Just using this mail to recall some general rules for the new
committers, and older ones. No blame, just trying to give some feedback
about such broken builds, as I already experienced the problem more than
once
- before committing, always run the build locally.
- just because it runs well on your computer, it does not mean it will
for other. If you are using Eclipse, you may have added a dependency on
your project class path and forgot to update the associated pom.xml (if
condition #1 is met, it's most often the reason for the failure). 'mvn
clean install' is the way to test that your build works, running tests
in the IDE is't enough.
- when you skip those steps, you may have other issues, like the one we
met on FtpServer, for those running the tests on a different platform.
Basically, Windows can lock some files when linux does not, file name
case sensitivity can be a issue on Mac OSX and Windows, etc. Sadly, it's
difficult to ask everyone to check on every platform :/
- When you have a broken build, just check the last commits, and provide
all the needed information (which test fails, logs, etc). As you have
svn upped before running the tests, try to add info about your last
revision and the commits that you have applied. It can help those who
have committed the code which broke the build
- You can also try to get the build fixed.
- Don't @ignore a failing test : pretty sure that doing so, the @ignore
test will remain @ignored and committed, and never fixed in the near
future. It's better to wait a bit to see what's going on...
In any case, remember that the more you work on the code, the more
likely you may break the build. Don't blame those who break it (unless
it's a pattern ;), they most certainly didn't intend to break it :)
Thanks !
Fernando Padilla wrote:
I just tried to do a normal build, and I suppose it runs tests by
default:
...
[INFO] Compilation failure
/Users/fern/work/vysper/server/extensions/xep0045-muc/src/test/java/org/apache/vysper/xmpp/modules/extension/xep0045_muc/AbstractServerInfoDiscoTestCase.java:[48,4]
method does not override or implement a method from a supertype
Anyone want to review and fix it?
--
--
cordialement, regards,
Emmanuel Lécharny
www.iktek.com
directory.apache.org
