This may or not be a bug in the code (though I think the nullpointerexception
is clearly a bug from a user perspective - the failure should be more
informative - I've submitted a bug about it in JIRA, MRELEASE-167).
But if it isn't a bug, it should be more clearly described in the online
documentation. This was an unnecessary waste of several hours, and I can
see from the other posts to the nabble forum, I wasn't the first. I'll
submit a JIRA issue about this.
Cheers,
Harold
Hi,
I've found the reason why release:prepare fails for my project (it's
an issue from Chas Douglass' post).
The scm section of my pom.xml was like this:
<scm>
<developerConnection>scm:svn:file:///usr/local/svnrep/xxx/trunk</
developerConnection>
</scm>
I downloaded the release plugin sources and checked the write method
in the PropertiesReleaseConfigurationStore.java class (this is where
the process fails).
And I've found, that the method assumes that a <connection> tag is
mandatory inside the scm section.
So the (temporary?) solution to this problem is to include both, the
<connection> and <developerConnection> in the scm section, eg:
<scm>
<connection>scm:svn:file:///usr/local/svnrep/xxx/trunk</connection>
<developerConnection>scm:svn:file:///usr/local/svnrep/xxx/trunk</
developerConnection>
</scm>
I think that this is a bug. The <developerConnection> should be
mandatory, but <connection> should be optional, because the
<developerConnection> is needed to perform
the release:perform. Or at least the process should stop and an
apropriate message should send to console.
If it's a bug, I will file a JIRA issue and send a patch.
What do you think?
Regards,
Jakub
--
View this message in context:
http://www.nabble.com/Release-Failure-tf1624778.html#a6574232
Sent from the Maven - Users mailing list archive at Nabble.com.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
