Track all deployments

Elevator pitch:

Right now, we have https://github.com/ApeWorX/ape/issues/530 which is a design to track deployments as such:

from ape import project 

my_contract = account.deploy(project.MyContract)
project.track_deployment(my_contract) 
my_contract = project.deployments(project.MyContract)[-1]

The line project.track_deployment(my_contract) would add information about this contract deployment to .build/deployments/<ecosystem>/<network>/track_deployments.json

This file would look as such:

{
  "MyContract": [
    "0x40Ee4CfF4d103509cEe0720362b8055Ad65f259b",
    .
    .
    .
  ],
  "NextContract:[
  .
  .
  ],
...
}

This is how the project would keep track of deployments.

Value:

This is for all developers who don’t want to have to write custom scripts to keep track of their deployments.

Dependencies:

I don’t think this has any dependencies.

Design approach:

The main design is in the elevator pitch. But to continue, we may have to adjust the api for local networks. We’d wouldn’t always want the saved deployment to be saved forever on local networks that get destroyed so often:

def track_deployment(contract, persist: bool=true):

Or, we could pull from the ape-config.yaml

name: apeworx-project
contracts_folder: contracts
networks:
  hardhat:
    persist: false

Task list:

• Deployments can be tracked
• Tests for deployments written
• Documentation updates

Estimated completion date:

2 weeks

Design review:

Do not signoff unless:

• 1. agreed the tasks and design approach will achieve acceptance, and
• 1. the work can be completed by one person within the SLA. Design reviewers should consider simpler approaches to achieve goals.

(Please leave a comment to sign off)

Let me know if this is something I can take a crack at and if the design looks good.

[fubuloubu]

So, how’s this for an API:

from ape import project, chain

my_contract = account.deploy(project.MyContract)
project.track_deployment(my_contract) 
my_contract = chain.contracts.MyContract[-1]

or better yet, we just remove the track_deployment line totally, and pull from the ape config on whether or not to track (testnets and public networks default to true, local networks default to false)

Yes, we don’t need track_deployment to use this feature (the use case there is for keeping track of final copies of contracts to be deployed e.g. official releases), maybe let’s do this instead:

from ape import project, chain

account.deploy(project.MyContract)  # Forgot to save the contract type
my_contract = chain.contracts.find_deployments(project.MyContract)[-1]  # But I can get it this way
# NOTE: Has to do a search over the contract cache
#       to find deployments with matching contract type

[unparalleled-js]

Renamed the issue because there is a separate ticket for tracking deployments for publishing reasons that goes in the .build folder.

This ticket ended up by closed by #856 and was just for all deployments.

I know the history in confusing but otherwise this is duplicate of https://github.com/ApeWorX/ape/issues/530