On 18/09/07 at 13:20 +0100, Matthew Johnson wrote: > reassign 443039 gjdoc > merge 439468 443039 > thanks > > Lucas, you already reported this as #439468, and I already reassigned > it and merged it with the other FTBFS that gjdoc is causing. I don't > suppose your scripts could refrain from filing further bugs on the > subject? If you feel there should be a serious bug filed against my > package, could you assign the existing ones back, with justification, > rather than filing new ones.
My script doesn't file any bug. I dunno why you think so. All bugs are filed manually. Your package failed to build again in a sid chroot updated yesterday. In the same chroot, libcsv-java, libgtk-java, and gcj-4.2 built fine (all of them used to fail because of gjdoc). I'm not sure if the bug is in gjdoc or libmatthew-java, but I'm sure that there was/is still a bug. Interestingly, I just rebuilt your package on my laptop, and it built fine *despite* a java.lang.NullPointerException from gjdoc. Looking at the failure log more closely: javadoc is called in 'debian/rules build', and this call resulted in a "java.net.ConnectException: Connection refused" (internet is not accessible from the systems where I run the build). But it seems that this exception is ignored somehow. Then javadoc is called again in 'debian/rules binary', this time resulting in a java.lang.NullPointerException. This one seems fatal. In the build that suceeded (on my laptop), javadoc is only called once in 'debian/rules build'. It's never called in 'binary'. Any idea why this behaviour differs like that? -- | Lucas Nussbaum | [EMAIL PROTECTED] http://www.lucas-nussbaum.net/ | | jabber: [EMAIL PROTECTED] GPG: 1024D/023B3F4F |
signature.asc
Description: Digital signature
