Feature/dapps staking root only registration

[shunsukew]

Pull Request Summary Dapps staking root-only contract registration.

Right now, when devs register a dapp for dapps-staking, they first need to be whitelisted. But only their account is whitelisted, we don’t know exactly which dapp they will deploy. If we make the entire process atomic without developer_pre_approval process, whoever wants to register can specify exactly what the dev account is and which contract will be used. Also, this is required for governance over dapps registration, because RootOrigin call will be scheduled to pallet-scheduler by pallet-democracy once referendums passed.

Check list

• [x] added unit tests
• [ ] updated documentation

Logs of try-runtime against Shibuya

[shunsukew]

next pallet version could be V3_1_0, not V4_0_0

[Dinonard]

One thing to also consider - the weights.rs are no longer valid since you've changed the interface. That should be updated.

We should re-run the benchmarks but we don't have a dedicated benchmarking machine. I'd suggest you just remove the weight consumed by pre-approval checks that have now been removed.

[shunsukew]

One thing to also consider - the weights.rs are no longer valid since you've changed the interface. That should be updated.

We should re-run the benchmarks but we don't have a dedicated benchmarking machine. I'd suggest you just remove the weight consumed by pre-approval checks that have now been removed.

I know my local machine is weaker than dedicated benchmarking machine satisfying requirements. but local execution had the same result this time.

========
Storage: DappsStaking PalletDisabled (r:1 w:0)
Storage: DappsStaking RegisteredDevelopers (r:1 w:1)
Storage: DappsStaking RegisteredDapps (r:1 w:1)

How did you define base weight 32_139_000?

[Dinonard]

One thing to also consider - the weights.rs are no longer valid since you've changed the interface. That should be updated. We should re-run the benchmarks but we don't have a dedicated benchmarking machine. I'd suggest you just remove the weight consumed by pre-approval checks that have now been removed.

I know my local machine is weaker than dedicated benchmarking machine satisfying requirements. but local execution had the same result this time.

========
Storage: DappsStaking PalletDisabled (r:1 w:0)
Storage: DappsStaking RegisteredDevelopers (r:1 w:1)
Storage: DappsStaking RegisteredDapps (r:1 w:1)

How did you define base weight 32_139_000?

I didn't define it, it was measured by benchmarking tool.

The part you pasted would be same on any machine, it's discrete measurement of number of DB operations.

It's strange that only one read is removed since the two reads for PreApproval have been removed. But I see now that benchmarking was updated to take into consideration both branches - one with pre-approval enabled, one without. Anyhow, the modification you did is fine. :+1:

[shunsukew]

One thing to also consider - the weights.rs are no longer valid since you've changed the interface. That should be updated. We should re-run the benchmarks but we don't have a dedicated benchmarking machine. I'd suggest you just remove the weight consumed by pre-approval checks that have now been removed.

I know my local machine is weaker than dedicated benchmarking machine satisfying requirements. but local execution had the same result this time.

========
Storage: DappsStaking PalletDisabled (r:1 w:0)
Storage: DappsStaking RegisteredDevelopers (r:1 w:1)
Storage: DappsStaking RegisteredDapps (r:1 w:1)

How did you define base weight 32_139_000?

I didn't define it, it was measured by benchmarking tool.

The part you pasted would be same on any machine, it's discrete measurement of number of DB operations.

It's strange that only one read is removed since the two reads for PreApproval have been removed. But I see now that benchmarking was updated to take into consideration both branches - one with pre-approval enabled, one without. Anyhow, the modification you did is fine. 👍

I see. Oh, I wonder why I thought so... Yes this will calculate actual weight it costs.

T::DbWeight::get().reads(3 as Weight)

[shunsukew]

next pallet version could be V3_1_0, not V4_0_0

We broke register API backward compatibility, so v4_0_0 should be fine.

[github-actions[bot]]

Package	Line Rate	Branch Rate	Health
precompiles/utils/src	84%	0%	✔
chain-extensions/types/dapps-staking/src	0%	0%	❌
precompiles/dapps-staking/src	95%	0%	✔
frame/dapps-staking/src/pallet	85%	0%	✔
precompiles/substrate-ecdsa/src	78%	0%	➖
precompiles/utils/macro/tests	100%	0%	✔
frame/collator-selection/src	89%	0%	✔
frame/dapps-staking/src	93%	0%	✔
precompiles/utils/src/data	72%	0%	➖
frame/custom-signatures/src	80%	0%	✔
chain-extensions/trait/src	0%	0%	❌
primitives/xcm/src	88%	0%	✔
precompiles/assets-erc20/src	91%	0%	✔
chain-extensions/impls/dapps-staking/src	0%	0%	❌
frame/block-reward/src	96%	0%	✔
frame/pallet-xcm/src	80%	0%	➖
frame/vesting/src	87%	0%	✔
precompiles/sr25519/src	78%	0%	➖
precompiles/utils/macro/src	0%	0%	❌
frame/xc-asset-config/src	88%	0%	✔
precompiles/xcm/src	80%	0%	➖
Summary	85% (8317 / 9801)	0% (0 / 0)	✔

Minimum allowed line rate is 60%