You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@commons.apache.org by Henri Yandell <fl...@gmail.com> on 2006/09/10 10:10:40 UTC
[lang] LANG-262
Any thoughts on this issue?
https://issues.apache.org/jira/browse/LANG-262
Says that the cEnumClasses Hashmap in the Enum class is keeping a
ClassLoader from being garbage collected and that switching it to a
WeakHashMap should fix it.
Seems fair to me, and an easy fix for 2.2; but not something I can
think of simple ways to unit test.
Any thoughts?
Hen
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [lang] LANG-262
Posted by Henri Yandell <fl...@gmail.com>.
On 9/12/06, Stephen Colebourne <sc...@btopenworld.com> wrote:
> I don't know whether a WeakHashMap works when the keys are Class objects
> (as Class objects aren't exactly run-of the-mill).
>
> This fix might work, or it might break things somewhere else. The
> [logging] team may be able to comment on WeakHashMap.
>
> ATM, I'm against fixing it in a rush.
I've committed this to trunk (2.2 is branched off) and asked the
reporter to give it a go. Long-run for 2.3 we can try to come up with
some code to test this.
Hen
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [lang] LANG-262
Posted by Stephen Colebourne <sc...@btopenworld.com>.
I don't know whether a WeakHashMap works when the keys are Class objects
(as Class objects aren't exactly run-of the-mill).
This fix might work, or it might break things somewhere else. The
[logging] team may be able to comment on WeakHashMap.
ATM, I'm against fixing it in a rush.
Stephen
Henri Yandell wrote:
> Any thoughts on this issue?
>
> https://issues.apache.org/jira/browse/LANG-262
>
> Says that the cEnumClasses Hashmap in the Enum class is keeping a
> ClassLoader from being garbage collected and that switching it to a
> WeakHashMap should fix it.
>
> Seems fair to me, and an easy fix for 2.2; but not something I can
> think of simple ways to unit test.
>
> Any thoughts?
>
> Hen
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [lang] LANG-279 HashCodeBuilder
Posted by Rahul Akolkar <ra...@gmail.com>.
On 9/29/06, Gary Gregory <gg...@seagullsoftware.com> wrote:
> Is SVN down? I am getting the following error:
>
<snip/>
Looks fine to me.
-Rahul
>
> Gary
>
> Problems reported while synchronizing SVNStatusSubscriber. 0 of 1
> resources were synchronized.
> An error occurred synchronizing /Apache Jakarta Commons
> trunks-proper/lang: Error getting status for resource F/Apache Jakarta
> Commons trunks-proper/lang org.tigris.subversion.javahl.ClientException:
> RA layer request failed
> svn: PROPFIND request failed on
> '/repos/asf/jakarta/commons/proper/lang/trunk'
> svn: PROPFIND of '/repos/asf/jakarta/commons/proper/lang/trunk': could
> not connect to server (https://svn.apache.org)
>
> Error getting status for resource F/Apache Jakarta Commons
> trunks-proper/lang org.tigris.subversion.javahl.ClientException: RA
> layer request failed
> svn: PROPFIND request failed on
> '/repos/asf/jakarta/commons/proper/lang/trunk'
> svn: PROPFIND of '/repos/asf/jakarta/commons/proper/lang/trunk': could
> not connect to server (https://svn.apache.org)
>
> org.tigris.subversion.javahl.ClientException: RA layer request
> failed
> svn: PROPFIND request failed on
> '/repos/asf/jakarta/commons/proper/lang/trunk'
> svn: PROPFIND of '/repos/asf/jakarta/commons/proper/lang/trunk': could
> not connect to server (https://svn.apache.org)
>
> org.tigris.subversion.javahl.ClientException: RA layer request
> failed
> svn: PROPFIND request failed on
> '/repos/asf/jakarta/commons/proper/lang/trunk'
> svn: PROPFIND of '/repos/asf/jakarta/commons/proper/lang/trunk': could
> not connect to server (https://svn.apache.org)
>
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
RE: [lang] LANG-279 HashCodeBuilder
Posted by Gary Gregory <gg...@seagullsoftware.com>.
Is SVN down? I am getting the following error:
Gary
Problems reported while synchronizing SVNStatusSubscriber. 0 of 1
resources were synchronized.
