Hello, Bastien <b...@gnu.org> writes:
> Nicolas Goaziou <n.goaz...@gmail.com> writes: > >>> than it is to automatically replace verbatim attribute code with plists, >>> especially if there are all sorts of html irregularities in there. >> >> I prefer not to mix the two methods as it would be fragile (e.g. what >> happens if an attribute is defined both outside and inside the :options >> keyword?). > > Sorry, catching up in this thread -- my understanding was that > :options was *always* available as a fallback, even with other > attributes set before it. Am I wrong? :options is more or less required in LaTeX export. Other keywords may go in different locations (optional arguments, mandatory arguments) so the back-end has to understand the keywords provided. It also means there will always be keywords it won't understand (i.e. which are not hardcoded). The :options bucket is a good place for them. On the other hand, I think we don't need it in HTML export. All attributes go in the same place, so the backend just transforms the plist into a string. Since it doesn't need to know about what it transforms, the bucket becomes useless, maybe confusing. Regards, -- Nicolas Goaziou
