You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
After updating my fork with the current state of the new branch, I realized the docker composition I added about a year ago no longer ran properly when following the instructions from the Docker section of the readme. Reason being, it appears the dotenv credentials loader I originally added to the echo_bot example has been replaced with a different yaml based loader pointed at examples/creds.yaml.
Notes:
After moving my creds from the previous .env file to examples/creds.yaml and slightly tweaking the Dockerfile, the example echo_bot works fine again when running via docker compose. No biggie.
It appears examples/creds.yaml is a VCS tracked file (not .gitignore'd), meaning that consumers of this API can easily make the mistake of checking in their credentials to source control for the world to see if they add their creds to the file and commit all local changes. This is why I had an .example.env file and the copy command in the readme as well as a dotenv .gitignore rule in my original docker PR.
I'm opening this issue just to gauge interest / opinions / preferences around standardizing the configuration of environment variables and settings such as client credentials for this project going forward. I personally don't have a strong preference as either approach seems to work, but certainly the readme will need to be updated and we should encourage consumers of the API to create a non-VCS tracked (.gitignore'd) copy of examples/creds.yaml if it's decided to stick with yaml based configuration.
I don't mind opening a PR containing suggested changes if yaml based configuration is chosen as the standard. If you'd like to tinker with this yourself using the echo_bot example or if you're just here trying to get your docker composition working again:
Add your creds to examples/creds.yaml (DO NOT COMMIT THESE CHANGES)
After updating my fork with the current state of the
new
branch, I realized the docker composition I added about a year ago no longer ran properly when following the instructions from the Docker section of the readme. Reason being, it appears the dotenv credentials loader I originally added to the echo_bot example has been replaced with a different yaml based loader pointed atexamples/creds.yaml
.Notes:
.env
file toexamples/creds.yaml
and slightly tweaking theDockerfile
, the example echo_bot works fine again when running via docker compose. No biggie.examples/creds.yaml
is a VCS tracked file (not .gitignore'd), meaning that consumers of this API can easily make the mistake of checking in their credentials to source control for the world to see if they add their creds to the file and commit all local changes. This is why I had an.example.env
file and the copy command in the readme as well as a dotenv .gitignore rule in my original docker PR.I'm opening this issue just to gauge interest / opinions / preferences around standardizing the configuration of environment variables and settings such as client credentials for this project going forward. I personally don't have a strong preference as either approach seems to work, but certainly the readme will need to be updated and we should encourage consumers of the API to create a non-VCS tracked (.gitignore'd) copy of
examples/creds.yaml
if it's decided to stick with yaml based configuration.I don't mind opening a PR containing suggested changes if yaml based configuration is chosen as the standard. If you'd like to tinker with this yourself using the echo_bot example or if you're just here trying to get your docker composition working again:
examples/creds.yaml
(DO NOT COMMIT THESE CHANGES)Dockerfile
:docker compose up --build
from the root of the repositoryThe text was updated successfully, but these errors were encountered: