On Thursday, 2001-09-27 at 10:25:14 +0200, Roland Giersig wrote: > > Test 10 fails because it exercises the OSes ability to pass long lines > > over a PTY connection. Solaris won't. I stops at 256 bytes. The > > newlineis not passed to the listening program because of the > > overflow.
> Right, this is a peculiarity of Solaris. AIX and Linux have a much > higher threshold (if they have one). Cygwin also has this limitation. Thinking about it, I believe it is in the STREAMS module that does line editing. I.e. it is not a limitation of Pseudo Ttys on Solaris per se. > > Please remove this test. > Hmm, I'm reluctant. Instead I will redefine the test goal: find out > the maximum line length for the given architecture. I'm already > thinking about setting the pty to raw to avoid such system > dependencies. "Portable Ptys" is an oxymoron... Well, the mechanisms already diverge IIRC, but the come the line disciplines and STREAMS modules... Do you mean you will set the tty side to non-canonical? This will remove the limitation on Solaris. Line handling is done by the ldterm STREAMS module. There might be a maximum number of bytes you can send without the tty side reading. Certainly the same for the tty->pty direction, too. See http://docs.sun.com:80/ab2/coll.45.13/STREAMS/@Ab2PageView/21553?Ab2Lang=C&Ab2Enc=iso-8859-1 > > BTW: "The quick brown fox jumped over the lazy dog." is similar to > > a sentence that contains every character at least once. > Right. This sentence was designed to find all those nitpickers out > there, and you're the first! ;-) Great! What do I win? ;-) Lupe Christoph -- | [EMAIL PROTECTED] | http://free.prohosting.com/~lupe | | I have challenged the entire ISO-9000 quality assurance team to a | | Bat-Leth contest on the holodeck. They will not concern us again. | | http://public.logica.com/~stepneys/joke/klingon.htm |
