Am Samstag, 5. Mai 2007 23:27 schrieb Dov Feldstern: > Georg Baum wrote: > > Am Dienstag, 10. April 2007 14:04 schrieb Jean-Marc Lasgouttes: > > > > I fixed this problem by not outputting the \inputencoding command in > > sections. The attached file works fine now, but if you uncomment the TOC > > you get some fontenc errors I don't understand. > > > > 1) This works, so thanks! > > 2) However, I now have a problem with the "default" encoding again! I > don't know if this is because of this fix you provided or not. The thing > is this: I think that "auto" and "default" should be identical, except > for the fact that in "default", we don't actually output the > \inputencoding commands, and in "auto" we do. Other than that, I don't > think there should be any differences between them.
Correct, that is how it has always been. > The attached patch fixes the problem I'm having, I'd be glad if you > could just confirm that it makes sense; also, are there other places > which you think should be fixed to achieve what I want, given the above > explanation? I don't know what problem you have, but the patch does not look right. It completely disables encoding switches (of the stream, not \inputencoding commands) for the "default" encoding. If you need to change something, then it should read - if (((bparams.inputenc == "auto" && !moving_arg) || - bparams.inputenc == "default") && + if (((bparams.inputenc == "auto" || + bparams.inputenc == "default") && !moving_arg) && But I really don't know why encoding switches inside a moving arg should be disabled for the "default" encoding. > 3) Regarding the TOC: I see that in 1.3.X we get the same errors (I've > never used the TOC)--- so this is of course something which should > eventually be fixed, but it's not a regression. I opened a new bug for > this at http://bugzilla.lyx.org/show_bug.cgi?id=3555 . > I actually think I know how to solve this, and it should be quite > simple, but I don't know how to implement it, so I'd be glad if you > could help (see the details in bugzilla). In order to find a good solution some knowledge of the internals of babel and inputenc is required: How does \inputencoding exactly work, and why does using \L or \R in sections has the same effect as \inputencoding. I don't have this knowledge, so I don't have an idea whether your proposed solution is Ok or whether it has side effects. Georg
