Hi Igor, > Minimal changes are always good idea. > However, we usually do not add new methods unless we actually use them. > Not sure what's the best approach in this case.
Maybe we should give you the FontManager work first (it's not finished yet, I have to test and fix on Windows first - this is aweful and takes a lot of time to get going...), then I could also implement some more alternative code paths on top this fonts-from-bytebuffer patch, as suggested. /Roman > > -igor > > Roman Kennke wrote: > >> Am i understanding this correctly and this adds new TrueType constructor > >> that is not used in the OpenJDK itself? > >> > > > > For now, yes. (I'm using this in my own FontManager implementation in > > the JamaicaVM, to load fonts from ROM instead of files. But this > > requires the FontManager refactoring in Caciocavallo too...) I will most > > likely add more alternative code paths on top of it later (for example, > > for Font.createFont(InputStream) to not create a temporary file, or > > generally for creating fonts using mmapped files instead of RAF, > > possibly activated by a system property or so). I was thinking that it > > would be easier to review if I send a minimal patch for the start. > > > > /Roman > > > > > >> -igor > >> > >> Roman Kennke wrote: > >> > >>> Hi there, > >>> > >>> What do you think about this patch (see webrev link below)? Can we get > >>> this into jdk7? It is very useful for me and I wouldn't like to maintain > >>> it myself... > >>> > >>> Cheers, Roman > >>> > >>> > >>> > >>>> http://kennke.org/~roman/truetypefrombuffer/webrev/ > >>>> > >>>> > >>> > >>> > -- http://kennke.org/blog/
signature.asc
Description: Dies ist ein digital signierter Nachrichtenteil
