Re: IVOA Footprint Service Protocol

From: Arnold Rots <arots-at-head.cfa.harvard.edu>
Date: Tue, 13 Nov 2007 17:04:28 -0500 (EST)


I understand that one of the questions that has come up is how to encode regions efficiently in exchange documents, in particular when there are many of them.

The STC standard provides for the general exchange of Region descriptions: a simple XML document that contains either a single Region (using an element with xsi:type="STCRegion") or a list of Regions (using an element with xsi:type="STCRegionList") This is the exchange format used in the JHU footprint service. Considering that there is a standard XML serialization of Regions, it makes sense to use this format when exchanging Region descriptions in XML.
An example of such a description may be found at:

   http://hea-www.cfa.harvard.edu/~arots/nvometa/STC/mytest_circle3.xml

As to the question what is the best way to incorporate Region descriptions in a larger document, I will give examples of three possible approaches. For convenience, they are all based on the little utility schema

   http://hea-www.cfa.harvard.edu/~arots/nvometa/STC/stcTab.xsd which allows putting all kinds of coordinate information in a simple table. This simple schema defines an STC metadata element and a plain table where components of the STC metadata can refer to specific columns in the table through ID/IDREF pairs; this ensures that the place of each datum in the metadata model is fully specified. The three examples implement the description of a 0.4x0.2 degree area in different locations on the sky, with different position angles. The table contains fiducial URLs to the supposed image data that should be ignored.

First, the brute-force way: simply embed the Region elements
(Polygons) in the table itself:

   http://hea-www.cfa.harvard.edu/~arots/nvometa/STC/My5EmbeddedRegions.xml It shows that it is perfectly feasible to embed this information in any document.

Next, referencing Regions that are recorded elsewhere, which makes the table much more compact:

   http://hea-www.cfa.harvard.edu/~arots/nvometa/STC/My5XlinkRegions.xml The actual Region descriptions may be found by following the XPath
(2.0) pointers in the Xlink hrefs to the XML document that contains
all the descriptions:

   http://hea-www.cfa.harvard.edu/~arots/nvometa/STC/My5Polygons.xml

Third, is there a way to compress the total volume of XML, by parameterizing the five congruent Regions? There is, in the third example:

   http://hea-www.cfa.harvard.edu/~arots/nvometa/STC/My5TransformRegions.xml Here the Region shape is specified once in the Table metadata and properly defined through a transformation element that has its parameters specified in each table row.

You may find this set of examples also on my STC web page, near the end:

    http://hea-www.cfa.harvard.edu/~arots/nvometa/STC/

Arnold H. Rots                                Chandra X-ray Science Center
Smithsonian Astrophysical Observatory                tel:  +1 617 496 7701
60 Garden Street, MS 67                              fax:  +1 617 495 7356
Cambridge, MA 02138                             arots-at-head.cfa.harvard.edu
USA                                     http://hea-www.harvard.edu/~arots/
--------------------------------------------------------------------------
Received on 2007-11-13Z23:05:09