February 2011 - macruby-tickets - lists.macosforge.org

[MacRuby] #1139: MacRuby inconsistent compared to CRuby when breaking from hash enumeration
by MacRuby 29 Mar '11

29 Mar '11

#1139: MacRuby inconsistent compared to CRuby when breaking from hash enumeration ------------------------------------+--------------------------------------- Reporter: mrada@… | Owner: lsansonetti@… Type: defect | Status: new Priority: blocker | Milestone: Component: MacRuby | Keywords: ------------------------------------+--------------------------------------- When I iterate over a hash in CRuby and `break` during one iteration the iteration will return value to be nil. If I try the same thing in MacRuby I will get the same return value that I would have gotten had the iteration finished. I found this problem with Hash objects, an Array has the same behaviour in this case on CRuby and MacRuby. With the CRuby irb: ruby-1.9.2-p136 :001 > { test:'pair' }.each_pair { |key,value| break } => nil ruby-1.9.2-p136 :002 > { test:'pair' }.each_pair { |key,value| key } => {:test=>"pair"} ruby-1.9.2-p136 :003 > With the MacRuby irb: irb(main):001:0> { test:'pair' }.each_pair { |key,value| break } => {:test=>"pair"} irb(main):002:0> { test:'pair' }.each_pair { |key,value| key } => {:test=>"pair"} irb(main):003:0> -- Ticket URL: <http://www.macruby.org/trac/ticket/1139> MacRuby <http://macruby.org/>

1 1

[MacRuby] #1141: NSArray.new.class should be "NSArray"
by MacRuby 09 Mar '11

09 Mar '11

#1141: NSArray.new.class should be "NSArray" ----------------------------------------+----------------------------------- Reporter: vincent.isambart@… | Owner: lsansonetti@… Type: defect | Status: new Priority: critical | Milestone: MacRuby 1.0 Component: MacRuby | Keywords: ----------------------------------------+----------------------------------- Even though NSArray's behavior is different from Array, Array.new.class, NSArray.new.class and NSMutableArray.new.class all return "Array". The user should be able to easily differentiate between Ruby arrays and Cocoa arrays so NSArray instances should not pretend to be Array instances. We should also consider doing it for NSDictionary and NSString. -- Ticket URL: <http://www.macruby.org/trac/ticket/1141> MacRuby <http://macruby.org/>

1 1

[MacRuby] #881: Abort occurs when calls Array#sort with array including Object, Using 32bit arch.
by MacRuby 08 Mar '11

08 Mar '11

#881: Abort occurs when calls Array#sort with array including Object, Using 32bit arch. ----------------------------------+----------------------------------------- Reporter: watson1978@… | Owner: lsansonetti@… Type: defect | Status: new Priority: blocker | Milestone: Component: MacRuby | Keywords: ----------------------------------+----------------------------------------- {{{ #!ruby # spec/frozen/core/array/sort_spec.rb # it "does not deal with exceptions raised by unimplemented or incorrect #<=>" do o = Object.new [o, 1].sort }}} Result: {{{ $ DYLD_LIBRARY_PATH=. ./macruby -v -I./lib t1.rb MacRuby 0.7 (ruby 1.9.2) [universal-darwin10.0, i386] terminate called without an active exception zsh: abort DYLD_LIBRARY_PATH=. ./macruby -v -I./lib t1.rb }}} -- Ticket URL: <http://www.macruby.org/trac/ticket/881> MacRuby <http://macruby.org/>

1 5

[MacRuby] #863: String#succ is broken.
by MacRuby 08 Mar '11

08 Mar '11

#863: String#succ is broken. ----------------------------------+----------------------------------------- Reporter: watson1978@… | Owner: lsansonetti@… Type: defect | Status: new Priority: minor | Milestone: Component: MacRuby | Keywords: ----------------------------------+----------------------------------------- Test Script: {{{ #!ruby $ cat test_succ.rb p "No.9".succ # => "No.10" p "-9".succ # => "-10" p "+9".succ # => "+10" p "\377".succ # => "\001\000" }}} Result of Ruby 1.9.1: {{{ $ ruby test_succ.rb "No.10" "-10" "+10" "\x01\x00" }}} Result of MacRuby Trunk: {{{ $ macruby test_succ.rb "Np.0" "1-0" "1+0" /Users/watson/src/MacRuby/test_succ.rb:1:in `<main>': cannot make receiver data as Unicode characters (ArgumentError) }}} -- Ticket URL: <http://www.macruby.org/trac/ticket/863> MacRuby <http://macruby.org/>

1 2

[MacRuby] #1168: Assertion failed: ((size_t)pos < current_exceptions.size()), function pop_current_exception, file vm.cpp, line 3426. -- using bundler
by MacRuby 07 Mar '11

07 Mar '11

#1168: Assertion failed: ((size_t)pos < current_exceptions.size()), function pop_current_exception, file vm.cpp, line 3426. -- using bundler ------------------------------+--------------------------------------------- Reporter: jhamor@… | Owner: lsansonetti@… Type: defect | Status: new Priority: critical | Milestone: Component: MacRuby | Keywords: ------------------------------+--------------------------------------------- While trying to invoke bundler the documented way, i get the following error: Assertion failed: ((size_t)pos < current_exceptions.size()), function pop_current_exception, file vm.cpp, line 3426. which resulted from the following snippet: $ macirb irb(main):001:0> require "rubygems" => true irb(main):002:0> require "bundler/setup" Assertion failed: ((size_t)pos < current_exceptions.size()), function pop_current_exception, file vm.cpp, line 3426. Abort trap $ This seems to only happen with a Gemfile (as one would expect). My gemfile is pretty sparse and looks like this: source 'http://rubygems.org' -- Ticket URL: <http://www.macruby.org/trac/ticket/1168> MacRuby <http://macruby.org/>

