Yes, it does need to be a build require. RC4 is ready from my perspective!

From: Chris Marshall 
Sent: Tuesday, May 05, 2015 9:25 PM
To: Ed 
Cc: pdl-devel 
Subject: Re: [Pdl-devel] build problems PDL-2.008 RC3

Hi Ed-


Are we good for an RC4 release?
Regarding File::Path, I think it needs to be a build requires.


--Chris



On Sat, May 2, 2015 at 11:29 AM, Ed <ej...@hotmail.com> wrote:

  This is some excellent research. My analysis of your results is there are two 
problems shown:

  1. A legitimate compile failure because of a clang flag supplied when 
compiler is not clang (fixed in branch buildfixes, 
https://github.com/PDLPorters/pdl/pull/102)

  2. A perl failure from a broken Perl installation, which doesn't have 
File::Path (which has been in Perl core since 5.001) and ignores the prereq now 
stated in Makefile.PL. Really nothing we can or should try to do about that.

  If PR #102 passes, I will merge it and there should probably be an RC4 
release.

  I favour leaving SF#378 until after 2.008.

  -----Original Message----- From: Chris Marshall
  Sent: Saturday, April 25, 2015 2:44 PM
  To: pdl-devel@lists.sourceforge.net
  Subject: [Pdl-devel] build problems PDL-2.008 RC3 


  I've just reviewed the CPAN Testers reports for
  CHM/PDL-2.007_16.tar.gz (a.k.a. PDL-2.008 RC3):

    PDL 2.007_16 (129 ALL, 5 FAILs, 6 NAs, 99 PASSes, 19 UNKNOWNs)

  These are considerably worse than the stats for
  PDL-2.007 where all non-PASS come in at less than
  3%.  Digging in the testers output by category...

     99 PASS:  Good to go

      5 FAIL:  These are all in t/pthread.t and for versions
               of perl 5.21.4, 5.21.5, 5.21.7, and 5.21.8.
               There are PASS for more recent versions of
               perl-5.21.x so I think these are perl issues
               and not PDL ones.

        6 NA:  3 of these are valid since we require perl 5.10.x
               but the remaining 3 are not since they are a
               result of the checking the prerequisites by
               hand and kicking out NA if something is inconsistent.
               I recommend we output a warning instead but let the
               test output stand.  That gives us information about
               PDL rather than hiding behind CPAN Testers issues.

  19 UNKNOWN:  All of these results are in the Mkbootstrap stage
               for either PDL::Ufunc or PDL::Core.  The messages
               are confusing to me but may be related to build
               glitches introduced by EUMM 7.04.

  Here are representative sections from the two types of
  UNKNOWN results:



    Running Mkbootstrap for PDL::Ufunc ()
    chmod 644 "Ufunc.bs"
    "/usr/perl5.16.3t/bin/perl"
    "/usr/perl5.16.3t/lib/site_perl/5.16.3/ExtUtils/xsubpp"  -typemap
    "/usr/perl5.16.3t/lib/5.16.3/ExtUtils/typemap" -typemap
    "/usr/home/eserte/.cpan/build/PDL-2.007_16-MJ8sAt/Basic/Core/typemap.pdl"
    Ufunc.xs > Ufunc.xsc && mv Ufunc.xsc Ufunc.c
    cc -c "-I/usr/home/eserte/.cpan/build/PDL-2.007_16-MJ8sAt/Basic/Core"
    -Wno-literal-conversion  -DHAS_FPSETMASK -DHAS_FLOATINGPOINT_H
    -fno-strict-aliasing -pipe -fstack-protector -I/usr/local/include
    -O    -DVERSION=\"2.007_16\"  -DXS_VERSION=\"2.007_16\" -DPIC -fPIC
    "-I/usr/perl5.16.3t/lib/5.16.3/amd64-freebsd-thread-multi/CORE" Ufunc.c
    cc1: error: unrecognized command line option "-Wno-literal-conversion"
    *** [Ufunc.o] Error code 1

    Running Mkbootstrap for PDL::Core ()
    chmod 644 "Core.bs"
    "/usr/home/cpan/pit/thr/perl-5.18.2/bin/perl" "-I../../blib/arch"
    "-I../../blib/lib" pdldataswitch.c.PL pdldataswitch.c
    (    case PDL_B: retval = PDL.bvals.Byte; break;     case PDL_S:
    retval = PDL.bvals.Short; break;     case PDL_US: retval =
    PDL.bvals.Ushort; break;     case PDL_L: retval = PDL.bvals.Long;
    break;     case PDL_IND: retval = PDL.bvals.Indx; break;     case
    PDL_LL: retval = PDL.bvals.LongLong; break;     case PDL_F: retval =
    PDL.bvals.Float; break;     case PDL_D: retval = PDL.bvals.Double;
    break;     case PDL_A: retval = PDL.bvals.Anyval; break;) at
    ../../blib/lib/PDL/Core/Dev.pm line 731.
    "/usr/home/cpan/pit/thr/perl-5.18.2/bin/perl" "-I../../blib/arch"
    "-I../../blib/lib" pdlbadvalinit.c.PL pdlbadvalinit.c
    "/usr/home/cpan/pit/thr/perl-5.18.2/bin/perl"
    "/usr/home/cpan/pit/thr/perl-5.18.2/lib/site_perl/5.18.2/ExtUtils/xsubpp"
    -typemap
    "/usr/home/cpan/pit/thr/perl-5.18.2/lib/5.18.2/ExtUtils/typemap"
    -typemap "typemap"  Core.xs > Core.xsc && mv Core.xsc Core.c
    Can't locate File/Path.pm in @INC (you may need to install the File::Path
    module) (@INC contains: many directories) at Dev.pm line 23.
    BEGIN failed--compilation aborted at Dev.pm line 23.
    Compilation failed in require at -e line 1.
    Error reading from pipe 'perl -e "require q{Dev.pm};
    PDL::Core::Dev::generate_core_flags()"':  in Core.xs, line 341
    *** Error code 1


  Hope this helps track these down.

  --Chris



  ------------------------------------------------------------------------------
  One dashboard for servers and applications across Physical-Virtual-Cloud
  Widest out-of-the-box monitoring support with 50+ applications
  Performance metrics, stats and reports that give you Actionable Insights
  Deep dive visibility with transaction tracing using APM Insight.
  http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
  _______________________________________________
  pdl-devel mailing list
  pdl-devel@lists.sourceforge.net
  https://lists.sourceforge.net/lists/listinfo/pdl-devel 

------------------------------------------------------------------------------
One dashboard for servers and applications across Physical-Virtual-Cloud 
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
_______________________________________________
pdl-devel mailing list
pdl-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/pdl-devel

Reply via email to