Re: [dita] DTD Formatting: Attribute list Alignment

[Eliot Kimber <ekimber@innodata-isogen.com>]

Deborah Aleyne Lapeyre wrote:

> 
> Ask yourself, quickly, how many attributes are these and
> how many are CDATA?
> 
> 
> <!ATTLIST foo
>   foo-type
>     CDATA
>     #IMPLIED
>   %dtd-vers;
>   xml:lang
>     NMTOKEN
>     "en"
>   status
>     (live |
>      dead |
>      phooie)
>     "dead"
>   foo-customization
>     CDATA
>     #REQUIRED

4 and 2: scan down the left side counting outdented tokens (4). Move in 
3 characters and scan down for "CDATA" (2). But then I'm used to this 
layout.

> Now ask yourself the same question:
> 
> <!ATTLIST  foo
>              foo-type   CDATA                          #IMPLIED
>              %dtd-vers;
>              xml:lang   NMTOKEN                            "EN"
>              status     (live | dead | phooie)           "dead"
>              foo-customization
>                         CDATA                         #REQUIRED

Now add an item to thet status list--requires lots of adjustment of 
white space and possibly breaking of lines.

I guess it partly comes down to optimizing for retrieval or 
modification. My one-line-per approach is optimized for 
modification--all changes require only adding or deleting lines.

For the DITA DTDs it may be that retrieval is the more important 
characteristic (ignoring the fact that I don't find the above format 
particularly readable, but that's because I'm used to the one-line-per 
format).

Cheers,

E.

-- 
W. Eliot Kimber
Professional Services
Innodata Isogen
9390 Research Blvd, #410
Austin, TX 78759
(512) 372-8122

eliot@innodata-isogen.com
www.innodata-isogen.com