Upgrade Preflight
Upgrade Preflight runs the same contract calls on two real local networks at different protocol versions and compares the results. This page shows real reports from that, and lets you re-run the tool's own comparison code on them.
This page does not run Docker, Stellar or any contract. The networks and contracts ran earlier, in GitHub Actions. What happens in your browser is the diff engine, on the recorded results.
Recorded report
Results at your threshold
GeneratedComputed in your browser by the repository's own diffScenario function (src/diff/engine.ts at commit b0e179b), from the recorded results above and the threshold you chose. The recorded verdicts used a 5% threshold.
| Scenario | Recorded verdict (5%) | Verdict at your threshold | Instructions, baseline | Instructions, target | Change |
|---|
What this shows, and what it doesn't
- The measured resources are
instructions,diskReadBytes,writeBytes, footprint entry counts and the minimum resource fee. There is no memory metric. - The contracts are four tiny examples, and the figures are simulation estimates, not on-ledger costs. Don't extrapolate them.
- The cause of the instruction changes has not been traced to a specific protocol change.
- On the 28 to 29 comparison, a threshold of 1 flags all five scenarios and a threshold of 2 flags none, because every instruction change is between 1.4% and 1.8%. The verdict depends on the threshold you choose.
Reproduce it yourself
The exact command and the workflow run that produced each report are in the provenance above. See the project page for setup.