Move remaining core specs to use fewer shared examples by Earlopain · Pull Request #1369 · ruby/spec

Earlopain · 2026-06-01T17:15:16Z

It's quite chonky but mostly more of the same for #1364. This converts the entirety of core specs.

Before:
3805 files, 35698 examples, 273269 expectations, 0 failure, 0 errors, 0 tagged

After:
3805 files, 34586 examples, 256944 expectations, 0 failure, 0 errors, 0 tagged

1112 fewer specs.

Copilot

Copilot wasn't able to review this pull request because it exceeds the maximum number of files (300). Try reducing the number of changed files and requesting a review from Copilot again.

Earlopain · 2026-06-01T19:29:03Z

Let me split this up in two PRs then

Copilot

Pull request overview

Copilot reviewed 150 out of 150 changed files in this pull request and generated 4 comments.

Copilot

Pull request overview

Copilot reviewed 150 out of 150 changed files in this pull request and generated 4 comments.

This shared example didn't even care about the method. So I just mentioned in one of the methods that tests are found in a different file

Copilot

Pull request overview

Copilot reviewed 150 out of 150 changed files in this pull request and generated 1 comment.

Comments suppressed due to low confidence (5)

core/io/tty_spec.rb:1

This spec can be flaky across CI/container environments because /dev/tty may be missing or inaccessible (e.g., Errno::ENOENT, Errno::EACCES, Errno::ENODEV), not just Errno::ENXIO. Consider broadening the rescued errors (or guarding by checking existence/permissions) so the spec reliably skips when TTY access is not available.
core/file/to_path_spec.rb:1
This platform_is/guard block defines before/after hooks but contains no examples that use @dir. Either add the missing TMPFILE-related examples within this scope or remove the unused hooks to avoid dead setup code.
core/float/fdiv_spec.rb:1
The second example’s description duplicates the first one verbatim, which makes failures harder to interpret. Rename the second description to reflect the distinct intent (e.g., division by a very large Integer).
core/kernel/dup_spec.rb:1
The spec that verified #dup calls #initialize_copy was removed, which drops coverage for an important and observable part of Ruby’s object-copying semantics. Consider re-adding an explicit example asserting initialize_copy is invoked (or adding equivalent coverage elsewhere in the same spec suite) so regressions in dup hooks are caught.
core/kernel/clone_spec.rb:1
Similarly to dup_spec, removing the initialize_copy expectation for #clone reduces coverage for clone’s copy hook behavior. Re-introduce an example that asserts initialize_copy is called during clone to keep coverage of this user-visible contract.

eregon

Great stuff and some of the deduplication notably for Marshal will be very nice when improving those specs.

eregon · 2026-06-02T19:36:27Z

 describe "ENV.merge!" do
-  it_behaves_like :env_update, :merge!
+  it "is an alias of ENV.update" do
+    ENV.method(:merge!).should == ENV.method(:update)


Minor: I think merge! is more common than update (for Hash and ENV)

eregon · 2026-06-02T19:38:39Z

 describe "File.delete" do
-  it_behaves_like :file_unlink, :delete
+  it "is an alias of File.unlink" do
+    File.method(:delete).should == File.method(:unlink)


Minor: I think File.delete is more common than File.unlink

eregon · 2026-06-02T19:46:23Z

-  it "calls #initialize_copy on the new instance" do
-    clone = @obj.clone
-    ScratchPad.recorded.should_not == @obj.object_id
-    ScratchPad.recorded.should == clone.object_id
-  end


Removed because it's a bad spec and in fact clone calls initialize_clone, which calls initialize_copy?
Also there are specs that initialize_clone is called below, so this is fine.

eregon · 2026-06-02T19:48:10Z

-  it "calls #initialize_copy on the new instance" do
-    dup = @obj.dup
-    ScratchPad.recorded.should_not == @obj.object_id
-    ScratchPad.recorded.should == dup.object_id
-  end


Removed because it's a bad spec and in fact clone calls initialize_dup, which calls initialize_copy?
Seems there is already another spec checking for initialize_copy but not for initialize_dup, but well that's out of scope.
I wonder why so many of these specs use custom objects instead of mocks though, seems unnecessarily obscure. Maybe because MSpec didn't have mocks back then (just a guess)?

eregon · 2026-06-02T19:53:37Z


 describe "MatchData#==" do
-  it_behaves_like :matchdata_eql, :==
+  it "is an alias of MatchData#eql?" do


Minor: == is more frequent than eql? (and eql? for core types is is separate than == for probably only numeric types and maybe 1-2 other classes)

eregon · 2026-06-02T19:54:32Z

 describe "Method#==" do
-  it_behaves_like :method_equal, :==
+  it "is an alias of Method#eql?" do
+    Method.instance_method(:==).should == Method.instance_method(:eql?)


