IVOA VOTable signature issue

[François Bonnarel]

Hi Pierre, all,

1 ) Having the standardID of the protocol in an INFO tag of the response 
is definitely a good idea.

2 ) When you want to use Datalink from a protocol response which is not 
SIAV2 or ObsTap: let's say, SCS, SSA, SIAV1, TAP apart Obstap, SLAP or 
whatever, the best way to call it is to add a Datalink descriptor as 
Carlos did in his ASTERICS presentation. The list of IDs used by the 
DataLink service must be in one of main table columns to make this 
solution work.

Cheers

François



Le 21/03/2017 à 12:08, Pierre Fernique a écrit :
>
>
> Dear DAL and App contributors,
> and especially SIAv2 and Datalink authors,
>
> In the contexte of the Aladin v10 developments, we are facing to a 
> recurrent issue in the IVOA protocol suite.
>
> *Most of IVOA protocols return VOTable, but without any signature or 
> magic code. *
> The consequence: IVOA clients can not easily recognize that this 
> VOTable stream comes from a CS, a SIAv1, a SIAv2, a SSA, a Datalink, a 
> TAP, an obsTAP or any other IVOA VOTable query result.
>
> Based on the presence of some UCDs or utypes or INFO tags, Aladin 
> tries to recognize the original protocol in order to have the adequat 
> actions (basic table display, fov supports, popup menus, column 
> display filtering...). But we fail to find such a method for SIAv2. We 
> are, more or less able to recognize obsCore results (based on utypes 
> list => "obscore:ObsDataset.dataProductType", 
> "obscore:Access.Reference" and "obscore:Access.Format" , but it is not 
> specifical to SIAv2.
>
> We are facing to the same difficulty for Datalink results. In this 
> case, it is worse in the sense that some providers have already 
> inserted Datalink links inside SIAv1 or CS results (see examples 
> below). So in these cases,*we cannot know "a priori" that such link 
> will return Datalink VOTable stream*. Without this knowledge, at the 
> end, Aladin can only try to display on the sky view the VOTable 
> Datalink result as a regular astronomical catalog (sad !).
>
> So, how can we do ? Thanks for any help.
>
> *For the future standard releases, may we just think to systematically 
> add the standardID of the original protocol inside an INFO tag, or 
> something equivalent ?*
>
> Pierre Fernique & Chaitra (in charge of SIAv2/Datalink/... 
> implementation in Aladin v10)
>
> ---
>
> Datalink examples:
>
> *GAVO SIAv1*
>
>   * http://dc.zah.uni-heidelberg.de/lswscans/res/positions/siap/siap.xml?POS=210.8431667,54.2961666667&SIZE=0.1323&FORMAT=image/fits
>   * => Example of Datalink URL inside the result :
>     http://dc.zah.uni-heidelberg.de/lswscans/res/positions/dl/dlmeta?ID=ivo%3A//org.gavo.dc/%7E%3Flswscans/data/part2/Schmidtplatten/FITS/A82.fits
>
> *VizieR CS*
>
>   * http://vizier.u-strasbg.fr/viz-bin/votable?-source=B%2Fcfht%2Fobscore&-c=09+55+33.17+%2B69+03+55.1&-out.add=_RAJ,_DEJ&-oc.form=dm&-out.meta=DhuL&-out.max=999999&-c.rm=10.81&-out=_VizieR,*Mime(image/fits),*&-mime=TSV&-out.all
>   * => Example of Datalink URL inside the result :
>     http://www.cadc-ccda.hia-iha.nrc-cnrc.gc.ca/caom2ops/datalink?runid=ejkaxv927qnvu4ef&ID=caom%3ACFHT%2F469601%2F469601o
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ivoa.net/pipermail/apps/attachments/20170323/6a797a4f/attachment.html>