tomcat-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mark Thomas <>
Subject Re: svn commit: r966882 - in /tomcat/trunk: build.xml checkstyle.xml webapps/docs/changelog.xml
Date Sat, 24 Jul 2010 08:12:50 GMT
On 24/07/2010 04:05, Bill Barker wrote:
>> I'm -1 (vote, not veto) on having validate be a dependency for
>> compile. Historically, Tomcat has been very lenient about what is
>> required just to build the project, and I can't see the reason to
>> force checkstyle just to be able to build. Mostly my concern is with
>> Gump (which I can fix the metadata when svn comes up again). But that
>> puts a transitive dependency on bcel that will get fixed who knows when.
>> I'm more than happy to add a tomcat-trunk-validate project to Gump if
>> it is split from compile.
> Having tried this (yes, I should have done this first), changing to -1
> veto. In the OS world of cat herding, we should allow for the fact that
> somebody mistakenly violates the checkstyle rules and commits without
> building. It happens all the time, and then they go off to sleep and
> nobody else can build the project with a fresh checkout for hours.
> As I've said, I'm happy to add this to Gump (assuming that bcel ever
> builds ever again) if it a separate target to get regular automated
> notices. But developers can run a separate target themselves (just like
> test).

I'll add a flag (defaults to off) that controls if the checkstyle target 
actually does anything. That'll give folks the option to enable it 
locally if they wish. We could do a similar thing with test too.


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message