Hi Suvayu, 2013ko martxoak 12an, Suvayu Ali <fatkasuvayu+-ek idatzi zuen: > > That is understandable, but if the aim is to support other kinds of > graphics that support :width, I expect the change to not interfere with > graphics that do not support it. I think there are two cases here: > > 1. the author targets the document for a specific backend that does not > support :width > > 2. the author expects to export to multiple backends and wants to be as > general as possible so uses and expects :width to work.
:width is an ATTR_LATEX keyword, as is :options. So they only apply to the LaTeX backend – and those which derive from it, such as beamer. If document authors wish to also export to e.g. HTML, they will have to make separate arrangements. > > So now the question is, is it reasonable to expect the author in case > (1) is aware of this subtlety and use :width even though her/his backend > does not support it, or would it be better to handle these deficient > cases "specially" and help the author by not clobbering their :options > argument. Eliminating subtleties is precisely the point of this change. All(-ish)* backends now use :width. * As far as I’ve checked, HTML(+ derived backends) and LaTeX(+derived backends). If there are any that don’t, they should probably be patched to do so as well. -- Aaron Ecay
