I will see if we can exclude the IDE directory from RAT.  I think we do want to 
include them in the source distribution.

In general, the source artifact includes everything in the source tree. The 
source zip should be buildable (in fact, that is one of the things we should be 
testing in the release vote). I believe that if log4j-perf was removed it would 
require a modification to some other pom, otherwise Maven would complain.  So 
no, it is not a problem.

The other tests jars are not included because nothing else needs them as 
dependencies as far as I am aware.  Several of the projects require the core 
tests jar to run their own tests.  Of course, that doesn’t mean the core-tests 
jar needs to be part of the distribution.

I would recommending opening a Jira for the documentation mistakes and pasting 
those there so that they do get fixed.

Before fixing checkstyle issues that show up in the report we should make sure 
that the checkstye configuration matches our style guidelines.

I’m not sure how to customize the issue tracking report. I believe the link to 
Atlassian is placed there by the plugin. To me though it makes sense that the 
word “Jira” would be linked to the Jira product web page and not the ASF Jira 
instance.

Ralph



On Jun 22, 2014, at 8:03 AM, Remko Popma <[email protected]> wrote:

> +1  Site looks good, tests all pass, artifacts look good.
> 
> Potential issue:
> RAT warns that src/ide/eclipse/4.3.2/organize-imports.importorder does not 
> have a license.
> 
> I could be wrong, but I don't think this is a showstopper. This file is not 
> in any of our binary distribution artifacts (although it is in the -src 
> artifact).
> This seems similar to the warning for the two js files, and we agreed with 
> previous releases that these are not a problem:
>   src/site/resources/js/jquery.js
>   src/site/resources/js/jquery.min.js
> 
> Also, the log4j-perf module is excluded from the binary distribution as 
> discussed on the ML, but the apache-log4j-2.0-rc2-src artifact contains the 
> source files for the log4j-perf module. Just to clarify: this is 
> intentionally, and not a problem, right?
> 
> 
> I noticed several other minor things, none of them showstoppers, just making 
> a note here so they can be fixed in an upcoming release:
> 
> Artifacts:
> Looks like only the JUnit tests for the log4j-core module are in the 
> apache-log4j-2.0-rc2-bin artifact.
> None of the other modules have a <module>-tests.jar file...
> Not a showstopper AFAIC, but worth looking into.
> 
> StyleGuide page:
> Says that block indentation should be +2 spaces, but we're using +4 spaces.
> Says to indent continuation lines with at least +4 spaces, we're using +8 
> spaces.
> Has several links to "Column limit:120)" with a closing bracket ')' but no 
> opening bracket.
> 
> JMX page:
> Description for starting JMX Client GUI is missing log4j-api-2.0.jar in 
> classpath.
> (last 2 command line examples at the bottom of the page)
> 
> Extending Log4j Configuration:
> Spelling mistake:
> "Programattically Adding ..." -> Programatically
> 
> Components side-nav: not all modules have all 6 Component Reports
> (Should have Rat Report, Checkstyle, JavaDocs, Source Xref, PMD, CPD)
> 
> * Commons Logging Bridge: missing PMD, CPD
> * Log4j 1.2 API: missing CPD
> * SLF4J Binding: missing CPD
> * Log4j2 to SLF4J Adapter: missing PMD, CPD
> * Tag Lib: missing PMD, CPD
> * JMX GUI:  missing CPD
> * Web Applications: missing Checkstyle, JavaDocs, Source Xref, PMD, CPD
> * Web Applications has unnecessary "JIRA Report" link
> * NoSQL Support/About: text now says: "Need something about 
> CouchDB/MongoDB"...
> * NoSQL Support: missing Checkstyle, JavaDocs, Source Xref, PMD, CPD
> * NoSQL Support has unnecessary "JIRA Report" link
> 
> Many Checkstyle warnings/errors for all modules.
> No showstoppers that I can see.
> 
