Preparing a release of Ruby 1.9.2

[luislavena]

Hi,

I am almost ready to release a new patch level of Ruby 1.9.2. (see the
HEAD of branches/ruby_1_9_2).

The remaining problems is a build failure of Tk on windows
environment. As Luis notified me in
http://redmine.ruby-lang.org/issues/4802,
I need to merge r31742. But automatic merge of r31742 against
ruby_1_9_2 failed. And it takes some time to merge it by hand.
I will release the new patch level release after I fix this problem.
Let me know if you have any other problem on the HEAD of ruby_1_9_2.

Dear platform maintainers, would you kindly test the HEAD of ruby_1_9_2?

[Yugui]

Let me know if you have any other problem on the HEAD of ruby_1_9_2.

This OpenSSL test case failure http://redmine.ruby-lang.org/issues/4679
still occurs for me on a fresh build of:

ruby 1.9.2p271 (2011-05-31 revision 31856) [i386-mingw32]

Jon

---

blog: http://jonforums.github.com/
twitter: @jonforums

“Anyone who can only think of one way to spell a word obviously lacks
imagination.” - Mark Twain

[Yugui]

On Tue, May 31, 2011 at 12:48 PM, Yugui [email protected] wrote:

Hi,

Hello,

I am almost ready to release a new patch level of Ruby 1.9.2. (see the
HEAD of branches/ruby_1_9_2).

The remaining problems is a build failure of Tk on windows
environment. As Luis notified me in
http://redmine.ruby-lang.org/issues/4802,
I need to merge r31742. But automatic merge of r31742 against
ruby_1_9_2 failed. And it takes some time to merge it by hand.

Thank you, I’ve tried to manually backport the patch using Git (which
works a bit better than svn for backporting) and due the complexity of
the modifications failed miserably. Sorry.

I will release the new patch level release after I fix this problem.
Let me know if you have any other problem on the HEAD of ruby_1_9_2.

Dear platform maintainers, would you kindly test the HEAD of ruby_1_9_2?

FYI:

Not the best results, but:

1.9.2-p180:

ruby 1.9.2p180 (2011-02-18) [i386-mingw32]
8592 tests, 1872709 assertions, 52 failures, 49 errors, 52 skips

https://gist.github.com/473d619ab3574f1c507c

Versus a fresh build:

ruby 1.9.2p271 (2011-05-31 revision 31856) [i386-mingw32]
8695 tests, 1875581 assertions, 57 failures, 48 errors, 52 skips

https://gist.github.com/bb877f2901acd08c59a6

-1 error, +5 failures

I’ve excluded form that gist all the skips since they added too much
noise.

[Yugui]

Hello,

From: Yugui <yugui_at_yugui.jp>
Date: Wed, 1 Jun 2011 01:48:04 +0900

Hi,

I am almost ready to release a new patch level of Ruby 1.9.2. (see the
HEAD of branches/ruby_1_9_2).

in addition to r31346, r31528 is needed to build openssl.so with
OpenSSL 1.0.0, I guess.

Furthermore, even if fix above, test-all failed on openssl.

$ make test-all TESTS=openssl/test_config.rb
./miniruby -I…/…/git/ruby/lib -I.ext/common -I./-
-r…/…/git/ruby/ext/purelib.rb …/…/git/ruby/tool/runruby.rb
–extout=.ext – “…/…/git/ruby/test/runner.rb” openssl/test_config.rb
Loaded suite …/…/git/ruby/test/runner
Started
./ruby: symbol lookup error:
/var/tmp/build/1.9.2/.ext/x86_64-linux/openssl.so: undefined symbol:
_CONF_new_data
make: *** [yes-test-all] Error 127

$ ./ruby -ropenssl -ve ‘OpenSSL::Config.new’
ruby 1.9.2p271 (2011-05-31 revision 31856) [x86_64-linux]
./ruby: symbol lookup error:
/tmp/1.9.2-head/lib/ruby/1.9.1/x86_64-linux/openssl.so: undefined
symbol: _CONF_new_data

r29048 ?

Thanks,