Peter et al.,
I've now completed SAMP in GAIA, and (if I've told git what
I think I've told it) pushed the changes back to the repository.
GAIA now communicates using SAMP, but not using PLASTIC. It would
have been possible to get it doing both, but somewhat more effort,
more complication and more chances for things to go wrong, so given
that I'm expecting PLASTIC to fall into disuse soon, I took the
easy route. Files which are no longer required have been removed.
I've made an update to SUN/214 which I think covers the current
situation in a similar level of detail to what was there before.
I haven't changed gaia.news, but I can supply something on request.
Any problems or queries, let me know.
Most of the MTypes implemented are the standard SAMP ones.
I've also added an MType gaia.execute.tcl as follows
(see GaiaSampAgent.tcl):
# MType specification:
# MType:
# gaia.execute.tcl
# Parameters:
# script (string) - executable Tcl
# Return values:
# value (string) - return value of successfully executed script
which allows remote control in much the same way as the PLASTIC
ivo://plastic.starlink.ac.uk/gaia/executeMd5 message did. The MD5
cookie business is no longer required because SAMP, unlike PLASTIC,
exercises reasonable control over who is able to send messages.
To see this working you can e.g. use JSAMP's MessageSender utility:
% starjava/bin/jsamp messagesender -mtype gaia.execute.tcl \
-param script "expr 1 + 2"
c1 (gaia)
samp.status: samp.ok
samp.result: value 3
Next up, JNIAST. I'm not sure if I'll complete that by the end
of the month (=tomorrow) as advertised; I need to refresh my memory
about what the actual issues are, but hopefully I should get
whatever it is that needs doing done by the end of this week.
Mark
--
Mark Taylor Astronomical Programmer Physics, Bristol University, UK
[log in to unmask] +44-117-928-8776 http://www.star.bris.ac.uk/~mbt/
|