Yes I noticed that but that's not the problem here! On Mar 13, 2018 11:50, "Jan Lahoda" <[email protected]> wrote:
> One thing to note is that binaries-list expects the hashes to be > upper-case, while the sum on the maven central is lower case. I.e.: > 4AE44A6157E1BC4C5B373BE0C274A8F1D9BADD76 > is expected instead of: > 4ae44a6157e1bc4c5b373be0c274a8f1d9badd76 > > Jan > > > On Tue, Mar 13, 2018 at 4:35 PM, Christian Bourque < > [email protected]> wrote: > > > Thanks Jaroslav, I'll have a look at this class and maybe I'll understand > > what is going on! > > > > On Tue, Mar 13, 2018 at 10:45 AM, Jaroslav Tulach < > > [email protected] > > > wrote: > > > > > Hard to say without debugging. You can find the code here: > > > https://github.com/apache/incubator-netbeans/blob/ > > > master/nbbuild/antsrc/org/netbeans/nbbuild/extlibs/ > > > DownloadBinaries.java#L368 > > > -jt > > > > > > > > > 2018-03-13 14:44 GMT+01:00 Christian Bourque < > > [email protected] > > > >: > > > > > > > Hey guys, > > > > > > > > Thanks for your replies! > > > > > > > > But I think I wasn't clear enough in my original post: basically my > > > > question is why is NetBeans expecting a different hash value than the > > one > > > > found on Maven Central? > > > > > > > > C. > > > > > > > > > > > > On Tue, Mar 13, 2018 at 4:29 AM, Jaroslav Tulach < > > > > [email protected]> > > > > wrote: > > > > > > > > > Did you built JGit from source and installed it into local Maven > > > > > repository? Then the locally built version takes precedence over > the > > > one > > > > in > > > > > the Maven central and has, of course, different SHA. Adjust the SHA > > or > > > > > delete the local copy from .m2/repository and let the build > download > > > the > > > > > real version from Maven central. > > > > > > > > > > -jt > > > > > > > > > > PS: My experience tells me this is painful, sorry for the hassle. > > > > > > > > > > > > > > > 2018-03-13 1:48 GMT+01:00 Christian Bourque < > > > [email protected] > > > > >: > > > > > > > > > > > Hi all, > > > > > > > > > > > > I'm currently trying to compile NetBeans 9.0 with a more recent > > > Eclipse > > > > > > JGit library (the one currently bundled with > > > > > > NetBeans is 3 years old!) but I'm having a hard time with the > hash > > > > (sha1) > > > > > > value that is used by the build system that's > > > > > > downloading the artefact from the Maven Central repository! When > I > > > > build > > > > > > NetBeans it says that the hash value > > > > > > that I put in the binaries-list file doesn't match the hash of > the > > > > > artefact > > > > > > found on Maven Central, but I took the hash > > > > > > value directly from the Maven Central repository so I expected it > > to > > > be > > > > > > correct: > > > > > > http://central.maven.org/maven2/org/eclipse/jgit/org. > > > > eclipse.jgit/4.11.0 > > > > > . > > > > > > 201803080745-r/ > > > > > > > > > > > > So then I decided to download the Eclipse JGit source code from > > > GitHub: > > > > > > https://github.com/eclipse/jgit/releases/tag/v4.11.0. > > 201803080745-r > > > > > > compile it and generate the sha1sum myself. To my surprise the > > > computed > > > > > > checksum was equal to the one expected by NetBeans during the > build > > > > > > process. > > > > > > > > > > > > So I'm a bit puzzled! If the build script is downloading an > > external > > > > > > dependency from Maven Central then the hash should > > > > > > match the one found in this file: > > > > > > http://central.maven.org/maven2/org/eclipse/jgit/org. > > > > eclipse.jgit/4.11.0 > > > > > . > > > > > > 201803080745-r/org.eclipse.jgit-4.11.0.201803080745-r.jar.sha1 > > > > > > > > > > > > Anyway, if anyone could shed some light on this situation I would > > > > really > > > > > > appreciate it! > > > > > > > > > > > > Thanks, > > > > > > > > > > > > Christian > > > > > > > > > > > > > > > > > > > > >
