The following example shows how to create a new database file.
Creating a new database file
      PROGRAM CDEXA1
*     ==============
*
*     Create a new, empty database
*
      PARAMETER   (NWPAW=100000)
      COMMON/PAWC/PAW(NWPAW)
*
*     Initialise Zebra, HBOOK and HEPDB
*
      CALL CDPAW(NWPAW,NHBOOK,IDIV,'USR-DIV',5000,50000,'ZPHU',IRC)
*
*     Unit for database access
*
      LUNCD  = 1
*
*     Database parameters
*
      NPAIR  = 1
      NREC   = 20000
      NTOP   = 1
*
*     Accept default record length (1024 words)
*
      LRECL  = 0
      CALL CDNEW(LUNCD,'HEPDB','HEPDB.DBS',IDIV,NPAIR,NREC,NTOP,
     +           LRECL,' ',IRC)
*
*     Set the log level
*
      CALL CDLOGL(' ',3,'A',IRC)
*
*     Terminate
*
      CALL CDEND(' ','A',IRC)
      END
The same result can be achieved by running the CDMAKE program, e.g. using the following script.
Script to run CDMAKE program
# # Make a new database # # export CDFILE='name of the database file' export CDFILE='test.dbs' # export CDPAIR='number of validity range pairs' export CDPAIR=1 # export CDPRE='number of records to be preformatted' export CDPRE=100 # export CDTOP='numeric ID for database' export CDTOP=1 # export CDQUO='number of records for database quota' export CDQUO=65000 # export CDRECL='record length of database file (words)' export CDRECL=1024 # # now run the job # /cern/pro/bin/cdmake