Steve Crossman wrote: > Furthermore, there are very many new toolbox calls that don't >accept FSSpecs. And there are some that don't accept FSRefs too -------------- H Not that I'm aware of _new_ ones! Or do we really face another transition to file path specifications that have also been annopunced a while ago in an Apple-TechNote? (I didn't hear much of this possible transition during the past two years, but it is standard e.g. in Java.) -------------- H > 4. > Last but not least, Apple's support of FSSpecs will decline rapidly, > because it simply isn't compatible with CF, as has been pointed out > already in a earlier contribution to this thread. Can I continue to use FSSpecs? Yes, they continue to be valid file references. emphasis mine, from Apple's docs you cite in reference. -------------- H The very article is from 2003! But you can still use FSSpecs, but I don't know of _new_ toolbox routines that accept them. -------------- H > 2. Another important issue is Unicode filenames. How about the handling of Unicode text editing in FB? It is sub-standard too, but I don't see anyone changing this anytime soon. -------------- H Well, this is another point of severe FB-critique. If there wasn't RP, we wouldn't even be able to deal with CFStrings... (And I don't know how one could avoid them when coding for OSX.) I'd say, not switching to CFStrings is substandard for the same reasons that not switching to FSRefs is substandard. Peu-a-peu you get problems with using modern toolbox calls. It starts with increasing overhead for converting FSSpecs to FSRefs and Pascal strings to CFStrings and ends with not being able to use new techniques introduced by Apple. How can you use CFPreferences without CFStrings and its CF-brothers and sisters? OK why should you use CFPreferences? This question has recently been discussed on this list. The short answer is: You need not -- yet. -------------- H Again, besides the file naming issues, where is the prolific problem with vrefs was my question? -------------- H I never used the term "prolific". I only tried to tell you that you will encounter increasing difficulties when using vrefs and FSSpecs for OSX. What's wrong with FSRefs that you don't consider them as part of the more modern way of coding for the Mac? -------------- H It almost sounds like the average FB coder cannot use them while migrating older generations of FB code to OS X, with out having a lot of problems. -------------- H Indeed this impression is not perfectly wrong. FSSpec is obsolete and the fact that you and others can still use them under OSX is mainly due to lucky circumstances. There are things you can't do with FSSpec in OSX but perhaps you don't need to do them -- yet. -------------- H This is particularly valid when it comes to Rosetta, which was another key aspect of all of this, not the undetermined future of FB and the need to rid ourself of deprecated code. -------------- H As I've said, Rosetta has little to do with the FSSpec/FSRef-issue. In any case, it won't help you with problems that you might encounter sooner or later if you stick with FSSpecs. -------------- H Even Microsoft Office continues to have file naming issues. -------------- H Well, if you accept MS as a model for good Mac coding... -------------- H However, I am in control of file naming for 99.5% of the functions in my program and only exported files suffer from vref-itis of its file naming limits. -------------- H Do what you think is best for you and your customers. -------------- H I don't see the things you mention as being prolific [ the key word in this discussion in my mind ] not the validity or negation of using FSRefs or even FSSpecs over vrefs. I am not arguing that point. -------------- H I never used the term "prolific". FSSpec is simply obsolete, that's all. -------------- H But i digress and need to get back to work. Steve Crossman ---------------------- Best to you and happy coding -- Herbie ------------------------ <http://www.gluender.de>