Thanks for the report. My thoughts below.... --Chris Henning Glawe wrote: > Moin, > here are the errors reported by the test suite on debian's autobuilders, with > the processor architecture where the error occured in brackets. > > will redo with verbose output. >
There was a problem with the PDL Data::Dumper interface as it constructs the objects to work with. The pointer is stringified as an integer in some earlier Data::Dumper releases which creates a string such as -123412344 and then an object is generated with a name like PDL-123412344 which is not a valid identifier. I believe the problem is fixed in a more recent Data::Dumper but I'm not sure which version. > dumper.t (ia64) > ================ > t/dumper.t .................. > # Perlmsgs: > Fatal error: argument is probably not a piddle, or magic no overwritten. > You're in trouble, guv: 178336 9657040 10148352 > (in cleanup) Fatal error: argument is probably not a piddle, or magic no > overwritten. You're in trouble, guv: 178048 9694464 7512456 > (in cleanup) Fatal error: argument is probably not a piddle, or magic no > overwritten. You're in trouble, guv: 178336 9657040 10148352 > (in cleanup) Fatal error: argument is probably not a piddle, or magic no > overwritten. You're in trouble, guv: 178168 9747344 10187824 > # Looks like you planned 16 tests but ran 5. > # Looks like your test exited with 255 just after 5. > # Testmsgs: > Dubious, test returned 255 (wstat 65280, 0xff00) > Failed 11/16 subtests My guess for the fftw.t failures is they come from size of double issues for the platforms. > fftw.t (alpha) - the t/fftw.t is after a string with a missing newline (!) > ==============------------------------------------------------------------ > # Failed test 'fftnd() and ifftw()' > # at t/fftw.t line 51. > > # Failed test 'ifftw() fftw() and original' > # at t/fftw.t line 59. > > # Failed test 'fftnd and infftw' > # at t/fftw.t line 69. > > # Failed test 'infftw nfftw and original' > # at t/fftw.t line 77. > fftw: rfftwnd_real_to_complex with complex-to-real plant/fftw.t > .................... > Dubious, test returned 1 (wstat 256, 0x100) > Failed 9/9 subtests > > fftw.t (ia64) > ============= > t/fftw.t .................... > # Testmsgs: > Failed 9/9 subtests I don't know. Maybe Judd has an idea. > hdf_sd.t (all) > ============== > t/hdf_sd.t .................. > # Perlmsgs: > PDL: Dimensions must be positive > Caught at file t/hdf_sd.t, line 306, pkg main > # Looks like you planned 32 tests but ran 31. > # Looks like your test exited with 2 just after 31. > # Testmsgs: > Dubious, test returned 2 (wstat 512, 0x200) > Failed 1/32 subtests These are warnings from reloading the definitions more than twice. The fix will be to load things only if needed but the clean up did not make the PDL-2.4.5 release. > opengl.t (all) > ============== > t/opengl.t .................. ok > # Perlmsgs: > Subroutine OpenGL::StructureNotifyMask redefined at > /build/buildd-pdl_2.4.5+dfsg-1-s390-XubPH0/pdl-2.4.5+dfsg/blib/lib/PDL/Graphics/OpenGL/Perl/OpenGL.pm > line 30. > Subroutine reshape redefined at > /build/buildd-pdl_2.4.5+dfsg-1-s390-XubPH0/pdl-2.4.5+dfsg/blib/lib/PDL/Graphics/TriD/GL.pm > line 706, <DATA> line 387 Don't know, I have yet to get plplot working on cygwin and I don't know if anyone has it working for win32 either. > plplot.t (ia64) > =============== > t/plplot.t .................. > # Testmsgs: > Failed 18/35 subtests _______________________________________________ Perldl mailing list [email protected] http://mailman.jach.hawaii.edu/mailman/listinfo/perldl
