Re: RTF styles from jade? (was Re: (dsssl) How to make a RTF styles-sheet with DSSSL?)

Subject: Re: RTF styles from jade? (was Re: (dsssl) How to make a RTF styles-sheet with DSSSL?)
From: "Paul Tyson" <paul@xxxxxxxxxxxxxxxxxxxxxx>
Date: Sun, 10 Mar 2002 09:09:11 -0800
----- Original Message ----- 
From: "Ron Ross" <ronross@xxxxxxxxx>
Sent: Saturday, March 09, 2002 9:47 PM

> How feasible would it be to implement wp-type styles in the RTF backend;
> that is, to encapsulate all the formating procedures for a particular
> type of element within an rtf-style whose name would be derived from the
> dsssl stylesheet?
> 
> I imagine that such an rtf-style backend would interface with OpenJade's
> SL either through an extension of JC's original "heading-level"
> extension (where the `heading-level' characteristic could take any name)
> or through standard dsssl by way of the `style' and `use' features (can
> a dsssl "style" specification "use" another "style"? That would provide
> a foundation on which to implement the "based-on" RTF style hierarchy).
> 

Exposing DSSSL 'style' objects would be a good thing, not just for the RTF backend, but others as well.  (A CSS backend, for instance.)  I don't think it would be useful to try it any other way--for instance, just gathering up all the characteristics that are applied in each construction rule, giving it a name, and writing it out as a 'style'.

One significant theoretical difficulty is that flow object characteristics inherit hierarchically, while the prevailing "inheritance" of word processing styles is linear.  (That is, certain characteristics of a paragraph in RTF are inherited from the preceding paragraph if not explicitly set or reset.)  If you have a deeply nested flow object whose characteristics have been set at 2, 3, or more different levels above, you find yourself with an exponentially increasing number of possible word processing styles.  Unless you solve this problem in a general way, you have to set all wp characteristics for each paragraph (like the current RTF backend does).  Even if you were able to calculate all possible 'style instances', the result would probably not be very useful in a word processing environment.

But, at a minimum, the ability to export named styles would be useful.   To make good use of this capability, you would have to write your style specifications very carefully, and the restrictions might prove too burdensome.  I can imagine some situations where this could be useful, though.

I wouldn't put it very high on the list of things openjade needs, but it should certainly be on the list.

Later,
--Paul

Paul Tyson, Principal Consultant                   Precision Documents
paul@xxxxxxxxxxxxxxxxxxxxxx              http://precisiondocuments.com
     "The art and science of document engineering."



 DSSSList info and archive:  http://www.mulberrytech.com/dsssl/dssslist

Current Thread