Re: Core failures from Eclipse

Well, clearly we need to figure out why the failures occur. I don’t believe that the tests “depend” on their order. It seems more likely that some tests are not cleaning up properly when they are done or are not cleaning up when they start. Ralph > On Dec 2, 2014, at 7:45 AM, Gary Gregory wr

Re: logging-log4j2 git commit: Test independence.

So clearMarkers is called both before and after? Then I don’t understand why the failure occurred in MarkerTest. Ralph > On Dec 1, 2014, at 1:38 PM, Gary Gregory wrote: > > Note that MarkerTest already had: > > @Before > @After > public void clearMarkers() { > MarkerManag

Re: logging-log4j2 git commit: Let the formatCause method have a null guard clause instead of each call site guarding the call.

No. I think the guidelines are pretty clear. It is just that making changes just because you think something would look better just creates more work for the rest of us as we review unnecessary changes. And in cases like this they might have a detrimental impact. Even then the code wouldn’t

Re: logging-log4j2 git commit: Let the formatCause method have a null guard clause instead of each call site guarding the call.

On Tue, Dec 2, 2014 at 9:49 PM, Ralph Goers wrote: > Please see http://logging.apache.org/log4j/2.x/guidelines.html#teamwork item > 4. > > “Fixing” things that only you have identified as a problem should have > been preceded by a discussion. > Wow, well, that seems like a pretty small change to

Re: logging-log4j2 git commit: Let the formatCause method have a null guard clause instead of each call site guarding the call.

Please see http://logging.apache.org/log4j/2.x/guidelines.html#teamwork item 4. “Fixing” things that only you have identified as a problem should have been preceded by a discussion. Ralph > On Dec 2, 2014, at 6:54 PM, Gary Gregory

Re: logging-log4j2 git commit: Let the formatCause method have a null guard clause instead of each call site guarding the call.

On Tue, Dec 2, 2014 at 6:44 PM, Remko Popma wrote: > Not sure that I agree with this not being a functional change: The null > check used to be on cause.getCauseProxy(), now it is on cause. > The old pattern was: if (cause != null) { foo(cause); } ... if (cause != null) { foo(cause); } v

Re: logging-log4j2 git commit: Let the formatCause method have a null guard clause instead of each call site guarding the call.

On Tue, Dec 2, 2014 at 6:37 PM, Ralph Goers wrote: > Why did you make the change? > Because I do not like code duplication and we had null checks before calling a method. It looks less cluttered now IMO. > Was something broken? > Not for the normal definition of "broken". But we don't have to

Re: logging-log4j2 git commit: Let the formatCause method have a null guard clause instead of each call site guarding the call.

Not sure that I agree with this not being a functional change: The null check used to be on cause.getCauseProxy(), now it is on cause. About unit tests: I believe the problem scenario is when the Throwable class does not exist on the process that deserialized the ThrowableProxy. It is difficult to

Re: logging-log4j2 git commit: Let the formatCause method have a null guard clause instead of each call site guarding the call.

Why did you make the change? Was something broken? Ralph > On Dec 2, 2014, at 3:08 PM, Gary Gregory wrote: > > Yes: This is a change that refactors guard clauses, no behavioral change. > Tests pass. > No: I do not know if the specific case you mention is covered by the unit > tests or the cod

Re: logging-log4j2 git commit: Let the formatCause method have a null guard clause instead of each call site guarding the call.

Yes: This is a change that refactors guard clauses, no behavioral change. Tests pass. No: I do not know if the specific case you mention is covered by the unit tests or the code for that matter. The tests do pass though. Thoughts? Gary On Tue, Dec 2, 2014 at 5:01 PM, Remko Popma wrote: > Are yo

Re: logging-log4j2 git commit: Let the formatCause method have a null guard clause instead of each call site guarding the call.

Are you sure this is correct? What about deserialized ThrowableProxies where the cause is null but the causeProxy is non-null? Sent from my iPhone > On 2014/12/03, at 3:10, [email protected] wrote: > > Repository: logging-log4j2 > Updated Branches: > refs/heads/master ab58a6d69 -> df71c934d

[jira] [Commented] (LOG4J2-904) RollingFile appender doesn't zip and create new files

[ https://issues.apache.org/jira/browse/LOG4J2-904?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14232097#comment-14232097 ] Nick Hardy commented on LOG4J2-904: --- I just ran into this bug as well, but not until i a

[jira] [Commented] (LOG4J2-891) AbstractLifecycle should not implement equals() and hashCode()

[ https://issues.apache.org/jira/browse/LOG4J2-891?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14231981#comment-14231981 ] Mariano Gonzalez commented on LOG4J2-891: - perfect, thank you! > AbstractLifecycl

