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
The worker should support adaptor versions @local, as in, @openfn/language-common@local, in which case it should load the adaptor version from the monorepo rather than trying to install it.
The CLI works the same way when the -m or --use-monorepo flag is passed. It just bypasses autoinstallation.
In practice I think all this really means is passing the monorepo flag and path into the compiler and runtime - which are already monorepo aware.
The text was updated successfully, but these errors were encountered:
There's a bit of a problem in the runtime associated with this
In the CLI, if you're running from the monorepo, then EVERYTHING runs from the monorepo
I'm not sure Lightning should work like that. It should only use the monorepo if you add @local explicitly.
Where this is a problem is that the current runtime design says: here's a map of module specifies and paths to load them from on disk. It's a global option.
What we really need to do is support this logic per step. Which means a runtime patch and a bunch of tests
The worker should support adaptor versions
@local
, as in,@openfn/language-common@local
, in which case it should load the adaptor version from the monorepo rather than trying to install it.The CLI works the same way when the
-m
or--use-monorepo
flag is passed. It just bypasses autoinstallation.In practice I think all this really means is passing the monorepo flag and path into the compiler and runtime - which are already monorepo aware.
The text was updated successfully, but these errors were encountered: