lrslib: Reverse search algorithm for vertex enumeration and convex hull problems#

Description#

lrslib implements the linear reverse search algorithm of Avis and Fukuda.

See the homepage (http://cgm.cs.mcgill.ca/~avis/C/lrs.html) for details.

We use an autotoolized version from https://github.com/mkoeppe/lrslib/tree/autoconfiscation

License#

lrslib is released under a GPL v2+ license.

Upstream Contact#

David Avis, avis at cs dot mcgill dot edu.

Dependencies#

To build and install the “plrs” binary, a multi-thread version of lrs, need to first install the full Boost package (“sage -i boost”).

If the package finds an MPI C++ compiler script (mpic++), it also builds and installs the “mplrs” binary, a distributed version of lrs using MPI.

(Sage currently does not make use of plrs and mplrs.)

Special Update/Build Instructions#

Type#

optional

Dependencies#

  • $(MP_LIBRARY)

Version Information#

package-version.txt:

071b+autotools-2021-07-13

Equivalent System Packages#

arch:

$ sudo pacman -S  lrs

conda:

$ conda install  lrslib

Debian/Ubuntu:

$ sudo apt-get install  lrslib

Fedora/Redhat/CentOS:

$ sudo yum install  lrslib

freebsd:

$ sudo pkg install math/lrslib

gentoo:

$ sudo emerge sci-libs/lrslib

nix:

$ nix-env --install lrs

opensuse:

$ sudo zypper install lrslib lrslib-devel

See https://repology.org/project/lrslib/versions

If the system package is installed, ./configure will check whether it can be used.