Fix transaction autobalancing when deregistering credential

[carbolymer]

Changelog

- description: |
    Fix transaction autobalancing when deregistering a credential
# uncomment types applicable to the change:
  type:
  # - feature        # introduces a new feature
  # - breaking       # the API has changed in a breaking way
  # - compatible     # the API has changed but is non-breaking
  # - optimisation   # measurable performance improvements
  # - refactoring    # QoL changes
   - bugfix         # fixes a defect
  # - test           # fixes/modifies tests
  # - maintenance    # not directly related to the code
  # - release        # related to a new release preparation
  # - documentation  # change in code docs, haddocks...

Context

The first step of autobalancing was not considering ADA inflow from deregistering a credential, so the resulting change could get negative and it was blowing up in ledger code.

Detailed cause explanation

The negative intermediate change value could be a result of the first call to:

calculateChangeValue sbe totalValueAtSpendableUTxO txbodycontent

calculateChangeValue considers only:

• tx inputs
• tx outputs
• minted assets

so if there's not enough input value, meaning tx outs > tx ins, the calculated intermediate change gets negative. Note, that's a valid scenario if enough funds are coming in from certificate deposit return.

If that change is then used as a new transaction output (the change output), it will throw an exception in the ledger code.

The fix

We're using the exact change value in the first step of autobalancing, for the execution units calculation.

Regression tests executed in:

• https://github.com/IntersectMBO/cardano-node-tests/actions/runs/13072587532

Checklist

• Commit sequence broadly makes sense and commits have useful messages
• New tests are added if needed and existing tests are updated. See Running tests for more details
• Self-reviewed the diff

[smelc]

LGTM. Ask me for approval when this is ready 👍

[Jimbo4350]

Can you add some more details as to exactly how you fixed this? It's not obvious to me what changes are the fix as they are overlapped with formatting changes.

[Jimbo4350]

Still unclear. Separate the formatting changes and the semantic changes with git add --patch I understand the change now after reading the PR description. However we need to confirm that the change value does not influence the execution units calculation. I think it may as a larger change value means a larger script context. Confirm this with the plutus team.

[Jimbo4350]

👍