Hi,

Thanks for your feedback, and sorry for taking a while to respond.

Your objections are very reasonable and parsing these lists as regular
paragraphs is a decent alternative, but it also has its drawbacks. So I 
came up
with a third solution. The main drawback was that these lists without 
markers
are often used to add paragraphs in a list item without visually adding a 
new
list element or ending the list. By using paragraphs, we would visually end 
the
list. My solution to this is to add the list items without markers as
paragraphs to the last item of the list immediately before if there is one. 
In
the other cases we propose to just use paragraphs as you suggested.

I have attached a docx file and an HTML file to demonstrate both cases and 
how
they would be rendered with my solution.

What are your thoughts on this solution? If you think it's good I'll send a
pull request either Friday or in two weeks.
On Tuesday, May 4, 2021 at 7:39:23 PM UTC+2 John MacFarlane wrote:

>
> I guess I do have objections to features that get implemented
> in just a couple of formats, leaving the work of implementing
> them in others to other people.
>
> There's also a question whether it makes sense to support
> this kind of list -- that depends on whether it has a reasonable
> rendering in enough of the formats we support. (Particularly
> in Markdown, which is supposed to be expressive enough for
> everything in the AST, with the current exception of some of
> the new table features.)
>
> What about parsing these docx lists as just lists of regular
> paragraphs? Then they'll render pretty well in every output
> format.
>
> Milan Bracke <milan....-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:
>
> > Hi,
> >
> > Both docx and HTML have a list type named "none", where no marker is 
> shown.
> > However, Pandoc doesn't support these types of lists and will put numbers
> > instead. See the attached docx file for an example.
> >
> > We would like to add a ListNumberStyle called None in pandoc-types and 
> use 
> > it in
> > the docx reader and HTML writer. For HTML5, we can't use the "type" 
> > attribute so
> > we will have to fall back on the "list-style-type" CSS as described here:
> > https://www.w3schools.com/tags/att_ol_type.asp, and for docx the change 
> > seems
> > simple since all the types can be set with w:numFmt.
> >
> > We'd prefer not to implement the usage of the new type in the other 
> readers 
> > and
> > writers yet. (We'll of course avoid breaking them.) We only need it in 
> the 
> > two
> > mentioned above and we don't master all the file types in Pandoc. Is 
> this 
> > OK and
> > should we make an issue to track the progress in the other readers and 
> > writers
> > when people implement the new type there?
> >
> > Do you have any feedback or objections? If there are no objections, 
> we'll 
> > make
> > a pull request soon.
> >
> > -- 
> > You received this message because you are subscribed to the Google 
> Groups "pandoc-discuss" group.
> > To unsubscribe from this group and stop receiving emails from it, send 
> an email to pandoc-discus...-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
> > To view this discussion on the web visit 
> https://groups.google.com/d/msgid/pandoc-discuss/4e936aa2-41d6-4d68-ba5a-bea916754922n%40googlegroups.com
> .
>

-- 
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe-/JYPxA39Uh5TLH3MbocFF+G/Ez6ZCGd0@public.gmane.org
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/11e20690-21cc-4287-81b0-e5f3e191cf1bn%40googlegroups.com.