feat: finish SealedTx cardano-api decommission (follow-up to #5271) by paolino · Pull Request #5272 · cardano-foundation/cardano-wallet

paolino · 2026-04-24T17:27:44Z

Summary

Follow-up to #5271. This PR finishes the Cardano.Wallet.Primitive.Types.Tx.SealedTx cardano-api decommission: the primitive SealedTx module no longer imports Cardano.Api, and the deprecated cardano-api-facing SealedTx exports are removed.

Base branch: feat/drop-cardano-api-sealedtx (#5271).
Rebased after #5270 merged, #5236 was rebased onto master, and #5271 was rebased onto #5236.

What changed

#	SHA	Purpose
1	`734f624`	Plan for finishing the SealedTx cardano-api decommission
2	`80a3de4`	Port metadata extraction in `Shared/Transactions` to ledger-native
3	`8faf76e`	Drop the dead cardano-api body in `addRequiredSigners`
4	`a929e8e`	Migrate `buildAndSignTransactionPure` to ledger-native seal
5	`e2e72af`	Drop the local `sealWriteTx` cardano-api bridge in Server.hs
6	`0b619dd`	Remove the `SealedTx` cardano-api surface and move remaining unit-test adapters local to tests
7	`b4da418`	Keep default sealed transaction byte decoding capped at the supported Conway era, and make signing/decoding re-read sealed bytes with the caller's preferred latest era

SealedTx surface removed

Removed from Cardano.Wallet.Primitive.Types.Tx.SealedTx and the umbrella Cardano.Wallet.Primitive.Types.Tx export list:

cardanoTxIdeallyNoLaterThan
sealedTxFromCardano
sealedTxFromCardano'
sealedTxFromCardanoBody
getSealedTxBody
getSealedTxWitnesses
internal cardanoTxFromBytes
internal cardanoApiTxToReadTx

sealedTxFromBytes' now takes Read.EraValue Read.Era instead of AnyCardanoEra, which lets the primitive module avoid Cardano.Api entirely.

CI fix after rerun

The first rerun of #5272 exposed a real Conway integration failure: sealed Conway transaction bytes could be decoded as Dijkstra by the new ledger-native default path, after which downstream code hit the still-incomplete Dijkstra transaction support.

This PR now keeps the public sealedTxFromBytes default capped at Conway, and the transaction layer re-decodes sealed bytes with the preferred latest era supplied by the caller before signing or decoding. That preserves the old "no later than current era" behavior without restoring the cardano-api bridge.

Remaining out of scope

Removing cardano-api from cardano-wallet-primitive.cabal; other primitive modules still import it.
Removing lib/cardano-api-extra/.
Full deletion of the old Shelley/Transaction.hs module.

Test plan

nix develop --command fourmolu --mode inplace ... on touched Haskell files
git diff --check
nix develop --command just build 'cardano-wallet-primitive cardano-wallet cardano-wallet-unit:unit'
nix develop --command cabal test cardano-wallet-unit:unit -O0 -v0 --test-options '--match="TransactionSpec"'
nix develop --command cabal test cardano-wallet-unit:unit -O0 -v0 --test-options '--match="SealedTx serialisation/deserialisation"'
nix develop --command cabal build cardano-wallet cardano-wallet-api cardano-wallet-unit:unit --enable-tests --enable-benchmarks --minimize-conflict-set -O0 -v0
rg 'Cardano\.Api|qualified Cardano.Api' lib/primitive/lib/Cardano/Wallet/Primitive/Types/Tx/SealedTx.hs returns no matches
CI green on the rebased tip

paolino · 2026-04-27T14:52:03Z

@copilot, pls review

Copilot · 2026-04-27T14:52:58Z

@copilot

Hi @paolino! It looks like your comment is just a ping with no specific request. Could you let me know what you'd like me to do or look at? Happy to help!

paolino · 2026-04-27T14:55:25Z

@copilot , pls review

paolino · 2026-04-28T07:13:43Z

@copilot, pls review

Copilot · 2026-04-28T07:18:02Z

@copilot , pls review

Here's my review of the 7 commits in this PR:

Overall assessment ✅

The PR is clean, well-structured, and achieves its stated goal: SealedTx no longer imports Cardano.Api and all deprecated cardano-api-facing functions are removed. The commits are logically ordered and each has a clear, accurate commit message.

Commit-by-commit notes

1 – docs: plan (734f624) — planning doc in specs/ only; no code concerns.

2 – port metadata extraction in Shared/Transactions (80a3de4) ✅
getMetadataFromTx via Meta.getEraMetadata is a direct, idiomatic replacement for the cardanoTxIdeallyNoLaterThan round-trip. No issues.

3 – drop dead code in addRequiredSigners (8faf76e) ✅
Removing the commented-out cardano-api body is the right call. The updated TODO comment (reqSignerHashesTxBodyL, Alonzo-and-later) gives the next implementer clear guidance.

4 – buildAndSignTransactionPure (a929e8e) ✅
Replacing toCardanoApiTx → inAnyCardanoEra → sealedTxFromCardano with sealWriteTx (Write.recentEra @era) is a straightforward improvement. The refreshed haddock on sealWriteTx is accurate.

5 – Server.hs bridge removal (e2e72af) ✅
Both call sites (constructSharedTransaction, toApiSerialisedTransaction) now pass Write.recentEra explicitly, which is correct since the era is known at the call sites. No issues.

6 – Remove SealedTx cardano-api surface (0b619dd) ✅

readTxFromBytes is cleanly ported from AnyCardanoEra to EraValue Read.Era.
withinEra uses a locally-defined numberEra ordering, which is clear and exhaustive.
Test helpers (sealedTxFromCardano, getSealedTxWitnesses, cardanoTx) are moved to the unit test files — appropriate, since they remain useful for roundtrip tests but don't belong in the primitive library.
cardanoTx in the test file has a partial match error for Dijkstra (error "cardanoTx: Dijkstra not yet supported"). This is intentional and reasonable for now — an explicit error is better than a silent failure — but it's worth tracking for when Dijkstra support lands.

7 – Keep decoding within supported era (b4da418) ✅
sealedTxFromBytes defaulting to EraValue Conway prevents Conway-serialised bytes from being decoded as Dijkstra before Dijkstra transaction support is complete. The readSealedTxIdeallyNoLaterThan helper in Transaction.hs mirrors the old cardanoTxIdeallyNoLaterThan semantics: try preferredLatestEra first, fall back to the already-cached unsafeReadTx if re-decoding fails. The new prop_sealedTxDefaultRecentEraRoundtrip test directly covers this invariant. The _era → preferredLatestEra rename in addVkWitnesses is a good clarity improvement.

Minor observations

readSealedTxIdeallyNoLaterThan in Transaction.hs is private (not exported). If the "re-decode or fall back" logic ever needs to be used from another module, consider promoting it. For now, keeping it local is fine.
sealedTxFromBytes' haddock says -- ^ Most recent era for the first argument; "maximum era" or "upper-bound era" would be slightly more precise given that it acts as a cap, not the actual current protocol era. Very minor.
The {-# OPTIONS_GHC -Wno-unused-imports #-} pragma added in commit 8 to TransactionsNew.hs (carried over from before) — this is a pre-existing issue and not introduced here, so fine.

No blocking concerns. The changes are correct, the fallback logic in readSealedTxIdeallyNoLaterThan is sound, and the test coverage is appropriate for the scope of the change.

@copilot, pls review

Review posted above on the earlier comment thread — please see the detailed notes there.

paweljakubas

lgtm, nothing suspicious here, pretty straightforward

…5236) ## Summary Remaining call-site migration after #5270 merged. This branch is rebased directly onto current `master` and contains the commits that switch existing wallet call sites onto the ledger-native surfaces introduced by #5270. Base branch: `master`. Follow-ups: #5271, then #5272. ## What is in this PR | # | SHA | Purpose | |---|---|---| | 1 | `84bd668` | Use `constructUnsignedTxLedger` in `balanceTx` paths | | 2 | `4c8b61b` | Replace `AnyCardanoEra` with `Read.EraValue` in NetworkLayer | | 3 | `55097e6` | Build delegation and voting certificates in `balanceTx` paths | | 4 | `f5b8ed2` | Replace `StakeAddress` with wallet-owned `RewardAccount` | | 5 | `26bad3e` | Migrate `TxMetadata` end-to-end to wallet-owned types | | 6 | `d543a6b` | Adapt singleton network conversions to #5270's `sNetworkIdToLedger` helper | The foundation work previously described here landed in #5270. In particular, the wallet-owned `NetworkId`, `TxMetadata`, `SealedTx`, ledger-native transaction builders, certificate helpers, and witness helpers are now part of `master`. ## What this PR does not do - It does not delete the remaining cardano-api bridges. - It does not remove `cardano-api` from cabal files. - It does not delete `lib/cardano-api-extra/`. - It does not finish the `SealedTx` decommission; that remains split across #5271 and #5272. ## Test plan - [x] Rebased cleanly onto `origin/master` after #5270 merged (2026-04-25). - [x] `nix develop --command cabal build cardano-wallet cardano-wallet-api cardano-wallet-unit:unit --enable-benchmarks --enable-tests --minimize-conflict-set -O0 -v0` - [ ] CI green on the rebased tip. Note: `just build 'cardano-wallet cardano-wallet-unit:unit'` reaches unrelated `cardano-wallet-read` deprecation/unused warnings promoted by `-Werror` on this branch; the same target without `-Werror` passes.

…5271) ## Summary Progress toward removing `Cardano.Api*` from `Cardano.Wallet.Primitive.Types.Tx.SealedTx`. Moves the first set of production and test call sites off the deprecated surface. Follow-up #5272 now finishes the remaining SealedTx cleanup. Base branch: `001-drop-cardano-api` (#5236). Rebased after #5270 merged and #5236 was rebased onto `master`. ## What this PR does Eight bisect-safe commits: | # | SHA | Effect | |---|---|---| | 1 | `04d65b2` | `docs`: plan document for the decommission series | | 2 | `b4b87af` | Test helper `compareOnCBOR` -> raw `serialisedTx` bytes, avoiding a cardano-api roundtrip | | 3 | `282807d` | Submission path `_postSealedTx` -> `EraValue Read.Tx` + `consensusGenTxFromTxRecent`; deletes `unsealShelleyTx` / `UnsealException` | | 4 | `a6d0c9d` | `parsePartialTx` -> `deserializeTx :: Either DecoderError (Read.Tx era)` per era | | 5 | `f1ba908` | Delete `cardanoTxInExactEra` after the callers are gone | | 6 | `55f2bd9` | Add `sealedTxFromLedgerTx :: Read.IsEra era => Read.Tx era -> SealedTx`; migrate `sealWriteTx` off `toCardanoApiTx`/`sealedTxFromCardano'` | | 7 | `e6feabc` | Add `sealedTxWitnessCount :: SealedTx -> Int` via ledger witnesses; migrate Server.hs away from `getSealedTxWitnesses` | | 8 | `c2ca434` | Integration test `getMetadataFromTx` -> `Meta.getMetadata . Meta.getEraMetadata` on `EraValue Read.Tx` | New ledger-native surface added to `SealedTx`: - `sealedTxFromLedgerTx :: Read.IsEra era => Read.Tx era -> SealedTx` - `sealedTxWitnessCount :: SealedTx -> Int` Deleted: - `cardanoTxInExactEra` - `Cardano.Wallet.Primitive.Ledger.Shelley.unsealShelleyTx` / `UnsealException` ## What this PR does not do The deprecated SealedTx cardano-api exports are removed in #5272, not here. Out of scope: - Removing `cardano-api` from `cardano-wallet-primitive.cabal`. - Removing `lib/cardano-api-extra/`. ## Test plan - [x] Rebased cleanly onto updated #5236 after #5270 merged (2026-04-25). - [ ] CI green on the rebased tip.

Enumerates every remaining caller of the deprecated surface after #5271: production (Cardano.Wallet.hs, Server.hs), old tx builder (Shelley/Transaction.hs, partially overlaps with #5243), integration tests (Shared/Transactions.hs, addRequiredSigners in TransactionsNew) and unit tests (TransactionSpec, TransactionLedgerSpec). Organises the work into three phases: A. reachable without #5243 (4 sites) B. unit-test migration (~20 sites, two files) C. delete the bridge functions and the three Cardano.Api* imports in SealedTx.hs

…ative Mirror the TransactionsNew.hs:getMetadataFromTx port: project metadata directly from Read.Tx era via Meta.getMetadata . Meta.getEraMetadata instead of round-tripping through cardano-api. Drops the two callers of cardanoTxIdeallyNoLaterThan in this file and the Cardano.Api / ApiEra qualified imports.

The function has been an ADP-3077 TODO stub (error "...") for a long time; the commented-out cardano-api body still referenced getSealedTxBody and sealedTxFromCardanoBody, which blocks their removal. Drop the dead block and note that a future implementation should operate on Read.Tx era via reqSignerHashesTxBodyL.

The signedTx wrapping in Cardano.Wallet went through toCardanoApiTx -> inAnyCardanoEra -> sealedTxFromCardano purely to smuggle the already-ledger tx back through cardano-api. Use the existing ledger-native sealWriteTx from Transaction.Ledger instead (now exported). Drops the two cardano-api bridges inAnyCardanoEra and sealedTxFromCardano from this module. Also refresh the sealWriteTx haddock (the function has been ledger- native since the last revamp; the stale comment claimed otherwise).

