Setup¶
Package managers¶
The easiest way to install Arb including all dependencies is via ready-made packages available for various distributions. Note that some package managers may not have the latest version of Arb.
Debian / Ubuntu / Linux Mint
Fedora
Arch Linux
Guix
Anaconda
Installing SageMath or Nemo (see below) will also create an installation of Arb local to those systems. It is possible to link user code to that installation by setting the proper paths.
Download¶
Tarballs of released versions can be downloaded from https://github.com/fredrik-johansson/arb/releases
Alternatively, you can simply install Arb from a git checkout of https://github.com/fredrik-johansson/arb/. The master branch is recommended for keeping up with the latest improvements and bug fixes and should be safe to use at all times (only stable code that passes the test suite gets merged into the git master).
Dependencies¶
Arb has the following dependencies:
Either MPIR (http://www.mpir.org) 2.6.0 or later, or GMP (http://www.gmplib.org) 5.1.0 or later. If MPIR is used instead of GMP, it must be compiled with the
--enable-gmpcompat
option.MPFR (http://www.mpfr.org) 3.0.0 or later.
FLINT (http://www.flintlib.org) version 2.5 or later. You may also use a git checkout of https://github.com/fredrik-johansson/flint2
Standalone installation¶
To compile, test and install Arb from source as a standalone library, first install FLINT. Then go to the Arb source directory and run:
./configure <options>
make
make check (optional)
make install
If GMP/MPIR, MPFR or FLINT is installed in some other location than
the default path /usr/local
, pass
--with-gmp=...
, --with-mpfr=...
or --with-flint=...
with
the correct path to configure (type ./configure --help
to show
more options).
After the installation, you may have to run ldconfig
to make sure that the system’s dynamic linker finds the library.
On a multicore system, make
can be run with the -j
flag to build
in parallel. For example, use make -j4
on a quad-core machine.
Running tests¶
After running make
, it is recommended to also run make check
to verify that all unit tests pass.
By default, the unit tests run a large number of iterations to improve
the chances of detecting subtle problems.
The whole test suite might take around 20 minutes on a single core
(make -jN check
if you have more cores to spare).
If you are in a hurry, you can adjust the number of test iterations via
the ARB_TEST_MULTIPLIER
environment variable. For example, the following
will only run 10% of the default iterations:
export ARB_TEST_MULTIPLIER=0.1
make check
It is also possible to run the unit tests for a single module, for instance:
make check MOD=arb_poly
Building with MSVC¶
To compile arb with MSVC, compile MPIR, MPFR, pthreads-win32 and FLINT using MSVC. Install CMake >=2.8.12 and make sure it is in the path. Then go to the Arb source directory and run:
mkdir build
cd build
cmake .. # configure
cmake --build . --config Release # build
cmake --build . --config Release --target install # install
To build a Debug build, create a new build directory and pass
-DCMAKE_BUILD_TYPE=Debug
to cmake
. To create a dll library, pass
-DBUILD_SHARED_LIBS=yes
to cmake
. Note that creating a dll library
requires CMake >= 3.5.0
If the dependencies are not found, pass -DCMAKE_PREFIX_PATH=/path/to/deps
to cmake
to find the dependencies.
To build tests add, pass -DBUILD_TESTING=yes
to cmake
and run ctest
to run the tests.
Running code¶
Here is an example program to get started using Arb:
#include "arb.h"
int main()
{
arb_t x;
arb_init(x);
arb_const_pi(x, 50 * 3.33);
arb_printn(x, 50, 0); flint_printf("\n");
flint_printf("Computed with arb-%s\n", arb_version);
arb_clear(x);
}
Compile it with:
gcc test.c -larb
Depending on the environment, you may also have to pass
the flags -lflint
, -lmpfr
, -lgmp
to the compiler.
On some Debian based systems, -larb
needs to be replaced
with -lflint-arb
.
If the Arb/FLINT header and library files are not in a standard location
(/usr/local
on most systems), you may also have to provide flags such as:
-I/path/to/arb -I/path/to/flint -L/path/to/flint -L/path/to/arb
Finally, to run the program, make sure that the linker
can find the FLINT (and Arb) libraries. If they are installed in a
nonstandard location, you can for example add this path to the
LD_LIBRARY_PATH
environment variable.
The output of the example program should be something like the following:
[3.1415926535897932384626433832795028841971693993751 +/- 6.28e-50]
Computed with arb-2.4.0
Computer algebra systems and wrappers¶
Python-FLINT (https://github.com/fredrik-johansson/python-flint) is a convenient Python interface to both FLINT and Arb.
SageMath (http://sagemath.org/) includes Arb as a standard package and contains a high-level Python interface. Refer to the SageMath documentation:
Nemo (http://nemocas.org/) is a computer algebra package for the Julia programming language which includes a high-level Julia interface to Arb. The Nemo installation script will create a local installation of Arb along with other dependencies.
Real balls: http://nemocas.github.io/Nemo.jl/latest/arb.html
Complex balls: http://nemocas.github.io/Nemo.jl/latest/acb.html
Arblib.jl (https://github.com/kalmarek/Arblib.jl) is a thin, efficient Julia wrapper around Arb.
Other wrappers include:
ArbNumerics (Julia): https://github.com/JeffreySarnoff/ArbNumerics.jl
ArbFloats (Julia): https://github.com/JuliaArbTypes/ArbFloats.jl