[Ringing-lib-discussion] Method database - getting there

[Martin B. <mjb...@li...>]

Hello everybody!  Over the last few months I and others have been gradually 
working on the method database, and it's looking pretty good at the moment.

Firstly, we have come up with a draft XML format for methods.  This was put 
together mostly by Richard Smith and me, with input from some others.  An 
XML Schema is in "method.xsd" on the CVS repository, and some text 
describing it in "method.xsd.txt".  If you've forgotten where the 
repository is, you can browse it here:
  <http://cvs.sourceforge.net/viewcvs.py/ringing-lib/methods/>

The good news is that the Methods Committee are taking an interest.  They 
want to have a standard XML reprentation of methods, and may be going to 
use our draft as a base for their proposals.  I think that we should make 
sure that the database supports whatever they finally come up with; in the 
meantime we can support our draft.

I'd appreciate any comments anybody might have on the draft, and so would 
the MC.


Secondly, I've just about finished the code for getting XML data out of the 
database.  I've abstracted the hard work into a Perl module, called 
DBIx::XMLServer, which I'll release on CPAN as soon as I've done a bit more 
documentation and testing.  You can find it at
  <http://sf.net/projects/dbix-xmlserver>.

Using this module, our web site only has to have an XML file describing the 
mapping from the database to the XML document.  You can see how it all 
works by looking at the code in CVS.  The only slight annoyance is that the 
generated XML has quite a few unnecessary namespace declarations, but 
that's really a cosmetic thing.


So, what now?

- Tidy up a few things like lead heads which don't quite work at the moment.

- Put an HTML front end on top of the XML, so that people can browse the 
database from the web site.

- Get scripts in place to keep the database up to date.

- Develop a few client-side interfaces so that programmers will use our 
database in their programs.

- Think about the `complicated' interface which will be a safe way for 
clients to execute almost arbitrary SQL statements on the database.

Probably loads more stuff as well.

Martin

-- 
Martin Bright
Department of Mathematical Sciences, University of Liverpool