey-patching branch 2 times, most recently from b330730 to 2926bc9 Compare December 25, 2020 14:02

pirj mentioned this pull request Dec 26, 2020

Skip config dependant specs rspec/rspec-core#2819

Merged

pirj force-pushed the remove-monkey-patching branch from 2926bc9 to bbaef45 Compare December 26, 2020 18:45

JonRowe added a commit that referenced this pull request Dec 27, 2020

Pre #1245 workaround

566311c

pirj force-pushed the remove-monkey-patching branch from bbaef45 to 6996e4f Compare December 27, 2020 19:42

pirj requested a review from JonRowe December 27, 2020 19:50

pirj mentioned this pull request Dec 30, 2020

Remove shared_context_metadata_behavior config option and make apply_to_host_groups the default rspec/rspec-core#2834

Merged

JonRowe previously requested changes Dec 30, 2020

View reviewed changes

pirj commented Dec 30, 2020

View reviewed changes

pirj requested a review from JonRowe December 30, 2020 12:18

JonRowe force-pushed the remove-monkey-patching branch 3 times, most recently from d983376 to 1ab4671 Compare December 31, 2020 14:20

pirj added 2 commits December 31, 2020 14:50

Remove non-monkey-patching should syntax, too

e8e8bfc

JonRowe force-pushed the remove-monkey-patching branch from 1ab4671 to e8e8bfc Compare December 31, 2020 14:50

JonRowe merged commit ebab7e2 into 4-0-dev Dec 31, 2020

JonRowe deleted the remove-monkey-patching branch December 31, 2020 15:33

benoittgt mentioned this pull request Jun 1, 2021

Add deprecations warnings for RSpec 4 #1302

Open

pirj mentioned this pull request Jul 18, 2021

Add deprecation warnings for features to be removed/changed in RSpec 4 #1301

Open

9 tasks

yujinakayama pushed a commit to yujinakayama/rspec-monorepo that referenced this pull request Oct 19, 2021

[expectations] Pre rspec/rspec-expectations#1245 workaround

2da8714

This commit was imported from rspec/rspec-expectations@566311c.

pirj mentioned this pull request Jul 11, 2022

Add RSpec 4 prerelease CI job and make the test suite compatible with it rubocop/rubocop#10806

Merged

5 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Scrape out monkey patching #1245

Scrape out monkey patching #1245

pirj commented Dec 18, 2020 •

edited

Loading

pirj commented Dec 25, 2020

pirj commented Dec 26, 2020

pirj commented Dec 26, 2020

JonRowe commented Dec 26, 2020

pirj commented Dec 26, 2020

pirj commented Dec 29, 2020

JonRowe Dec 30, 2020

pirj Dec 30, 2020

pirj Dec 30, 2020

pirj Dec 30, 2020

pirj Dec 30, 2020

pirj Dec 30, 2020

pirj Dec 30, 2020

Scrape out monkey patching #1245

Scrape out monkey patching #1245

Conversation

pirj commented Dec 18, 2020 • edited Loading

pirj commented Dec 25, 2020

pirj commented Dec 26, 2020

pirj commented Dec 26, 2020

JonRowe commented Dec 26, 2020

pirj commented Dec 26, 2020

pirj commented Dec 29, 2020

JonRowe Dec 30, 2020

Choose a reason for hiding this comment

pirj Dec 30, 2020

Choose a reason for hiding this comment

pirj Dec 30, 2020

Choose a reason for hiding this comment

pirj Dec 30, 2020

Choose a reason for hiding this comment

pirj Dec 30, 2020

Choose a reason for hiding this comment

pirj Dec 30, 2020

Choose a reason for hiding this comment

pirj Dec 30, 2020

Choose a reason for hiding this comment

pirj commented Dec 18, 2020 •

edited

Loading