In our previous episode, Michael Van Canneyt said: > > I'm new to FPC. > > I've used TP for many years up to TP7 but not Delphi. > > Then I came to Ada. > > FPC seems to bring many enhancements to TP. > > First I'm reading some documentation ;-) > > I'm looking for a way to program with Cocoa on Mac. > > Thank you for the remarks. > I'm currently updating the documentation and I will take your remarks into > account.
As we are talking about the ref docs: I'm mostly ready with the keyword updates to ref.tex btw. and the textmode IDE now supports cross-chm links. I yesterday updated the CHM snapshot at http://www.stack.nl/~marcov/doc-chm.zip While working on this, I stumbled on a new problem, how to deal with multiple occurances of keywords in different CHMs. Within (fpdoc) CHMs these are disambiguated with child nodes in the index, but this isn't possible cross-CHM. For the normal docs, I don't know how \keywordlink works, but it would great if it could also dump the topic title to the .kwd file. Even though the prog/user/tex index is small, there are duplicates (like CLASS, VAR and CONST which have multiple meanings) The problem is worse with FPDOC. If you merge multiple CHMs into a general index, and two CHMs contain the same keyword, there is not much info to disambiguate. Maybe all keywords in fpdoc should have context, afaik one can set a visibility on sub-items in cases where this is annoying. Pierre probably also encountered this, and loaded topic titles to disambiguate, but this delays loading the main index noticably (several seconds) on modern hardware. Apparantly the number of topics grew harder than the hardware became faster :-) A different solution would be to simply cache the index. More generic, more powerful, but also more work :-) _______________________________________________ fpc-pascal maillist - fpc-pascal@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-pascal