Server.hs carried a local copy of sealWriteTx that re-entered cardano-api (toCardanoApiTx + sealedTxFromCardano) to wrap an already ledger-native balanceTx result. Import the ledger-native sealWriteTx from Transaction.Ledger instead; the local helper goes away. Two call sites now pass Write.recentEra explicitly. Drops the cardano-api bridges cardanoEraFromRecentEra, toCardanoApiTx and W.sealedTxFromCardano from this module's imports.

paolino self-assigned this Apr 24, 2026

paolino added the Refactoring No functional changes label Apr 24, 2026

paolino force-pushed the feat/drop-cardano-api-sealedtx branch from b530ab1 to 65a5529 Compare April 25, 2026 09:41

paolino force-pushed the feat/drop-cardano-api-sealedtx-remainder branch from 80ec3ce to dd6ae54 Compare April 25, 2026 09:42

This was referenced Apr 25, 2026

feat: drop cardano-api dependency (additive migration, 47/75 files) #5236

Merged

feat: decommission SealedTx cardano-api surface (follow-up to #5236) #5271

Merged

paolino force-pushed the feat/drop-cardano-api-sealedtx branch from 65a5529 to c2ca434 Compare April 25, 2026 10:11

paolino force-pushed the feat/drop-cardano-api-sealedtx-remainder branch from dd6ae54 to 0b619dd Compare April 25, 2026 10:16

paolino requested review from paweljakubas and removed request for paweljakubas April 25, 2026 10:18

Copilot started work on behalf of paolino April 27, 2026 14:52 View session

Copilot finished work on behalf of paolino April 27, 2026 14:53

Copilot started work on behalf of paolino April 28, 2026 07:13 View session

Copilot finished work on behalf of paolino April 28, 2026 07:19

paweljakubas approved these changes May 5, 2026

View reviewed changes

paolino force-pushed the feat/drop-cardano-api-sealedtx branch from 18a23ac to fdfadf0 Compare May 8, 2026 10:21

Base automatically changed from feat/drop-cardano-api-sealedtx to master May 8, 2026 11:00

paolino added 5 commits May 8, 2026 12:14

paolino added 2 commits May 8, 2026 12:14

refactor: remove SealedTx cardano-api surface

80428fe

fix: keep SealedTx decoding within supported era

eb80853

paolino force-pushed the feat/drop-cardano-api-sealedtx-remainder branch from b4da418 to eb80853 Compare May 8, 2026 11:18

paolino merged commit fbcb6ad into master May 8, 2026
112 of 113 checks passed

paolino deleted the feat/drop-cardano-api-sealedtx-remainder branch May 8, 2026 13:53

paolino mentioned this pull request May 8, 2026

Upgrade dependencies to cardano-node 11.0.1 #5275

Open

This was referenced May 18, 2026

cardano-api removal: transaction-layer remaining work #5243

Open

Remove cardano-api dependency #5237

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat: finish SealedTx cardano-api decommission (follow-up to #5271)#5272

feat: finish SealedTx cardano-api decommission (follow-up to #5271)#5272
paolino merged 7 commits into
masterfrom
feat/drop-cardano-api-sealedtx-remainder

paolino commented Apr 24, 2026 •

edited

Loading

Uh oh!

paolino commented Apr 27, 2026 •

edited

Loading

Uh oh!

Copilot AI commented Apr 27, 2026

Uh oh!

paolino commented Apr 27, 2026 •

edited

Loading

Uh oh!

paolino commented Apr 28, 2026

Uh oh!

Copilot AI commented Apr 28, 2026 •

edited

Loading

Uh oh!

paweljakubas left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

paolino commented Apr 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

What changed

SealedTx surface removed

CI fix after rerun

Remaining out of scope

Test plan

Uh oh!

paolino commented Apr 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Copilot AI commented Apr 27, 2026

Uh oh!

paolino commented Apr 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

paolino commented Apr 28, 2026

Uh oh!

Copilot AI commented Apr 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Overall assessment ✅

Commit-by-commit notes

Minor observations

Uh oh!

paweljakubas left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

paolino commented Apr 24, 2026 •

edited

Loading

paolino commented Apr 27, 2026 •

edited

Loading

paolino commented Apr 27, 2026 •

edited

Loading

Copilot AI commented Apr 28, 2026 •

edited

Loading