Hello,
I have a question on the use of a parameter files with DX which I like to
address to this group.
I have used IBM-DX 3.1.4 for a few years to visualise fisheries acoustic
data
that show fish aggregations and bathymetry in their appropriate spatial
context. I like to upgrade an existing visual program to include a much
larger number of user defined parameters and are looking for a sample DX
program that reads numeric constants (depth range, echosounder gain, etc)
and
strings (paths, file name, fish species names, etc) from an annotated
parameter ascii file rather than from the GUI. A typical parameter file will
consist of a numeric and string entries that are followed by a comment. The
comment here is proceeded by a quote. It might include:
d:\0098roc1fr ' Path to data folder
tr3045.dat ' Data file
5.34 ' Gain, dB
50,100 ' Depth range, m
... etc
I had the following suggestions from Peter Daniel Kirchner [EMAIL PROTECTED]:
1. parse the file into a .net file which you "include" in the main net (or
append to it).
this .net file would just be assignments e.g.
depth_range= {50,100};
These values would enter your visual program via receivers named for
the variable, in this example "depth_range". You do not need or want
corresponding transmitters. Avoid reserved words like "sequence" (not all
reserved words are well documented either... if your net doesn't work,
suspect your choice of variable names.
2. edit (or generate) a single-data-line spreadsheet (tabular) format file
and use ImportSpreadsheet to grab it. The column name and the
corresponding value are read in.
3. make a shell script that parses your file and echoes to stdout a .dx
file for one or more variable names. use Import( "!your_script v1 v2 v3")
to import that .dx format information. concerns: variable-at-a-time could
be slow, many-but-not-all-variables could exceed the available number of
characters that can be given to the shell. If you ever go to using a
database, this would approximate the database query and how you would get
the information into dx from the database (i.e. an odbc query echoed to
stdout).
I am looking forward to any thoughts o this, Robert
Dr. Robert Kieser
Research Scientist
Pacific Biological Station
Department of Fisheries and Oceans
Hammond Bay Road
Nanaimo, BC, V9R 5K6, Canada
Tel 250-756-7181 Fax 250-756-7053
Email [EMAIL PROTECTED]
http://www.pac.dfo-mpo.gc.ca/sci/sa-mfpd/
