in reply to prompt and timeout in tests

Regarding the -t test; as far as I know it's (at least conceptually, and probably in implementation too) based on the POSIX mechanisms for determining if a stream is bound to a TTY. In other words, if there is an "interactive terminal" bounded to the stream.

That means it should work on POSIX, may work on other systems that have a POSIXy concept of what a TTY is and what streams are, and may not work on other systems. I haven't tested it, but I wouldn't be surprised if it didn't work in windows.

Replies are listed 'Best First'.
Re^2: prompt and timeout in tests
by leocharre (Priest) on Jan 15, 2008 at 23:04 UTC
    Oh heck yeah it works wonderful.. :-) It's really cute.
    Screw M$- I just don't want my code to break on a real operating system. If it has a mental breakdown and kills itself on a M$ machine, who can blame it. Poor little script..