Rocket I
S
D
A
T

Page maintained by
Reine Gill

 

Release notes for ISDAT 2.5.

Documentation is available from ftp.irfu.se by anonymous ftp
in directory pub/isdat/doc, the file pub/isdat/doc/README
lists the documents.
The information can also be accessed using your favorite
web browser to connect to http://www.irfu.se/isdat.

!!! NOTE !!! The web pages have not been updated yet and contains
the old ISDAT version. We will inform you when it is updated.
So please use ftp.



ISDAT package (isdat-2.5.tar.gz)
==================================

Available from ftp://ftp.irfu.se/pub/isdat.

To install:
gzcat isdat-2.5.tar.gz | tar xf -
install any extra instrument modules, see below
cd isdat
make isdat


1)
You may want to customize the config file for your computer:
isdat/config/cf/sitesun5.cf for Solaris 2.x
isdat/config/cf/sitehp.cf for HP-UX
isdat/config/cf/siteosf1.cf for DEC OSF1
isdat/config/cf/sitelinux.cf for Linux
isdat/config/cf/siteibm.cf for IBM AIX

The features you may need to disable is:
HasCplusplus
HasFortran
HasSunPhigs
HasSiPhigs
HasIdl
HasMatlab
HasCdf
HasPurify

It is no longer necessary to install xgks to build isdat. It is
built as part of the normal build and has been moved to isdat/lib/xgks.


2)
You may want to install the CDF library to build ISDAT, the IGR
client can use it to generate CDF files.
The default configuration files expects it in /opt/cdf on Sun
and /usr/local/cdf on other platforms.
You can either get the source from ncgl.gsfc.nasa.gov
or get the precompiled archive cdf-25-sol2.5-sparc.tar.gz
or cdf-25-sol2.5-x86.tar.gz from the anonymous ftp server at
ftp.irfu.se in the pub/isdat directory (same place as the
public ISDAT distribution).
To unpack, do:

cd /opt
gzcat cdf-2.5-sol2.5.tar.gz | tar xf -


3)
The time manager is renamed to ctm (old name was cuitm).


4)
You need to set the XUSERFILESEARCHPATH environment variable:

XUSERFILESEARCHPATH=$ISDAT/cfg/app-defaults/%N%S; export XUSERFILESEARCHPATH

If you don't you will get bad labels on buttons and menues in
some clients and ctm.


5)
ISDAT have been built on the following platforms.

Sun Solaris 2.5.1 (SPARC and x86)
Sun Solaris 2.5.1 with Motif 2.0.1 (SPARC and x86)
HP HPUX 9.01
HP HPUX 10.01
Digital UNIX V3.2C (alpha)
Linux (RedHat 4.0) with Motif 2.0.0
IBM AIX 3.2 (builds but fails to start ctm, will check that later)


6)
New features in igr.

Igr now allows more than one spectrogram, one in each panel is allowed.

The igr calculator now supports R(2,0){di,*} and R(2,0){*,di}
to pick out rows and columns of multidimensional quantities,
the form R(2,0){di} have been removed. This is mainly of interest
to STAFF.
For a complete list of available subscripts see the file isdat/doc/SUBSCRIPT.

The calculator now has temporary variables, eg.
v0 = q0 + 10
v1 = 10
p0y = v0 + 10
p4y = v0 - v1
p8y = q8 * v0
Subscripts can be applied to variables
v0 = fft(q0)
p0y = v0{5}

Both ctm and igr now saves the state of the application at exit
in the files init.ctm and init.igr. This sometimes leads to
unexpected behaviour (although correct) which we will try to
provide a workaround for in the next release.
Ctm can be started without reading init.ctm if started as
ctm -clean
To start igr without reading init.igr you can either start it from
the command line as
igr -clean
or select 'Options -> clean client start' in ctm before starting
igr from the Clients menu.


7)
The isdat directory structure have been modified slightly.

All user specific directories like sci, liza, tech, eiscat, ...
have been moved to the isdat/contrib directory.
The oplib directory is renamed to operators.
The contents of scilib is moved to isdat/lib.



Freja F4 module (Freja4-2.tar.gz)
=====================================

Available from ftp.irfu.se in frejaftp ftp account in directory isdat.

To install:
cd isdat
gzcat Freja4-2.tar.gz | tar xf -
make isdat


1)
Pressing Content in ctm doesn't generate any index files as
previous versions did. To generate index files you have to
select 'Clients -> admin -> indexgen' in ctm and then press OK.
The indexgen user interface is a bit rough and will need some
makeup in future releases.


