Re: [jruby-dev] benefits of the Rake-based build system

2007-10-23 Thread Evan Weaver
I didn't realize that the current Ant system was so troublesome for people. Personally, I don't think being self-hosted has much real benefit especially since the JVM will remain as a dependency :). Any reason plain old 'make' couldn't be used? Evan On 10/23/07, Charles Oliver Nutter <[EMAIL PR

Re: [jruby-dev] benefits of the Rake-based build system

2007-10-23 Thread Charles Oliver Nutter
Evan Weaver wrote: Not to step out of my place, but I saw the tickets and patch for the Rake build system ( http://jira.codehaus.org/browse/JRUBY-1449 ) and don't understand the benefit of the additional complexity. What do other people think? If it's *more* complicated than building with Ant,

[jruby-dev] benefits of the Rake-based build system

2007-10-23 Thread Evan Weaver
Not to step out of my place, but I saw the tickets and patch for the Rake build system ( http://jira.codehaus.org/browse/JRUBY-1449 ) and don't understand the benefit of the additional complexity. What do other people think? Evan -- Evan Weaver Cloudburst, LLC -

[jruby-dev] Any objections to auto-converting BigDecimal to/from Java/Ruby types?

2007-10-23 Thread Charles Oliver Nutter
Well we have a decent working patch to autoconvert BigDecimal to/from the Java/Ruby types when calling Java APIs that accept or return Java's BigDecimal. Are there any objections to applying it? http://jira.codehaus.org/browse/JRUBY-939 - Charlie -

Re: [jruby-dev] [jira] Created: (JRUBY-1449) rake-based build system for JRuby

2007-10-23 Thread Alexey Verkhovsky
On 10/23/07, Eric Armstrong <[EMAIL PROTECTED]> wrote: >How did you get tasks named "safe:clean", etc? namespace :foo do task :bar do ... end end -- Alexey Verkhovsky CruiseControl.rb [http://cruisecontrolrb.thoughtworks.com] RubyWorks [http://rubyworks.thoughtworks.com] ---

[jruby-dev] [jira] Created: (JRUBY-1468) blocks with var args does not work the same as MRI or JRuby 1.0.1 when the argument supplied is an array

2007-10-23 Thread Pat Cheng (JIRA)
blocks with var args does not work the same as MRI or JRuby 1.0.1 when the argument supplied is an array Key: JRUBY-1468 URL: http://jira.codehaus.org/browse/

Re: [jruby-dev] Default bug value is "don't watch"

2007-10-23 Thread Charles Oliver Nutter
Eric Armstrong wrote: Gad. If I hadn't gone to add some additional info to my bug report, I would never have know that Ola commented on it. The default setting for a bug I've filed is "don't watch". I don't know why that would be--it's not like I saw any option at the time I filed the bug. Is th

[jruby-dev] Default bug value is "don't watch"

2007-10-23 Thread Eric Armstrong
Gad. If I hadn't gone to add some additional info to my bug report, I would never have know that Ola commented on it. The default setting for a bug I've filed is "don't watch". I don't know why that would be--it's not like I saw any option at the time I filed the bug. Is there a profile setting t

Re: [jruby-dev] [jira] Created: (JRUBY-1449) rake-based build system for JRuby

2007-10-23 Thread Eric Armstrong
This sounds really terrific. A basic Rake question: How did you get tasks named "safe:clean", etc? Is ":" a legal character in a name, and you've done some really clever naming, or is there some sort of implied hierarchy? Travis Tilley (JIRA) wrote: rake-based build system for JRuby

Re: [jruby-dev] Bug cleanup in progress

2007-10-23 Thread Charles Oliver Nutter
The first pass is complete. Here's where we stand: JRuby 1.0.2: 54 scheduled JRuby 1.1: 147 scheduled JRuby 1.x: 7 scheduled 71 unscheduled Many of the open bugs including most of the unscheduled bugs, are awaiting followup from submitters. They'll likely get punted if there's no followup, so

[jruby-dev] [jira] Created: (JRUBY-1467) DB Connection Leak using rails-integration in Resin

2007-10-23 Thread Atif Khan (JIRA)
DB Connection Leak using rails-integration in Resin --- Key: JRUBY-1467 URL: http://jira.codehaus.org/browse/JRUBY-1467 Project: JRuby Issue Type: Bug Components: ActiveRecord-JDBC, G

[jruby-dev] [jira] Created: (JRUBY-1466) Object methods defined in Kernel should not be module methods too

2007-10-23 Thread MenTaLguY (JIRA)
Object methods defined in Kernel should not be module methods too - Key: JRUBY-1466 URL: http://jira.codehaus.org/browse/JRUBY-1466 Project: JRuby Issue Type: Bug Affects Ve

[jruby-dev] [jira] Created: (JRUBY-1465) Test failures under Windows

2007-10-23 Thread Bill Dortch (JIRA)
Test failures under Windows --- Key: JRUBY-1465 URL: http://jira.codehaus.org/browse/JRUBY-1465 Project: JRuby Issue Type: Bug Affects Versions: JRuby 1.1.0 Environment: trunk (4678), WinXp, Java 1.6

Re: [jruby-dev] Problem with test result

2007-10-23 Thread Bill Dortch
I've been getting the same failure (until a new one cropped up today). JIRA filed (JRUBY-1465). -Bill

Re: [jruby-dev] jruby-debug stability, enabled in NetBeans (was Re: [jruby-dev] Obscure test/unit interaction problems)

2007-10-23 Thread Peter Brant
It fails there too. My off the cuff guess is that we're not always creating a debugger frame when we should and the debugger ends up thinking it's in the same frame. I'll take a look after work. Pete On 10/23/07, Martin Krauskopf <[EMAIL PROTECTED]> wrote: > > I suppose you are trying everythin

Re: [jruby-dev] jruby-debug stability, enabled in NetBeans (was Re: [jruby-dev] Obscure test/unit interaction problems)

2007-10-23 Thread Martin Krauskopf
Chris Nelson wrote: Peter, The problem, it turns out, is not with stepping into it's with stepping over. When trying to step over the Car.find line the debugger ends up going into dependencies.rb, when it should go to the next line into the controller. It works correctly in MRI, and I'm (no

Re: [jruby-dev] jruby-debug stability, enabled in NetBeans (was Re: [jruby-dev] Obscure test/unit interaction problems)

2007-10-23 Thread Chris Nelson
Peter, The problem, it turns out, is not with stepping into it's with stepping over. When trying to step over the Car.find line the debugger ends up going into dependencies.rb, when it should go to the next line into the controller. It works correctly in MRI, and I'm (now) at Rails 1.2.5 in both

Re: [jruby-dev] Re: Slowness of attr sets

2007-10-23 Thread Charles Oliver Nutter
Charles Oliver Nutter wrote: Quite. I'm looking at it now. I haven't found anything obvious. In both stack and heap-based compiled methods, self is a simple ALOAD. It shouldn't perform any differently than an arbitrary reference. - Charlie --