On Fri, Mar 02, 2007 at 11:34:06AM +0100, Jean-Marc Lasgouttes wrote: > >>>>> "Martin" == Martin Vermeer <[EMAIL PROTECTED]> writes: > > Martin> On Fri, Mar 02, 2007 at 04:13:14AM +0100, Juergen Spitzmueller > Martin> wrote: > >> José Matos wrote: > >> > >> > What is Jürgen's opinion on the subject? > >> > >> Without having a closer look at the patch, my opinion is that > >> Martin's approach is good (and necessary) in general. However, it > >> should enclose numerical citations in one way or another. > >> > >> Jürgen > > Martin> That would be like the attached... but note that I don't think > Martin> that is right. I agree that numeric should be handled, but > Martin> what happens here is that, also when natbib is pre-loaded by > Martin> the document class, the _user_ is allowed to choose between > Martin> numeric and author-year. I don't think that is right. > > I have no idea on the actual logic, but nevertheless: > > - biblio::CiteEngine const cite_engine = buffer.params().cite_engine; > + BufferParams const & p = buffer.params(); > + biblio::CiteEngine cite_engine; > + if (p.getLyXTextClass().provides(LyXTextClass::natbib) > + && p.cite_engine != biblio::ENGINE_NATBIB_NUMERICAL) > + cite_engine = biblio::ENGINE_NATBIB_AUTHORYEAR; > + else > + cite_engine = p.cite_engine; > > Should getEngine from biblio.C be moved to BufferParams and used > there? I'd like to factor code whenever possible.
Hmm. What would be the advantage? Sure I can do it... won't be next
Tuesday though.
> - features.require("natbib");
> + if
> (!features.bufferParams().getLyXTextClass().provides(LyXTextClass::natbib))
> + features.require("natbib");
>
> I think it is not necessary because LaTeXFeatures handles it.
Damn you're right... and I actually looked at that code when preparing
for this. This was the only thing the old code got right ;-/
- Martin
pgppQm6Vk8BMy.pgp
Description: PGP signature
