Change of behaviour for sleep(0) in 1.9


#1

Hi,

Just thought I’d share this to save anyone else the bafflement it caused
me.

In ruby 1.8.x and jruby, sleep(0) means do not sleep at all. In 1.9,
it appears to mean sleep forever.

In both cases, sleep with no arguments means sleep forever.

Perhaps differing interpretations of this ambiguous documentation from
ri?

“Zero arguments causes sleep to sleep forever.”

Regards,
Sean


#2

El Martes 31 Marzo 2009, Sean O’Halpin
escribió:> Hi,

Just thought I’d share this to save anyone else the bafflement it caused
me.

In ruby 1.8.x and jruby, sleep(0) means do not sleep at all. In 1.9,
it appears to mean sleep forever.

I’ve tested it in Ruby 1.9.1 and sleep(0) means do not sleep at all.

Perhaps differing interpretations of this ambiguous documentation from ri?

“Zero arguments causes sleep to sleep forever.”

This must be a doc bug.


#3

On Tue, Mar 31, 2009 at 6:41 PM, Iñaki Baz C. removed_email_address@domain.invalid wrote:

El Martes 31 Marzo 2009, Sean O’Halpin escribió:

Hi,

Just thought I’d share this to save anyone else the bafflement it caused
me.

In ruby 1.8.x and jruby, sleep(0) means do not sleep at all. In 1.9,
it appears to mean sleep forever.

I’ve tested it in Ruby 1.9.1 and sleep(0) means do not sleep at all.

Hmmm. Just tested on Ubuntu 8.04 and sleep(0) works as expected in
all versions (i.e. it doesn’t wait). However, this is what I get on
Mac OS X 10.4:

$ multiruby -rtimeout -e ‘Timeout::timeout(2) { sleep(0) }’

[snip working versions]

VERSION = v1_9_1_0
CMD = ~/.multiruby/install/v1_9_1_0/bin/ruby -rtimeout -e
Timeout::timeout(2) { sleep(0) }

-e:1:in sleep': execution expired (Timeout::Error) from -e:1:inblock in ’
from -e:1:in `’

RESULT = 256

TOTAL RESULT = 1 failures out of 5

Passed: jruby-1.2.0RC2, jruby-1.1.6, v1_8_6_110, 1.8.7-p72
Failed: v1_9_1_0
$ uname -a
Darwin xxxx 8.11.1 Darwin Kernel Version 8.11.1: Wed Oct 10 18:23:28
PDT 2007; root:xnu-792.25.20~1/RELEASE_I386 i386 i386

I guess I should file a bug report.

Regards,
Sean


#4

Hi,

In message “Re: Change of behaviour for sleep(0) in 1.9”
on Wed, 1 Apr 2009 00:02:20 +0900, “Sean O’Halpin”
removed_email_address@domain.invalid writes:

|Perhaps differing interpretations of this ambiguous documentation from ri?
|“Zero arguments causes sleep to sleep forever.”

“Zero arguments” means “sleep”, not “sleep(0)”. Disambiguation welcome.

          matz.

#5

On Mar 31, 2009, at 11:17 PM, Yukihiro M. wrote:

“Zero arguments” means “sleep”, not “sleep(0)”. Disambiguation
welcome.

          matz.

“Called without an argument, sleep() will sleep forever.”

The fastri output colors the first “sleep” as a Ruby item and leaves
the second “sleep” as a Human item. It also doesn’t use the empty
parentheses, but they certainly help in email.

-Rob

Rob B. http://agileconsultingllc.com
removed_email_address@domain.invalid


#6

Hi,

In message “Re: Change of behaviour for sleep(0) in 1.9”
on Wed, 1 Apr 2009 13:02:58 +0900, Rob B.
removed_email_address@domain.invalid writes:

|“Called without an argument, sleep() will sleep forever.”

That sounds better. Thank you.

          matz.

#7

2009/4/1 Sean O’Halpin removed_email_address@domain.invalid:

Timeout::timeout(2) { sleep(0) }
Failed: v1_9_1_0
Hm, works for me:

15:49:23 bas$ allruby -e ‘puts Time.now; sleep 0; puts Time.now’
CYGWIN_NT-5.1 padrklemme1 1.5.25(0.156/4/2) 2008-06-12 19:34 i686 Cygwin
ruby 1.8.7 (2008-08-11 patchlevel 72) [i386-cygwin]
Wed Apr 01 15:49:24 +0200 2009
Wed Apr 01 15:49:24 +0200 2009
ruby 1.9.1p0 (2009-01-30 revision 21907) [i386-cygwin]
2009-04-01 15:49:25 +0100
2009-04-01 15:49:25 +0100
15:49:25 bas$

Cheers

robert


#8

Sean O’halpin wrote:

On Wed, Apr 1, 2009 at 2:50 PM, Robert K.
removed_email_address@domain.invalid wrote:

Hm, works for me:

15:49:23 bas$ allruby -e ‘puts Time.now; sleep 0; puts Time.now’
CYGWIN_NT-5.1 padrklemme1 1.5.25(0.156/4/2) 2008-06-12 19:34 i686 Cygwin

Looks like a Mac OSX thing then. I’ve filed a bug report. Is there
anyone else out there with ruby 1.9 on OSX (10.4 or 10.5) who can
confirm that this is a problem?

$ uname -a
Darwin Machine.local 8.11.0 Darwin Kernel Version 8.11.0: Wed Oct 10
18:26:00 PDT 2007; root:xnu-792.24.17~1/RELEASE_PPC Power Macintosh
powerpc

$ ruby-1.9 -v
ruby 1.9.1p0 (2009-01-30 revision 21907) [powerpc-darwin8.11.0]

$ ruby-1.9 -r Timeout -e ‘Timeout.timeout(1) {sleep(0)}’
-e:1:in sleep': execution expired (Timeout::Error) from -e:1:inblock in ’
from -e:1:in `’
$

Seems to be a problem here, too.


#9

On Wed, Apr 1, 2009 at 2:50 PM, Robert K.
removed_email_address@domain.invalid wrote:

Hm, works for me:

15:49:23 bas$ allruby -e ‘puts Time.now; sleep 0; puts Time.now’
CYGWIN_NT-5.1 padrklemme1 1.5.25(0.156/4/2) 2008-06-12 19:34 i686 Cygwin

Looks like a Mac OSX thing then. I’ve filed a bug report. Is there
anyone else out there with ruby 1.9 on OSX (10.4 or 10.5) who can
confirm that this is a problem?