This is a guide I wrote up as requested by Bodo for the setup of the 
analyzer package for use with N20 analysis and expanded into the
GEn experiment.  If you have any questions or suggestions for improvement, 
let me know.

Seamus
riordan@jlab.org

----------------------------------------------------------------------

June 9, 2005:   updated the file to reflect changes in THaCodaDecoder.C
                Bodo Reitz

----------------------------------------------------------------------

----------------------------------------------------------------------

May 18, 2006:   Added in special tag for GEn classes in the analyzer
		and the separation of tracking and non tracking libraries
                -Seamus

----------------------------------------------------------------------


To get analyzer installed, you must first have ROOT installed.  
Assuming this is not done, the latest ROOT can be obtained from 
http://root.cern.ch/root/Availability.html in the form of precompiled 
binaries or source.  Once these are unpacked add the path to the top 
level ROOT directory to the environment variable ROOTSYS (for example, 
this is /apps/root/4.00-08/root on ifarml1), add $ROOTSYS/bin to PATH, 
and $ROOTSYS/lib to LD_LIBRARY_PATH.  If ROOT is already installed, 
this may or may not be already done.  Check to make sure these 
variables are correct, because ROOTSYS must be defined before installing analyzer.

Before using CVS set CVS_RSH to ssh.  The standard analyzer package can be 
checked out from the CVS depository using:

cvs -d :ext:username@cvs.jlab.org:/group/halla/analysis/cvs co -d myanalyzer analyzer

Where $ANALYZER is the directory where analyzer is installed.  

The GEn experiment has a special version tagged with classes for
DecData and beam.  To get this (if you're reading this you probably want to)
use the following line:

cvs -d :ext:username@cvs.jlab.org:/group/halla/analysis/cvs co -rgen-e02013 -d myanalyzer  analyzer

Once this is downloaded, go into the analyzer directory and (assuming 
you're putting this on Linux) run gmake and grab some coffee.  

If you downloaded the analyzer with the -rgen-e02013 option, you also must build
the libraries for GenDecData, GenBeam, etc. by going into the src directory
in the analyzer and running make:

cd src
make

This will generate these optional libraries for you.

The top level directory analyzer (also now containing the program analzyer) 
should be added to the environment variable ANALYZER and then $ANALYZER to 
both your PATH and LD_LIBRARY_PATH.  Once all this is done, you should be 
able to execute analzyer from the command line.  


The agen package may be checked out from CVS by:

cvs -d :ext:username@cvs.jlab.org:/group/halla/analysis/cvs co -d myagenlib agen

There are two separate ways to build the libarary.  The first is done by:

make

which will create a version with drift chamber tracking disabled.  To 
enable tracking:

make DO_TRACKING=1

which will create a library called libAGen_trk.so

For the most part, tracking slows down analysis time substantially, so unless 
you need that functionality, it is best to leave it out.

The library libAGen.so can now be accessed by adding 
the agen directory to LD_LIBRARY_PATH or, since it is a single library, 
it may be more convenient to load it on a case by case basis by 
gSystem->Load(<path>/libAGen.so) or gSystem->Load(<path>/libAGen_trk.so) in 
ROOT/analyzer where <path> is the path in the directory structure.

Before running analzyer, you should set the DB_DIR environment variable to the 
directory containing the necessary databases for your analysis.

Here is an example of what I have in my .login in my ifarml1 account (csh):

 setenv CVS_RSH ssh
 setenv ROOTSYS /apps/root/4.00-08/root
 setenv ANALYZER /work/halla/e02013/disk1/analyzer
 setenv PATH $PATH':'$ANALYZER':'$ROOTSYS/bin
 setenv LD_LIBRARY_PATH $ANALYZER':'$ANALYZER/src:'$ROOTSYS/lib
 setenv DB_DIR /work/halla/e02013/disk1/db

and for a machine with a bash account in my .bashrc:

 export CVS_RSH=ssh
 export ROOTSYS=/home/riordan/root
 export ANALYZER=/home/riordan/analyzer
 export PATH=$PATH:$ROOTSYS/bin:$ANALYZER
 export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$ROOTSYS/lib:$ANALYZER:$ANALYZER/src
 export DB_DIR=/home/riordan/DB


The analzyer should now be ready to run.  However, one should keep in mind 
that the analyzer should not be run in the analyzer directory.  It is best 
to create a directory for this purpose, usually containing the scripts one 
wishes to run.  To execute a script you may go into that directory and from
the command line run:

analzyer <script>

You may also run analyzer by itself to work from the ROOT/C interpreter. 
To execute a script within the interpreter run:

.x <script>

or, if the script requires or takes arguments:

.x <script>+(args... )

where args is the standard comma delimited set of arguments. 
To exit the interpreter, simply use

.q

For more information on analyzer or ROOT in general:

Analyzer Quick Start Guide	    http://hallaweb.jlab.org/root/doc/quickstart.html
Hall A C++ Analyzer Documentation   http://hallaweb.jlab.org/root/doc/index.html
ROOT Documentation		    ftp://root.cern.ch/root/doc/Users_Guide_4_08.pdf

To obtain the latest version of the software after it has been checked out as above,
simply go into the directory with the package you wish to update and run:

cvs update

This will sync the copy of source you have with the copy in the CVS repository.  
This only updates the source, however, and you MUST run make again for the latest binaries.