> Issue Tracking: first link on that page should be to the Log4J issue tracker,
> not to the Atlassian Jira product page.
> 
> 
> 
> On Sun, Jun 22, 2014 at 9:19 AM, Ralph Goers <[email protected]> 
> wrote:
> This is a vote to release Log4j 2.0-rc2, which will hopefully be the last 
> release candidate.  
> 
> Please note that I did not add the new logo to the web site. I will add that 
> after the release.
> 
> Please test and cast your votes.
> [] +1, release the artifacts
> [] -1, don't release because…
> 
> The vote will remain open for 72 hours (or more if required).
> 
> 
> New features:
> o LOG4J2-675:  RollingFile and RollingRandomAccessFile now write the layout 
> footer before rollover. 
> o LOG4J2-674:  Made RollingFileAppender buffer size configurable. 
> o LOG4J2-42:  Create an appender to route log events to the ServletContext 
> log. 
> o LOG4J2-655:  Add Vagrantfile for testing in GNU+Linux. 
> o LOG4J2-654:  Add log4j-perf module to provide a home for all log4j 
> performance tests.
>         Add support for JMH microbenchmark performance tests. 
> o LOG4J2-652:  Add support for default plugin values and attributes. 
> o LOG4J2-598:  Add support for types other than String for plugin factory 
> values/attributes. 
> o LOG4J2-644:  Implement a SecureSocketAppender and secure server (SSL/TLS). 
> o LOG4J2-629:  Document the system properties used in Log4J 2. 
> o LOG4J2-595:  Introduce Java annotation processor as alternative plugin 
> pre-caching mechanism. This is available in the
>           log4j-plugin-processor module. 
> o LOG4J2-576:  Add org.apache.logging.log4j.Logger.getLevel(). 
> o LOG4J2-439:  Add EncodingPatternConverter to escape newlines and HTML 
> special characters. Thanks to Bruce Brouwer. 
> o LOG4J2-427:  Add support for configuration via YAML. Thanks to Alexander 
> Reelsen. 
> 
> Fixed Bugs:
> o LOG4J2-581:  RollingRandomAccessFile now writes the layout header after 
> rollover. Thanks to Alexander Khokhlov. 
> o LOG4J2-622:  RollingFileManager now correctly honours the bufferedIO 
> configuration after rollover. Thanks to Farooq Khan. 
> o LOG4J2-141:  Improved documentation regarding log4j status logger. Thanks 
> to Joern Huxhorn. 
> o LOG4J2-539:  Fixed issue with "Reconfigure using XML below" function in JMX 
> Client GUI.
>         ConfigurationSource is now a top-level class and can be obtained with 
> Configuration.getConfigurationSource().
>         LoggerContext.getConfiguration().getConfigurationSource() 
>         provides a reliable public method for obtaining a logger context's 
> configuration location and content. Thanks to Colin Froggatt. 
> o LOG4J2-619:  Invalid XML configuration files do not prevent the config file 
> from being checked again. Thanks to Scott Harrington. 
> o LOG4J2-637:  JMX: Updating a Logger's level via jConsole now correctly 
> takes effect. Thanks to Mansoor Sajjad, Jon Wilmoth. 
> o LOG4J2-668:  Correctly process log events when combining AsyncLoggers with 
> AsyncAppender. 
> o LOG4J2-669:  Prevent NPE when combining AsyncLoggers with 
> AsyncLoggerConfigs. 
> o LOG4J2-563:  FlumeAvroManager now always uses a client type of 
> default_failover. Thanks to Michael Friedmann. 
> o LOG4J2-535:  Reset rollover time when size rollover is triggered. 
> o LOG4J2-664:  Moved plugin cache file to META-INF for OSGi compatibility. 
> o LOG4J2-640:  Fix NPE that can be caused by a null ThreadContextClassLoader. 
> o LOG4J2-651:  Log4j 2 throws ArrayIndexOutOfBoundsException. 
> o LOG4J2-620:  Perform reconfiguration in a separate thread to prevent 
> deadlocks. 
> o LOG4J2-639:  Prevent NPE in AsyncLogger and AsyncLoggerConfig if logger is 
> used after log4j has been shut down. Thanks to Mck SembWever. 
> o LOG4J2-469:  FailoverAppender was not resetting its status after the 
> primary appender recovered. 
> o LOG4J2-623:  Generate MDC properties as a JSON map in JSONLayout. 
