Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!mstar!mstar.morningstar.com!bob From: bob@MorningStar.Com (Bob Sutterfield) Newsgroups: comp.sys.pyramid Subject: Re: read from /dev/tty problems Message-ID: Date: 24 Jul 90 21:37:04 GMT References: <64234@yarra.oz.au> Sender: usenet@MorningStar.COM (USENET Administrator) Reply-To: bob@MorningStar.Com (Bob Sutterfield) Organization: Morning Star Technologies Lines: 30 In-Reply-To: karl_kleinpaste@cis.ohio-state.edu's message of 24 Jul 90 13:42:56 GMT In article karl_kleinpaste@cis.ohio-state.edu writes: kenj@yarra.oz.au (Ken McDonell) writes: Could you please provide some more information on the version of OSx in which you see the problem, and the precise conditions under which you can make it hapen? We have the same problem under OSx 4.anything (currently 4.4c) and under both X10 and X11. Just start an xterm, any xterm, and you'll find that /dev/tty is (evidently) not attached to it. The quickest test is to say `cat < /dev/tty` to a shell running in an xterm. If it returns immediately (as is the case in xterms on Pyramids running the MIT X distributions), you've found the bug. If it waits for tty input (as is the case on lots of other things running MIT X), you're clean. The presence or absence of the "-ls" switch seems to make no difference. The X server host type seems to make no difference. I read news regularly, but have not seen you comments previously (is it possible they are not escaping for worldwide distribution?) This has been discussed here occasionally -- I've mentioned it once or twice myself. This has happened at least since OSx 4.0 (perhaps 4.0beta, but I'm not positive I tried it then) and X10R3, and has been occasionally reported in comp.sys.pyramid and comp.windows.x for several years. I've got no particular beef with Pyramid on this one, since MIT X is not their product. It's just monotonous to explain it to the users...