116226 2005-12-20 19:02 0000 version bump: dev-java/commons-net-1.4.1 2005-12-21 20:32:06 0000 1 1 1 Unclassified Gentoo Linux Ebuilds unspecified All Linux RESOLVED FIXED http://jakarta.apache.org/commons/net/ InSVN P3 enhancement --- 1 nichoj@gentoo.org java@gentoo.org nichoj@gentoo.org 2005-12-20 19:02:08 0000 We are currently at 1.3.0... and 1.4.1 was relased a week or so ago. nichoj@gentoo.org 2005-12-20 21:29:16 0000 A simple bump doesn't seem to entire cut it. With USE=junit, I found the tests fail: [junit] Running org.apache.commons.net.time.TimeTCPClientTest [junit] Tests run: 2, Failures: 0, Errors: 1, Time elapsed: 1.093 sec [junit] Testsuite: org.apache.commons.net.time.TimeTCPClientTest [junit] Tests run: 2, Failures: 0, Errors: 1, Time elapsed: 1.093 sec [junit] Testcase: testInitial took 0.009 sec [junit] Testcase: testCompareTimes took 1.074 sec [junit] Caused an ERROR [junit] null [junit] java.lang.NullPointerException [junit] at org.apache.commons.net.SocketClient.disconnect(SocketClient.java:266) [junit] at org.apache.commons.net.time.TimeTCPClientTest.testCompareTimes(TimeTCPClientTest.java:133) I put an ebuild up on experimental at http://gentooexperimental.org/svn/java/gentoo-java-experimental/dev-java/commons-net nichoj@gentoo.org 2005-12-21 08:59:41 0000 I managed to fix the null pointers, and filed a bug upstream: http://issues.apache.org/bugzilla/show_bug.cgi?id=37985 The next issue is that another test in the same TestCase attempts to lookup the local machine's hostname. If the host doens't have a fully qualified domain name and doesn't have an entry in /etc/hosts for the hostname, then the test will fail. nichoj@gentoo.org 2005-12-21 20:26:46 0000 So I looked into it a little more, and apparently you don't get the null pointers in the first place if you have your hostname in your /etc/hosts file. nichoj@gentoo.org 2005-12-21 20:32:06 0000 Fixed in CVS. I've disabled unit tests for now, until we it can get fixed upstream...