Well I am shamed right now. But I still need help figuring out where this JAI break occurred from ...
Talking to Justin he has it building better with the "nojai" profile (just added the docs to the root pom.xml), and is now stuck on some CRS failure - I am going to follow in his footsteps and hunt down the failures and rollback as needed. If someone else has a clue about the JAI conflict that would be great. Jody > Justin Deoliveira ha scritto: >> Activating the nojai profile makes it go away. But this leads to more >> problems in the referencing module. The following tests fail: >> >> testEnvelopeTransformation(org.geotools.referencing.CrsTest) >> >> testTransverseMercator(org.geotools.referencing.operation.projection.SouthOrientedTest) >> >> >> >> testKrovak(org.geotools.referencing.operation.projection.SouthOrientedTest) >> >> >> With the following error from the test logs: >> >> org.geotools.factory.FactoryNotFoundException: No factory of kind >> "CoordinateOperationAuthorityFactory" found. >> >> Jody, you have been playing around with FactoryFinder stuff latley no? >> >> Aside, I have about had it with trying to maintain a continuous for >> geotools. Its a huge time sync and our build *never* works. > > This is bad, because not having a continuous build opens the door > for more breakage. Yet, it seems the continuous build server did > not scare people enough. > > We can try a few options, before calling the trunk on trunk experiment > failed and go back to the old way of doing things. > One would be to build a sort of "hall of shame" with a score that > goes up for each developer that breaks the build. > Another is stronger, but I guess it would work: if some commit breaks > the build, and does not get fixed within a few hours, revert it > without mercy. Unfortunately this does not work for geoapi changes... > On "Joel on Software" they suggested that a very good way was to make > the one who break the build the build manager, that is, the one that > has to take care of the build system. Unfortunately I don't think > we can apply this. > > All in all, I'd say just revert changes that break the build, we have > a versioning system, people that break the build for long times will > go thru the pain of resurrecting their changes and fixing them. > If the changes people are doing are so big that they make the build > consistently break, then I'd say a branch is needed. > > Cheers > Andrea ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Geotools-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/geotools-devel
