Jurgen Doreleijers, Madison,
Ton Rullmann, Utrecht




  2. INSTALLATION (Unix only)

1. File organization

The top directory contains the following files:
  • version history
  • installation hints (this document)
  • instructions for users
  • list of new features and changes
  • general introduction (recommended to read first)
  • script for defining aliases and variables
  • idem; intended for interactive use

and the following subdirectories:
  • bin
  • cgi
  • data 
  • doc 
  • exmpls 
  • extras
  • html
  • scripts
  • server
  • src 
  • sub 
  • executables (to be made by 'make'; see below)
  • cgi server software (please ignore)
  • name-library files
  • documentation
  • example jobs and output (used by 'make test')
  • some useful non-Aqua tools
  • html server documentation and examples (please ignore)
  • conversion and analysis scripts
  • server forms (please ignore)
  • main source code files and makefile
  • auxiliary source code

2. Installation (Unix only)

1. Uncompress and unpack the archive (replacing X.Y by the version number that you down loaded):
        tar -xvzf aquaX.Y.tar.Z # or leave out the uncompressing if already done:
        tar -xvf  aquaX.Y.tar
This will create a subdirectory called 'aquaX.Y' in the present working directory. This subdirectory will be referred to as the aquaroot throughout this text. The organization of the aquaroot directory is described above, and should not be changed.

2. Go to the 'src' subdirectory and modify the file 'makefile':

3. Build the executables: (There's an auxiliary Fortran program 'exth.f' which is not really needed. If you cannot compile that file, follow the instructions in the makefile).
The installation/compilation may be done by root or by any user having write permission for the aquaroot directory tree. There is no "make install". All binaries and data files reside in the aquaroot directory tree. You may move the tree to a different location afterwards without recompiling. Just make sure to keep the subdirectory structure intact.

4. Localization:

5. Test the installation, and cleanup (first go to the 'src' subdirectory):
    make test  #optional: test whether installation was succesfull
If the test fails on a nawk script with messages about "unknown option -v" or "undefined function toupper", you have to install gawk, the GNU version of nawk. See the file 'doc/gawk.txt'.
If all is well, you might retrieve some disk space by issuing:
    make strip # optional: reduce size of executables by removing symbols
    make clean # optional: remove object files
6. Define the environment variable 'aquaroot' to point to the Aqua top directory, and initialize:
    cd ..                 # assume we are now back in the Aqua top directory
    setenv aquaroot $cwd
    source $aquaroot/aqsetupi
Now you are ready to use Aqua! See the file 'HOW_TO_USE' for more details.

3. Finishing touch

1. You may want to store the 'aquaroot' definition in a system-wide definition file, such as '/etc/cshrc'. An alias for initializing the Aqua setup may also be defined there (again replacing X.Y by the current version number):
    setenv aquaroot $YOUR_LOCATION/aquaX.Y
    alias aqua 'source $aquaroot/aqsetupi'
I advise you not to import definitions from the 'aqsetup' script into places like '/etc/cshrc' or '.cshrc'.
2. Procheck upgrade: for Aqua version 1.0 and above to work correctly in conjunction with Procheck-NMR you have to upgrade to Procheck version 3.4 or higher.

Contact the author or webmaster for help, when required.