Re: [docbook] Re: Followup to Norm's write up on numbering.

[nico <nicolas.marsgui@libertysurf.fr>]

On Wed, 19 Jul 2006 12:51:32 -0400, Norman Walsh <ndw@nwalsh.com> wrote:

> |  2) Add an optional @linenumberstep (or something similar) that allows
> |     the author to override the global setting for line number
> |     increment.  This is not to say that the stylesheet should not be
>
> I dunno. This seems so uncommon that a PI should be sufficient.

PIs are good, but specific to a processor. IMHO line numbering step is not  
specific to a processor, and the document should not be aware of all the  
processors that can be applied on it.

Maybe that a generic PI processor should be defined by the DocBook  
Project, like <?dbrender?> with an official predefined set of global  
supported PI parameters. The benefits would be that:

- a document would not need to put <?dbhtml?>, <?dbfo?>, for each  
processor, improving its portability for other processors,
- everyone interested in providing its own processor (it's my case ;-)  
could try to handle the same set of instructions.

But it's another subject.

> On the
> other hand, if your world view is that the line numbers are every bit
> a part of the content as the rest of the listing, then maybe it's
> reasonable to provide direct control to the other. But on balance, I'd
> rather not.

Yes, it's rather my view, such like the starting number, continuation,  
etc. are part of the listing. All those pieces of information describe how  
to map and show the numbers of the listing lines.

BTW, there are other listing line numbering attributes that I would  
suggest to add:

@linenumstart, @linenumstop: usefull when handling an external file, it  
would tell what portion of the file should be displayed by the document.  
These are not rendering attributes, are they?

But again it's another subject, and I guess I should file another RFE :-)

Regards,
BG