[jira] [Comment Edited] (LOG4J2-891) AbstractLifecycle should not implement equals() and hashCode()

[ https://issues.apache.org/jira/browse/LOG4J2-891?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14231967#comment-14231967 ] Gary Gregory edited comment on LOG4J2-891 at 12/2/14 7:28 PM: --

[jira] [Comment Edited] (LOG4J2-891) AbstractLifecycle should not implement equals() and hashCode()

[ https://issues.apache.org/jira/browse/LOG4J2-891?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14231967#comment-14231967 ] Gary Gregory edited comment on LOG4J2-891 at 12/2/14 7:28 PM: --

[jira] [Resolved] (LOG4J2-891) AbstractLifecycle should not implement equals() and hashCode()

[ https://issues.apache.org/jira/browse/LOG4J2-891?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Gary Gregory resolved LOG4J2-891. - Resolution: Fixed Fix Version/s: 2.2 Setting version to 2.2. Code has been in git master fo

[jira] [Commented] (LOG4J2-891) AbstractLifecycle should not implement equals() and hashCode()

[ https://issues.apache.org/jira/browse/LOG4J2-891?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14231950#comment-14231950 ] Mariano Gonzalez commented on LOG4J2-891: - hello. Can you confirm this fix will be

Core failures from Eclipse

If I execute all the Core tests in one go from Eclipse, I get a lot of failures. I fixed one but we have a lot of tests that depends on test order. They just happen to work from Maven most of the time. See http://pastebin.com/CkdUYZLs Gary -- E-Mail: [email protected] | [email protected]

[jira] [Commented] (LOG4J2-913) Allow to load log4j2 configuration from a centralized repository, like a database or a webservice

[ https://issues.apache.org/jira/browse/LOG4J2-913?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14231550#comment-14231550 ] Gary Gregory commented on LOG4J2-913: - Patches welcome ;-) > Allow to load log4j2 con

[jira] [Commented] (LOG4J2-911) Configuration monitorInterval is not honored for XInclude hrefs

[ https://issues.apache.org/jira/browse/LOG4J2-911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14231541#comment-14231541 ] Edwin F. López commented on LOG4J2-911: --- Thanks a lot. I have oppened the new featur

[jira] [Created] (LOG4J2-913) Allow to load log4j2 configuration from a centralized repository, like a database or a webservice

Edwin F. López created LOG4J2-913: - Summary: Allow to load log4j2 configuration from a centralized repository, like a database or a webservice Key: LOG4J2-913 URL: https://issues.apache.org/jira/browse/LOG4J2-913

[jira] [Closed] (LOG4J2-911) Configuration monitorInterval is not honored for XInclude hrefs

[ https://issues.apache.org/jira/browse/LOG4J2-911?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Gary Gregory closed LOG4J2-911. --- Resolution: Won't Fix Closing as requested. Feel free to contribute to Log4j through Jira, the mailin

[jira] [Commented] (LOG4J2-911) Configuration monitorInterval is not honored for XInclude hrefs

[ https://issues.apache.org/jira/browse/LOG4J2-911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14231497#comment-14231497 ] Edwin F. López commented on LOG4J2-911: --- Hi Gary. Well, I thought to do that, but di

[jira] [Updated] (LOG4J2-911) Configuration monitorInterval is not honored for XInclude hrefs

[ https://issues.apache.org/jira/browse/LOG4J2-911?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Gary Gregory updated LOG4J2-911: Summary: Configuration monitorInterval is not honored for XInclude hrefs (was: monitorInterval is n

[jira] [Updated] (LOG4J2-911) monitorInterval is not honored for XInclude hrefs

[ https://issues.apache.org/jira/browse/LOG4J2-911?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Gary Gregory updated LOG4J2-911: Summary: monitorInterval is not honored for XInclude hrefs (was: monitorInterval is not honored whe

[jira] [Commented] (LOG4J2-911) monitorInterval is not honored when using XInclude

[ https://issues.apache.org/jira/browse/LOG4J2-911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14231489#comment-14231489 ] Gary Gregory commented on LOG4J2-911: - bq. I can't "touch" my config file since it is

[jira] [Commented] (LOG4J2-911) monitorInterval is not honored when using XInclude

[ https://issues.apache.org/jira/browse/LOG4J2-911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14231469#comment-14231469 ] Edwin F. López commented on LOG4J2-911: --- Thank you very much Gary and Remko for your

[jira] [Updated] (LOG4J2-911) monitorInterval is not honored when using XInclude

[ https://issues.apache.org/jira/browse/LOG4J2-911?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Edwin F. López updated LOG4J2-911: -- Description: Hi there. I'm using Log4J2 version 2.1 and while doing a proof of concept for cent