On Sun, Oct 24, 2010 at 08:38:21PM +0200, silverdr@wfmh.org.pl wrote: > > If the library uses a certain lfn, it cannot prevent the user from using the > > same one in his own code. > > How about "forgetting" the numbers? I mean if we use POSIX style I/O, we > shouldn't care about this. You are right, as long as you just used POSIX file I/O, you don't have to care. But the question was caused by mixing CBM file I/O with POSIX file I/O. > Some time ago - I wrote a short code that scans the file table and returns > available "filehandle", unless the limit of ten concurrent files is reached. [...] > Maybe something like that could go into lib? I should still have it somewhere > but if memory serves, it was quite trivial. As always it's balancing the advantages and disadvantages of the possible solutions. Yours works and is not difficult to implement, but it is larger, and platform dependent, because there is no "official" access to the file handle table of the kernal. Given the case in question, it would offer an advantage, but in my eyes the better solution is to make CBM file I/O unnecessary (by adding POSIX directory routines). Regards Uz -- Ullrich von Bassewitz uz@musoftware.de ---------------------------------------------------------------------- To unsubscribe from the list send mail to majordomo@musoftware.de with the string "unsubscribe cc65" in the body(!) of the mail.Received on Sun Oct 24 20:48:48 2010
This archive was generated by hypermail 2.1.8 : 2010-10-24 20:48:50 CEST