Sometimes you need to do a tail -f, grep the results, and then nuke the file and start all over again... wouldn't it be nice if you could just run a single command to do it, and optionally have it serve the output up on a port?
Hotgrep does just that! And the file can be deleted, moved, recreated, truncated... all during a single hotGrep session.
Example command line:
hotgrep /that/file/over.there '.+' 3
Which will tail -f the file, returning all new lines, and checking every 3 seconds for new data.
Argument 1 is the filename.
Argument 2 is the regexp pat applied to lines, only lines that match are returned (deafult is .+, which matches all lines).
Argument 3 is the delay in seconds between updates (if not provided defaults to 2 seconds).
Argument 4, if provided, is the port to run the server on.
UNIX only.
http://inferno.slug.org/wiki/HotGrep
The recipe [L1 ] in the Book ActiveState Tcl Cookbook shows how to emulate "tail -f" in Tcl. It achieves platform independence by doing as little as possible. Combining that with hotGrep would provide a platform-independent solution.