Re: [MacRuby-devel] RubySpec updated

2009-08-03 Thread Eloy Duran

Looks great, only a few notes:
* Ruby version guards inlince in the example are not allowed. You  
should always use the version guard and in its block define the  
example outlining the specific behaviour on that version. This means  
some examples will have to broken up into smaller ones.
* To check for a method, use the have_method matchers. Unfortunately,  
there's no have_private_method matcher yet, so one should be added to  
MSpec, which is very straightforward. You can do so in our frozen  
copy, which I will then merge upstream later on. Or you can do it  
directly against MSpec, which might be easier as you can safely run  
the specs of MSpec without worrying about MacRuby hacks.


Cheers,
Eloy

On Aug 2, 2009, at 9:17 PM, Matt Aimonetti wrote:

Done, would you mind checking that I did it according to the  
rubyspec standards?


Thanks,

- Matt

On Sun, Aug 2, 2009 at 11:30 AM, Matt Aimonetti > wrote:
Thanks for the tip, I'll go back and fix strscan and will make the  
modifications before pushing stringio.


- Matt


On Sun, Aug 2, 2009 at 11:22 AM, Eloy Duran  
 wrote:

Thanks Matt :)

Indeed, in order to specify the API changes between 1.8 and 1.9 you  
should use version guards:


 ruby_version_is "" ... "1.9" do
   it "works as such on all versions prior to 1.9" do
   end
 end

 ruby_version_is "1.9" do
   it "works as such on 1.9" do
   end
 end

To guard against compile errors you should use the language_version  
guard. See the rubyspec wiki for more info.


The 1.9.x specs should target HEAD. There's no specific 1.9 target  
yet set for either RubySpec or MacRuby.


HTH,
Eloy


On 2 aug 2009, at 19:49, Matt Aimonetti wrote:

Hi Eloy,

Welcome back. Unfortunatelly, due to API changes, the specs can't  
all pass on 1.8 and 1.9 unless we use a version check mechanism.


My understanding was that we should focus on 1.9.2 preview 1.

What do you want me to do?

- Matt

Sent from my iPhone

On Aug 2, 2009, at 6:42, Eloy Duran  wrote:

Hi,

I'm un-jet lagging a bit, so I thought I'd update the ruby specs  
again. We are now passing: 18160 examples.


@Matt: Great work on the StringScanner! Could you please make sure  
the specs run on 1.8 as well? Currently 4 fail:


$ mspec -B ruby.1.8.mspec library/stringscanner

StringScanner#getch is multi-byte character sensitive FAILED
StringScanner#getch should keep the encoding ERROR
StringScanner#initialize_copy is a private method FAILED
StringScanner#initialize is a private method FAILED

@Laurent: There were quite some changes to complex that you might  
want to look into since you recently worked on those :)


Cheers,
Eloy
___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel
___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel

___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


[MacRuby-devel] Information about the new YAML module

2009-08-03 Thread Patrick Thomson

Hi everyone,

I've just pushed my work on a new YAML module to the experimental branch 
(revision 2184). Rather than being backed by the old syck code that 
1.8/1.9 use, this is backed by the libYAML library (BSD licensed). As of 
right now it appears to be about 4x slower than 1.9's YAML module, but 
there's a lot of room for optimizing my code.


This new YAML module aims to be wholly compatible with the old API 
inside yaml.rb. *Please *test this module out, and let me know if you 
find any inconsistencies or false results. As of right now it passes the 
RubySpecs for YAML::dump() and YAML::load().


(Oh, and sorry for spamming the macruby-changes list; git-svn applied 
something like 40 commits at one time. Oops!)


-- Patrick Thomson
___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


[MacRuby-devel] Miniruby Causing Trouble with SVN Edge

2009-08-03 Thread Maxwell Hawkins
I've been trying to compile 0.5 from SVN with no luck. I installed  
LLVM revision 72741 as outlined in the readme but when I ran MacRuby's  
rake file I got the following error:


../../miniruby -I../.. -I../../lib -r rbconfig -e  
"RbConfig::CONFIG['libdir'] = '../..'; require './extconf.rb'"
Assertion failed: (castIsValid(getOpcode(), S, Ty) && "Illegal  
BitCast"), function BitCastInst, file Instructions.cpp, line 2493.


Thinking my environment might be the problem, I installed a fresh copy  
of Leopard and reinstalled LLVM. I still get the same error message.


