Re: [Oorexx-devel] why oorexx.org can't be found

2008-09-11 Thread Bill Turner, WB4ALM
I get the same results as Gil. The domain name (oorexx.org) is not known, but the subdomain of (www.oorexx.org) is cached. The DNS authoritive servers are NS1.PLANETDOMAIN.COM and NS2.PLANETDOMAIN.COM The domain is registered to Mark Hessling, so he should contact PLANETDOMAIN.COM to get t

[Oorexx-devel] API tests are working great!

2008-09-11 Thread Rick McGuire
I just wanted to give a shout out to David Ashley and Mark Miesfeld for putting together the API test framework so quickly. It's already paying dividends in uncovering the sort of fringe errors that are much more difficult to troubleshoot once they ship. Rick

Re: [Oorexx-devel] why oorexx.org can't be found

2008-09-11 Thread Jack Woehr
Gil Barmwater wrote: > I just tried the link in Jack's message and IE couldn't find it. But > when I put the www. in the link, all was well. Is this something we can > fix? > http://www.orexx.org/ fails for me also. > Sahananda (Jon) Wolfers wrote: > >> I can see it >> I think there

Re: [Oorexx-devel] Allow a class to be "do over"ed

2008-09-11 Thread David Ruggles
Ok, I had done it right, it was another part of my code I had messed up. Thanks, David Ruggles CCNA MCSE (NT) CNA A+ Network EngineerSafe Data, Inc. (910) 285-7200 [EMAIL PROTECTED] -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Rick McGuire

Re: [Oorexx-devel] Allow a class to be "do over"ed

2008-09-11 Thread Rick McGuire
All you need is a makearray method. Here's a short example: do i over .tester~new say i end ::class tester ::method makearray return .array~of(1,2,3) Rick On Thu, Sep 11, 2008 at 3:15 PM, David Ruggles <[EMAIL PROTECTED]> wrote: > ooRexx 3.2 > > I would like to make a class that I can d

[Oorexx-devel] Allow a class to be "do over"ed

2008-09-11 Thread David Ruggles
ooRexx 3.2 I would like to make a class that I can do over. I thought do over sent the toarray or makearray methods, but I can't figure how how to make it work. Thanks, David Ruggles CCNA MCSE (NT) CNA A+ Network EngineerSafe Data, Inc. (910) 285-7200 [EMAIL PROTECTED] -

Re: [Oorexx-devel] why oorexx.org can't be found

2008-09-11 Thread Gil Barmwater
I just tried the link in Jack's message and IE couldn't find it. But when I put the www. in the link, all was well. Is this something we can fix? Sahananda (Jon) Wolfers wrote: > I can see it > > 2008/9/11 Jack Woehr <[EMAIL PROTECTED] > > > Mark H. & al: > >

Re: [Oorexx-devel] why oorexx.org can't be found

2008-09-11 Thread Sahananda (Jon) Wolfers
I can see it 2008/9/11 Jack Woehr <[EMAIL PROTECTED]> > Mark H. & al: > > I've figured out why oorexx.org can't be found. Your domain servers don't > answer. Probably > your domain registrar from whom you get your DNS is going to Hades in a > Handbasket > or something: > > bash-3.00$ nslookup oo

[Oorexx-devel] why oorexx.org can't be found

2008-09-11 Thread Jack Woehr
Mark H. & al: I've figured out why oorexx.org can't be found. Your domain servers don't answer. Probably your domain registrar from whom you get your DNS is going to Hades in a Handbasket or something: bash-3.00$ nslookup oorexx.org NS1.PLANETDOMAIN.COM Server: NS1.PLANETDOMAIN.

Re: [Oorexx-devel] To infinity, and beyond....

2008-09-11 Thread Mike Cowlishaw
> So, to sumerize, the questions are: > > 1) How should out of range double and float arguments be handled? > 2) How should we handle nan, +infinity and -infinity as return values. Infinities are fairly important to some functions (even some very basic ones, such as tan & arctan). Mike

Re: [Oorexx-devel] To infinity, and beyond....

