Adjust transaction retry from timestamp calculation

[dhedey]

Summary

• Adjusts the logic for calculation of how long to wait before allowing a recalculation of a transaction which wants to be in the mempool (or is already in the mempool, but got kicked out).
• Refactors the mempool and cached to ensure we evict from mempool on re-evaluation

Testing

Test was added; related code was also refactored.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 41.49660% with 172 lines in your changes missing coverage. Please review.

Project coverage is 42.5%. Comparing base (3eb74c0) to head (ef6d856).
Report is 3 commits behind head on release/cuttlefish.

Files with missing lines	Patch %	Lines
...-rust/state-manager/src/mempool/mempool_manager.rs	9.3%	107 Missing ⚠️
...er/src/mempool/pending_transaction_result_cache.rs	62.1%	56 Missing ⚠️
...e-rust/state-manager/src/transaction/validation.rs	0.0%	3 Missing ⚠️
core-rust/state-manager/src/state_manager.rs	77.7%	2 Missing ⚠️
...-rust/state-manager/src/transaction/preparation.rs	85.7%	2 Missing ⚠️
...-server/src/core_api/handlers/transaction_parse.rs	0.0%	1 Missing ⚠️
core-rust/state-manager/src/jni/mempool.rs	0.0%	1 Missing ⚠️

Additional details and impacted files

@@                 Coverage Diff                  @@
##             release/cuttlefish   #1044   +/-   ##
====================================================
  Coverage                  42.4%   42.5%           
- Complexity                 4624    4627    +3     
====================================================
  Files                      1782    1782           
  Lines                     55167   55175    +8     
  Branches                   1524    1524           
====================================================
+ Hits                      23406   23450   +44     
+ Misses                    31282   31248   -34     
+ Partials                    479     477    -2     

Flag	Coverage Δ
rust	42.5% <41.4%> (+<0.1%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
...-server/src/core_api/handlers/transaction_parse.rs	0.0% <0.0%> (ø)
core-rust/state-manager/src/jni/mempool.rs	0.0% <0.0%> (ø)
core-rust/state-manager/src/state_manager.rs	86.3% <77.7%> (-0.7%)	⬇️
...-rust/state-manager/src/transaction/preparation.rs	71.0% <85.7%> (+0.1%)	⬆️
...e-rust/state-manager/src/transaction/validation.rs	13.1% <0.0%> (+2.9%)	⬆️
...er/src/mempool/pending_transaction_result_cache.rs	71.3% <62.1%> (-0.8%)	⬇️
...-rust/state-manager/src/mempool/mempool_manager.rs	12.4% <9.3%> (-4.3%)	⬇️

... and 3 files with indirect coverage changes

[github-actions]

Docker tags
docker.io/radixdlt/private-babylon-node:pr-1044
docker.io/radixdlt/private-babylon-node:b805b3b1fd
docker.io/radixdlt/private-babylon-node:sha-b805b3b

[talekhinezh]

maybe better to clamp exponent instead?

[dhedey]

Out of interest, why do you think it would be better?

I actually originally had a clamp of the exponent, but I changed to clamping the multiplier as it felt more declarative / easier to understand; but maybe that assumption is wrong! I'm happy to change back though if you prefer?

For what it's worth, (2^100) = +Inf, which clamps to multiplier_range.end - will add some unit tests for this though.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud