New VOTable 1.4 Working Draft

[Mark Taylor]

Tom,

I hope to add some TIMESYS capability to STIL/STILTS/TOPCAT before
the Paris interop, I'm not sure yet if this would be in released or
pre-release code.  The intention would be to do what I've done for
COOSYS (since topcat 4.5): model the TIMESYS element in STIL's
internal model of a table and update the VOTable input and output
handlers accordingly, so that topcat/stilts can round-trip VOTables
while preserving TIMESYS elements, and allow display of the TIMESYS
attributes in topcat's column metadata window.  On this timescale
I probably will not attempt anything smarter than that,
e.g. functions to actually make use of this time system metadata.

I guess I should prototype TIMESYS validation in taplint too,
though coping with a new VOTable version *may* make that more
complicated than it sounds ... I'll see how I go.

I'll report back if I make progress.

Mark


On Thu, 4 Apr 2019, Tom Donaldson wrote:

> Hi All,
> 
> A new VOTable 1.4 working draft has been published at:
> http://www.ivoa.net/documents/VOTable/20190403/
> 
> The changes this time were fairly minor.  Please see the complete change logs for all the working drafts at: 
> https://wiki.ivoa.net/twiki/bin/view/IVOA/VOTable14WorkingDraftNotes
> 
> At this point, major changes are unlikely unless we need to address issues found during implementation.  In order to move to a "Proposed Recommendation" status, we should have at least 2 reference implementations as well as provisions for validation.
> 
> Exactly what that means is up for debate, but ideally I would like to see (at least) one data provider that can utilize the TIMESYS element, and (at least) one client that can not only parse the VOTable, but also make it clear that it "understands" the TIMESYS semantics.
> 
> I will try to work on the parsing side in both the Astropy VOTable parser and probably the MAST Portal parser.  Neither of those will be able to use the TIMESYS data by themselves, so I'd be interested in hearing some possible use cases as well as getting some realistic VOTables to test on.
> 
> Please let me know if you plan on, or may get to, working on an implementation or validator.
> 
> Thanks,
> Tom
> 
> 
> 

--
Mark Taylor   Astronomical Programmer   Physics, Bristol University, UK
m.b.taylor at bris.ac.uk +44-117-9288776  http://www.star.bris.ac.uk/~mbt/