NeXus java implementation

[Mark Koennecke]

  Ray,

On Sat, 4 Nov 2000, Ray Osborn wrote:

> Mark,
> I'm only just getting around to updating the web pages and FTP site to
> include your java implementation.  Thanks for doing this so quickly.  I had
> one immediate concern that I thought I'd better raise quickly.  You have
> named the java package neutron.nexus.  Is there a reason for not just
> calling the package nexus?  My main concern is that NeXus is for neutron and
> xray scatterers, so it seems a discriminatory name.  It also means that some
> of the javadoc file names became too long for Macs (at least before OSX
> takes over) so it's screwing up my CVS repository.
> 
  The reason I did this is to avoid name clashes. NeXus is not an uncommon
  word these days (I now have a bicylce with a nexus hub!)  and I'am aware
  of at least another Java package using the name nexus. 

  I had no problems putting the sources and docs into the NAPI CVS
  repository at ISIS. The Mac problem: we have no shared library for 
  Mac's yet. This may also be a difficult thing to do, I believe Apple
  went its own way with some of the JNI-stuff. Then the Java documentation
  is generated from the source. I wonder if there isn't a Macintosh tool
  which takes care of the filename problem. Especially as you would have
  problems with filename length with all the standard  Java classes. 
  java.awt.event.ActionEvent is even longer then neutron.nexus. 

> If there is some name clash, then I would recommend that we call the package
> just NX, i.e. NX.NeXusFile etc.
> 
  I see your point about offending the X-ray community. Unfortunately
  changing the package name now is a lot of work. All source files, all
  documentation and all the signatures in the JNI-interface implementation 
  would need to be changed. I would rather prefer to add a line of
  apology in the documentation, perhaps stating that we named it
  neutron.nexus because it was written at a neutron scattering site. BTW:
  I do not think that X-ray scatterers are so faint-hearted that they
  would not use a package named neutron.nexus.  
>
> I'll try and integrate the documentation over the next week or so, although
> I don't have very much free time at the moment.
> 
  
  Another thing: did you get my email with the suggestion for a date for 
  the joint NeXus/SCANS meeting at PSI? I feel that this room and time 
  issue is more pressing at the moment.


               Regards,

                       Mark