Re: tee wishlist: --immediate option?
Jim Meyering <[email protected]> Sat, 20 Dec 2003 10:49:56 +0100
| Newsgroups | gmane.comp.gnu.sh-utils.bugs |
|---|---|
| Message-ID | <[email protected]> |
"Steven G. Johnson" <[email protected]> wrote: > This is just a wishlist item of mine that I thought I would pass on for > your consideration. It would be nice if tee had an option like > --immediate that would flush its output immediately instead of writing it > in large chunks (8k, I think). > > The reason is that I sometimes like to use tee to save the output of slow > numerical computations, while still viewing the output in the terminal to > monitor the progress. This doesn't work very well, though, because the > of tee's buffering--I may not get any output at all until the program > finishes. > > In other words, it would be nice if tee's output could come at (at least) > the same rate as the output of the program without tee. Thanks for the report. tee (at least in coreutils) already arranges not to do buffering on standard output, as POSIX requires. Since you're writing to the semi-obsolete bug-sh-utils address, I suspect you're using an older version of tee. [email protected] is the preferred bug-reporting address. Can you reproduce your problem with the tee from one of these newer packages? STABLE ftp://ftp.gnu.org/gnu/coreutils/coreutils-5.0.tar.gz ftp://ftp.gnu.org/gnu/coreutils/coreutils-5.0.tar.bz2 (coreutils is the union of fileutils, textutils, and sh-utils) BETA ftp://alpha.gnu.org/gnu/coreutils/coreutils-5.0.91.tar.gz ftp://alpha.gnu.org/gnu/coreutils/coreutils-5.0.91.tar.bz2 If so, and if you can describe your system and how to reproduce the problem well enough that I can duplicate the failure, then we should be able to fix it.