Same, == is the main one I'd say and eql? the alias

eregon · 2026-06-02T19:55:30Z

-  it_behaves_like :method_to_s, :to_s
-  it_behaves_like :method_to_s_aliased, :to_s, -> meth { meth }
+  it "is an alias of Method#inspect" do
+    Method.instance_method(:to_s).should == Method.instance_method(:inspect)


For Float we have inspect is alias of to_s, I think that's good because to_s is more frequent than inspect, could you follow the pattern used for Float for other classes then?

Earlopain force-pushed the fewer-shared-specs-2 branch 2 times, most recently from f311e61 to f51cf1d Compare June 1, 2026 17:50

eregon requested a review from Copilot June 1, 2026 19:15

Copilot AI reviewed Jun 1, 2026

View reviewed changes

Earlopain mentioned this pull request Jun 1, 2026

Move remaining core specs to use fewer shared examples - Part 2 #1370

Open

Earlopain force-pushed the fewer-shared-specs-2 branch from f51cf1d to 304b83a Compare June 1, 2026 19:34

eregon requested a review from Copilot June 1, 2026 19:52

Copilot started reviewing on behalf of eregon June 1, 2026 19:52 View session

Copilot AI reviewed Jun 1, 2026

View reviewed changes

Comment thread core/enumerable/find_spec.rb

Comment thread core/matchdata/size_spec.rb Outdated

Comment thread core/file/fnmatch_spec.rb Outdated

Comment thread core/matchdata/equal_value_spec.rb Outdated

eregon reviewed Jun 1, 2026

View reviewed changes

Comment thread core/enumerable/map_spec.rb Outdated

Earlopain force-pushed the fewer-shared-specs-2 branch from 304b83a to 83885f1 Compare June 1, 2026 20:11

eregon requested a review from Copilot June 1, 2026 20:19

Copilot started reviewing on behalf of eregon June 1, 2026 20:19 View session

Copilot AI reviewed Jun 1, 2026

View reviewed changes

Comment thread core/kernel/to_enum_spec.rb

Comment thread core/kernel/to_enum_spec.rb

Comment thread core/env/select_spec.rb Outdated

Comment thread core/env/select_spec.rb Outdated

Earlopain added 13 commits June 2, 2026 13:20

Move Enumerable to rely less on shared examples

b66405a

Move Enumerator to rely less on shared examples

24d340a

Move ENV to rely less on shared examples

61b3a39

Move Fiber to rely less on shared examples

01df840

This shared example didn't even care about the method. So I just mentioned in one of the methods that tests are found in a different file

Move File to rely less on shared examples

69a3574

Move Float to rely less on shared examples

4599ae2

Move Integer to rely less on shared examples

5a7cabe

Move IO to rely less on shared examples

21ab592

Move Kernel to rely less on shared examples

4c6ed26

Move Marshal to rely less on shared examples

c2da21f

Move MatchData to rely less on shared examples

f25d4b4

Move Method to rely less on shared examples

1b95566

Move Module to rely less on shared examples

a62129f

Earlopain force-pushed the fewer-shared-specs-2 branch from 83885f1 to a62129f Compare June 2, 2026 11:20

Earlopain requested a review from Copilot June 2, 2026 11:20

Copilot AI reviewed Jun 2, 2026

View reviewed changes

Comment thread core/enumerable/reduce_spec.rb

eregon reviewed Jun 2, 2026

View reviewed changes

Uh oh!

Conversation

Earlopain commented Jun 1, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

Earlopain commented Jun 1, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

eregon left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

eregon Jun 2, 2026

Choose a reason for hiding this comment

Uh oh!

eregon Jun 2, 2026

Choose a reason for hiding this comment

Uh oh!

eregon Jun 2, 2026

Choose a reason for hiding this comment

Uh oh!

eregon Jun 2, 2026

Choose a reason for hiding this comment

Uh oh!

eregon Jun 2, 2026

Choose a reason for hiding this comment

Uh oh!

eregon Jun 2, 2026

Choose a reason for hiding this comment

Uh oh!

eregon Jun 2, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants