Re: interval 1.4.0 released

[Marius Schamschula]

Oliver,

On Feb 9, 2016, at 11:37 AM, Oliver Heimlich <address@hidden> wrote:

> Hi,
> 
> a new release of the interval package for real-valued interval
> arithmetic [1] is out, version 1.4.0.
> 
> A summary of important user-visible changes is also available online
> [2]. I am particularly excited about the new function “expm” for
> interval matrix exponentiation, which can be used for initial value
> problems of linear ODEs with constant coefficients. More interesting use
> cases of the package can be found in the manual [3].
> 
> The interval package for real-valued interval arithmetic allows one to
> evaluate functions over subsets of their domain.  All results are
> verified, because interval computations automatically keep track of any
> errors.
> 
> These concepts can be used to handle uncertainties, estimate arithmetic
> errors and produce reliable results.  Also it can be applied to
> computer-assisted proofs, constraint programming, and verified computing.
> 
> The implementation is based on interval boundaries represented by
> binary64 numbers and is conforming to IEEE Std 1788-2015, IEEE standard
> for interval arithmetic.
> 
> Enjoy Octave and free your intervals,
> Oliver
> 
> [1] http://octave.sourceforge.net/interval/
> [2] http://octave.sourceforge.net/interval/NEWS.html
> [3] http://octave.sourceforge.net/interval/package_doc/Examples.html
> 
> _______________________________________________
> Help-octave mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/help-octave


While building interval-1.4.0 under MacPorts (against octave 4.0.0), I get the 
following error:

/usr/bin/clang++ -c -I/opt/local/include -fPIC 
-I/opt/local/include/octave-4.0.0/octave/.. 
-I/opt/local/include/octave-4.0.0/octave -I/opt/local/include -D_THREAD_SAFE 
-pthread -pipe -Os -arch x86_64 -stdlib=libc++    mpfr_function_d.cc -o 
mpfr_function_d.o
mpfr_function_d.cc:184:24: error: conversion from 'const octave_int<unsigned 
long long>' to 'unsigned long' is ambiguous
      mpfr_fac_ui (mp, arg.elem (i), rnd);
                       ^~~~~~~~~~~~

Maybe clang++ is more picky than g++...

Marius
--
Marius Schamschula