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
I think it should be like webpack-dev-server; for most scenarios, a self-signed certificate should suffice.
Moreover, I believe that mkcert presents a certain level of initial learning curve for developers' local environments. Therefore, I think integrating a self-signed certificate might also be a better approach.
What does the proposed API look like?
The external API remains as usual, except that the server.https configuration supports a boolean.
When set to true or when cert and key do not exist, it will automatically create and use a self-signed certificate.
Of course, self-signed certificates indeed have issues with browser trust, but I think the framework leans more towards being usable out of the box. We can still recommend users to use mkcert, but internally, there should be a fallback solution.
The text was updated successfully, but these errors were encountered:
I might not quite agree with the view that it should be made into a separate plugin.
First of all, npm packages like selfsigned already have the capability to accomplish similar functions, so creating a separate plugin might not offer significant benefits.
Moreover, for rsbuild, this seems to be only related to the dev-server, hence I believe it should simply be implemented as a util function within the dev-server itself.
What problem does this feature solve?
I think it should be like webpack-dev-server; for most scenarios, a self-signed certificate should suffice.
Moreover, I believe that mkcert presents a certain level of initial learning curve for developers' local environments. Therefore, I think integrating a self-signed certificate might also be a better approach.
What does the proposed API look like?
The external API remains as usual, except that the server.https configuration supports a boolean.
When set to true or when cert and key do not exist, it will automatically create and use a self-signed certificate.
Of course, self-signed certificates indeed have issues with browser trust, but I think the framework leans more towards being usable out of the box. We can still recommend users to use mkcert, but internally, there should be a fallback solution.
The text was updated successfully, but these errors were encountered: