[jira] [Comment Edited] (LOG4J2-1348) Add an AutoCloseable ThreadContext class: CloseableThreadContext

2016-04-16 Thread Remko Popma (JIRA)
[ https://issues.apache.org/jira/browse/LOG4J2-1348?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15244495#comment-15244495 ] Remko Popma edited comment on LOG4J2-1348 at 4/17/16 12:59 AM:

[jira] [Commented] (LOG4J2-1348) Add an AutoCloseable ThreadContext class: CloseableThreadContext

2016-04-16 Thread Remko Popma (JIRA)
[ https://issues.apache.org/jira/browse/LOG4J2-1348?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15244495#comment-15244495 ] Remko Popma commented on LOG4J2-1348: - Ralph's comment made me realize: the static me

Re: [4/5] logging-log4j2 git commit: LOG4J2-1334 Jackson-based layouts somehow filter out Message.getFormat() for Log4jLogEvent. Need to set up the same filters for MutableLogEvent but don't know how.

2016-04-16 Thread Gary Gregory
Ah! Thank you for the clarification. Gary On Apr 16, 2016 1:56 PM, "Remko Popma" wrote: > Well, many existing unit tests fail without this. That's how I found there > was a problem. > > Sent from my iPhone > > On 2016/04/16, at 23:28, Gary Gregory wrote: > > Again: do we a matching unit test? >

Re: [2/5] logging-log4j2 git commit: LOG4J2-1334 MutableLogEvent bugfix: don't clear thread name

2016-04-16 Thread Remko Popma
You are right. Unit tests are on the way. I started with fixing the issues brought to light by the existing unit tests, but I still need to add tests for MutableLogEvent and its factory. About that comment: thread name is initialized once by the factory when the MutableLogEvent is created and

[jira] [Commented] (LOG4J2-1348) Add an AutoCloseable ThreadContext class: CloseableThreadContext

2016-04-16 Thread Matt Sicker (JIRA)
[ https://issues.apache.org/jira/browse/LOG4J2-1348?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15244456#comment-15244456 ] Matt Sicker commented on LOG4J2-1348: - I also like the use of just push and put. I'v

Re: [4/5] logging-log4j2 git commit: LOG4J2-1334 Jackson-based layouts somehow filter out Message.getFormat() for Log4jLogEvent. Need to set up the same filters for MutableLogEvent but don't know how.

2016-04-16 Thread Remko Popma
Well, many existing unit tests fail without this. That's how I found there was a problem. Sent from my iPhone > On 2016/04/16, at 23:28, Gary Gregory wrote: > > Again: do we a matching unit test? > > Gary > > -- Forwarded message -- > From: > Date: Apr 16, 2016 5:55 AM > Su

Re: [3/5] logging-log4j2 git commit: LOG4J2-1334 ListAppender must add snapshot of MutableLogEvent to the list, not the MutableLogEvent itself (since it will change)

2016-04-16 Thread Remko Popma
Makes sense. Sent from my iPhone > On 2016/04/16, at 23:20, Gary Gregory wrote: > > Hi, > > The ser+deser sequence feels like something that should be refactored into a > serializedCopy() or deepCopy() method depending on whether or not you want to > publicize the copying technique. > > Ga

[jira] [Commented] (LOG4J2-1348) Add an AutoCloseable ThreadContext class: CloseableThreadContext

2016-04-16 Thread Ralph Goers (JIRA)
[ https://issues.apache.org/jira/browse/LOG4J2-1348?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15244316#comment-15244316 ] Ralph Goers commented on LOG4J2-1348: - I don't think I understand why there the "with

[jira] [Updated] (LOG4J2-1348) Add an AutoCloseable ThreadContext class: CloseableThreadContext

2016-04-16 Thread Greg Thomas (JIRA)
[ https://issues.apache.org/jira/browse/LOG4J2-1348?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Greg Thomas updated LOG4J2-1348: Attachment: ctc-also.patch This patch implements the .push() / .withPush() / .put() / .withPut() AP

[jira] [Commented] (LOG4J2-1348) Add an AutoCloseable ThreadContext class: CloseableThreadContext

2016-04-16 Thread Greg Thomas (JIRA)
[ https://issues.apache.org/jira/browse/LOG4J2-1348?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15244280#comment-15244280 ] Greg Thomas commented on LOG4J2-1348: - OK, I'll stick with the withPush / withPut >

[jira] [Commented] (LOG4J2-1348) Add an AutoCloseable ThreadContext class: CloseableThreadContext

2016-04-16 Thread Gary Gregory (JIRA)
[ https://issues.apache.org/jira/browse/LOG4J2-1348?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15244277#comment-15244277 ] Gary Gregory commented on LOG4J2-1348: -- I really do not like alsoPut(); with() would

[jira] [Commented] (LOG4J2-1348) Add an AutoCloseable ThreadContext class: CloseableThreadContext

2016-04-16 Thread Greg Thomas (JIRA)
[ https://issues.apache.org/jira/browse/LOG4J2-1348?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15244274#comment-15244274 ] Greg Thomas commented on LOG4J2-1348: - Hmm, I quite like the {{CloseableThreadContext

Fwd: [4/5] logging-log4j2 git commit: LOG4J2-1334 Jackson-based layouts somehow filter out Message.getFormat() for Log4jLogEvent. Need to set up the same filters for MutableLogEvent but don't know how

2016-04-16 Thread Gary Gregory
Again: do we a matching unit test? Gary -- Forwarded message -- From: Date: Apr 16, 2016 5:55 AM Subject: [4/5] logging-log4j2 git commit: LOG4J2-1334 Jackson-based layouts somehow filter out Message.getFormat() for Log4jLogEvent. Need to set up the same filters for MutableLogEven

Fwd: [2/5] logging-log4j2 git commit: LOG4J2-1334 MutableLogEvent bugfix: don't clear thread name

2016-04-16 Thread Gary Gregory
1) I'm seeing some commits fly by sans unit tests. Are setting our selves up for regressions? 2) Also, better comments would help (me). For example, this comments states" THreadName should not be cleared" but why? 3) Typo in capitalization: THreadName Gary -- Forwarded message --

Fwd: [3/5] logging-log4j2 git commit: LOG4J2-1334 ListAppender must add snapshot of MutableLogEvent to the list, not the MutableLogEvent itself (since it will change)

2016-04-16 Thread Gary Gregory
Hi, The ser+deser sequence feels like something that should be refactored into a serializedCopy() or deepCopy() method depending on whether or not you want to publicize the copying technique. Gary -- Forwarded message -- From: Date: Apr 16, 2016 5:55 AM Subject: [3/5] logging-log