The overriding of signals by the interpreter does not restore the siginfo setting.

Patches item #6418, was opened at 2006-10-31 22:50
You can respond by visiting: 
http://rubyforge.org/tracke
r/?func=detail&atid=1700&aid=6418&group_id=426

Category: Ruby1.8
Group: None
Status: Open
Resolution: None
Priority: 3
Submitted By: Arnold Gutter (agutter)
Assigned to: Nobody (None)
Summary: The overriding of signals by the interpreter does
not restore the siginfo setting.

Initial Comment:
When embedding the ruby interpreter, the overriding of
signals by the interpreter does not restore the siginfo
setting. Enclosed is a short example program effe.c
demonstrating the phenomenon.

------------------------------------------------------------
----------

You can respond by visiting: 
http://rubyforge.org/tracke
r/?func=detail&atid=1700&aid=6418&group_id=426

Hi,

At Wed, 1 Nov 2006 06:50:31 +0900,
Arnold Gutter (agutter) wrote in [ruby-core:09363]:
> When embedding the ruby interpreter, the overriding of
signals by
> the interpreter does not restore the siginfo setting.
Enclosed is a
> short example program effe.c demonstrating the
phenomenon.

ruby_init() is for initialization of the interpreter,
including signal
handlers.  Although I'm not sure what and how you expect,
pause()
should be called via rb_protect() to catch Interrupt
exception from
the interpreter.

-- 
Nobu Nakada