2008-09-11 Thread Rick McGuire
We don't really have any special kind of run time object we can use, other than maybe just making that object a character string. Once Mike's decimal package is in place, things would end up getting mapped to the real numeric values for these entities. I'm not sure it makes sense to start bending

Re: [Oorexx-devel] To infinity, and beyond....

2008-09-11 Thread David Ashley
Speaking of the decimal math package, should the values like nan be mapped to a run time object like .nan, .infinity, etc? This seems to me to be a more object oriented approach to these exception values rather than a character string representation. Just a thought. Feel free to shoot it down if

Re: [Oorexx-devel] To infinity, and beyond....

2008-09-11 Thread Mark Miesfeld
On Thu, Sep 11, 2008 at 6:49 AM, Rick McGuire <[EMAIL PROTECTED]> wrote: > ... There are > basically two issues. 1) How to deal with values getting passed out as > arguments and 2) Dealing with the edge cases of values getting passed back > in. > > ... There's no checking for the infinities at

Re: [Oorexx-devel] To infinity, and beyond....

2008-09-11 Thread Gil Barmwater
Don't have any great insight here but just wanted to be sure we keep in mind Mike's decimal arithmetic changes that we will be supporting eventually. I.e. let's not do something here that makes that support more difficult or incompatible when we do it. Rick McGuire wrote: > I was just playing

[Oorexx-devel] To infinity, and beyond....

2008-09-11 Thread Rick McGuire
I was just playing around with some edge cases on the float/double argument support. There are some problems involved with out-of-range values that need to be addressed and they should be addressed consistently. There are basically two issues. 1) How to deal with values getting passed out as ar

[Oorexx-devel] ooTest work from my sandbox merged back to test trunk

2008-09-11 Thread Mark Miesfeld
A quick note, I have to run. In case anyone missed the commit message I merged the work I was doing for the native API tests back into test/trunk. David, if you look at the current, either METHOD.testGroup or FUNCTION.testGroup, you can see the changes to the boilerplate. I will create a templat

Re: [Oorexx-devel] Building the API tests on Windows.

2008-09-11 Thread Mark Miesfeld
On Thu, Sep 11, 2008 at 5:07 AM, Rick McGuire <[EMAIL PROTECTED]> wrote: > We're getting a lot of different switches and options on the test runner. I > wonder if we should do something like maven does with different options. > Maven uses the java convention of -Dname=value to define different sw

Re: [Oorexx-devel] Building the API tests on Windows.

2008-09-11 Thread Rick McGuire
Mark, We're getting a lot of different switches and options on the test runner. I wonder if we should do something like maven does with different options. Maven uses the java convention of -Dname=value to define different switches, then the defined variables are made available within the build sy

Re: [Oorexx-devel] Building the API tests on Windows.

2008-09-11 Thread Mark Miesfeld
On Thu, Sep 11, 2008 at 3:53 AM, Rick McGuire <[EMAIL PROTECTED]> wrote: > Since I'm generally going to be building this from a window where I've > already got things set up for building ooRexx, is there any chance that > REXX_BUILD_HOME could be inferred from the SRC_DRV/SRC_DIR variables > requi

[Oorexx-devel] Building routine and method test stubs from common code.

2008-09-11 Thread Rick McGuire
Here's a simple solution to the problem of duplication of method/routine native tests. Step 1, create a header file with a set of definitions like this: #define TestUnit1(ret, name, t1, n1) UNITTYPE1(ret, TstRoutine##name, t1, n1) We'll need one for each argument count from 0 to 10. Step 2, in

[Oorexx-devel] A suggestion for the API tests.

2008-09-11 Thread Rick McGuire
David, I think the library-resident methods and routines should be given names that reflect their usage. For example, rather than tst022 for the float argument test, something like tstFloatArg() would be much better. When writing the different test variations, you're not going to want to scan th

Re: [Oorexx-devel] Building the API tests on Windows.

2008-09-11 Thread Rick McGuire
Mark, Since I'm generally going to be building this from a window where I've already got things set up for building ooRexx, is there any chance that REXX_BUILD_HOME could be inferred from the SRC_DRV/SRC_DIR variables required by makeorx? Rick On Wed, Sep 10, 2008 at 7:39 PM, Mark Miesfeld <[EMA