Skip to content

paulmking/analyzer

 
 

Repository files navigation

Hall A C++ Analyzer

This is the standard data analysis software for Hall A at Jefferson Lab. The current stable version is 1.7.0. This software is also known as "Podd".

Overview

The analyzer is an object-oriented, modular and extensible physics event processing framework built on top of the ROOT data analysis framework from CERN. Classes are available for the most common analysis tasks involving data from the standard Hall A experimental equipment, in particular the HRS spectrometers and detectors. Standard physics calculations for single arm (e,e'), conincidence (e,e'X) and photoproduction reactions are available, as well as for auxiliary tasks such as energy loss corrections, vertex position calculations, etc.

A Software Development Kit (SDK) is included that provides users with a rapid development environment for building experiment-specific extension libraries. One can quickly implement new detectors, physics computation modules and even entire spectrometers.

The analyzer core libraries are also used by the Hall C analysis software hcana as well as the analysis package for the SuperBigBite series of experiments in Hall A, SBS-offline.

More documentation can be found in the Wiki.

System Requirements

  • Linux or macOS with a C++11-capable compiler (gcc 4.8+ or clang 9+).
  • Essentially any hardware powerful enough to support ROOT 6 will support the analyzer as well.

Dependencies

  • ROOT version 6 or higher. The latest version is recommended. For the analyzer to find ROOT, either $ROOTSYS must point to the top of the ROOT installation or the root-config script must be in your PATH.

  • EVIO version 4.0 or higher. If not installed, the build system will download and build a copy of EVIO automatically. This requires Internet access. If using a preinstalled version, $CODA must point to the top of its installation location.

  • CMake 3.5 or higher (3.15 or newer recommended).

  • GNU make or ninja.

  • Instead of CMake and make/ninja, SCons 2.3 or newer (3.0 or newer recommended) in combination with Python 2.7 or newer can be used (see below).

Building

The build system is a standard CMake setup. In the top-level analyzer directory, the following commands will configure, build and install the software:

cmake -DCMAKE_INSTALL_PREFIX=$HOME/local/analyzer -B builddir -S .
cmake --build builddir -j8
cmake --install builddir

Here, $HOME/local/analyzer is an example installation destination. Modify this variable as appropriate. If omitted, it defaults to /usr/local. Additional CMake variables and options can be set as desired, for example -DCMAKE_BUILD_TYPE=Debug or -GNinja. Consult the CMake documentation for details.

The analyzer can be run directly form the build location (builddir/apps/analyzer in the example above) without any further configuration. However, installation is recommended since it is required for building analyzer plugin modules. To work most effectively with the installed version, several environment variables should be set appropriately. This can be done conveniently with the included setup scripts, for example, for the bash shell:

source $HOME/local/analyzer/bin/setup.sh

Alternatively, if environment modules are available on your system, the following instructions will set up the analyzer environment:

module use $HOME/local/analyzer/share/modulefiles
module load analyzer

The modulefiles may also be copied to a central location for such files, e.g. $HOME/local/modulefiles.

The binaries are built with RPATH set for the build or install location, as applicable, so including the library directory in (DY)LD_LIBRARY_PATH should not be necessary.

Building with SCons (deprecated)

For backward compatibility, the SCons build system is still supported in this version of the analyzer, although it is no longer recommended and will be removed in a future version. SCons requires Python. The following steps will accomplish the same as the CMake example above:

export SCONS_INSTALL_PREFIX=$HOME/local/analyzer
scons -j8
scons install

Documentation

Please see the Wiki

Contributing

To participate in development, please contact the developers

Bug reports and other issues may be submitted to the Redmine issue tracker by anyone. You are strongly encouraged to use the issue tracker system. In this way, all active developers are notified and able to respond quickly.

Packages

No packages published

Languages

  • C++ 94.6%
  • CMake 2.6%
  • Python 2.1%
  • Other 0.7%