Re: [cml/ccml-discuss] Fwd: Re: CML and JSpecView (from Robert Lancashire)

[peter murray-r. <pm...@ca...>]

At 09:08 03/06/2006, Robert Lancashire wrote:


> >Hi all,
> >
> >I have now added CML spectra support to JSpecView
> >(http://jspecview.sf.net)
> >the JARS are under FILES not CVS since that has not recovered
> >from the sourceforge disk problems.

FWIW the SF CVS has changed the Root name (to (say) 
cml.cvs.sourceforge.net rather than cvs.sourceforge.net. I only found 
this out yesterday and am now re-updating the CML CVS

> >
> >The error checking done is minimal at present and so it
> >may crash if it comes across something it doesn't understand!

Understood. We need to agree on the semantic constraints for 
CMLSpect. e.g. can there be a y axis without an x-xaxis - or vice 
versa. These constraints can then be added to the schema

> >
> >The standalone and applet JARS originally designed for JCAMP-DX
> >can both read and write CML as well as AnIML files and it is possible
> >to convert between JCAMP-DX and AnIML and CML.

Excellent Excellent

> >
> >When importing, AnIML files are recognised by file extensions of
> >aml or xml, while CML files are distinguished by the cml extension.
> >This will need to be rewritten to test more substantially since
> >if you attempt to load a CML saved with an XML extension it will
> >probably crash!  Sorry but I'm in the middle of exam chaos so it
> >may take a while to find some time.

This is a difficult problem. In general I now use XML for all my 
files and use the namespace to disambiguate. The main reason for 
*.cml would be for legacy tools that don't have namespace tools. In 
any case we can expect compound documents that contain CML, HTML, 
AnIML, etc. and *.xml (or *.xhtml) seems the only reasonable option.

> >
> >good luck, try it out.

Thanks - I will try to do so soon.



Peter Murray-Rust
Unilever Centre for Molecular Sciences Informatics
University of Cambridge,
Lensfield Road,  Cambridge CB2 1EW, UK
+44-1223-763069