Re: Preceding list function

[Brandon Ibach <bibach@xxxxxxxxxxxxxx>]

Quoting Norman Walsh <ndw@xxxxxxxxxx>:
> Thanks, Brandon. In considering your solution, I couldn't convince myself
> that it handled all of the various nesting cases correctly.
> 
> I wasn't, for example, sure that simply parent was looking far enough
> back, since we could be nested arbitrarily deep.
> 
   Hmm... perhaps I didn't completely understand the semantics of the
continuation feature.  My assumption was that a continuation can not
span certain ancestor elements.  The version of your function that I
modified demonstrated this by only searching the current component
level element's descendants for "candidate previous lists" (CPLs :).
I assumed that an ancestor list would fall into this same category,
thus any list outside of the ancestor list would not be a valid
candidate.  To demonstrate:

<orderedlist>
  <listitem>Group I
    <orderedlist>
      <listitem>First</listitem>
      <listitem>Second</listitem>
      <listitem>Third</listitem>
    </orderedlist>
  </listitem>
</orderedlist>
<para>Some intervening text</para>
<orderedlist continuation=continues>
  <listitem>Group II
    <orderedlist continuation=continues>
      <listitem>Fourth</listitem>
      <listitem>Fifth</listitem>
      <listitem>Sixth</listitem>
    </orderedlist>
  </listitem>
</orderedlist>

   With this fragment, should the "Fourth", "Fifth" and "Sixth" items
be numbered 4, 5, 6 or 1, 2, 3?  This example also raises another
issue.  Shouldn't the continuation look for a preceding list at the
same nesting depth?  If not, then the example would end up looking
like:

   1. Group I
      1. First
      2. Second
      3. Third
   Some intervening text
   2. Group II
      2. Fourth
      3. Fifth
      4. Sixth

   Also, on a related, but not yet discussed, issue, if the numbering
of the inner lists should be continuable, shouldn't it be possible for
the outer lists to be unordered (ie, bullets), such that you could
produce:

   o Group I
     1. First
     2. Second
     3. Third
   Some intervening text
   o Group II
     4. Fourth
     5. Fifth
     6. Sixth

   Finally, and perhaps most importantly, is there something just a
little weird about the fact that I consider working this sort of stuff
out "fun"? ;)

-Brandon :)


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