Re: forwarded message from RUS-CERT, University of Stuttgart

Hrvoje Niksic wrote on 19-April-2000: ->Don't we? I distinctly remember Unix98 PTY patches for XEmacs, some ->time ago. I found this in process.h: I think I remember Neil Becker adding support, but a simple (naive) grep for a helpful comment didn't show anything (apart from a line in configure.in). ->#ifdef HAVE_GETPT ->#define PTY_ITERATION ->#define PTY_OPEN \ -> if ((fd = getpt()) < 0 || grantpt (fd) < 0 || unlockpt (fd) < 0) \ -> return -1; ->#define PTY_NAME_SPRINTF ->#define PTY_TTY_NAME_SPRINTF strcpy (pty_name, ptsname (fd)); ->#endif -> ->> 2. Unsafe creation of temporary files ->> ->> Stating-the-obvious. If FSF Emacs 21 has make-temp-file we could ->> steal it. We do support the TMPDIR variable, ->> though. cf. temp-directory. -> ->The semantics of make-temp-file is unclear to me. Emacs doesn't ->really support a concept of "file descriptors" to which you write. We ->have buffers which may or may not be visiting a file. So what will ->make-temp-file do, exactly? Call mkstemp? Or exec mktemp? Or emulate mkstemp in Lisp? I believe we only need this to provide an API for package writers. Thus if efs, vm, any-new-package, etc need to use a temporary file we can do so "securely". Comments? nic -- Dr N.J.Doye, Systems Programmer, NISS, PO Box 2674, Bath. BA2 7XY