pyre-bot / pyre-bot Goto Github PK

Describe the bug
When using on_command_error, exceptions that aren't specifically called don't output.

To Reproduce
Break something

Expected behavior
Show me exceptions

With the new use of JSON files for data storing, the setup script should create these files on first launch

Documentation needs updated to match current release

Before regular restart, I want Discord message like
'Server is restarting for maintenance. It takes less than 3 minutes'.

(My)Discord Members can understand that The server will come back soon.

Describe the bug
Stat tracking is reading wrong values and putting negatives into Discord

Pretty straightforward, just need to change the way the chat string is pulled from the log to prevent cutting off anything with regex.

Since it doesn't verify the existence of the item or equip, it passes along a command which features (none) items or equips if given an argument that doesn't match anything in the game. The solution for this is to add dictionaries for items and equips that can be referenced. Keep in mind that DebugToolkit, which is the plugin that handles the give_item and give_equip commands, has autocomplete, i.e. you can say the item name is "shield" and it will know you meant to give the "PersonalShield" item and carry that out.

At the very least, the bot could wait for the server output, because if it doesn't find the item or equip it will give this as part of the output:
[Info : Unity Log] The requested object could not be found:

We could set that up the same way we set up the chat command to inform the sender on Discord that the item / equip is not found and to try again, and not passing the command to botcmd.txt.

Also add a dictionary for stages so that their "real" names can be referenced rather than their dev names (see: https://github.com/risk-of-thunder/R2Wiki/wiki/List-of-scene-names)

Since Discord updated their bot API to expect '/' as a prefix, this is something that requires an update

Is your feature request related to a problem? Please describe.
Running multiple bot instances is excessive

Describe the solution you'd like
Single bot for several servers

Describe alternatives you've considered
Using the current channel dictionary, determine which bot is being used

This branch is for developing a singular version of the bot, one that only requires one instance to control multiple servers. This requires the bot to read from server log files in a central directory for output to discord, as well as for the servers to read commands to ingest. Previously, each bot had its own installation, and had to be located on the same machine as the server, in order to read and write to the necessary files.

As of now, we have seq working to ingest logs from all servers simultaneously, filter out irrelevant events, and forward the resulting events to logs stored in centralized directory that can be (theoretically) read by one bot instance. I want to explore putting additional tags to the logs that signify to the bot which server wrote it, rather than the file name prefix notation I'm using now (which requires 1 log file per server).

The major challenge from this point is how to ingest commands to the game servers using our database system. I know that taking commands from discord and outputting them to another service with the Python bot is easy, and I just need to move that to its own feed, but how we go about having the server read from that feed is what we have to figure out now.

Access the player.duration variable in players.py, any time a player connects / disconnects (detected by the server log output already), add the duration of their time spent to a log that is tracked across sessions and can be output to the discord.

In the future, be able to assign Steam players to their Discord IDs so they can track their own stats. Can be expanded to things like runs cleared, times joined, etc.

Possible issue I forsee: May not be able to retrieve player.duration if the player just disconnected. Will have to test this out. If this is a problem, you could just compare the real time at which they joined to the real time at which they left to get their stay duration, instead of using player.duration.

May involve moving stat tracking to C# plugin

Describe the bug
When using the reload commands, autochat gets duplicated

To Reproduce
Steps to reproduce the behavior:

1. Use >reload on the bot

Expected behavior
The bot will only output the chat once.

Tentative solution: Execute fixed_time on the server at a fixed interval (every few seconds or so) and reference the last value if this situation occurs.

Create an admin command which signals to the bot that next time the respective server hits the lobby, pass the "disconnect" cmd to the game server to shut it down and ping the admin user who used this command. I want this because I'm constantly checking to see which servers are in lobby when I want to shut down a server to update a plugin or something else. This would make it so I don't have to monitor them manually.

Is your feature request related to a problem? Please describe.

• The setup script sources an old version of BotCommands, please have it grab the latest at any given time
• Allow setup script to read server configuration to set port automatically
• Allow setup script to read IP to setup server address automatically
• Rename server_config to match server.cfg

Change stat tracking to disabled by default

Before this is completed the following must be done:

• Disable stat tracking by default
• Create config entries to enable or disable
• Update documentation to reflect stat tracking
• Update setup script to configure AWS credentials and create tables to ease setup

Is your feature request related to a problem? Please describe.
Create update command for admins that actually updates the server.

Describe the bug
Server is constantly outputting "Unable to remove log file" when it is trying to auto-restart

Screenshots

Is your feature request related to a problem? Please describe.
Allow bot and relevant cogs to auto-update