[Octave-bug-tracker] [bug #46406] parallel-3.0.3 failed with Octave 4.0.0 and 64 bit integers on Linux

[Martin K]

URL:
  <http://savannah.gnu.org/bugs/?46406>

                 Summary: parallel-3.0.3 failed with Octave 4.0.0 and 64 bit
integers on Linux
                 Project: GNU Octave
            Submitted by: grisu86
            Submitted on: Mo 09 Nov 2015 13:06:17 GMT
                Category: Octave Forge Package
                Severity: 3 - Normal
                Priority: 5 - Normal
              Item Group: Build Failure
                  Status: None
             Assigned to: None
         Originator Name: 
        Originator Email: 
             Open/Closed: Open
         Discussion Lock: Any
                 Release: 4.0.0
        Operating System: GNU/Linux

    _______________________________________________________

Details:

I cannot compile the current version of the parallel package on Linux(amd64)
with a 64-bit enabled Octave.  The C++ compiler used is gcc 4.8.5. 

When installing parallel-3.0.3 using 

pkg install -verbose -forge parallel 


I get the following error 

g++ -c  -fPIC
-I/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave/..
-I/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave
-I/scratch/koehlerm/local_install/conda-2.7/include -pthread -fopenmp -g -O2
-I/scratch/koehlerm/local_install/conda-2.7/include   -Wno-int-to-pointer-cast
   network_get_info.cc -o network_get_info.o
network_get_info.cc: In function ‘octave_value_list Fnetwork_get_info(const
octave_value_list&, int)’:
network_get_info.cc:138:34: error: conversion from ‘int’ to ‘const
idx_vector’ is ambiguous
       retval.assign (i, node_info);
                                  ^
network_get_info.cc:120:12: note: candidates are:
   for (int i = 0; i < nconns; i++)
            ^
In file included from
/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave/../octave/Array.h:36:0,
                 from
/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave/../octave/boolMatrix.h:27,
                 from
/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave/../octave/mx-base.h:32,
                 from
/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave/../octave/Matrix.h:30,
                 from
/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave/../octave/oct.h:33,
                 from network_get_info.cc:24:
/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave/../octave/idx-vector.h:507:3:
note: idx_vector::idx_vector(bool)
   idx_vector (bool x) : rep (new idx_mask_rep (x)) { chkerr (); }
   ^
/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave/../octave/idx-vector.h:504:3:
note: idx_vector::idx_vector(float)
   idx_vector (float x) : rep (new idx_scalar_rep (x)) { chkerr (); }
   ^
/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave/../octave/idx-vector.h:502:3:
note: idx_vector::idx_vector(double)
   idx_vector (double x) : rep (new idx_scalar_rep (x)) { chkerr (); }
   ^
/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave/../octave/idx-vector.h:495:3:
note: idx_vector::idx_vector(char)
   idx_vector (char c) : rep (new idx_colon_rep (c)) { chkerr (); }
   ^
/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave/../octave/idx-vector.h:467:3:
note: idx_vector::idx_vector(octave_idx_type)
   idx_vector (octave_idx_type i) : rep (new idx_scalar_rep (i))
   ^
/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave/../octave/idx-vector.h:430:3:
note: idx_vector::idx_vector(idx_vector::idx_base_rep*) <near match>
   idx_vector (idx_base_rep *r) : rep (r) { }
   ^
/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave/../octave/idx-vector.h:430:3:
note:   no known conversion for argument 1 from ‘int’ to
‘idx_vector::idx_base_rep*’
In file included from network_get_info.cc:30:0:
/scratch/koehlerm/local_install/conda-2.7/include/octave-4.0.0/octave/oct-map.h:429:8:
error:   initializing argument 1 of ‘void octave_map::assign(const
idx_vector&, const octave_map&)’
   void assign (const idx_vector& i, const octave_map& rhs);
 

There are two solution for this problem. On the one hand, extending the
idx_vector class by a function 

idx_vector(int x) 
 
or changing the (incorrect) usage of and standard int in network_get_info.cc
which is done by the attached patch. 



    _______________________________________________________

File Attachments:


-------------------------------------------------------
Date: Mo 09 Nov 2015 13:06:17 GMT  Name: parallel-3.0.3-wrong-int.patch  Size:
532B   By: grisu86
Patch mentioned in the bug report. 
<http://savannah.gnu.org/bugs/download.php?file_id=35421>

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?46406>

_______________________________________________
  Nachricht gesendet von/durch Savannah
  http://savannah.gnu.org/