Faast-3 Reference Manual Addendum # 4
DUMP Command
Effective with Release 3.3D
1. Overview:
Release 3.3D of Faast-3 introduces a new command which provides a machine-
readable data file containing Faast data and results.
2. Purpose of DUMP Command:
The data files produced by the DUMP command are intended
to be used by user-written post-processor programs. Examples of such programs
would be programs to import Faast-3 results into CAD programs, and programs to
automatically alter the Faast-3 model to achieve a design goal (such as
changing pump curves, pipe sizes, or tank sizes) followed by a new session to
solve the modified model.
3. Format of DUMP Files:
Faast-3 DUMP files are plain-text files with fixed-length
data fields and variable length data records. As such, these files may be
examined with text display and edit programs, and these files may be read by
programs written in languages such as FORTRAN, BASIC and
C. The detail format of DUMP files is
discussed in a separate Addendum (provided on request). It is intended that
the format of DUMP files will not change except to add
new types of data, thus your post-processor programs will never need changing
unless you need to use some future new data type.
The format of DUMP files is designed to be identical on
all computers and fully-portable between differing computer systems.
Exceptions to portability (for which users will need to take action) will
occur when transporting data between computer systems using
ASCII and computer systems not using
ASCII (for example, IBM mainframe- and mini-computers
which use EBCDIC), as well as between computers using
variants of UNIX operating system, which do not use carriage return control
codes at the end of text lines and more conventional computers (such as
MS-DOS and VAX/VMS) which use both
carriage return and line-feed control codes at the end of text lines.
4. Using DUMP Versus SAVE:
The DUMP command and the SAVE command
do similar things but with three main differences:
a. The SAVE command produces a binary (not text) file,
which is not human-readable, but takes less space on disk, and may be loaded
into Faast very quickly with the RECALL command. The
format of SAVE files is not published.
b. Files produced by the DUMP command MAY NOT
be used as input to Faast-3, they are intended for post-processing only.
c. The SAVE command sets a flag to save after the
next solution or group of solutions, whereas the DUMP
command either dumps right now (the CASE. option) or it
sets a flag to dump after all subsequent solutions (the ON=AUTO
option) until reset (the OFF=AUTO option). A single use
of the DUMP command does not produce a dump after the next
single solution.
5. DUMP Command Syntax:
The DUMP command should be used (interactively or in a
data file) prior to the SOLVE command for which results
are to be dumped. If ON=AUTO is in effect, the
DUMP file is written immediately after a solution,
similarly to the way in which a SAVE file is written and
the way in which printable output is written to the output file.
If you forget to use the DUMP command with the
ON=AUTO option before a solution, then you may use the
DUMP command with the CASE. option
after the solution to dump the results.
The DUMP command has the following syntax:
DUMP ( FILE= filename , DFREQ= frequency , CASE. ,
ON= / OFF= [ DPF , AUTO ] )
where:
filename is the desired name of the file to which
data will be written; the default file name is FSTDMP.DAT
and will include a "cycle number" to avoid overwriting a previous file of the
same (the exact functioning of cycle numbers varies with the computer system
in use),
frequency is the desired frequency of dumps (default
is 1); only applies to STEP and
SOLVE LINK solutions and time simulations,
CASE. is a verb that implies that you want an immediate
DUMP of current results, not waiting until the next
solution (use this option if you forgot to set ON= AUTO
before the most recent solution); this switch does not stay set - it resets
after one use,
DPF is a toggle switch that implies "DUMP-Per-File" meaning
that only one DUMP will be placed in any single data file
(a new file will be created for each set of data being dumped, default is
OFF=DPF), and
AUTO is a toggle switch that implies each subsequent
solution will be written to the DUMP file until the switch
is set OFF (default is OFF). If this
switch is OFF then the DUMP file will
not be written after the next solution (in which case you would have to use
the CASE. option to obtain a dump).
Remember that, unlike the SAVE command, the
DUMP command does not produce a dump unless you use either
the CASE. option (for an immediate dump) or the
ON=AUTO option (for dumps after all subsequent solutions
until turned off).
# # #
Copyright (c) 1998 Faast Software. All rights Reserved.