Spec failures with the rails.yaml profile
Reported by Brian Ford | May 6th, 2008 @ 02:47 PM | in 1.0 preview
Running this command: bin/mspec -w rails.yaml spec/ruby gives the following failures (summary line provided for those following along at home). Anyone can tackle these. Suggested approach would be to open a ticket for a particular failure and edit this body to add strike-through (or add a comment) when fixed.
(Note: the profile data was provided by John Lam. It has not been verified. The errors/failures below are not necessarily failures blocking Rails from running since they may describing aspects of the method's behavior that are not relied on or may be using data from the domain of the method outside of that used by Rails.)
rubinius 0.8.0 (ruby 1.8.6 compatible) (af3407251) (05/06/2008) [i686-apple-darwin9.2.2]
1958 files, 1887 examples, 6772 expectations, 20 failures, 3 errors
DONE 1)
File.expand_path raises an ArgumentError if the path is not valid FAILED
Expected ArgumentError but no exception was raised
DONE 2)
File.expand_path expands ~ENV['USER'] to the user's home directory FAILED
Expected "/Users/brian/devel/rubinius/~brian"
to equal "/Users/brian"
3) (Eric Hodel)
File.fnmatch matches multiple directories with ** when flags includes File::FNM_PATHNAME FAILED
Expected false
to equal true
DONE 4)
IO#close makes the stream unavailable for any further data operations FAILED
Expected IOError but no exception was raised
DONE 5)
IO#close sets $? if the stream is opened by IO.popen FAILED
Expected #<Process::Status:0x2c79 @pid=18793 @status=1>
to equal 0
DONE 6)
IO#eof? raises IOError on stream closed for reading by close_read ERROR
No method 'close_read' on an instance of File.
DONE 7)
IO#gets returns tainted strings FAILED
Expected false
to equal true
DONE 8)
IO#gets updates $. with each invocation FAILED
Expected 0
to equal 1
9) (ignore for now)
IO.new emits a warning if given a block FAILED
Expected warning to match: /IO::new.*does not take block.*IO::open.*instead/
10)
IO#sync= sets the sync mode to true or false FAILED
Expected true
to equal false
11)
IO#sync= accepts non-boolean arguments FAILED
Expected true
to equal false
12)
IO#write on a file writes all of the string's bytes but buffers them FAILED
Expected "abcde56789"
not to equal "abcde56789"
13)
Kernel#instance_eval raises a TypeError when defining methods on numerics FAILED
Expected TypeError but no exception was raised
DONE 14)
Module#const_defined? returns true if constant with the given String is defined in its parent FAILED
Expected false
to equal true
DONE 15)
Module#const_get returns the value of the constant when a scoped constant name ERROR
Missing or uninitialized constant: Super::SuperChild
DONE 16)
Module#module_function as a toggle (no arguments) in a Module body affects evaled method definitions also even when outside the eval itself FAILED
Expected false
to equal true
DONE 17)
Module#module_function as a toggle (no arguments) in a Module body affects definitions when inside an eval even if the definitions are outside of it FAILED
Expected false
to equal true
DONE 18)
Module#private_class_method makes an existing class method private FAILED
Expected NoMethodError but no exception was raised
DONE 19)
Module#private_class_method makes an existing class method private up the inheritance tree FAILED
Expected NoMethodError but no exception was raised
DONE 20)
Module#private should make the target method uncallable from other types FAILED
Expected NoMethodError but no exception was raised
21)
Proc.new returns from within a method ERROR
LongReturnException
22)
String#inspect can handle malformed UTF-8 string when $KCODE is UTF-8 FAILED
Expected ""äöüzÃ""
to equal ""\aäöüz\303""
DONE 23)
Marshal.load on singleton object previously serialised in this process returns the singleton instance FAILED
Expected false
to equal true
Comments and changes to this ticket
-
Ryan Davis May 7th, 2008 @ 04:36 PM
- → Assigned user changed from to Ryan Davis
-
-
-
Ryan Davis May 12th, 2008 @ 03:15 PM
I'm leaving all the IO bugs to eric.
I'm currently working up from the bottom.
-
Ryan Davis May 12th, 2008 @ 03:21 PM
23/Marshal.load fixed by virtue of removing the offending spec.
-
Ryan Davis May 12th, 2008 @ 03:25 PM
Hrm... when I run with -tr I get:
1) IO#read can read lots of data ERROR Exception occurred during: after :each Bad file descriptor - test.txt ./spec/ruby/1.8/core/io/read_spec.rb:92:in `close' ./spec/ruby/1.8/core/io/read_spec.rb:92 -
-
-
-
-
-
-
Wilson Bilkovich May 16th, 2008 @ 06:32 PM
I get:
2027 files, 1902 examples, 6842 expectations, 7 failures, 2 errors
right now.. does that match everybody else?
-
Ryan Davis July 10th, 2008 @ 02:07 PM
- → Tag changed from to bug core rails spec
- → State changed from open to resolved
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 »
