On Mon, Jan 26, 2009 at 09:21:31AM -0800, Alan Irwin wrote:
> On 2009-01-26 10:33-0000 Andrew Ross wrote:
> > On Thu, Jan 22, 2009 at 12:56:53PM -0800, Alan Irwin wrote:
> >> I asked the guy to try cmake-2.6.2 since I was unable to confirm the issue
> >> for cmake-2.6.2, and it is possible some inconsistency with cmake-2.4.x has
> >> crept in.
> >> If it does turn out to be a cmake-2.4.x issue, I won't fix it since
> >> cmake minimum version for our project will shortly be 2.6.0.
> > Alan,
> > Any progress on this? I've been using cmake-2.4.x until this week when
> > you made the changes to plplot and I upgraded to the latest Ubuntu
> > release. Based on this it is unlikely to be a cmake issue. plcol0Cmd is
> > defined in the automatically generated tclgen.h, which is why the bug
> > reporter couldn't find it in the src tree. The fact that this is the
> > first entry after the tclgen generated tclgen_s.h file suggests either a
> > path problem or a tclgen problem.
> Thanks, Andrew, for the reminder. I had assumed I would be notified by
> e-mail if they guy said anything more, but it turns out that does not
> work with the new SF system, and you have to keep checking (UGH!).
> The guy said 2.6.2 made no difference (just like you predicted). Would you
> like to take over at
> now? I am not at all familiar with our tcl bindings implementation so I am
> not quite sure what next debugging step to recommend to this guy.
I've posted a response ask for more details (specifically the tclgen
generated files) and cmake output.
His other problem appears to be gcj related. The error looks like gcj is
being called (without the -C option) instead of javac. On my machine at
least gcj comes with a javac script to properly emulate the standard