-
Notifications
You must be signed in to change notification settings - Fork 209
wip system deploys #3021
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
wip system deploys #3021
Conversation
|
a thing I fiddled with in 5c7ba76 is a "pure" representation of a given system and all its dependent bytecode, where references to public libs in the bytecode are replaced by the library bytecode that the deployer would "unwind" and deploy deepest first, swap with address, then continue outward until the system is deployed this is ~double the filesize as the current state where we now reference "deployables" and have a top-level "deployables" object with bytecode of libraries even though it's larger, I still like the former because it's more "pure" and easier to merge without naming conflicts across projects ("deployable" source file structures and contracts may be similarly named within a namespace), but hard to each opton until we get to the step where we start combine configs/manifests for the entire world deploy |
this is almost a representation of the things needed to deploy but aren't represented by config output, just missing system functions to be registered (wouldn't be too hard to add) then would need to bake this new JSON into deployer and figure out how to build a metadata package with this new format + bake in as a default during world deploy but worried about the number of unknowns still here when we get to the config combining step that I am thinking of pausing this work for now to just unblock metadata via regular world module then revisit this later |
sort of continuing this work in #3050 closing this for now |
We're exploring what it might look like for package-exportable representation of MUD resources, where the current MUD config is only a partial representation since systems are defined by what's on the filesystem. This adds an intermediate build step and JSON representation of those systems so that we can then export a complete view of what's to be deployed.
TODO
accessList
to a set of addresses + system IDs? (could still be done by deployer but might be nice to move here)