feat: refactor `getContractMetadata` and `getContractClassMetadata` in `Wallet` #19457

Thunkar · 2026-01-09T11:23:21Z

This PR aims to improve getContractMetadata and getContractClassMetadata in the Wallet interface, avoiding the expensive back-and-forth of sending artifacts and classes.

Main change is that there was a weird roundabout thing when accessing protocol contracts. These are preloaded in PXE and fixed for a given aztec version, so pulling them from the wallet just to compute the ABI was incredibly wasteful.
This PR pregenerates the ABI for them because we can count on the contracts being available inside PXE. It does add a generate step to aztec.js but I think the tradeoff is well worth it. It also hints that we should separate bytecode and ABI soon.

It also removes getContractMetadata and getContractClassMetadata from PXE. These are from a time when PXE was accessed directly by apps and it obscured information, you could never be sure if you were pulling from PXE's internal storage or were asking the node directly. They are replaced by simple instance artifact getters, and it's the wallet's job to assemble the metadata. This metadata has been refactored:

`ContractMetadata`

instance: very lightweight, essentially tells you about the "version" of the contract PXE has registered if at all. An undefined value would mean PXE does not know about it (no node proxying nonsense). This is a potential privacy leak, however:
- Wallets need a comprehensive authorization framework anyways: "do you want to let app X know information about stored contract Y?"
- A malicious app could always attempt to simulate an see if it gets a "contract not registered in PXE" error
isContractInitialized : even if PXE does not know about the contract, we can compute the initialization nullifier and check if it's published. Not leaky, public info.
isContractPublished: does the contract exist in the node?
isContractUpdated: Since updates are forcibly public, we can get the contract from the node and check its currentContractClassId and originalContractClassId. If mismatched, the contract has been updated.
updatedContractClassId: If the previous is true, we put nodeContract.currentContractClassId here, allowing the app to see if instance.currentContractClassId matches and creating a convenient way to know if it's necessary to registerContract(instance, newArtifact)

`ContractClassMetadata`

isContractClassPubliclyRegistered: This makes deployments extremely convenient, as we can know in the TS wrapper for the contract with just the wallet whether we need to include class registration in the payload or not.
isArtifactRegistered: Does PXE know about this artifact?

AztecBot · 2026-01-12T13:38:21Z

Flakey Tests

🤖 says: This CI run detected 4 tests that failed, but were tolerated due to a .test_patterns.yml entry.

\033FLAKED\033 (8;;http://ci.aztec-labs.com/9e4b0af83bee49a0�9e4b0af83bee49a08;;�):  yarn-project/end-to-end/scripts/run_test.sh simple src/e2e_epochs/epochs_invalidate_block.parallel.test.ts "committee member invalidates a block if proposer does not come through" (96s) (code: 1) group:e2e-p2p-epoch-flakes (\033Gregorio Juliana\033: feat: refactor `getContractMetadata` and `getContractClassMetadata` in `Wallet` (#19457))
\033FLAKED\033 (8;;http://ci.aztec-labs.com/9a8124864c57af4d�9a8124864c57af4d8;;�): yarn-project/end-to-end/scripts/run_test.sh web3signer src/composed/web3signer/e2e_multi_validator_node_key_store.test.ts (36s) (code: 1) (\033Gregorio Juliana\033: feat: refactor `getContractMetadata` and `getContractClassMetadata` in `Wallet` (#19457))
\033FLAKED\033 (8;;http://ci.aztec-labs.com/b6fb4b1e246b8140�b6fb4b1e246b81408;;�):  yarn-project/end-to-end/scripts/run_test.sh simple src/e2e_epochs/epochs_high_tps_block_building.test.ts (289s) (code: 1) group:e2e-p2p-epoch-flakes (\033Gregorio Juliana\033: feat: refactor `getContractMetadata` and `getContractClassMetadata` in `Wallet` (#19457))
\033FLAKED\033 (8;;http://ci.aztec-labs.com/5a12ed38509aed39�5a12ed38509aed398;;�):  yarn-project/end-to-end/scripts/run_test.sh simple src/e2e_p2p/gossip_network.test.ts (451s) (code: 1) group:e2e-p2p-epoch-flakes (\033Gregorio Juliana\033: feat: refactor `getContractMetadata` and `getContractClassMetadata` in `Wallet` (#19457))

benesjan

Overall looks great and would approve if it was for the duplicate code between builder and aztec.js.

BTW any reasons why the protocol contract generation could not be in a separate PR? It seems quite independent from the other changes in this PR and it makes the change somewhat hidden (it's not in the PR title) and the PR harder to review.

benesjan · 2026-01-12T23:19:55Z

yarn-project/cli-wallet/src/cmds/check_tx.ts

-      : undefined,
-  );
+    await Promise.all(classIds.map(classId => (classId ? wallet.getContractArtifact(classId) : undefined)))
+  ).map((artifact, index) => (artifact ? { ...artifact, classId: classIds[index] } : undefined));


This transformation is quite hard to grok.

Why not first filter out the undefineds so that we don't need to do classId ? ... and artifact ? ...?

Because we would lose the indexing information, and the classId assignment would get messed up. TBH I'm still very much in favor of killing this command...

yarn-project/pxe/src/pxe.ts

yarn-project/aztec.js/src/scripts/generate_protocol_contract_types.ts

Thunkar · 2026-01-13T06:15:22Z

BTW any reasons why the protocol contract generation could not be in a separate PR? It seems quite independent from the other changes in this PR and it makes the change somewhat hidden (it's not in the PR title) and the PR harder to review.

The main objective was to unblock the removal of getContractMetadata and getContractClassMetadata, which required the protocol contract change and not much else. Unfortunately ran into the problems I detailed on slack, so it became a bigger refactor 😢

…n `Wallet` This PR aims to improve `getContractMetadata` and `getContractClassMetadata` in the `Wallet` interface, avoiding the expensive back-and-forth of sending artifacts and classes. Main change is that there was a weird roundabout thing when accessing protocol contracts. These are preloaded in PXE and fixed for a given aztec version, so pulling them from the wallet just to compute the ABI was incredibly wasteful. This PR pregenerates the ABI for them because we can count on the contracts being available inside PXE. It does add a `generate` step to `aztec.js` but I think the tradeoff is well worth it. It also hints that we should separate bytecode and ABI soon. It also removes `getContractMetadata` and `getContractClassMetadata` from PXE. These are from a time when PXE was accessed directly by apps and it obscured information, you could never be sure if you were pulling from PXE's internal storage or were asking the node directly. They are replaced by simple instance artifact getters, and it's the wallet's job to assemble the metadata. This metadata has been refactored: ## `ContractMetadata` * `instance`: very lightweight, essentially tells you about the "version" of the contract PXE has registered if at all. An undefined value would mean PXE does not know about it (no node proxying nonsense). This is a potential privacy leak, however: - Wallets need a comprehensive authorization framework anyways: "do you want to let app X know information about stored contract Y?" - A malicious app could always attempt to simulate an see if it gets a "contract not registered in PXE" error * `isContractInitialized` : even if PXE does not know about the contract, we can compute the initialization nullifier and check if it's published. Not leaky, public info. * `isContractPublished`: does the contract exist in the node? * `isContractUpdated`: Since updates are forcibly public, we can get the contract from the node and check its currentContractClassId and originalContractClassId. If mismatched, the contract has been updated. * `updatedContractClassId`: If the previous is true, we put nodeContract.currentContractClassId here, allowing the app to see if instance.currentContractClassId matches and creating a convenient way to know if it's necessary to `registerContract(instance, newArtifact)` ## `ContractClassMetadata` * `isContractClassPubliclyRegistered`: This makes deployments extremely convenient, as we can know in the TS wrapper for the contract with just the wallet whether we need to include class registration in the payload or not. * `isArtifactRegistered`: Does PXE know about this artifact?

mverzilli

Always nice to see some code go away from PXE :D

Allows ALL methods in the wallet to be batched (because why not). Initially I was going for allowing them on a case-by-case basis, but integrating the Wallet SDK has made apparent apps can have many interaction flows and this extra flexibility is nice. Updated the migration notes and in the process realized I missed exporting some artifacts in #19457, so here they are (protocol contract wrappers) Co-authored-by: thunkar <gregojquiros@gmail.com>

Allows ALL methods in the wallet to be batched (because why not). Initially I was going for allowing them on a case-by-case basis, but integrating the Wallet SDK has made apparent apps can have many interaction flows and this extra flexibility is nice. Updated the migration notes and in the process realized I missed exporting some artifacts in #19457, so here they are (protocol contract wrappers)

Allows ALL methods in the wallet to be batched (because why not). Initially I was going for allowing them on a case-by-case basis, but integrating the Wallet SDK has made apparent apps can have many interaction flows and this extra flexibility is nice. Updated the migration notes and in the process realized I missed exporting some artifacts in #19457, so here they are (protocol contract wrappers) Co-authored-by: thunkar <gregojquiros@gmail.com>

Thunkar self-assigned this Jan 9, 2026

Thunkar added the ci-no-fail-fast Sets NO_FAIL_FAST in the CI so the run is not aborted on the first failure label Jan 9, 2026

Thunkar force-pushed the gj/cleanup_wallet_interface branch from 9ed5b8f to 0b94a99 Compare January 12, 2026 06:14

Thunkar changed the title ~~feat: remove getContractMetadata and getContractClassMetadata from Wallet~~ feat: refactor getContractMetadata and getContractClassMetadata in Wallet Jan 12, 2026

Thunkar requested review from benesjan, mverzilli and nventuro January 12, 2026 13:20

benesjan reviewed Jan 12, 2026

View reviewed changes

Thunkar added the ci-squash-and-merge label Jan 13, 2026

AztecBot force-pushed the gj/cleanup_wallet_interface branch from 2c2d7dc to 4e46222 Compare January 13, 2026 07:34

AztecBot removed the ci-squash-and-merge label Jan 13, 2026

AztecBot enabled auto-merge January 13, 2026 07:34

AztecBot force-pushed the gj/cleanup_wallet_interface branch from 4e46222 to cb46a9d Compare January 13, 2026 07:37

mverzilli approved these changes Jan 13, 2026

View reviewed changes

AztecBot added this pull request to the merge queue Jan 13, 2026

github-merge-queue bot removed this pull request from the merge queue due to failed status checks Jan 13, 2026

Thunkar added this pull request to the merge queue Jan 13, 2026

Merged via the queue into next with commit b37c199 Jan 13, 2026
17 checks passed

Thunkar deleted the gj/cleanup_wallet_interface branch January 13, 2026 11:11

This was referenced Jan 14, 2026

SDK is not abstract enough which degrades Aztec design #18791

Closed

feat: wallet full batchability #19652

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

feat: refactor `getContractMetadata` and `getContractClassMetadata` in `Wallet` #19457

feat: refactor `getContractMetadata` and `getContractClassMetadata` in `Wallet` #19457

Uh oh!

Thunkar commented Jan 9, 2026 •

edited

Loading

Uh oh!

AztecBot commented Jan 12, 2026 •

edited

Loading

Uh oh!

benesjan left a comment

Uh oh!

benesjan Jan 12, 2026 •

edited

Loading

Uh oh!

Thunkar Jan 13, 2026 •

edited

Loading

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Thunkar commented Jan 13, 2026

Uh oh!

mverzilli left a comment

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

feat: refactor getContractMetadata and getContractClassMetadata in Wallet #19457

feat: refactor getContractMetadata and getContractClassMetadata in Wallet #19457

Uh oh!

Conversation

Thunkar commented Jan 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

ContractMetadata

ContractClassMetadata

Uh oh!

AztecBot commented Jan 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Flakey Tests

Uh oh!

benesjan left a comment

Choose a reason for hiding this comment

Uh oh!

benesjan Jan 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Thunkar Jan 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Thunkar commented Jan 13, 2026

Uh oh!

mverzilli left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

feat: refactor `getContractMetadata` and `getContractClassMetadata` in `Wallet` #19457

feat: refactor `getContractMetadata` and `getContractClassMetadata` in `Wallet` #19457

Thunkar commented Jan 9, 2026 •

edited

Loading

`ContractMetadata`

`ContractClassMetadata`

AztecBot commented Jan 12, 2026 •

edited

Loading

benesjan Jan 12, 2026 •

edited

Loading

Thunkar Jan 13, 2026 •

edited

Loading