Re: [office] List styles

[David Faure <faure@kde.org>]

On Friday 22 October 2004 16:38, Michael Brauer - Sun Germany - ham02 - Hamburg wrote:
> Dear TC members,
> 
> I did some investigations for list styles.
> 
> First of all I checked how a list styles can be switched off in a style 
> derived from a style that specifies a list style. It seems to me that 
> this is not possible at the moment. There is no attribute to do so, nor 
> can the style:list-style-name attribute be empty. Interestingly, the 
> OpenOffice.org application in fact has the issue that one may switch of 
> a list style in the UI for paragraph styles, but this situation 
> currently is lost when saving and reloading a file.
> 
> To resolve this issue in our specification, we either
> a) could allow a style:list-style-name to be empty, or
> b) we could add an attribute similar to the "text:enable-numbering" we 
> discussed last week, that in fact may have a better name.
> 
> I personally would prefer solution a) here.
I agree.

> We remaining issue is: If a graphic style contains a list style, and if 
> this list style is switched off in a derived style, can this style by 
> switched on in a further derived style again, and if so how. The 
> following solution came into my mind here:
> 
> c) The derived style may contain an empty <text:list-style> element. If 
> it is present, all style settings are inherited from the ancestor style 
> that defines a list style already.
> d) The list style can be referenced by a "special" value of the 
> "style:list-style-name" attribute, like "*".
> e) The list style can be referenced by a "special" value of the 
> "style:list-style-name" attribute, like "*", and in addition to this, 
> the embedded list style in general will only be evaluated if a 
> "style:list-style-name" attribute is present.
> f) like a), that is, we add an attribute similar to the 
> "text:enable-numbering".

All of those seem to go against the general mechanisms for style inheritance.
Given that for everything else there is no way to "switch on" settings that were 
"switched off" by inheritance, other than re-stating the exact values of the 
required settings, why should this be different for list styles?
I cannot see the need for this "exception".

-- 
David Faure, faure@kde.org, sponsored by Trolltech to work on KDE,
Konqueror (http://www.konqueror.org), and KOffice (http://www.koffice.org).