1 2

[MacRuby] #1171: `macruby_deploy --gem` problem with directories that have a space
by MacRuby 07 Mar '11

07 Mar '11

#1171: `macruby_deploy --gem` problem with directories that have a space ------------------------------------+--------------------------------------- Reporter: mrada@… | Owner: lsansonetti@… Type: defect | Status: new Priority: blocker | Milestone: Component: MacRuby | Keywords: ------------------------------------+--------------------------------------- If there is a space in the path to the app bundle then the gem copying will fail. The quick fix is to quote the paths. A better fix might be to use NSFileManager, but then #execute would not get called. I patched the problem in my fork of MacRuby on github: https://github.com/ferrous26/MacRuby/commit/06e66d8ae9bf28085bd1254142de5f6… -- Ticket URL: <http://www.macruby.org/trac/ticket/1171> MacRuby <http://macruby.org/>

1 1

[MacRuby] #938: StringScanner Performance
by MacRuby 07 Mar '11

07 Mar '11

#938: StringScanner Performance --------------------------------+------------------------------------------- Reporter: murphy@… | Owner: lsansonetti@… Type: defect | Status: new Priority: major | Milestone: Component: MacRuby | Keywords: StringScanner, benchmark --------------------------------+------------------------------------------- The StringScanner seems to have quadratic, rather than linear, running time (based on the input's length). I've performed a series of simple benchmark with MacRuby 0.7 and MRI 1.9.2: {{{ time [mac]ruby -rstrscan -e 's = StringScanner.new("abc," * XX_000); s.scan(/./) until s.eos?' }}} The results are shown in the attached diagram. The issue is problematic because the time it takes to scan Strings bigger than a few KiB is too long. This makes the tests for the CodeRay library (see http://redmine.rubychan.de/projects/coderay/wiki/Test_Suite) effectively fail. -- Ticket URL: <http://www.macruby.org/trac/ticket/938> MacRuby <http://macruby.org/>

1 2

[MacRuby] #1101: Rake test tasks for minitest and rspec2 cause a segfault
by MacRuby 05 Mar '11

05 Mar '11

#1101: Rake test tasks for minitest and rspec2 cause a segfault ------------------------------------+--------------------------------------- Reporter: mrada@… | Owner: lsansonetti@… Type: defect | Status: new Priority: blocker | Milestone: Component: MacRuby | Keywords: ------------------------------------+--------------------------------------- These are the two cases I found so far. If I have a minitest rake task like this (generated by jeweler): require 'rake/testtask' Rake::TestTask.new(:test) do |test| test.libs << 'lib' << 'test' test.pattern = 'test/**/test_*.rb' test.verbose = true end And then `rake test` I see: ± rake test (in /Users/mrada/Developer/test) /Library/Frameworks/MacRuby.framework/Versions/0.9/usr/bin/macruby -I"lib:lib:test" "/Users/mrada/.rvm/gems/macruby- nightly/gems/rake-0.8.7/lib/rake/rake_test_loader.rb" "test/test_test.rb" [1] 77800 segmentation fault rake test I will also attach a crash report. I am using MACRUBY_REVISION 5120 (nightly build from Jan 5). -- Ticket URL: <http://www.macruby.org/trac/ticket/1101> MacRuby <http://macruby.org/>

1 2

[MacRuby] #1147: Kramdown does not work
by MacRuby 05 Mar '11

05 Mar '11

#1147: Kramdown does not work ----------------------------------+----------------------------------------- Reporter: kunc.filip@… | Owner: lsansonetti@… Type: defect | Status: new Priority: blocker | Milestone: MacRuby 1.0 Component: MacRuby | Keywords: Kramdown ----------------------------------+----------------------------------------- When installing kramdown via macgems, everything seems to be working. But on more complex markdown stuff, MacRuby behaves a lot differently than standard ruby. For example this string: # header test A list * first bullet * second bullet Renders via puts Kramdown::Document.new(text).to_html only list and ignores header. -- Ticket URL: <http://www.macruby.org/trac/ticket/1147> MacRuby <http://macruby.org/>

1 5

[MacRuby] #1174: Defining TopLevel#copy causes crashes
by MacRuby 28 Feb '11

28 Feb '11

#1174: Defining TopLevel#copy causes crashes --------------------------+------------------------------------------------- Reporter: sema@… | Owner: lsansonetti@… Type: defect | Status: new Priority: blocker | Milestone: Component: MacRuby | Keywords: --------------------------+------------------------------------------------- {{{ puts "string 1".copy # ok def copy(arg) p arg end puts "string 2".copy # crash: segmentation fault }}} {{{ puts "string 1".copy # ok def copy; end puts "string 2".copy # crash: trace trap }}} -- Ticket URL: <http://www.macruby.org/trac/ticket/1174> MacRuby <http://macruby.org/>

1 0