On Thursday 13 March 2003 21:10, Daniel Vogelheim wrote:
Very interesting reasoning and development of the issue.
The next suggestion (Michael) uses that as base, but adds an 'escape'
[i.e. declaring an individual paragraph to be listed at some level] to
take away some of the burden from the filter people
As a general guideline, I would say that 'clean' XML solutions are to be
preferred _if_ they can represent existing documents.
I think this raises the following question: shouldn't discontinued lists,
(like the numbering of tables, pictures, etc.) be modelled with the
above solution? This would be, I think, "cleaner" in terms of structure
than the current OO solution (of autonumbered variables).
1. pict one
2. pict two
1) table one
2) table two
3. pict three
(1). equation one
3) table three
(2). equation two
4. pict four
Each paragraph belongs to one of the 3 discontinued lists going on here
(picts, tables, equations).
I think this would even allow some more features than simple autonumbered
counters, like being able to refer to a given picture twice without getting
the number increased twice. It would also give all the features available for
list counters (without having to redevelop or refactor the code for auto-
For such lists there is obviously no "XML-structured list" solution. However
the closest XML-clean solution is simply to associate each paragraph with
a list style.... and to set the continue-numbering flag. Hmm, I guess this means
that the OO file format already has support for this, it's just not being used for
this case by OO itself AFAIK.