Further investigation reveals that the problem occurs when rbconfig is  
required by miniruby in line 325 of builder.rake. It throws the error  
when when it encounters the "==" in "RUBY_VERSION == "1.9.0"" at the  
top of rbconfig.rb. I encounter similar problems (and the same error  
message) concatenating strings (-e "'string' + 'string'") with miniruby.


Both experiments share a similar gdb backtrace:


#0  0x9309de42 in __kill ()
#1  0x9309de34 in kill$UNIX2003 ()
#2  0x9311023a in raise ()
#3  0x9311c679 in abort ()
#4  0x931113db in __assert_rtn ()
#5  0x00969cff in llvm::BitCastInst::BitCastInst () at vm.h:582
#6  0x0012efa9 in RoxorCompiler::compile_double_coercion  
(this=0xa029200, val=0xe6578c, mask=0xe65fe8, fallback_bb=0xe65ee0,  
f=0xe641f0) at compiler.cpp:1929
#7  0x00138231 in RoxorCompiler::compile_optimized_dispatch_call  
(this=0xa029200, sel=optimizations>, argc=1, para...@0xbfffcb4c) at compiler.cpp:2254
#8  0x00149276 in RoxorCompiler::compile_node (this=0xa029200,  
node=0x181a150) at compiler.cpp:3813
#9  0x0013f5e6 in RoxorCompiler::compile_node (this=0xa029200,  
node=0x181a070) at compiler.cpp:4029
#10 0x001427f0 in RoxorCompiler::compile_node (this=0xa029200,  
node=0x1811fe0) at compiler.cpp:3598
#11 0x00146833 in RoxorCompiler::compile_node (this=0xa029200,  
node=0x1859c00) at compiler.cpp:2876
#12 0x0014f770 in RoxorCompiler::compile_main_function  
(this=0xa029200, node=0x1859c00) at compiler.cpp:4717
#13 0x0017ca65 in rb_vm_run (fname=0x0, node=0x0, binding=0x0,  
inside_eval=false) at vm.cpp:4132
#14 0x00040a01 in rb_load (fname=to optimizations>, wrap=0) at load.c:293
#15 0x000415c3 in rb_require_safe (fname=25483856, safe=0) at load.c: 
488
#16 0x001701a8 in rb_vm_dispatch (cache=0xe61b30, self=25258368,  
sel=0x3cbf, block=0x0, opt=0 '\0', argc=unavailable, due to optimizations>) at vm.cpp:2026

#17 0x0a880062 in ?? ()
#18 0x0003879b in ruby_run_node (n=0x1816980) at eval.c:198
#19 0x0011d2a8 in main (argc=3, argv=0xe04c30, envp=0xb724) at  
main.cpp:38


Is there any way I can get this edge release working?

I'm on a MacBook Pro with a 2.16GHz Core Duo running 10.5.7.

Thank you,
Max Hawkins
___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


Re: [MacRuby-devel] Miniruby Causing Trouble with SVN Edge

2009-08-03 Thread M. Scott Ford

> I'm on a MacBook Pro with a 2.16GHz Core Duo running 10.5.7.

Core Duo is 32 bit. If I had to guess, that would be the source of the issue.

___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


Re: [MacRuby-devel] Information about the new YAML module

2009-08-03 Thread Eloy Duran

Awesome work Patrick! Thanks a lot for your work on this.

Eloy

On 3 aug 2009, at 19:51, Patrick Thomson wrote:


Hi everyone,

I've just pushed my work on a new YAML module to the experimental  
branch (revision 2184). Rather than being backed by the old syck  
code that 1.8/1.9 use, this is backed by the libYAML library (BSD  
licensed). As of right now it appears to be about 4x slower than  
1.9's YAML module, but there's a lot of room for optimizing my code.


This new YAML module aims to be wholly compatible with the old API  
inside yaml.rb. Please test this module out, and let me know if you  
find any inconsistencies or false results. As of right now it passes  
the RubySpecs for YAML::dump() and YAML::load().


(Oh, and sorry for spamming the macruby-changes list; git-svn  
applied something like 40 commits at one time. Oops!)


