|
6 | 6 | - [ ] Repository health script passes |
7 | 7 | - [ ] No nested repo, RnD solution, build output, logs, captures, or local settings are tracked |
8 | 8 | - [ ] Version matches `Directory.Build.props`, README, landing page, release workflow, and release notes |
9 | | -- [ ] Changelog and tested-configuration notes are updated |
| 9 | +- [ ] Changelog, architecture notes, and tested-configuration notes are updated |
10 | 10 |
|
11 | 11 | ## Automated validation |
12 | 12 |
|
|
15 | 15 | - [ ] All tests pass |
16 | 16 | - [ ] Runtime Stability workflow passes all repeated `Category=RuntimeStability` iterations |
17 | 17 | - [ ] Runtime Stability TRX artifact is retained with the release evidence |
| 18 | +- [ ] Runtime Architecture workflow passes all repeated `Category=RuntimeArchitecture` iterations |
| 19 | +- [ ] Runtime Architecture TRX artifact is retained with the release evidence |
| 20 | +- [ ] Classic PCAP little/big-endian and micro/nanosecond variants pass |
| 21 | +- [ ] Truncated, unsupported-link-type, invalid-timestamp, and oversized-record cases are rejected |
18 | 22 | - [ ] CodeQL completes |
19 | 23 | - [ ] Dependency review has no unresolved high-severity finding |
20 | | -- [ ] Portable package verification succeeds |
21 | | -- [ ] ZIP checksum and release manifest are generated |
| 24 | +- [ ] Portable candidate package verification succeeds |
| 25 | +- [ ] ZIP checksum and candidate/release manifest are generated |
22 | 26 |
|
23 | 27 | ## Runtime smoke test |
24 | 28 |
|
|
33 | 37 | - [ ] Duplicate, forward-gap, and publisher-restart behavior is exercised |
34 | 38 | - [ ] GOOSE, PTP, and SCL workspaces open and update |
35 | 39 | - [ ] 30–60 minute soak test shows no material memory growth or UI freeze |
36 | | -- [ ] `docs/validation/V1.3.0_BETA2_FIELD_EVIDENCE.md` is completed with sanitized evidence |
| 40 | +- [ ] Maintained live evidence remains available in `docs/validation/V1.3.0_BETA2_FIELD_EVIDENCE.md` |
| 41 | + |
| 42 | +## Replay and immutable snapshot checks |
| 43 | + |
| 44 | +- [ ] Sanitized classic Ethernet PCAP replays through `RawProcessBusAnalyzer.ObserveOwnedFrame` |
| 45 | +- [ ] Recorded frame order and capture duration are preserved |
| 46 | +- [ ] Selected-stream runtime generation contains copied identity, samples, analog, phasor, shape, and diagnostics |
| 47 | +- [ ] Advancing the analyzer does not mutate a previously published generation |
| 48 | +- [ ] Interleaved replay streams remain isolated when selected and published |
| 49 | +- [ ] Replay is file-input only and does not open a transmit path |
| 50 | +- [ ] PCAPNG is listed as unsupported until implemented and validated |
37 | 51 |
|
38 | 52 | ## Claims and evidence |
39 | 53 |
|
40 | 54 | - [ ] Timing wording remains screening-level unless hardware evidence exists |
| 55 | +- [ ] Replay is not described as recreating NIC, switch, or hardware timestamp behavior |
41 | 56 | - [ ] Automated deterministic stress is not described as a replacement for live/replay soak validation |
| 57 | +- [ ] Full internal analyzer decomposition is not claimed before it is complete |
42 | 58 | - [ ] Screenshots match the released UI |
43 | 59 | - [ ] No customer or project-sensitive evidence is included |
44 | 60 | - [ ] Known limitations are listed in release notes |
|
0 commit comments