[Ex-plain] Change 3: Types list

Robert Sanderson azaroth at liverpool.ac.uk
Mon Jan 26 02:08:02 CET 2004


> > Yes. And Ralph already suggested 'sru'.
> > Same applies to the list of configInfo types.
> OK.  I shall leave you craft perfect prose describing this situation

Yup.

> > > > sortSchema       The default schema used in sorting, in short name
> > > >                  form
> > Path is the only mandatory field.
> Sorry, I'm off the Internet at the moment so I can't read this, but
> ...  If path is specified then schema is irrelevant ... isn't it?

You don't know what schema the path is part of. For example:

/record/title  ... Really what we need to know is the semantics of 
'title'. But just /record/ before it helps not an iota.

It's a bit of a hack though, as you should be able to say:
  /dcloc:dc/dc:title

But there's no namespace assignments to take these from.
I'm not losing sleep over it.

 
> > > > stylesheet       Default stylesheet URL, or if stylesheets are
> > > >                  supported 
> > > Remind me what this is used for?
> > 
> > The xml-stylesheet processing instruction before the xml, such that 
> > terminally braindead clients can render the response using XSLT.
> 
> OK, so the server's not going to _do_ anything with this?  It's just
> the echoed-back-to-the-client stylesheet parameter?  So the explain
> element is saying "If you don't tell me what stylesheet to tell you to
> use, I'm going to tell you to use this one"?

Yup.  So if your client will barf on a stylesheet, then you know to 
request nothing.
(Can you request no stylesheet??)


> > But neither are very amenable to nailing down as the size of a term
> > could conceivably be monstrous, just like the size of a record.  For
> Hmm, yeah, OK, I buy it.  (And don't forget Les Wibberly's Z39.50
> search-terms that are 10,000-node chemical formulae.)

Yup.

> > > > extraSearchData    A type of extraRequestData available in the 
> > > What's the value?  An {identifier URI, element-name} pair?
> > I guess so.  That needs further nailing down if we want to specify every 
> > single element name rather than just the identifier?

> I think we _have_ to so this on a per-element rather than
> per-namespace basis.  It's quite possible that someone will define two
> extensions in a single namespace and I'll support only one of them.
> Is there a standard syntax for representing {namespace, element-name}
> pairs?  You'd think there would be, wouldn't you?

dc:title ? 
But you can't expand dc out to the full namespace URI.
Ahhh.... Don't know.


> > > > charSet            A character set supported, or the default character 
> > > Remind me where comes into play?
> > Ahh ... Z39.50 only I guess. As you can't request a character set in SRW.
> OK.  Then let's either remove this, or (better) you could build an
> initial list for describing Z39.50 support, too.

It could describe the character set in which you're going to get the 
records back, but that's able to be carried on the xml version 
instruction.

Rob

-- 
      ,'/:.          Dr Robert Sanderson (azaroth at liverpool.ac.uk)
    ,'-/::::.        http://www.o-r-g.org/~azaroth/
  ,'--/::(@)::.      Special Collections and Archives, extension 3142
,'---/::::::::::.    Nebmedes:  http://nebmedes.o-r-g.org:8000/
____/:::::::::::::.  
I L L U M I N A T I  








More information about the Ex-plain mailing list