octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: have headers & libs, but cannot configure/make


From: Chris Roat
Subject: Re: have headers & libs, but cannot configure/make
Date: Thu, 12 Jan 2017 16:33:13 -0800



On Fri, Jan 6, 2017 at 1:23 PM, Mike Miller <address@hidden> wrote:
I'm moving this to the maintainers list, please include it in future
replies instead of the help list.

On Fri, Jan 06, 2017 at 11:43:01 -0800, Chris Roat wrote:
> I'm happy to help where I can. It's easy for me to test things, and suggest
> improvements.  I've use makefiles/autotools/configure in the past, so might
> be able to suggest things here and there.

If you are able to try to build from the current hg development branch,
the problems you encountered with PCRE and CXSPARSE should now be fixed,
no workarounds needed.

I started to build from the current HEAD.  So far, so good -- the configure step worked once I re-arranged some of my variables.  Though on make (which I run in parallel with -j) I had to install bison, flex, and gperf due to a couple of errors.  I don't think I needed those before.

The png and zlib libraries had to be appended to FT2_LIBS.  I've always had to add png, but zlib is new, even though I am still specifing it via --with-z="...".
 
Nothing has been done yet about fontconfig/freetype or GLPK yet. Can you
please test with the default branch and see whether you still need
workarounds for those libraries (and whether any other problems have
arisen)?

I need to compile in the complex long support for UMFPACK, and I'm running into build errors that might be related to that.  See this one:

liboctave/array/CSparse.cc: In member function 'ComplexDET SparseComplexMatrix::determinant(octave_idx_type&, double&, bool) const':
liboctave/array/CSparse.cc:1119:51: error: invalid conversion from 'const octave_idx_type* {aka const long int*}' to 'const long long int*' [-fpermissive]
                                      0, 1, control);
<elided>
umfpack_report_matrix.h:43:18: note: initializing argument 3 of 'long long int umfpack_zl_report_matrix(long long int, long long int, const long long int*, const long long int*, const double*, const double*, long long int, const double*)'
 SuiteSparse_long umfpack_zl_report_matrix

Can you tell if that's a config error?
 

I notice that, on my system, `pkg-config --libs fontconfig` yields
"-lfontconfig -lfreetype". So it may be that Octave is relying on that
implicit dependency, while in your case you need to add -lfreetype.

Thanks - make sense.  I appended ${FT2_LIBS} to --with-fontconfig.
 

I also just spotted a possible typo, configure.ac refers to
FREETYPE_LIBS, which should probably be FT2_LIBS. Can you test whether
that helps your build?

Without making a change, I've been making progress.  But until I'm fully done, I won't be able to comment on this.
 

--
mike


reply via email to

[Prev in Thread] Current Thread [Next in Thread]