Build awk commands interactively with live result sets.
Make sure live-awk.el is on your load path, then add a hook like so:
(add-hook 'awk-mode-hook (lambda() (require 'live-awk) (live-awk-mode 1)))
Open an awk-mode buffer, and live-awk will prompt you for an input file, this is the file your awk command will be run on.
On editing the awk command, when the buffer changes and doesn't produce anything to stderr, the buffer is updated with the latest output sent to stdout.
It (sort of) works!
- It's messy.
- It requires
headbe installed on the machine.
- The results pop up in an obtrusive window.
- It uses
live-awk-max-linesbecause large data sets would cause emacs to halt.
- The input file <-> awk buffer relationship is one to one.
- Setting it up to use proper emacs processes (
- Possible integration with vlfi for large result sets.
- Cleaning up popup window, maybe using
popwinor at least, read-only buffers.
- Perhaps highlighting in the result set of where the cursor in the
awk-modebuffer corresponds to.