validation of SIA 2

[Mark Taylor]

On Thu, 25 Feb 2016, James.Dempsey at csiro.au wrote:

> 6. It would be nice to have lack of a schema location as a warning rather than showing as an invalid VO table.

I'm guessing this characteristic of the implementation might be
related to the fact that it seems to be shockingly difficult to
write java code that validates an XML document against a known XSD,
regardless of any schema declaration in that document.
I spent a lot of time banging my head against this in taplint;
Pierre if you want to take a look at the way I do it, the relevant
code (with fairly extensive comments) is mostly here:

   https://github.com/Starlink/starjava/blob/master/ttools/src/main/uk/ac/starlink/ttools/taplint/XsdStage.java

If you don't need to mix standard and private namespaces then
it's less painful, you can see the previous version (prior to
76db3dc66a9ab5c82) e.g. here:

   https://github.com/Starlink/starjava/blob/df07e0d45d8f814ce4fe32a564bbedf37adfe2b7/ttools/src/main/uk/ac/starlink/ttools/taplint/XsdStage.java

You're welcome to use or abuse anything there if you think it would
help, though it still might take some implementation effort to adapt
it to your purposes.

If this lot is not relevant - sorry!  just ignore it.

Mark

--
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/