Ruby on OpenVMS needs a maintainer

[Ben Armstrong]

For some years, now, we have nursed along an OpenVMS port of Ruby 1.8.2 when its original maintainer from HP in Japan stopped working on it. It is far behind the latest stable release (1.8.7), does not have a proper build system, is not packaged in a proper distribution kit, but works well enough for our purposes. (We have successfully deployed some Rails applications with it.) The code is here:

http://rubyforge.org/scm/?group_id=2483

We do not have the resources to be the maintainers. We are doing the job merely out of necessity. We would like to contact someone at HP to see if anyone there has an interest in supporting this project.

Here are the Ruby committer's thoughts about OpenVMS support from their developer's camp this summer:

http://www.ruby-forum.com/topic/162565

In short, OpenVMS is currently on their "not supported" list, but could be at "best effort" support level if it had a maintainer.

Is anyone here interested in the job? Would someone please let me know who I should contact at HP about this?

Thanks,
Ben

[Robert Brooks_1 · Accepted solution]

I'd try Leo Demers and Gaitan Dantoni

(both email addresses in the form of first.last@hp.com)


-- Rob

[J Gessling]

I may be interested. I'm semi-retired and have the time. And have done some porting, like a Python release, in that past. Can you point me where I could look to get an idea of the effort, like state of code, tools, releases, community etc.

Thanks, Jim

email is my forum name at cowboy_yell_place.com

[Ben Armstrong]

Rob, thanks for the contact info. I'd like to see what kind of interest I can stir up in this thread before proceeding, now.

Thanks for your interest, Jim. I'll answer your questions the best I can.

State of the code:

I ported rubicon to try to get an idea of what works and what doesn't. Unfortunately, the suite depends on some 'nix tools (could possibly be satisified with GNV but I never pursued it) so several tests fail. Nevertheless, about 90% of the tests pass. Here's the latest rubicon output:

========================================================================
All Tests Test Results V0.3.5
========================================================================
Name OK? Tests Asserts Failures Errors
------------------------------------------------------------
Rubicon::TestCase 1 0
AccessControl 3 18
Array 69 614
Arrays 1 0
Assignment 4 75
BasicExpressions 3 15
Bignum 29 1598
Binding 1 0
BlocksProcs 3 61
BooleanExpressions 1 2
CatchThrow 1 1
Class 5 16
Classes 1 0
Comparable 7 26
Constants 2 15
Continuation 5 5
Dir FAIL 21 54 13
Enumerable 23 176
Eval 8 39
Exception 6 16
Exceptions 6 6
FalseClass 6 17
File FAIL 33 33 33
FileTest FAIL 4 4 4
File__Stat FAIL 32 32 32
Fixnum 35 414
Flip 6 28
Float 28 178
Floats 3 34
GC 4 1
Hash 44 296
Hashes 1 0
IO FAIL 53 1174 2 9
IfUnless 1 2
Integer 9 347
Integers 1 0
Invocation FAIL 5 5 5
KanjiIndex 20 190
Kernel FAIL 109 32234 14 2
LoopStuff 15 65
Marshal 6 31
MatchData 11 18
Math FAIL 2 59 1
Method 4 11
Methods 2 6
Module 25 101
ModulePrivate 15 41
Modules 1 0
Names 1 0
NilClass 8 21
Numeric 10 21
Object 2 11
ObjectSpace FAIL 4 6 1
PredefinedVariables 1 3
Proc 4 5
Range 14 159
Ranges 1 0
Regexp 14 38
RegularExpressions 3 2993
Scope 12 27
SourceLayout FAIL 7 15 3
String 87 1252
Strings 5 27
Struct 11 47
Struct__Tms FAIL 1 0 1
Symbol 8 20
Symbols 1 0
Thread FAIL 33 98 1 2
ThreadGroup 4 7
Time 41 1813
TrueClass 6 17
========================================================================
All 71 files FAIL 953 44638 105 18
========================================================================

Sorry about the formatting. Cut-and-paste this table and render it in a monospaced font for easier viewing.

Here are some things we have working and use either for development or in various deployed applications:

- irb (but no readline support, so no command history)
- rubicon
- webrick
- drb
- rails
- ferret
- PDF::Writer

State of the tools:

There is a ruby.mss and some build*.com files to build the code and the beginnings of a directory structure to make into a release kit. The instructions are in [.vmsbuild.com]0.readme.

All [.ext] modules have been hand-ported. We need a tool to automatically generate the necessary .mms and .opt files so we can support more modules.

State of releases:

As stated in my intro, there is no distribution kit, so there is no release.

State of the community:

While I would love to be able to point you at the community, I'm afraid it's been very lonely. I have had a few interested parties in the past, but nobody who stuck with it. So if there *are* others out there, they haven't been talking to me.

It's a bit of a chicken-and-egg problem. Until a proper release is made, we're unlikely to attract a community. Without a community, I'm finding it hard to gather the resources to make a release. And with neither, I'm going to have a hard time making my case to HP that they should care about this.

[Robert Brooks_1]

Another place to look for help is the usenet newsgroup comp.os.vms, although the noise-to-signal ratio has been painfully high lately.

-- Rob

[Ian Miller.]

You may wish to post a request on the openvmshobbyist web site.

I posted something here
http://www.openvms.org/stories.php?story=08/09/07/6842771

to publicise your request. Let me know if its OK

[labadie_1]

Something obvious, but the maintainer of Ruby for OpenVMS should contact his equivalent for Perl (Craig Berry ?) and Python (JF Pieronne), as they will have to solve similar problems, and they could share a fix (humm, may be I should say it another way...) when a problem arises.