a few additional notes on the fuseki build issue. I upgrade to the latest Maven and as expected, that didn't make a difference. I also tried to build in cygwin but that didn't help either, i.e. same error message. None of my paths have spaces btw, something that often interferes with scripts. For now, we can assume the problem is unique to my machine/installation. If there is someone else here who can try to build on Windows, that would help confirm this of course. I agree it is entirely uncritical for the release itself
From: Andy Seaborne <[email protected]> To: [email protected] Date: 10/15/2012 05:21 AM Subject: Re: Preparation for Jena 2.7.4 release Having got the SDB release out of the way, I'll start on a Jena 2.7.4 On 08/10/12 21:34, Andy Seaborne wrote: > There are a few things to be sorted before the release process can start. > > 1/ JENA-333 :: javadoc production > (from the DAMLVocabulary issue) > > Do we need to do anything here? > I'd rather use Java 7 (generating Java 6 class files) as before. The two javadoc files DAMLVocabulary and DAML_OIL are zero bytes but nothing else seems to be affected. As DAML+OIL is all rather old, I propose ignoring the problem. > > 2/ JENA-333 ; DAML+OIL > > Are we going to deprecate vocabularies DAML_OIL and DAMLVocabulary? > Should anything else be deprecated as well? > > My default is to deprecate the two vocabularies. > Ian's done this and also deprecated many related things. > 3/ JENA-296 :: tdb.TS_TDBLoader3 test fails in Windows > > Unless someone has some inspiration, I propose to simple remove the > tests. The problem seems to be the tests can't clean up across tests in > the same run, rather than being anything to do with tdbloader3 Proposal: Avoid the test problem by building on Linux (I may find time to crank up a pure Windows environment to take a look; not-critical) It may be better to remove tdbloader3 from the command line documentation on the website for the moment. It already does require significant tuning to get it running faster than the other loaders. 4/ There is a potential build problem with Fuseki encountered by Simon - but it does not seem to affect me (yet :-). Either it's windows dependent, the incremental version of maven or bad luck. Proposal: If it does not interfere with building, it is not critical. Andy
