[
http://issues.apache.org/jira/browse/POOL-87?page=comments#action_12441638 ]
Sandy McArthur commented on POOL-87:
If poolable objects are being made simply to ensure a minIdle threshold, then
they would go directly into the idle object
It might do; not sure what the performance of ThreadLocal is. However
the price is extra memory usage: a DateFormatter for every thread in the
system that ever logs a message.
On Wed, 2006-10-11 at 17:42 -0400, Kenneth Xu wrote:
Hi,
I think a thread local formatter will give us better
Author: niallp
Date: Thu Oct 12 01:24:23 2006
New Revision: 463154
URL: http://svn.apache.org/viewvc?view=revrev=463154
Log:
IO-84 Modify copy() methods that return an int to throw an ArithmeticException
if the returned size is too large for an int to handle. Add two new equivalent
copyLarge()
[ http://issues.apache.org/jira/browse/IO-84?page=comments#action_12441663
]
Niall Pemberton commented on IO-84:
---
Committed those changes to IOUtils to add copyLarge() methods and throw and
ArithmeticException in the original copy() methods.
Simon Kitching wrote on Thursday, October 12, 2006 9:43 AM:
It might do; not sure what the performance of ThreadLocal is. However
the price is extra memory usage: a DateFormatter for every
thread in the
system that ever logs a message.
Hm. Two scenarios:
1/ The Thread dies -- the
[ http://issues.apache.org/jira/browse/IO-94?page=comments#action_12441670
]
Joerg Schaible commented on IO-94:
--
Personally I have always used (CGLIB-enhanced) jMock to create mocks of streams
on the fly. Therefore I agree with Stephen, that
[ http://issues.apache.org/jira/browse/IO-94?page=comments#action_12441672
]
Niall Pemberton commented on IO-94:
---
OK but can you create a mock on the fly that emulates a file of a specified
size - such as a 2GB file?
New Mock InputStream
[ http://issues.apache.org/jira/browse/IO-94?page=comments#action_12441674
]
Joerg Schaible commented on IO-94:
--
Yes, but not *that* easy :)
But what I really needed and wrote once was a NullInputStream of a specific
capacity delivering bytes
How do you know when to clean it up? Just instantiate a new one each time.
It might do; not sure what the performance of ThreadLocal is. However
the price is extra memory usage: a DateFormatter for every thread in the
system that ever logs a message.
On Wed, 2006-10-11 at 17:42 -0400,
To whom it may engage...
This is an automated request, but not an unsolicited one. For
more information please visit http://gump.apache.org/nagged.html,
and/or contact the folk at [EMAIL PROTECTED]
Project commons-jelly-tags-jsl-test has an issue affecting its community
integration.
To whom it may engage...
This is an automated request, but not an unsolicited one. For
more information please visit http://gump.apache.org/nagged.html,
and/or contact the folk at [EMAIL PROTECTED]
Project commons-jelly-tags-jsl-test has an issue affecting its community
integration.
[ http://issues.apache.org/jira/browse/POOL-87?page=all ]
Bernie McGourty updated POOL-87:
Attachment: TestGenericObjectPool.java
I've added a new method: testMinIdlePoolSequence() to the
TestGenericObjectPool.java.
Here's the result ofr the test:
[ http://issues.apache.org/jira/browse/IO-82?page=comments#action_12441734
]
Henri Yandell commented on IO-82:
-
Much, much faster:
[junit] Running org.apache.commons.io.FileCleanerTestCase
[junit] Tests run: 5, Failures: 0, Errors: 0,
[ http://issues.apache.org/jira/browse/IO-82?page=comments#action_12441736
]
Henri Yandell commented on IO-82:
-
So I can commit the fix - but any reason why it doesn't just sleep in there?
Are we just doing busywork there to create a pause, or
[
http://issues.apache.org/jira/browse/POOL-87?page=comments#action_12441750 ]
Sandy McArthur commented on POOL-87:
Poolable Objects created by makeObject() are considered to be an active state
so they do not need to be activated again.
[ http://issues.apache.org/jira/browse/POOL-87?page=all ]
Sandy McArthur updated POOL-87:
---
Assignee: Sandy McArthur
Commons-Pool does not always calling activateObject on newly created Objects
Hi,
as already discussed on
http://marc.theaimsgroup.com/?t=11601226581
I would like to see a release of the parent POM for the Jakarta Commons
Proper projects. The current version 1-SNAPSHOT can be found on
I guess it will clean itself up when the thread is garbage collected. I'd
play around with it just to be sure, though. Application servers typically
use a thread pool, so they won't be garbage collected. But, that's okay,
because you want to be able to reuse them if you can to avoid having to
[
http://issues.apache.org/jira/browse/POOL-87?page=comments#action_12441771 ]
Bernie McGourty commented on POOL-87:
-
OK, I guess I misunderstood. I've modified my PoolableObjectFactory to call
activateObject after constructing the object
[ http://issues.apache.org/jira/browse/IO-82?page=comments#action_12441785
]
Stephen Colebourne commented on IO-82:
--
We need to force a gc to start the cleaner working
Test takes a long time to run in Mustang
Author: bayard
Date: Thu Oct 12 09:34:06 2006
New Revision: 463310
URL: http://svn.apache.org/viewvc?view=revrev=463310
Log:
Fixing #IO-82, in JDK 1.6 the gc call does not seem to be giving us the garbage
collection, so Stephen came up with this to force that to happen. Speed seems
good in JDK
[ http://issues.apache.org/jira/browse/IO-82?page=all ]
Henri Yandell resolved IO-82.
