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
It would be helpful to have automated testing that Puter can build and run correctly. This can help prevent issues appearing in setups that the dev team don't use. (eg, Windows and Docker.)
The general process would be something like:
Get a copy of the repo
npm ci, and check for errors
npm start, and check for errors
Access the hosted Puter website
This could start out as a simple curl call to check the response code, but ideally we'd run some basic tests using something like selenium-webdriver to interact with the page.
End the puter process
Ideally there would also be a variation of this that creates a Docker container from the repo and runs that - we get some issues when running in Docker that don't occur otherwise.
At a basic level, this would ensure that Puter at least can be installed and run, on whichever OSes we configure the CI task for. If we want to get fancy though, we could test that individual parts of the UI, like signing up, logging in, and creating files all work as expected, and across different web browsers.
The text was updated successfully, but these errors were encountered:
It would be helpful to have automated testing that Puter can build and run correctly. This can help prevent issues appearing in setups that the dev team don't use. (eg, Windows and Docker.)
The general process would be something like:
npm ci
, and check for errorsnpm start
, and check for errorscurl
call to check the response code, but ideally we'd run some basic tests using something likeselenium-webdriver
to interact with the page.Ideally there would also be a variation of this that creates a Docker container from the repo and runs that - we get some issues when running in Docker that don't occur otherwise.
At a basic level, this would ensure that Puter at least can be installed and run, on whichever OSes we configure the CI task for. If we want to get fancy though, we could test that individual parts of the UI, like signing up, logging in, and creating files all work as expected, and across different web browsers.
The text was updated successfully, but these errors were encountered: