Bug 225073
| Summary: | math/octave: port update to v4.2.1 | ||||||
|---|---|---|---|---|---|---|---|
| Product: | Ports & Packages | Reporter: | Ben Lauwens <ben.lauwens> | ||||
| Component: | Individual Port(s) | Assignee: | Stephen Montgomery-Smith <stephen> | ||||
| Status: | Closed FIXED | ||||||
| Severity: | Affects Only Me | CC: | emaste, kargl, maho, sgk, stephen | ||||
| Priority: | --- | Flags: | bugzilla:
maintainer-feedback?
(maho) |
||||
| Version: | Latest | ||||||
| Hardware: | Any | ||||||
| OS: | Any | ||||||
| Attachments: |
|
||||||
|
Description
Ben Lauwens
2018-01-11 14:05:28 UTC
Created attachment 189614 [details]
Updated port
Ben, Thanks for creating an update. I have/had a math/octave42 sitting in my tree for a bit, and finally got around to cleaning it up. Decided to check to see if a bug report existed. Guess I can toss my update. Any chance that this will get committed. I think the maintainer has to take action. I have no commit rights... Ben, yes, I know we need to get the maintainer to commit the patch. I should have addressed Maho with my ping. Apologies for any confusion. sorry for not responding. currently I don't have a machine installed FreeBSD. Could someone commit instead? I'd like to drop maintainership as well. I'll take it. I'll take maintainership as well. I am trying to build this port. I am getting the weirdest error. Any ideas?
This is with a recent build of FreeBSD 10-stable.
/bin/sh ./libtool --tag=CXX --mode=compile c++ -std=gnu++11 -DHAVE_CONFIG_H -I. -Iliboctave -I./liboctave -I./liboctave/cruft/misc -I./liboctave/array -Iliboctave/numeric -I./liboctave/numeric -Iliboctave/operators -I./liboctave/operators -I./liboctave/system -I./liboctave/util -I./libinterp/octave-value -Ilibinterp -I./libinterp -I./libinterp/operators -Ilibinterp/parse-tree -I./libinterp/parse-tree -Ilibinterp/corefcn -I./libinterp/corefcn -I./liboctave/wrappers -I/usr/local/include/GraphicsMagick -I/usr/local/include -D_THREAD_SAFE -fPIC -D_THREAD_SAFE -pthread -Wall -W -Wshadow -Wold-style-cast -Wformat -Wpointer-arith -Wwrite-strings -Wcast-align -Wcast-qual -O2 -pipe -fstack-protector -fno-strict-aliasing -MT libinterp/libinterp_liboctinterp_la-liboctinterp-build-info.lo -MD -MP -MF libinterp/.deps/libinterp_liboctinterp_la-liboctinterp-build-info.Tpo -c -o libinterp/libinterp_liboctinterp_la-liboctinterp-build-info.lo `test -f 'libinterp/liboctinterp-build-info.cc' || echo './'`libinterp/liboctinterp-build-info.cc
In file included from libinterp/template-inst/Array-tc.cc:33:
In file included from ./libinterp/octave-value/ov-classdef.h:28:
In file included from /usr/include/c++/v1/map:423:
/usr/include/c++/v1/__tree:747:59: error: static_cast from '__node_pointer' (aka
'std::__1::__tree_node<std::__1::__value_type<cdef_class,
std::__1::list<cdef_class, std::__1::allocator<cdef_class> > >, void *>
*') to '__node_base_pointer' (aka 'std::__1::__tree_node_base<void *> *')
is not allowed
...static_cast<__node_base_pointer>(__ptr_)));
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
(In reply to Stephen Montgomery-Smith from comment #8) The stupidity of no being able to reply to emails for bugzilla is making this difficult. Anyway, I don't use FreeBSD 10 or 11, only current. Can you drop the -std=gnu++11 or change it to gnu++14? This looks like yet another clang incompatibility with gcc. Adding USES=compiler:c++14-lang seemed to fix it. Thanks. Other changes needed:
Need to tell configure to look in ${LOCALBASE}/openjdk8/include/freebsd instead of ${LOCALBASE}/openjdk8/include/linux.
texi2dvi didn't work unless print/tex-formats was installed, so added USE_TEX=formats:build
Added the following lines to pkg-plist:
%%PORTDOCS%%%%DOCSDIR%%/liboctave.ps
%%PORTDOCS%%%%DOCSDIR%%/octave.ps
%%PORTDOCS%%%%DOCSDIR%%/refcard-a4.ps
%%PORTDOCS%%%%DOCSDIR%%/refcard-legal.ps
%%PORTDOCS%%%%DOCSDIR%%/refcard-letter.ps
Now I am seeing if octave-forge builds on top of it.
(In reply to Stephen Montgomery-Smith from comment #11) Stephen, I know a few of the forge packages works. I have octave-4.2.1 High-level interactive language for numerical computations octave-forge-base-1.4_1 Octave-forge baseport for all packages octave-forge-control-3.0.0_4 Octave-forge package control octave-forge-image-2.6.2 Octave-forge package image octave-forge-netcdf-1.0.11_2 Octave-forge package netcdf octave-forge-signal-1.3.2_4 Octave-forge package signal A commit references this bug: Author: stephen Date: Sat Feb 24 21:52:33 UTC 2018 New revision: 462865 URL: https://svnweb.freebsd.org/changeset/ports/462865 Log: - Update to 4.2.1. - Maintainership to stephen@. PR: ports/225073 Submitted by: Ben Lauwens <ben.lauwens@gmail.com> Changes: head/math/octave/Makefile head/math/octave/distinfo head/math/octave/files/patch-build-aux_find-defun-files.sh head/math/octave/files/patch-build-aux_find-files-with-tests.sh head/math/octave/files/patch-build-aux_mk-builtins.sh head/math/octave/files/patch-build-aux_mk-octave-config-h.sh head/math/octave/files/patch-configure head/math/octave/files/patch-gsed head/math/octave/files/patch-libgnu-math.in.h head/math/octave/files/patch-libinterp_mk-pkg-add head/math/octave/files/patch-libinterp_mkops head/math/octave/files/patch-liboctave-numeric-eigs-base.cc head/math/octave/files/patch-liboctave_system_file-stat.cc head/math/octave/files/patch-scripts+pkg+private+configure__make.m head/math/octave/files/patch-scripts_mk-pkg-add head/math/octave/files/patch-scripts_pkg_private_configure__make.m head/math/octave/pkg-plist Committed, and appropriate changes made to the octave-forge ports. Thank you. thanks! |