An error occurred synchronizing /Apache Jakarta Commons
trunks-proper/lang: Error getting status for resource F/Apache Jakarta
Commons trunks-proper/lang org.tigris.subversion.javahl.ClientException:
RA layer request failed
svn: PROPFIND request failed on
'/repos/asf/jakarta/commons/proper/lang/trunk'
svn: PROPFIND of '/repos/asf/jakarta/commons/proper/lang/trunk': could
not connect to server (https://svn.apache.org)
Error getting status for resource F/Apache Jakarta Commons
trunks-proper/lang org.tigris.subversion.javahl.ClientException: RA
layer request failed
svn: PROPFIND request failed on
'/repos/asf/jakarta/commons/proper/lang/trunk'
svn: PROPFIND of '/repos/asf/jakarta/commons/proper/lang/trunk': could
not connect to server (https://svn.apache.org)
org.tigris.subversion.javahl.ClientException: RA layer request
failed
svn: PROPFIND request failed on
'/repos/asf/jakarta/commons/proper/lang/trunk'
svn: PROPFIND of '/repos/asf/jakarta/commons/proper/lang/trunk': could
not connect to server (https://svn.apache.org)
org.tigris.subversion.javahl.ClientException: RA layer request
failed
svn: PROPFIND request failed on
'/repos/asf/jakarta/commons/proper/lang/trunk'
svn: PROPFIND of '/repos/asf/jakarta/commons/proper/lang/trunk': could
not connect to server (https://svn.apache.org)
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [lang] LANG-279 HashCodeBuilder
Posted by Henri Yandell <fl...@gmail.com>.
On 9/19/06, Henri Yandell <fl...@gmail.com> wrote:
> We also have:
>
> https://issues.apache.org/jira/browse/LANG-69
I've assigned both of the ObjectCycle bugs to you Gary under the
assumption that your previous email means that you have a particular
set of fixes in mind.
Feel free to unassign them again if that's wrong and I'll take a stab at a fix.
Hen
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [lang] LANG-279 HashCodeBuilder
Posted by Henri Yandell <fl...@gmail.com>.
We also have:
https://issues.apache.org/jira/browse/LANG-69
ToStringBuilder has cycle problems, but the supplied patch messed
things up and needs either ignoring and the fix being done from
scratch, or ripping apart and applying.
I've branched 2.2.x off of the trunk, so trunk is now 2.3-SNAPSHOT and
you can dive in and start working on this on trunk (and hopefully
LANG-69 if it's the same concept and you're willing). I'm going to
start doing 2.3 stuff on trunk as well.
We can then backport to 2.2 when it gets committed if 2.2's not
released. Unless anyone thinks we should hold 2.2 up for these
particular bugs.
Hen
On 9/19/06, Gary Gregory <gg...@seagullsoftware.com> wrote:
> Arg, I just ran into a case in our application where I get a stack
> overflow with HashCodeBuilder and cyclical object trees:
>
> https://issues.apache.org/jira/browse/LANG-279
>
> I can fix this the same why I fixed ReflectionToStringBuilder.
>
> The question is: do this now or after 2.2?
>
> Gary
>
> > -----Original Message-----
> > From: Henri Yandell [mailto:flamefew@gmail.com]
> > Sent: Sunday, September 10, 2006 1:11 AM
> > To: Jakarta Commons Developers List
> > Subject: [lang] LANG-262
> >
> > Any thoughts on this issue?
> >
> > https://issues.apache.org/jira/browse/LANG-262
> >
> > Says that the cEnumClasses Hashmap in the Enum class is keeping a
> > ClassLoader from being garbage collected and that switching it to a
> > WeakHashMap should fix it.
> >
> > Seems fair to me, and an easy fix for 2.2; but not something I can
> > think of simple ways to unit test.
> >
> > Any thoughts?
> >
> > Hen
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> >
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
[lang] LANG-279 HashCodeBuilder
Posted by Gary Gregory <gg...@seagullsoftware.com>.
Arg, I just ran into a case in our application where I get a stack
overflow with HashCodeBuilder and cyclical object trees:
https://issues.apache.org/jira/browse/LANG-279
I can fix this the same why I fixed ReflectionToStringBuilder.
The question is: do this now or after 2.2?
Gary
> -----Original Message-----
> From: Henri Yandell [mailto:flamefew@gmail.com]
> Sent: Sunday, September 10, 2006 1:11 AM
> To: Jakarta Commons Developers List
> Subject: [lang] LANG-262
>
> Any thoughts on this issue?
>
> https://issues.apache.org/jira/browse/LANG-262
>
> Says that the cEnumClasses Hashmap in the Enum class is keeping a
> ClassLoader from being garbage collected and that switching it to a
> WeakHashMap should fix it.
>
> Seems fair to me, and an easy fix for 2.2; but not something I can
> think of simple ways to unit test.
>
> Any thoughts?
>
> Hen
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org