> o LOG4J2-520:  Resolved issue where AsyncAppender dropped events if queue 
> still contained
>         events when application is stopped. Thanks to JavaTech, Andre Bogus. 
> o LOG4J2-392:  Resolved a problem with the previous solution for LOG4J2-392 
> that resulted in dropped events
>         when using AsyncLoggerConfig with slow appenders when application is 
> stopped. Thanks to Andre Bogus. 
> o LOG4J2-613:  The OSGi version of log4j-web imports Servlet 2.5 at minimum 
> instead of 3.0. 
> o LOG4J2-602:  Unit tests are now less verbose during the build process. 
> o LOG4J2-570:  Fix shutdown thread memory leak in servlet containers. 
> o LOG4J2-542:  Make Throwable transient in ThrowableProxy. 
> o LOG4J2-440:  During shutdown, a NullPointerException could be thrown due to 
> the NullConfiguration class no longer being
>         available to the ClassLoader. 
> o LOG4J2-346:  Cyclic dependency with log4j-slf4j-impl in OSGi. 
> o LOG4J2-345:  The log4j-1.2-api module didn't export any packages to OSGi. 
> o LOG4J2-605:  Password data from the NoSQL plugins no longer shows up in 
> cleartext in debug logging. 
> o LOG4J2-448:  A StringIndexOutOfBounds exception could occur during property 
> substitution. Thanks to X86core. 
> o LOG4J2-597:  StatusLogger was not skipping multiple instances of the FQCN 
> class, causing messages from classes in
>         the Verbose list to be printed. 
> o LOG4J2-564:  Renamed SLF4J logger class to Log4jLogger. 
> o LOG4J2-579:  Rework Level comparison APIs. 
> o LOG4J2-560:  SyslogAppenderTest and RFC5424LayoutTest were failing in Java 
> 8. 
> o LOG4J2-499:  Add equals and hashcode to Log4jLogEvent. 
> o LOG4J2-378:  Add DateLookup and ThreadContextLookup to default lookups. 
> 
> Changes:
> o LOG4J2-419:  Support default value for missing key in look ups with 
> fallbacking to looking in the properties map. Thanks to Woonsan Ko. 
> o LOG4J2-554:  Allow configuration files to be located as Servlet Context 
> resources. 
> o LOG4J2-250:  Refactor Log4jLogEvent to lazily create ThrowableProxy. 
> o LOG4J2-647:  Upgrade to Flume 1.5.0. 
> o LOG4J2-646:  Merge the TLS Syslog appender into the Syslog appender. 
> o LOG4J2-641:  Override commons-logging dependency version in tests. 
> o LOG4J2-566:  Made RollingRandomAccessFileAppender buffer size configurable. 
> Thanks to Luigi Alice. 
> o LOG4J2-628:  Use Clock to generate all log event timestamps, not just for 
> Async Loggers. 
> o LOG4J2-617:  Update SLF4J to 1.7.7. 
> o LOG4J2-616:  Update Jackson to 2.3.3. 
> o LOG4J2-585:  Add support for multiple parents to Markers. Thanks to Bruce 
> Brouwer. 
> o LOG4J2-574:  Make Blocking the default WaitStrategy for Async Loggers. 
> o LOG4J2-555:  Introduce ExtendedLogger interface to facilitate implementing 
> and extending Loggers. Thanks to Bruce Brouwer. 
> o LOG4J2-561:  Allow spaces around commas in Configuration's package 
> attribute. Thanks to vibin. 
> o LOG4J2-547:  Have Logger API expose a PrintWriter instead of custom 
> LoggerStream. Thanks to Bruce Brouwer. 
> o LOG4J2-496:  Allow header and footer to be specified as lookup patterns in 
> PatternLayout. 
> o LOG4J2-410:  SLf4JLogger is now Serializable. Thanks to Ivlin Zeng. 
> o LOG4J2-468:  Add support to add a LoggerConfig. Document two ways to modify 
> the configuration. 
> o LOG4J2-582:  Rename org.apache.logging.log4j.core.net.SocketServer to 
> TCPSocketServer and refactor with UDP. 
> o LOG4J2-592:  Update Jackson to 2.3.2 from 2.2.2. 
> 
> Tag:  
> https://svn.apache.org/repos/asf/logging/log4j/log4j2/tags/log4j-2.0-rc2/
> 
> SVN revision: 1604493
> 
> Web Site: http://people.apache.org/~rgoers/log4j2/
> 
> Artifacts: 
> https://repository.apache.org/content/repositories/orgapachelogging-1003
> 
> Nexus did not send an email. The list of artifacts can be found at the link 
> above. 
> 

Reply via email to