~aleteoryx/employ

dfacbadf28e14a57814c05b758c919b02eefeccb — Aleteoryx 9 days ago 2a99729
readme
1 files changed, 34 insertions(+), 1 deletions(-)

M README.md
M README.md => README.md +34 -1
@@ 1,3 1,36 @@
# this software is NOT safe to use for ANY purpose

## employ

a simple job server. 
a simple job server. it works by shitting shell scripts over the
network. designed to be easy to write shell scripts for, and easy to
work with in any other language, too

the protocol is somewhat documented in /PLAN

### files

- employer.py keeps track of jobs. it takes a database path, port, and
optionally a host to listen on

- employee.py runs jobs. it takes a host and port, and polls the server
for new jobs, reporting a done or failed status based on the exit code
of the server. it's not the most reliable thing atm

- taken.py connects to a server, and outputs all the jobs that are in
progress

- queuers/employ.sh takes a given file and adds it to the job queue,
outputting the job's id

- queuers/reenc.py is what i actually built this for. it provides a way
to queue ffmpeg encoding jobs, with some organization and stuff handled
for you. it takes a host and port, and a directory to re-encode. it is
specifically for tv shows, and other things released in a
season/episode fashion. it recursively finds all media files in the
given directory, and generates a file of metadata. then, it opens the
file for the user to edit. episodes each have a season, episode, and
title. once you've corrected the info in the file, re-run the script
with the same args, and it'll queue jobs for every resolution in the
RESOLUTIONS array. you'll probably want to change the actual encoding
instructions if you use this, mine is for low-power devices