This started as a collection of scripts to Bridge Orcaslicer with Elegoo Centauri Carbon.

But as I added more and more stuff, I eventually rewrote the whole thing. What was once a single daemon.py file. Has turned into something much more.

Not everything is perfect. It works mostly for my own setup. BUT this can work for anyone else as long as the config file is filled properly.

I am also attempting to make this work within a docker container to allow for a persistent "always on" type mode. That is more of a WIP, so default, we will be running this more as a one and done script whenever you click on "Print" in Orcaslicer.

To accomplish this, included is another file "copy_to_watch.py". ATM, this file needs to be edited to your needs. But the purpose of this file is to act as a intermediatory between the daemon and Orca for the purpose of exporting / saving the GCODE to a watch folder of your choice that the daemon can read. Because, while print sends the GCODE to the printer, we don't have the file. YET!

(as I have been looking throught the sdcp api, there are ways I can list / delete file on the printer. I may work out a way to perhpas copy the currently printing GCODE across to a "watch" folder, to then extract the information required for the daemon. EG filament color, gram usage estimate. Perhaps a work flow, instead of the watch_folder, could be: 1. Detect Print. 2. Copy GCODE / Read GCODE of current print. 3. Extract info from GCODE to variables. 4. When print is complete, instead of deleting file from watch_folder, we can instead have the option to delete it from the printer.

This would remove all need of a copy_to_watch.py at all, it would also remove the need of post processing in orca as well. Which would make setup easier.) Forgive my rambling, my brainstorming is just typing my thoughts. This doesn't belong in a README, so I will transfer this into issues and features when I have time.

Why not include it within the daemon? I would like to, and to be honest, I am not sure why it isn't. There was some reason I had to have it seperate. Ideally I'd like it to be all in one. Perhaps, it was because I was trying to make this universal between network mounts and stuff.

Now, why have I made this?

I like the idea of having an inventory of the spool I have and what I have left, even if its just a decent estimate. Unfortunately the Centauri Carbon isn't able to integrate with spoolman directly, and the newer versions of orcaslicer has no way to link with spoolman either.

Sure, we could input all the data manually if we wanted. But I like when things happen automatically!

You can run it in 2 ways:

1. Always Running Mode:

  you can run this script in the background as a service on Linux or windows by calling either run.bat or run.sh.

  This will allow it to be monitoring the defined watch folder for new gcode entries, and then when a print job matching that gcode is started, it will subtract the filament used

  to a matching spool in spoolman.

2. One Time Mode:

  If you'd prefer, this can be run once when you click print in orcaslicer. Only modification required is to add a post-process script entry to copy_to_watch.py where you should save

  locally. EG in my orcaslicer dir, I created a folder "Scripts" and put the copy_to_watch.py file there.

  This will instead run the daemon when you click print and will self close the daemon when the print is finished. You can see when the filament has been subtracted via the console,

  so if you choose, you can close the terminal window when you see that.

When exporting g-code, it will also run the daemon. So if you aren't planning on printing, just close the daemon.

*/ Installation /*

Just extract all files to a place you want to run the daemon from. And then place the copy_to_watch.py file into a location of your choice that orcaslicer can access.

Then you must setup post-process scripts within orcaslicer, pointing to this python file.

***NOTE*** You will need to have python installed to run the initial script! But the daemon will install its required packages into a venv via the requirements.txt file included.

You must open config_example.yaml and enter all fields to ensure that the daemon can access all required parameters. Save it as config.yaml for the daemon to work.

  1. "sdcp_ws_url": "ws://:3030/websocket?command=subscribe"

 

  This is the access point to the printer for the daemon, it will subscribe and listen to Topic statuses. EG Printing / Idle / Paused.

  I have only tried this with OPEN CENTAURI and not stock firmware.

  2. "watch_folder": ""

  This is the folder that will contain a copy of the GCODE that is sent to the printer. This is so the daemon can open and parse for filament information.

 

  3. "spoolman_local_url": "http://localhost:7912"

  Rare that you would need to change this as it is a fall back point if it is unreachable at spoolman_url. If both are unreachable, than spoolman isn't setup currently.

