On Tue, Feb 26, 2002 at 05:06:57PM +0100, Jean-Marc Lasgouttes wrote: > >>>>> "Martin" == Martin Vermeer <[EMAIL PROTECTED]> writes: > > Martin> Actually, having thought about this, I have come up with a > Martin> much simpler and more elegant approach, that ought to work, is > Martin> sufficiently general and for which we already have the > Martin> infrastructure. > [...] > > Martin> The optional parameter could be entered in a box similar to > Martin> that used by the label width string, and also otherwise > Martin> treated similarly. > > There is a very big problem with this: your optional argument will be > limited to pure text, without any formatting of inset. This is not > acceptable IMO, and I think we should be trying to get rid of this > kind of constructs where we have them (citation optional argument, for > example).
And don't forget the preamble! > So the right way is probably to use a collapsable insettext to do what > we want. But this will not happen in 1.2.0 frame. Perhaps. That will take time though. Considerable time, if you ask me :-) > That's why I sent this proposal for extending layout format: find an > interim solution for 1.2.0. However, it is not clear that this > solution would allow clean upgrade of documents later. My proposal looks slightly better in this respect. Cleaner. But yes, there will be extra work for converting old docs, once \latexoptionstring my string becomes \begin_inset LatexOption my string \end_inset. Still fairly straightforward. > Martin> ...but apart from this practical detail: Does this look like a > Martin> stategy proposal that can be accepted? > > I don't think so, but maybe others have other ideas... > > JMarc Still... this is important functionality that has been lacking for ages. Do we let the perfect be the enemy of the good? 95% of users will be perfectly happy with this, and the other 5% will use raw LaTeX in the string anyway. Martin
msg33480/pgp00000.pgp
Description: PGP signature
