OP25 Boatbod API?

[xtraorange]

Hello - I'm working on a fairly complex project which would benefit from an API for OP25. I'm just wondering if one exists before I build one? Also, if I build an API, is there a way to pass terminal commands to OP25 (boatbod) from a separate app?

Thanks!

 

[boatbod]

Hello - I'm working on a fairly complex project which would benefit from an API for OP25. I'm just wondering if one exists before I build one? Also, if I build an API, is there a way to pass terminal commands to OP25 (boatbod) from a separate app?

Thanks!

Ahhhh, well... maybe... but bear in mind the only documentation is the code itself.

Really the practical way of achieving your goal is to configure op25 to use the web terminal and then send it json formatted HTTP POST messages. You do need to be aware that there are differences in the web interface for rx.py and multi_rx.py versions of the application. Please do NOT build anything around the legacy rx.py version of the app because I would really like that to go die a quiet death in a corner somewhere.

If you find that there is something you really need changed in the existing terminal interface, just reach out via a DM or email and we can discuss further.

 

[xtraorange]

Ahhhh, well... maybe... but bear in mind the only documentation is the code itself.

Really the practical way of achieving your goal is to configure op25 to use the web terminal and then send it json formatted HTTP POST messages. You do need to be aware that there are differences in the web interface for rx.py and multi_rx.py versions of the application. Please do NOT build anything around the legacy rx.py version of the app because I would really like that to go die a quiet death in a corner somewhere.

If you find that there is something you really need changed in the existing terminal interface, just reach out via a DM or email and we can discuss further.

Thanks, that's helpful info. I've been building it around rx.py, so I will switch that over.
So I'm trying to essentially pull all the info out of OP25 and also create functionality to control it. I'm wondering if perhaps there's a better way.
Right now, to get information about the current talk group, etc, I'm reading logs and using SSE to push it to my server since I didn't see an obvious way to hook into the webui. Do you have a better recommendation?

 

[boatbod]

Thanks, that's helpful info. I've been building it around rx.py, so I will switch that over.
So I'm trying to essentially pull all the info out of OP25 and also create functionality to control it. I'm wondering if perhaps there's a better way.
Right now, to get information about the current talk group, etc, I'm reading logs and using SSE to push it to my server since I didn't see an obvious way to hook into the webui. Do you have a better recommendation?

Operational information is requested by the terminal using an "update" query, and it results in a "channel_update" message being returned. If you explore the javascript file main.js (found in ~/op25/op25/gr-op25_repeater/www/www-static) you can add some console logging to dump the contents of the response to see if it has the info you need.

Right now, no log information is passed to the terminal. I wish it were, but for that to happen I have to make major changes to how op25 logs stuff. It's on my todo list, but hasn't happened yet.

 

[xtraorange]

The main reason I was using the terminal output (just upping verbosity to 3) was to avoid polling the server. I can react to a new console message and push it out with an SSE. However what I can capture from that is pretty limited, and I don't get any solid clues on when a transmission completes. Any suggestions on how I might accomplish something similar from the webui? Or is polling really the only option?

I was trying to avoid modifying the source code since that makes updates a pain and makes my package much less useful to others.

 

[xtraorange]

Sorry, I feel like it seems like I'm trying to by mysterious about what I'm doing: I'm not, it just got so unreasonably complicated because of my overthinking I was unsure if it was worth explaining. Let me give some context:
I've built 3 tools. The first tool is a wrapper around op25 - it executes it, and currently captures terminal output looking for particular messages and forwarding them via SSE to subscribers.
The second tool is a janus proxy, which allows me to keep my janus server in my lan and expose a reduced attack surface for Janus. I use cloudflared and nginx to provide access to the proxy. This also receives authorization tokens from my third project so a user must be authorized to get a feed.
The third tool (the main one) is a PHP Laravel based (it's just what I'm most familiar with) webserver. It sort of draws everything together, providing a player and authentication for the janus feed, and receiving status updates from the op25 wrapper. It will also call the op25 wrapper's API to execute actions on the server. The idea is that this will be built out with several features for interacting with OP25, managing users and permissions, etc.
I considered putting the third tool on the same server as op25, but I really wanted to keep my op25 and janus servers tucked inside my local lan so that the attack surface could be narrowed.



To clarify, the reason for all this was I wanted to create a way to expose my op25 feed and controls over the web, and keep the latency very low. I'm at a 4-500ms delay right now, which works for my purposes.
Hopefully that all makes sense... I know it's unnecessarily complicated, but I guess so am I.

 

[boatbod]

The op25 terminal - both curses and http versions - are polling based by design. If you don't have a terminal instance running there is nothing polling the op25 server and hence less overhead on a headless system.

From your description it sounds like you just need to implement a more secure terminal, so the existing http api should be fine as long as you disallow external 'unfriendly' entities to send http commands directly to the op25 server.