test: cover submitpackage other-wtxid for same-txid-diff-witness #35893

pull mercie-ux wants to merge 1 commits into bitcoin:master from mercie-ux:test-submitpackage-other-wtxid changing 1 files +31 −0

mercie-ux commented at 6:50 AM on August 5, 2026: contributor

The specific behaviour tied to the // TODO: allow witness replacement in packages. had no functional test anywhere, just the unit test coverage, showing the underlying mempool logic and not the RPC's actual JSON output. This test submits a same-txid-different-witness transaction via submitpackage while its counterpart is already in the mempool, and asserts submitpackage correctly reports other-wtxid pointing at the transaction already in the mempool, its existing, documented behavior, not something this change introduces. Verified the test can actually fail by temporarily asserting the wrong wtxid and confirmed it fails as expected.
DrahtBot added the label Tests on Aug 5, 2026
DrahtBot commented at 6:50 AM on August 5, 2026: contributor



The following sections might be updated with supplementary metadata relevant to reviewers and maintainers.



Code Coverage & Benchmarks

For details see: https://corecheck.dev/bitcoin/bitcoin/pulls/35893.



Reviews

See the guideline and AI policy for information on the review process.

Type Reviewers

ACK jeanpablojp, sedited, instagibbs

If your review is incorrectly listed, please copy-paste <code></code> into the comment that the bot should ignore.
instagibbs commented at 1:55 PM on August 9, 2026: member

concept ack
jeanpablojp commented at 1:04 AM on August 12, 2026: contributor

tACK ec8a53618ebb4c71f4f71ba282fb990e45e413fc

Type	Reviewers
ACK	jeanpablojp, sedited, instagibbs

test: cover submitpackage other-wtxid for same-txid-diff-witness

The specific behaviour tied to the `// TODO: allow witness replacement in packages.` had no functional test anywhere, just the unit test coverage, showing the underlying mempool logic and not the RPC's actual JSON output.
This test submits a same-txid-different-witness transaction via `submitpackage` while its counterpart is already in the mempool, and asserts `submitpackage` correctly reports `other-wtxid` pointing at the transaction already in the mempool, its existing, documented behavior, not something this change introduces.
Verified the test can actually fail by temporarily asserting the wrong wtxid and confirmed it fails as expected.

9054f7cde4

in test/functional/rpc_packages.py:113 in ec8a53618e
```
 108 | +            rebalance_parent_output_amount=parent_amount,
 109 | +            child_amount=child_amount,
 110 | +        )
 111 | +
 112 | +        self.wallet.sendrawtransaction(from_node=node, tx_hex=parent.serialize().hex())
 113 | +        self.generate(node, 1)
```
jeanpablojp commented at 1:04 AM on August 12, 2026:

this block isn't needed, the mempool takes child_one on the unconfirmed parent; deleted the line and re-ran to check. feel free to ignore

mercie-ux commented at 3:37 AM on August 12, 2026:

removed it, thanks

mercie-ux commented at 3:38 AM on August 12, 2026:

removed it, thanks
mercie-ux force-pushed on Aug 12, 2026
jeanpablojp commented at 8:59 AM on August 12, 2026: contributor

reACK 9054f7cde43a264300bb0567215bda5f55803ae5
sedited approved
sedited commented at 2:06 PM on September 23, 2026: contributor

ACK 9054f7cde43a264300bb0567215bda5f55803ae5
fanquake requested review from instagibbs on Sep 23, 2026

in test/functional/rpc_packages.py:122 in 9054f7cde4

 117 | +        self.log.info("Submit child_one to the mempool directly")
 118 | +        node.sendrawtransaction(child_one.serialize().hex())
 119 | +
 120 | +        self.log.info("Submit child_two via submitpackage; it should be reported as other-wtxid, not an error")
 121 | +        submitres = node.submitpackage([child_two.serialize().hex()])
 122 | +        result = submitres["tx-results"][child_two.wtxid_hex]

instagibbs commented at 4:13 PM on September 23, 2026:

when checking results it would be best to check that the mempool was not actually mealleated (child_two wasnt swapped in), and for return fields other things could be asserted

instagibbs approved
instagibbs commented at 4:13 PM on September 23, 2026: member

ACK 9054f7cde43a264300bb0567215bda5f55803ae5

non-blocking comments, coverage is better
fanquake merged this on Sep 23, 2026
fanquake closed this on Sep 23, 2026

Contributors

Labels

Linked (view graph)

#36317 mempool, fees: stop treating different witness variants as mined

test: cover submitpackage other-wtxid for same-txid-diff-witness #35893

Code Coverage & Benchmarks

Reviews