home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.amiga.programmer
- Path: sparky!uunet!mcsun!Germany.EU.net!Urmel.Informatik.RWTH-Aachen.DE!dfv.rwth-aachen.de!maho
- From: maho@dfv.rwth-aachen.de (Martin Horneffer)
- Subject: May a DOS-Handler exit? (was: "Re: How to program a DOS: handler?")
- Message-ID: <1992Sep14.183153.26302@dfv.rwth-aachen.de>
- Organization: Communication Networks
- References: <1992Sep13.200005.20531@spuddy.uucp> <35073@cbmvax.commodore.com>
- Date: Mon, 14 Sep 1992 18:31:53 GMT
- Lines: 24
-
- jesup@cbmvax.commodore.com (Randell Jesup) writes:
-
- > This is discussed in the packet docs in the 3rd Ed Bantam manual
- >(though it does have some mistakes, notably in the ACTION_SET_DATE packet and
- >one or two others).
-
- Can I get or find a list of these mistakes somewhere?
-
- >normally exit on the last Close(). Filesystems (that set dol_Task) normally
- >never exit (one day they may have the ability, not now).
-
- Would it be acceptable for a non-filing handler (like pipe:), that sets
- dol_Task (unlike pipe:, I suspect), to exit when it's guaranteed that all
- filehandle have been correctly closed? (The handler would support no
- locks or directory packets anyway.)
-
- I know - in theory it's not allowed if dol_Task ist set.
- But who should (could?) GetDeviceProc() (or so) on such a handler?
-
- Martin
- --
- Martin Horneffer - maho@dfv.rwth-aachen.de;
- Fido: 2:242/7.9; Martin_Horneffer@mowgli.fido.de;
- Maus: AC, AC2; Martin_Horneffer@ac.maus.de
-