Re: [Geotools-devel] Java 9 test build

Ah, that actually makes sense, looks like I forgot to use OS-specific newlines for that test; I suspect windows is expecting /r/n for that assert. Should be a simple fix, just an error in the test case. Thanks for testing. Torben On Tue, Jan 16, 2018 at 4:54 PM, Vincent Privat wrote: > Hi Torb

Re: [Geotools-devel] Java 9 test build

Hi Torben, I have a test failure: Failed tests: SLDTransformerTest.testLabelCDataNewline:1440 null Tests run: 1642, Failures: 1, Errors: 0, Skipped: 1 ... [INFO] GeoTools ... SUCCESS [ 7.843 s] ... [INFO] Main module ..

Re: [Geotools-devel] Java 9 test build

GeoTools can now pass all tests (including unsupported modules) on Java 9, while retaining backwards-compatibility with Java 8. See: https://github.com/tbarsballe/geotools/tree/java9-compat-1.0 The caveat to this is that it requires using --add-modules=java.xml.bind --add-modules=java.activation t

Re: [Geotools-devel] Java 9 test build

On Mon, Jan 15, 2018 at 11:29 PM, Andrea Aime wrote: > On Tue, Jan 16, 2018 at 2:19 AM, Torben Barsballe < > tbarsba...@boundlessgeo.com> wrote: > >> After this was added to the test argLine, there was only one other >> failure in gt-imagemosaic - HeterogenousCRSTest.testHeterogeneousCRS(). >> Th

Re: [Geotools-devel] Java 9 test build

On Tue, Jan 16, 2018 at 2:19 AM, Torben Barsballe < tbarsba...@boundlessgeo.com> wrote: > After this was added to the test argLine, there was only one other failure > in gt-imagemosaic - HeterogenousCRSTest.testHeterogeneousCRS(). This test > is known to be unstable, and may be as much a test conf

Re: [Geotools-devel] Java 9 test build

Wow what a lot of research: - With Java EE being sent to the eclipse foundation do you know if jaxb is also? Oh wow they renamed it. - it will be interesting to see if this removes the need for OSGi. > >1. *No two modules may export the same packag

Re: [Geotools-devel] Java 9 test build

Hello All, The SLDTransformer tests have been merged. Everything seems to be working fine for those. The next issue that has come up is a host of test failures in gt-imagemosaic, related to JAXB, Activation, and JAI / ImageIO I have tried a couple approaches for these failures. *TL;DR:* JAXB and

Re: [Geotools-devel] Java 9 test build

Ok, seems like everyone is happy with accepting both inputs. PR here: https://github.com/geotools/geotools/pull/1790 Ian - I noticed a comment on some functionality missing from the Java XPATH processor, so that might be why XMLUnit isn't being used here. Several of the test do use it, just not al

Re: [Geotools-devel] Java 9 test build

This may be a good time to update our XML tests to use XMLUnit ( http://www.xmlunit.org/) instead of simple string comparisons to avoid this sort of style issue. Ian On 13 January 2018 at 11:03, Ian Turton wrote: > I did some work to tidy up CDATA handling in https://github.com/ > geotools/geot

Re: [Geotools-devel] Java 9 test build

I did some work to tidy up CDATA handling in https://github.com/geotools/geotools/pull/1735 which could have helped there. So it may be possible to close that issue off too. Ian On 13 January 2018 at 00:07, Torben Barsballe wrote: > > > On Fri, Jan 12, 2018 at 2:13 PM, Jody Garnett > wrote: >

Re: [Geotools-devel] Java 9 test build

Hi Torben, can we just accept both outputs for the time being? The CDATA section contents are the same so all good I'd say. Cheers Andrea On Fri, Jan 12, 2018 at 10:59 PM, Torben Barsballe < tbarsba...@boundlessgeo.com> wrote: > I took a deeper look at the SLDTransformer failures. As noted in t

Re: [Geotools-devel] Java 9 test build

So to continue .. I would be content to relax the tests so that the whitespace differences between Java 8 and Java 9 pretty printing do not result in failure. The XML machine to machine functionality is not impacted. -- Jody Garnett On 12 January 2018 at 16:07, Torben Barsballe wrote: > > > On

Re: [Geotools-devel] Java 9 test build

On Fri, Jan 12, 2018 at 2:13 PM, Jody Garnett wrote: > That is great news Torben, some caution/testing is required when making > whitespace changes around CDATA sections. > > A couple of developers (Alex and David) went in here to try and untangle > how whitespace either side of a CDATA section w

Re: [Geotools-devel] Java 9 test build

That is great news Torben, some caution/testing is required when making whitespace changes around CDATA sections. A couple of developers (Alex and David) went in here to try and untangle how whitespace either side of a CDATA section was getting "lost" when round tripping from SLD --> (geotools sty

Re: [Geotools-devel] Java 9 test build

I took a deeper look at the SLDTransformer failures. As noted in the PMC Meeting notes thread, the failures are related changes in XML output, particularly surrounding pretty-print / indentation. Looking at the test cases, the Java 9 result actually seems better than the Java 8 result. For example:

Re: [Geotools-devel] Java 9 test build

I've updated the epsg-wkt tests with a threshold, and they now work with both Java 8 and Java 9 (locally at least). This definitely seems like an issue to be aware of (and something that should probably be mentioned in any initial blog post discussing GeoTools Java 9 support), but I don't think it

Re: [Geotools-devel] Java 9 test build

We are an LGPL library and cannot back port Java 9 code... -- Jody Garnett On 3 January 2018 at 14:31, Vincent Privat wrote: > Hi Torben, > Thank you for testing Java 9. > We faced similar reprojection issues in JOSM unit tests two years ago. > You can find valuable information (why it happens

Re: [Geotools-devel] Java 9 test build

Hi Torben, Thank you for testing Java 9. We faced similar reprojection issues in JOSM unit tests two years ago. You can find valuable information (why it happens and how we solved the issues) in our bugtracker: https://josm.openstreetmap.de/ticket/11889 https://josm.openstreetmap.de/ticket/13387 ht

[Geotools-devel] Java 9 test build

I have set up a Java 9 build on build.geoserver.org: https://build.geoserver.org/view/geotools/job/geotools-java9 (also set up builds for geowebcache and geoserver, but those aren't likely to be relevant until geotools builds on java 9) The build is set up on a shared workspace "/var/jenkins/works