aboutsummaryrefslogtreecommitdiffstats
path: root/README.md
diff options
context:
space:
mode:
Diffstat (limited to 'README.md')
-rw-r--r--README.md38
1 files changed, 25 insertions, 13 deletions
diff --git a/README.md b/README.md
index 2848e26..a17d9ce 100644
--- a/README.md
+++ b/README.md
@@ -6,7 +6,19 @@ Currently the only way to run Sound Chan is to build the bot from source. As suc
## Instructions For Use
-To run the sound bot on your Server, you'll need a developer's bot token from Discord. Full instructions for use will be coming, but in the meantime for development purposes, place your token and the corresponding settings in a file named, `soundchan.properties`. You can see an example of how the file should be formatted in `soundchan.properties.example`.
+To run the sound bot on your Server, you'll need a developer's bot token from Discord.
+As the exact way to get a token may change over time, checkout the [Discord documentation](https://discordapp.com/developers/docs/reference).
+Once you get a token, place your token and the corresponding settings in a file named, `soundchan.properties`.
+You can see an example of how the file should be formatted in `soundchan.properties.example`.
+
+In your `soundchan.properties`, you can specify all sorts of settings that SoundChan will read on startup.
+An example of a setting you may want to set is the directory to sounds you want SoundChan to play as an interjection.
+Other things you can set is if SoundChan should watch this directory for changes so you can add and remove files on the fly.
+See `soundchan.properties.example` for a full list of currently adjustable things.
+
+Another file you can optionally add is `usersounds.properties`.
+Here you can specify a sound from the sound directory (set in `soundchan.properties`) that will play when a user joins the voice channel SoundChan is currently in.
+See `usersounds.properties.example` for more information about how to set this up.
## Requirements to Run
@@ -23,15 +35,15 @@ More information to come.
## Todo
-* Move the ExtendedListenerAdapter outside of main
-* Add a monitor for updated files (refresh every few minutes or on command?)
-* Add pre-emption support for sound effects (without disturbing the queue)
-* Support seeking (starting a track at any point, or skipping ahead in the track)
-* User roles
-* Help command
-* Document run steps properly
-* Package a batch file / bash file for easy use
-* List currently playing song
-* Like or dislike support for currently playing track
-* Have SoundChan reply with a voice bot
-* Turn features on and off using config files
+- [x] Move the ExtendedListenerAdapter outside of main
+- [x] Add a monitor for updated files (refresh every few minutes)
+- [x] Add pre-emption support for sound effects (without disturbing the queue)
+- [ ] Support seeking (starting a track at any point, or skipping ahead in the track)
+- [ ] User roles
+- [x] Help command
+- [ ] Document run steps properly
+- [ ] Package a batch file / bash file for easy use
+- [x] List currently playing song
+- [ ] Like or dislike support for currently playing track
+- [ ] Have SoundChan reply with a voice bot
+- [x] Turn features on and off using config files