  4. "spoolman_url": "http://<SPOOLMAN_IP>:7912"

 

  IP to spoolman service / container.

  5. "delete_after_print": true

  Great for 1 time prints. Work flow is Click Print in Orca --> File is created in Watch Folder --> Metadata is parsed --> Printer starts printing matching file -->

  daemon uses filament info to call spoolman api and subtract filament from spool --> deletes gcode file.

  But if you set this to false, the gcode will stay in the folder. This can be useful if you run the script on always on and you have a file on the printer you print often

  from the printers interface. Then this should still work and subtract the filament as it will still match the gcode files.

  6. "always_running": true

  keeps the websocket alive continues to watch the folder for new gcodes to parse. This is great when you want it to just run in the background and just send print jobs when needed. I am even considering creating a companion docker container that you could spin up in a stack with spoolman so it just runs in its own.

  But you can also have this as false if you'd prefer to spin it up on demand.

  7. "hide_one_time_mode_terminal": false

  This is to work with always running set to true. When this is true, when you print the file it will hide the terminal and run it in the background. Only set this

  if you are determined to declutter you PC. I prefer to have the console shown as then I know for sure the process closes / ends when the print job is finished

  and if not, I can just close the terminal window.

*/ Additional Setup /*

You need to name your filament in Orcaslicer a certain way for this to work.

- - == ELEGOO - PLA - Black

as long as you name the filament preset like this, then the parser can identify and then match your spools in spoolman. If you create any of your own custom spools, you need to copy how all the external filaments are named to keep consistency.

I decided on this format so that you can expand your spool inventory across many vendors, types and colors. And so that the daemon won't subtract the wrong filament from the wrong spool.

*/ Troubleshooting /*

If you are going to utilize the hide_one_time_terminal, you should know this line in powershell to help you identify id's to stop if for some reason something is stuck.

Get-CimInstance Win32\_Process -Filter "Name = 'python.exe'" | Select-Object ProcessId, CommandLine

This will display the process running and their script names. EG /path/to/daemon.py. If needed, you can force close the daemon by:

Stop-Process -Id <ID>

/ Testing /

The project includes automated unit tests to ensure everything works correctly.

Quick Test:

./test.sh

Manual Commands:

# Run all tests
pytest

# Run with verbose output
pytest -v

# Run specific test file
pytest spooler/tests/test_gcode.py

# Run with coverage report
pytest --cov=spooler --cov-report=html

Test Files:

• spooler/tests/test_utils.py - Unit conversion tests
• spooler/tests/test_gcode.py - G-code parsing tests
• spooler/tests/test_spoolman.py - Spool matching tests
• spooler/tests/test_sdcp.py - SDCP message parsing tests
• spooler/tests/test_watchers.py - File watching tests

See TESTING.md for complete testing documentation.

3. Web GUI Mode:

 \tRun run_webgui.sh (Linux) or run_webgui.bat (Windows) from the spooler/ folder to launch a browser-based setup wizard and YAML config editor.

 \tThe GUI runs on http://localhost:8949 and edits config/config.yaml.

 \tOptional: pass daemon host (and optional port) directly to target a remote daemon forwarder.

# Linux
./run_webgui.sh 192.168.1.102
./run_webgui.sh 192.168.1.102 8765

REM Windows
run_webgui.bat 192.168.1.102
run_webgui.bat 192.168.1.102 8765

4. Split Host Mode (WebGUI and Daemon on Different Machines):

 \tIf WebGUI runs on one machine and daemon runs on another, set the daemon forwarder bind host and the WebGUI forwarder target explicitly.

 \tOn the daemon host:

export SPOOLER_FORWARDER_HOST=0.0.0.0
export SPOOLER_FORWARDER_PORT=8765
cd spooler && ./run.sh

REM Windows daemon host (same effect)
run.bat 0.0.0.0 8765

 \tOn the WebGUI host:

export SPOOLER_FORWARDER_URL=ws://DAEMON_HOST_IP:8765
cd spooler && ./run_webgui.sh

 \tMake sure TCP port 8765 is open between hosts.

REM Windows firewall example (run as Administrator)
netsh advfirewall firewall add rule name="Spooler Forwarder 8765" dir=in action=allow protocol=TCP localport=8765