-
Resolution: Fixed
svn ci -m Fixing #IO-82, in JDK 1.6 the gc call does not seem to be giving us
the garbage collection, so Stephen came up with this to force that to
[ http://issues.apache.org/jira/browse/IO-94?page=comments#action_12441792
]
Niall Pemberton commented on IO-94:
---
From what you describe this is what MockInputStream and MockReader do out of
the box:
MockInputStream:
[ http://issues.apache.org/jira/browse/IO-86?page=comments#action_12441815
]
Niall Pemberton commented on IO-86:
---
I see using an exception as a better mechanism for stopping the processing
(unwinding from the recursively called walk method) -
[ http://issues.apache.org/jira/browse/POOL-87?page=all ]
Henri Yandell closed POOL-87.
-
Resolution: Invalid
User requested the issue be closed.
Commons-Pool does not always calling activateObject on newly created Objects
Can you summarise in a paragraph why we need a publicly published POM
for those of us maven luddites who haven't followed the discussion, so
we know what is being voted on and why?
thanks
Stephen
Jochen Wiedmann wrote:
Hi,
as already discussed on
On 10/12/06, Stephen Colebourne [EMAIL PROTECTED] wrote:
Can you summarise in a paragraph why we need a publicly published POM
for those of us maven luddites who haven't followed the discussion, so
we know what is being voted on and why?
Think of the POM as a project configuration, it's a
I'd like to remove the issueManagement section. It does not provide
anything useful to the project that inherits from the parent pom. The
project still has to define this for itself.
Other than that I'm +1
--
Dennis Lundberg
Jochen Wiedmann wrote:
Hi,
as already discussed on
On 10/12/06, Dennis Lundberg [EMAIL PROTECTED] wrote:
I'd like to remove the issueManagement section. It does not provide
anything useful to the project that inherits from the parent pom. The
project still has to define this for itself.
Other than that I'm +1
I agree with you, but for in
[ http://issues.apache.org/jira/browse/IO-94?page=comments#action_12441851
]
Stephen Colebourne commented on IO-94:
--
As I said before, I'm unsure of the use case outside mocking, but this is a
strange world, and someone will probably think
[ http://issues.apache.org/jira/browse/IO-86?page=comments#action_12441855
]
Stephen Colebourne commented on IO-86:
--
It sounds like you have a use case for this which I don't have at present. I
suggest that you commit your code and we make
Author: niallp
Date: Thu Oct 12 17:37:09 2006
New Revision: 463529
URL: http://svn.apache.org/viewvc?view=revrev=463529
Log:
IO-94 - Re-name MockInputStream/MockReader to NullInputStream/NullReader
Added:
Author: niallp
Date: Thu Oct 12 17:54:17 2006
New Revision: 463536
URL: http://svn.apache.org/viewvc?view=revrev=463536
Log:
javadoc only
Modified:
jakarta/commons/proper/io/trunk/src/java/org/apache/commons/io/output/CountingOutputStream.java
Modified:
Author: niallp
Date: Thu Oct 12 17:56:18 2006
New Revision: 463537
URL: http://svn.apache.org/viewvc?view=revrev=463537
Log:
IO-84 - add tests for files 2GB to IOUtils.copy(), CountingInputStream and
CountingOutputStream
Modified:
Yes, it'll be GC'ed when thread is. And initialized when first use in a new
thread. Here is the test code:
public class TestMe {
private static String pattern = -mm-dd;
private static ThreadLocal formatter = new ThreadLocal() {
protected Object
XPathExpressionEngine nodeKey method create a wrong key for attribute node
--
Key: CONFIGURATION-230
URL: http://issues.apache.org/jira/browse/CONFIGURATION-230
Project: Commons
OTOH - how much time costs the creation and usage of such an object?
Good question. I wrote a rough test code and posted in my earlier email, the
result on my P-M 1.2G laptop is 2797 v.s. 22656 in 1M calls, so really not
much.
Author: niallp
Date: Thu Oct 12 21:46:16 2006
New Revision: 463560
URL: http://svn.apache.org/viewvc?view=revrev=463560
Log:
IO-86 - Update cancellation behaviour in DirectoryWalker to use a new
CancelException, also add IOException to lifecycle method definitions.
Modified:
[ http://issues.apache.org/jira/browse/IO-94?page=all ]
Niall Pemberton updated IO-94:
--
Summary: New NullnputStream NullReader implementations (was: New Mock
InputStream Writer implementations)
New NullnputStream NullReader implementations
[ http://issues.apache.org/jira/browse/IO-94?page=comments#action_12441911
]
Niall Pemberton commented on IO-94:
---
MockInputStream and MockReader have been renamed to NullInputStream and
NullReader
New NullnputStream NullReader
[ http://issues.apache.org/jira/browse/IO-86?page=comments#action_12441912
]
Niall Pemberton commented on IO-86:
---
OK, changed committed. IMO internal and external cancellation are now well
documented and the exception is called
[ http://issues.apache.org/jira/browse/IO-84?page=all ]
Niall Pemberton resolved IO-84.
---
Resolution: Fixed
Closing as fixed.
Many classes are limited to length of stream 2 GB, and behave incorrectly
on larger streams
close stream when creating a signature
--
Key: SANDBOX-177
URL: http://issues.apache.org/jira/browse/SANDBOX-177
Project: Commons Sandbox
Issue Type: Bug
Components: OpenPGP
Affects
[ http://issues.apache.org/jira/browse/SANDBOX-177?page=all ]
Antoine Levy-Lambert updated SANDBOX-177:
-
Attachment: patch.txt
Patch
close stream when creating a signature
--
Key:
[PATCH] ant task to generate signatures
---
Key: SANDBOX-178
URL: http://issues.apache.org/jira/browse/SANDBOX-178
Project: Commons Sandbox
Issue Type: New Feature
Components: OpenPGP
45 matches
Mail list logo