-- Patrick Thomson
___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


Re: [MacRuby-devel] Information about the new YAML module

2009-08-03 Thread Eloy Duran
Btw: About git-svn creating many commits, I wouldn't worry about it :)  
But if you'd like to normalize, I'd do something like:


$ git checkout -b yaml_branch

# work on it

$ git checkout master
$ git diff yaml_branch | patch -p1

$ git add .
$ git commit -v

Eloy

On 3 aug 2009, at 19:51, Patrick Thomson wrote:


Hi everyone,

I've just pushed my work on a new YAML module to the experimental  
branch (revision 2184). Rather than being backed by the old syck  
code that 1.8/1.9 use, this is backed by the libYAML library (BSD  
licensed). As of right now it appears to be about 4x slower than  
1.9's YAML module, but there's a lot of room for optimizing my code.


This new YAML module aims to be wholly compatible with the old API  
inside yaml.rb. Please test this module out, and let me know if you  
find any inconsistencies or false results. As of right now it passes  
the RubySpecs for YAML::dump() and YAML::load().


(Oh, and sorry for spamming the macruby-changes list; git-svn  
applied something like 40 commits at one time. Oops!)


-- Patrick Thomson
___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


Re: [MacRuby-devel] Information about the new YAML module

2009-08-03 Thread Matt Aimonetti
hmm maybe I should give git-svn another chance. (it annoyed me so much I
went back to svn)
Eloy, would you mind giving a quick rundown of your setup and workflow?

Thanks,

- Matt


On Mon, Aug 3, 2009 at 1:01 PM, Eloy Duran  wrote:

> Btw: About git-svn creating many commits, I wouldn't worry about it :) But
> if you'd like to normalize, I'd do something like:
> $ git checkout -b yaml_branch
>
> # work on it
>
> $ git checkout master
> $ git diff yaml_branch | patch -p1
>
> $ git add .
> $ git commit -v
>
> Eloy
>
> On 3 aug 2009, at 19:51, Patrick Thomson wrote:
>
>  Hi everyone,
>
> I've just pushed my work on a new YAML module to the experimental branch
> (revision 2184). Rather than being backed by the old syck code that 1.8/1.9
> use, this is backed by the libYAML library (BSD licensed). As of right now
> it appears to be about 4x slower than 1.9's YAML module, but there's a lot
> of room for optimizing my code.
>
> This new YAML module aims to be wholly compatible with the old API inside
> yaml.rb. *Please *test this module out, and let me know if you find any
> inconsistencies or false results. As of right now it passes the RubySpecs
> for YAML::dump() and YAML::load().
>
> (Oh, and sorry for spamming the macruby-changes list; git-svn applied
> something like 40 commits at one time. Oops!)
>
> -- Patrick Thomson
>  ___
> MacRuby-devel mailing list
> MacRuby-devel@lists.macosforge.org
> http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel
>
>
>
> ___
> MacRuby-devel mailing list
> MacRuby-devel@lists.macosforge.org
> http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel
>
>
___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


Re: [MacRuby-devel] Information about the new YAML module

2009-08-03 Thread Brian Mitchell
You could also use:

$ git merge --squash yaml_branch

A little easier than piping a diff.

Brian.

On Mon, Aug 3, 2009 at 16:01, Eloy Duran wrote:
> Btw: About git-svn creating many commits, I wouldn't worry about it :) But
> if you'd like to normalize, I'd do something like:
> $ git checkout -b yaml_branch
> # work on it
> $ git checkout master
> $ git diff yaml_branch | patch -p1
> $ git add .
> $ git commit -v
> Eloy
>
> On 3 aug 2009, at 19:51, Patrick Thomson wrote:
>
> Hi everyone,
>
> I've just pushed my work on a new YAML module to the experimental branch
> (revision 2184). Rather than being backed by the old syck code that 1.8/1.9
> use, this is backed by the libYAML library (BSD licensed). As of right now
> it appears to be about 4x slower than 1.9's YAML module, but there's a lot
> of room for optimizing my code.
>
> This new YAML module aims to be wholly compatible with the old API inside
> yaml.rb. Please test this module out, and let me know if you find any
> inconsistencies or false results. As of right now it passes the RubySpecs
> for YAML::dump() and YAML::load().
>
> (Oh, and sorry for spamming the macruby-changes list; git-svn applied
> something like 40 commits at one time. Oops!)
>
> -- Patrick Thomson
> ___
> MacRuby-devel mailing list
> MacRuby-devel@lists.macosforge.org
> http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel
>
>
> ___
> MacRuby-devel mailing list
> MacRuby-devel@lists.macosforge.org
> http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel
>
>
___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


Re: [MacRuby-devel] Information about the new YAML module

2009-08-03 Thread Eloy Duran

Oh… Why did it annoy you?

