On Wed, 30 Mar 2005, Tim Jenness wrote:

> CADC are very interested in an alternative to aladin for SIAP image
> access. I was under the impression that SoG supported SIAP but now I can't
> find any mention of it on the SoG web page (www.starlink.ac.uk/sog) so I'm
> worrying that I oversold the Starlink software to CADC.
>
> Can someone clarify the SIAP support in starjava?

Tim,

I can't see any reference to SIAP in SoG either, but it might be
there - I think Peter is back from holiday today so can clarify.

It is actually possible to use TOPCAT v1.4 or later (+SoG) as a SIAP
client:

   1. Run 'topcat -siap' (the flag isn't strictly necessary but it
      makes the SIAP option more prominent in the load dialogue).

   2. From the load dialogue hit the 'SIAP Query' button

   3. Select the SIAP service from those listed (the list is obtained
      from a registry somewhere).

   4. Enter the coordinates & radius of the region you're interested in.
      This window doesn't currently have a name resolution box,
      clearly it should.

   5. Hit the OK button

   6. TOPCAT will (attempt to) contact the server with the query you've
      made and load in a table with the resulting available images

   7. On the TOPCAT control window for the newly loaded table hit
      the 'Activation Action' button and select 'Display Named Image'
      in the resulting dialogue (the column selected for this purpose
      should automatically be the column in the SIAP return table
      which contains an image URL).  Hit OK on that dialogue.

   8. View either a plot of the resulting table or the table data
      viewer.  If you select one of the rows (click on a point in the
      plot or a row in the table) then it will download the image
      from the server and fire up SoG with the result.

As you can see this is a bit of a hoo-hah.  Although reworking within
TOPCAT could improve matters, what we concluded at the last programmers'
meeting was that trying to get TOPCAT working as a SIAP client is
probably going to result in both clunkifying of TOPCAT and a second-rate
SIAP client, and it really needed a new application - this will be
one of the things that VO{RTEX,GON} is for.  For this reason the
SIAP functionality in TOPCAT isn't something I'm working on improving
or have advertised much (it's not really documented in SUN/253).

Although I'd still stand by my estimate of a couple of months work
to get something worthwhile working for VORTEX, I haven't started
work on it yet, and probably won't until after Kyoto (May) - after
various bits of this and that I'm of the opinion that command-line
table tools is probably a higher priority to get working.  But
certainly I'd be interested in any definite requirements that CADC
have for this, why they want alternatives to Aladin, what timescale
their requirements have and so on.  If they're keen to have something
soon which we can provide with a small amount of effort, as may be
the case, I could maybe bring forward the start of the VORTEX work.

Mark

--
Mark Taylor    Starlink Programmer     Physics,  Bristol University, UK
[log in to unmask] +44-117-928-8776 http://www.star.bris.ac.uk/~mbt/