2)
To locate your data files you need to edit the lines

*.freja4.data:          $HOME/data/freja/data
*.freja4.index:         $HOME/data/freja/index
*.freja4.cal:           $HOME/data/freja/cal
*.freja4.log:           $HOME/data/freja/log
*.freja4.debug:         $HOME/data/freja/debug

in the file isdat/config/isdat.server.
The default directory for the data files is $HOME/data/freja/data but
can be changed by editing isdat/config/isdat.server.
The default directory for index files is $HOME/data/freja/index, if
it doesn't exist you need to create it.
You also need to create $HOME/data/freja/cal and $HOME/data/freja/log.
The time calibration file should be named $HOME/data/freja/cal/time.
The geometry information file should be named $HOME/data/freja/cal/geometry.
The time and geometry files are provided with the distribution of
the module (in isdat/server/Freja4).


3)
The module now supports all F2 and F4 modes:

F2 all modes
F4 standard mode
F4 burst mode
F4 continuous mode LF and LFMF
F4 SMM mode (also called survey medium)
F4 SMM snapshot mode (also called survey coarse)



Wec module (Wec-2.tar.gz)
=========================

Available from http://ftp.irfu.se/isdat/wec.
can also be fetched here.

To install:
cd isdat
gzcat Wec-2.tar.gz | tar xf -
make isdat


1)
Pressing Content in ctm doesn't generate any index files as
previous versions did. To generate index files you have to
select 'Clients -> admin -> indexgen' in ctm and then press OK.
The indexgen user interface is a bit rough and will need some
makeup in future releases.


2)
Make sure that the directory pointed to by *.cluster.index
is empty before starting the server the first time using this
release.
Old index files from previous versions are not compatible with this version.


3)
To locate your data files you need to edit the lines

*.cluster.rdm:          $HOME/data/cluster/RDM
*.cluster.dds:          $HOME/data/cluster/DDS

in the file isdat/config/isdat.server.
The value given to *.cluster.rdm needs to be a directory
with subdirectories named after the CDROM volume label.
Doing an ls -l on it should give:

drwxrwxr-x   7 al       isdat        512 May  8 23:49 950323_1_2a/
drwxrwxr-x   8 al       isdat        512 May 24 14:31 950323_2_2a/
drwxrwxr-x   8 al       isdat        512 May 24 14:31 950324_1_1a/

if all three test CDROMS from the MCT are loaded or copied.

The value given to *.cluster.dds needs to be a directory
with files named according to the specification by Per=Arne
Lindqvist. Doing an ls -l on it should give:

-rw-rw-r--   1 al       isdat     3687220 Apr 21 14:42 9511080000sh.03
-rw-rw-r--   1 al       isdat    49347219 Apr 21 14:43 9511080000wb.03
-rw-rw-r--   1 al       isdat     1426644 Apr 21 14:42 9511080000wh.03
-rw-rw-r--   1 al       isdat    14715000 Apr 21 14:42 9511080000wn.03

To look at test files from Velizy/Dornier you need to convert them
to the DDS format using the TED sistodds/tmc programs.


4)
The decommutation software is based on TED 2.3.


5)
When requesting HK data from a client, the DDS header is now
passed in the map array. An example is provided in
isdat/tech/wstat/main.c in the EFWHK sections.


6)
Access to the HK parameter definition (HPD) files are implemented.

An example of how to access the HPD data in a client is
provided in isdat/tech/wstat/main.c in the EFWHPD sections.
The content of the HPD file with the DDS header stripped off is
passed to the client as data. The DDS header (15 bytes) is passed
in the map array.

Access to the HPD data from within the server is possible
from the Orbit structure, the whole HPD file is read into a string
by the WEC module before any data extraction function gets called.

pOrbit->fs[hpd_file].buf	DDS packet header followed by null terminated
				HPD string
pOrbit->fs[hpd_file].ptr	null terminated HPD string
pOrbit->fs[hpd_file].size	size of string excluding terminating null
pOrbit->fs[hpd_file].n		size of string including terminating null
pOrbit->fs[hpd_file].time	DDS header time stamp

If size or n is zero no HPD data is available.
The hpd_file must be one of FILE_ED, FILE_FD, FILE_CD,
FILE_PD, FILE_RD, FILE_WD, FILE_AD, FILE_SD
for EDI, FGM, CIS, PEACE, RAPID, WEC, ASPOC and SPACECRAFT respectively.


7)
There is a status client to display WEC housekeeping data written
by Simon Walker. It is started from ctm as Clients->cluster->hkeep.
The WEC housekeeping can also be viewed in igr by selecting the hkeep
instrument.


8)
There is a status client to display EFW status data. It is started
from ctm as Clients->cluster->efwstat.
If no status is available the text area is cleared.