My normal workflow is:

$ git svn rebase
$ git checkout -b topic

# work

$ git checkout master
$ git svn rebase
$ git checkout topic
$ git rebase master
$ git checkout master
# And since 10 seconds ago:
$ git merge --squash topic

# sanity check:
$ git svn rebase

# push
$ git svn dcommit

I wouldn't do the squashing actually, as I see no problem with many  
commits. Actually I find it handier, as you can follow the development  
better with narrative commit messages etc. The "real" problem IMO is  
that the commit emails can be annoying and what we miss is a good  
browser. For this purpose I use gitx locally, which is way faster  
anyways, or github (to where I push my git svn checkout).


Hope this answers your questions.

Cheers,
Eloy

On 3 aug 2009, at 22:15, Matt Aimonetti wrote:

hmm maybe I should give git-svn another chance. (it annoyed me so  
much I went back to svn)


Eloy, would you mind giving a quick rundown of your setup and  
workflow?


Thanks,

- Matt


On Mon, Aug 3, 2009 at 1:01 PM, Eloy Duran   
wrote:
Btw: About git-svn creating many commits, I wouldn't worry about  
it :) But if you'd like to normalize, I'd do something like:


$ git checkout -b yaml_branch

# work on it

$ git checkout master
$ git diff yaml_branch | patch -p1

$ git add .
$ git commit -v

Eloy

On 3 aug 2009, at 19:51, Patrick Thomson wrote:


Hi everyone,

I've just pushed my work on a new YAML module to the experimental  
branch (revision 2184). Rather than being backed by the old syck  
code that 1.8/1.9 use, this is backed by the libYAML library (BSD  
licensed). As of right now it appears to be about 4x slower than  
1.9's YAML module, but there's a lot of room for optimizing my code.


This new YAML module aims to be wholly compatible with the old API  
inside yaml.rb. Please test this module out, and let me know if you  
find any inconsistencies or false results. As of right now it  
passes the RubySpecs for YAML::dump() and YAML::load().


(Oh, and sorry for spamming the macruby-changes list; git-svn  
applied something like 40 commits at one time. Oops!)


-- Patrick Thomson
___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel



___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


Re: [MacRuby-devel] Information about the new YAML module

2009-08-03 Thread Eloy Duran

Nice! :)

On 3 aug 2009, at 22:23, Brian Mitchell wrote:


You could also use:

$ git merge --squash yaml_branch

A little easier than piping a diff.

Brian.

On Mon, Aug 3, 2009 at 16:01, Eloy Duran  
wrote:
Btw: About git-svn creating many commits, I wouldn't worry about  
it :) But

if you'd like to normalize, I'd do something like:
$ git checkout -b yaml_branch
# work on it
$ git checkout master
$ git diff yaml_branch | patch -p1
$ git add .
$ git commit -v
Eloy

On 3 aug 2009, at 19:51, Patrick Thomson wrote:

Hi everyone,

I've just pushed my work on a new YAML module to the experimental  
branch
(revision 2184). Rather than being backed by the old syck code that  
1.8/1.9
use, this is backed by the libYAML library (BSD licensed). As of  
right now
it appears to be about 4x slower than 1.9's YAML module, but  
there's a lot

of room for optimizing my code.

This new YAML module aims to be wholly compatible with the old API  
inside

yaml.rb. Please test this module out, and let me know if you find any
inconsistencies or false results. As of right now it passes the  
RubySpecs

for YAML::dump() and YAML::load().

(Oh, and sorry for spamming the macruby-changes list; git-svn applied
something like 40 commits at one time. Oops!)

-- Patrick Thomson
___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel



___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


Re: [MacRuby-devel] Information about the new YAML module

2009-08-03 Thread Josh Ballanco

Hey Matt,

Just to add another data point, I've been using git-svn to interface  
with a svn repo where some of my git branches are matched to svn  
branches and others are not. In general, my workflow is:


# Make a branch in svn:
$ svn cp svn+ssh://path/to/trunk svn+ssh://path/to/branch/yaml_branch

# Retrieve the new branch metadata:
$ git svn fetch

# Creates a new git branch that mirrors the svn branch:
$ git checkout -b git_yaml_branch yaml_branch

# work on it

# Try a little experiment:
$ git checkout -b experimental_yaml_branch

# work some more

# Bring it all together again:
$ git checkout git_yaml_branch
$ git merge --squash experimental_yaml_branch
$ git diff (* just to inspect the changes *)
$ git commit -a
$ git svn rebase (* don't worry if there was a change before your  
commit, git will replay yours on top of it *)

$ git svn dcommit

Also, one more tip on OS X:
$ git config --global merge.tool opendiff

Then, if there's ever a conflict during a merge or rebase, using "git  
mergetool" let's you resolve conflicts in FileMerge.app's nice GUI.


