Jean-Marc Lasgouttes wrote: >>>>>> "Martin" == Martin Vermeer >>>>>> <[EMAIL PROTECTED]> writes: > >>> swedish if the inset is one with noFontChange() == true, else >>> norwegian. At least I think that it works like that. > > Martin> Yes, that's how I think it *ought* to work. What else is > Martin> noFontChange for.
Does this patch restore 1.3 behaviour of noFontChange, or did you change it wrt. languages because the sentence above? I did not want you to do the latter, I agree with Jean-Marc that noFontChange should not do that, I only described above what I thought it actually does. > Dekel Tsur has always fought against an inherit_language value, on the > ground that a text does have a language and is not supposed to inherit > anything when you do some cut-and-paste. I can see some value in the > argument, but it causes many problems. He is right IMO. I think the problem is that language is currently a font attribute. We need to do all sorts of crazy stuff with fonts (realize, reduce) which does not work very well for languages. I believe that the problems would go away if language was a text attribute independent from the font. Georg
