Singleton.new fails tests during CI
Reported by Marnen Laibow-Koser | April 18th, 2008 @ 10:24 AM | in 0.10
When I run rake spec:ci, about every second time or so I get one failed test, as follows:
1)
Singleton.new is a private method FAILED
Expected NoMethodError but no exception was raised
Expectation.fail_with at ./mspec/expectations.rb:10
Object(Proc::Function)#should at ./mspec/expectations/should.rb:6
Object#__script__ {} at ./spec/ruby/1.8/library/singleton/new_spec.rb:6
Proc#call at kernel/core/proc.rb:84
MSpec.protect at ./mspec/runner/mspec.rb:124
RunState#protect {} at ./mspec/runner/state.rb:46
Array#each at kernel/core/array.rb:573
RunState#protect at ./mspec/runner/state.rb:46
RunState#process {} at ./mspec/runner/state.rb:64
Array#each at kernel/core/array.rb:573
RunState#process at ./mspec/runner/state.rb:68
MSpec.describe at ./mspec/runner/mspec.rb:29
Object#describe at ./mspec/runner/object.rb:11
Object#__script__ at ./spec/ruby/1.8/library/singleton/new_spec.rb:4
CompiledMethod#as_script at kernel/core/compiled_method.rb:326
Compile.single_load at kernel/core/compile.rb:233
Compile.unified_load {} at kernel/core/compile.rb:150
Array#each at kernel/core/array.rb:573
Compile.unified_load at kernel/core/compile.rb:120
Kernel.load at kernel/core/compile.rb:390
MSpec.files {} at ./mspec/runner/mspec.rb:45
Proc#call at kernel/core/proc.rb:84
MSpec.protect at ./mspec/runner/mspec.rb:120
MSpec.files {} at ./mspec/runner/mspec.rb:46
Array#each at kernel/core/array.rb:573
MSpec.files at ./mspec/runner/mspec.rb:42
MSpec.process at ./mspec/runner/mspec.rb:35
MSpecCI#run at mspec/bin/mspec-ci:66
MSpecCI.main at ./mspec/bin/script.rb:71
Object#__script__ at mspec/bin/mspec-ci:73
CompiledMethod#as_script at kernel/core/compiled_method.rb:326
Compile.single_load at kernel/core/compile.rb:233
Compile.load_from_extension at kernel/core/compile.rb:310
Object#__script__ at kernel/loader.rb:200
I'm using Mac OS X 10.5.2 (Intel). So far I haven't tried it in another build environment. I plan to investigate over the next few days, but if anyone wants to lend a hand, I'd appreciate it.
Comments and changes to this ticket
-
Federico Builes April 18th, 2008 @ 07:32 PM
I'm not able to reproduce this under 10.5.2 running HEAD. Have you tried running "rake distclean" before running the specs? If you haven't, can you try it and update us?
Thanks.
-

Marnen Laibow-Koser April 21st, 2008 @ 04:52 PM
rake distclean has not helped, but it certainly doesn't look like these failures are reproducible in any other build environment. Weird.
-
Wilson Bilkovich April 25th, 2008 @ 05:54 PM
- → Assigned user changed from to Wilson Bilkovich
I made the Singleton.new spec use a different class to avoid cross-contamination.
Anybody still seeing this?
-

-

Marnen Laibow-Koser April 25th, 2008 @ 08:50 PM
Apparently I lied. I got the error just after posting that. Let me pull and rebuild once more.
-

Marnen Laibow-Koser April 25th, 2008 @ 08:51 PM
Ah, OK. I just got your changes in 71fe2d45d147fe2c41937ae5ef6dbb8814f491c4. We'll see what happens.
-

Marnen Laibow-Koser April 25th, 2008 @ 09:04 PM
Even with your changes, followed by a distclean, I got the same issue on the second run of rake spec:ci. Oy.
-
-
Wilson Bilkovich June 25th, 2008 @ 10:25 AM
- → Tag changed from to build ci spec
- → Milestone changed from 1.0 preview to 0.10
- → Assigned user changed from Wilson Bilkovich to Brian Ford
-
Brian Ford July 15th, 2008 @ 12:27 AM
- → State changed from open to invalid
I have been unable to reproduce this. I'm on OS X 10.5.4 intel.
If you encounter this failure, please try running with
bin/mspec ci -B full, andbin/mspec ci spec/ruby/1.8/library/singletonand indicate whether it fails for either of those commands or only with the rake command.Closing for now; reopen if necessary.
Please Login or create a free account to add a new comment.
You can update this ticket by sending an email to from your email client. (help)
Create your profile
Help contribute to this project by taking a few moments to create your personal profile. Create your profile »