Hope that helps!

- Josh

On Aug 3, 2009, at 1:15 PM, Matt Aimonetti wrote:

hmm maybe I should give git-svn another chance. (it annoyed me so  
much I went back to svn)


Eloy, would you mind giving a quick rundown of your setup and  
workflow?


Thanks,

- Matt


On Mon, Aug 3, 2009 at 1:01 PM, Eloy Duran   
wrote:
Btw: About git-svn creating many commits, I wouldn't worry about  
it :) But if you'd like to normalize, I'd do something like:


$ git checkout -b yaml_branch

# work on it

$ git checkout master
$ git diff yaml_branch | patch -p1

$ git add .
$ git commit -v

Eloy

On 3 aug 2009, at 19:51, Patrick Thomson wrote:


Hi everyone,

I've just pushed my work on a new YAML module to the experimental  
branch (revision 2184). Rather than being backed by the old syck  
code that 1.8/1.9 use, this is backed by the libYAML library (BSD  
licensed). As of right now it appears to be about 4x slower than  
1.9's YAML module, but there's a lot of room for optimizing my code.


This new YAML module aims to be wholly compatible with the old API  
inside yaml.rb. Please test this module out, and let me know if you  
find any inconsistencies or false results. As of right now it  
passes the RubySpecs for YAML::dump() and YAML::load().


(Oh, and sorry for spamming the macruby-changes list; git-svn  
applied something like 40 commits at one time. Oops!)


-- Patrick Thomson
___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel



___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


Re: [MacRuby-devel] Miniruby Causing Trouble with SVN Edge

2009-08-03 Thread Vincent Isambart

I'm on a MacBook Pro with a 2.16GHz Core Duo running 10.5.7.


Core Duo is 32 bit. If I had to guess, that would be the source of  
the issue.



Indeed, the 0.5 branch currently does not work in 32 bits. Well, I  
fixed the BitCast error in 32 bits yesterday, but there are other big  
problems left.
You will have to wait a bit since 32 bits support is currently not a  
high priority (but it's scheduled for the 0.5 release).


Cheers,
Vincent
___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


Re: [MacRuby-devel] Miniruby Causing Trouble with SVN Edge

2009-08-03 Thread Maxwell Hawkins

I'm on a MacBook Pro with a 2.16GHz Core Duo running 10.5.7.


Core Duo is 32 bit. If I had to guess, that would be the source of  
the issue.



Indeed, the 0.5 branch currently does not work in 32 bits. Well, I  
fixed the BitCast error in 32 bits yesterday, but there are other  
big problems left.
You will have to wait a bit since 32 bits support is currently not a  
high priority (but it's scheduled for the 0.5 release).


Ah, I'll need to be patient then. Looking forward to 0.5!

Thanks,
Max
___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


[MacRuby-devel] Add Integer#ord

2009-08-03 Thread dan sinclair

Attached patch ports Integer#ord from Ruby 1.9.

dan



integer_ord.diff
Description: Binary data


___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


Re: [MacRuby-devel] Add Integer#ord

2009-08-03 Thread Laurent Sansonetti

Committed in r2203, thanks!

Laurent

On Aug 3, 2009, at 6:54 PM, dan sinclair wrote:


Attached patch ports Integer#ord from Ruby 1.9.

dan


___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel


[MacRuby-devel] NSNumber boolean spec

2009-08-03 Thread dan sinclair
The attached patch gets the NSNumber boolean conversion spec working.   
I'm not sure if the change is correct but I changed the spec from  
should != to should_not == and it appears to be working correctly now.


dan



numeric_spec.diff
Description: Binary data


___
MacRuby-devel mailing list
MacRuby-devel@lists.macosforge.org
http://lists.macosforge.org/mailman/listinfo.cgi/macruby-devel