Fix for JRUBY-3563 introduces regression in Socket#getnameinfo
--------------------------------------------------------------
Key: JRUBY-3725
URL: http://jira.codehaus.org/browse/JRUBY-3725
Project: JRuby
Issue Type: Bug
Components: Core Classes/Modules
Affects Versions: JRuby 1.3RC2
Environment: jruby 1.3.0RC2 (ruby 1.8.6p287) (2009-05-31 98f0d4b)
(Java HotSpot(TM) 64-Bit Server VM 1.6.0_07) [x86_64-java]
Reporter: Brice Figureau
Attachments:
0001-Fix-JRUBY-3563-fix-Socket.getnameinfo-to-accept-ge.patch,
0001-Implements-Socket-getnameinfo-specs.patch
Unfortunately JRUBY-3563 introduced a regression where Socket#getnameinfo is
now unable to parse the output of Socket#getpeername.
The issue is that Socket.getnameinfo is still waiting a string and not a
sock_addr string encoded as getpeername uses.
For the background issue, see the comment in JRUBY-3385:
http://jira.codehaus.org/browse/JRUBY-3385?focusedCommentId=176201&page=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#action_176201
I'm enclosing a patch fixing this issue (and adding also the support for the
non numeric service variants).
Also I'm attaching a rubyspec for testing, which I sent to the RubySpec project
(hoping they'll merge it).
The whole patch is also available in my github repository (masterzen), branch:
jruby-3563:
http://github.com/masterzen/jruby/tree/jruby-3563
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe from this list, please visit:
http://xircles.codehaus.org/manage_email
