iotaledger / iota / 36655154579

72%

Build Type

push

github

Committed by web-flow

Commit Message

fix(iota-transactional-test-runner,move-transactional-test-runner): handle transactions that are only simulated (#13035)

# Description of change

A simulated transaction — `--dry-run` or `--dev-inspect` — writes
nothing. The transactional test framework assumed a task that produced
effects also wrote state and returned modules, in three places.

Naming created objects. Names like `object(2,0`) stay stable because new
objects are sorted by their contents, read back from storage. A
simulation wrote none, so `get_object_sorting_key` panicked - only once
a task created two, since sorting one compares nothing. The sort now
takes the objects from `SimulateTransactionResult.output_objects`,
keeping the ordering by type. The issue records this against
`--dry-run`; `--dev-inspect` shares the path and was broken the same
way.
Publishing nothing. A simulated publish correctly returns an empty
module list, and the `.mvir` branch of `store_modules` did
`pop().unwrap()` on it.
A module left visible to the compiler. `store_modules` registers the
task's temp source file before looking at the modules, so an empty list
registered it anyway. `source_files()` then offered that module to every
later task, which compiled against something the chain never got and
failed at run time.
An early return from `store_modules` on an empty list covers the last
two. A real publish always returns at least one module, so nothing else
changes.

Changed: `iota-transactional-test-runner` for the first problem,
`move-transactional-test-runner` for the other two.

## Links to any relevant issues

Fixes #12810 .

## How the change has been tested

- [x] Basic tests (linting, compilation, formatting, unit/integration
tests)
- [x] Patch-specific tests (correctness, functionality coverage)
- [x] I have added tests that prove my fix is effective or that my
feature works
- [x] I have checked that new and existing unit tests pass locally with
my changes

Coverage Stats

10 of 11 new or added lines in 1 file covered. (90.91%)

1117 existing lines in 65 files now uncovered.

185110 of 263038 relevant lines covered (70.37%)

440648.46 hits per line