RE: [dita] syntax for domain usage declaration

["Bruce Nevin (bnevin)" <bnevin@cisco.com>]

I guess I don't understand the reason for using regular expression
syntax to describe the syntax for entering an attribute value in a
specialization. Does some tool for creating domain usage declarations
require regular expressions as input? Is XSL pattern matching required?
What audience is the regular expression syntax directed to, and for what
purpose?

Wouldn't it be more straightforward to say each parenthesized module
ancestry specification is a space-separated sequence of module names
beginning with the root module name (topic or map) and ending with the
name of the module whose ancestry is being declared? (The phrase
"parenthesized module ancestry specification" occurs just prior to this
definition.) 

Jeff said something similar:

> It doesn't fully describe the domains attribute value since 
> attribute domain declarations take the form:
>    a(module1 moduleN)
> where module1 is always either "props" or "base".

(But did you mean "where module1 is always either "topic" or "map"? And
I don't understand the leading 'a', or did you mean a=(module1
moduleN)?)

Robert added:

> I wonder if it might be clearer if the example also allowed (topic
pr-d), 
> the programming domain; the topic then contains 3 domains coming off
of topic, 
> and one domain that further specializes one of those three.

It would be just as informative to point out that the example includes
cpp but excludes pr-d as a whole. 

But my question on reading this is, having declared pr-d, why would I
want to still declare cpp-d? All the elements in cpp-d are already
available as part of pr-d, aren't they? Is the ancestry of pr-d not
recoverable from that of cpp-d?

	/Bruce


> -----Original Message-----
> From: Ogden, Jeff [mailto:jogden@ptc.com] 
> Sent: Thursday, September 24, 2009 2:34 PM
> To: Bruce Nevin (bnevin); DITA TC
> Subject: RE: [dita] syntax for domain usage declaration
> 
> I think it is correct as far as it goes. It is hard to read 
> because of the different usages of paren.
> 
> It doesn't fully describe the domains attribute value since 
> attribute domain declarations take the form:
> 
>    a(module1 moduleN) where module1 is always either "props" 
> or "base".
> 
> I don't think we have any real attribute domain declarations 
> anywhere in the DITA DTDs or XSDs. That is too bad since it 
> makes finding a real example to look at hard.
> 
>    -Jeff
> 
> > -----Original Message-----
> > From: Bruce Nevin (bnevin) [mailto:bnevin@cisco.com]
> > Sent: Thursday, September 24, 2009 2:20 PM
> > To: DITA TC
> > Subject: [dita] syntax for domain usage declaration
> > 
> > Is the following description correct? It's in 
> > ArchSpec/Base/Specialization/"Domain usage declaration (the domains 
> > attribute)". I'm having trouble making sense of it:
> > 
> > 	Each domain defines its module ancestry as a parenthesized
> > 	sequence of space-separated module names from root module
> > 	to provided module: '(', modulename, (' ', modulename)+, ')'.
> > 
> > The example given there is
> > 
> > 	domains="(topic ui-d) (topic sw-d) (topic pr-d cpp-d)"
> > 
> > 
> > 
> > 
> ---------------------------------------------------------------------
> > To unsubscribe from this mail list, you must leave the 
> OASIS TC that 
> > generates this mail.  Follow this link to all your TCs in OASIS at:
> > 
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
> 
>