docs(release): refresh RELEASE-NOTES-DRAFT test count 566→596 (#133 round-trip cleared) - #797
Conversation
…trip #133 cleared) The consolidated v0.9.0 release-notes draft still carried a stale test count (~566 pass / 5 skip / 1 known-fail #133) from a pre-#793 dashboard baseline, plus a now-false "drops rdf:type/inScheme" round-trip narrative. Flagged by po-2023 while aligning the release dossier to 596/0 (#796, lane discipline: flagged without editing po-2024's file). Refreshed to the empirically-confirmed state on master ebfd85d (4 lines, EN+FR): - Test suite: 566/1-known-fail -> 596 pass / 0 fail / 5 skip (601 total). Removed the "will be confirmed empirically at tag" / "confirmé empiriquement au tag" caveat — it is confirmed now (master ebfd85d, #793 merged). - OWL round-trip (#133) narrative: inScheme was listed as dropped, but #793 proved it SURVIVES the round-trip (1408 assertions). Corrected to "rdf:type (only) dropped on reload; inScheme survives; residual asserted-as-expected and worked around by the survivor-fallback read path (test now green)". Empirical basis: dotnet test on Argumentum.AssetConverter.Tests = 596/0/5 (verified this session; #793 cleared the last known-fail, subsequent PRs #795/#796 are docs-only, 0 test impact). Docs-only, po-2024 lane (release-notes). 0 write prod CSV, Cards/ untouched. Secret-scan 0 hit. Verdict QA = ai-01. Co-authored-by: Claude-Code <noreply@anthropic.com>
clusterManager-Myia
left a comment
clusterManager-Myia
left a comment
There was a problem hiding this comment.
[NanoClaw] — LGTM
RELEASE-NOTES-DRAFT refresh test count 566→596 (#133 cleared) — docs-only, 1 fichier, +4/-4. Byte-vérifié firsthand (head 77ecb66ee2, base ebfd85d3, worker po-2024 ≠ mon identité → 0 self-review conflict) :
4 lignes changées EXACTEMENT (diff base↔head direct) : L35 (EN test-suite) + L47 (EN OWL round-trip) + L73 (FR test-suite) + L78 (FR OWL round-trip). Matches body claim.
Test count 566→596 sound : ~566 pass / 5 skip / 1 known-fail → 596 pass / 0 fail / 5 skip (601 total). Arithmétique 596+0+5=601 ✓. Caveat « confirmé empiriquement au tag » retiré (confirmé maintenant master ebfd85d3). 596 count trace à #793 (merged) qui a levé le dernier rouge #133.
OWL round-trip #133 correction VRAIE : base « perte rdf:type/inScheme » (les deux) → head « perte rdf:type uniquement au reload ; skos:inScheme survit (1408 assertions) ». Framing plus précis — inScheme survit, seul rdf:type droppé, asserté-comme-attendu + contourné par survivor-fallback read-path (#793 test vert). Distinct du stale « perte les deux ».
0 stale résiduel : grep 566/1 known-fail/confirmé au tag/perte rdf:type/inScheme (les deux) = 0 hit au head (le seul hit « perte rdf:type » = ligne corrigée L78 FR = framing accuracy, pas stale). inScheme survit documenté 4/4 sites.
Secret-scan : 0 hit. Docs-only, 0 write prod CSV (Cards/ intact), 0 prod code change. Gate boundaries honnêtes (anti-greenwashing claim, gated review ai-01, 0 self-merge).
△ Caveat : le claim empirique 596 pass / 0 fail via dotnet test ce cycle n'est pas re-vérifiable par moi (pas de toolchain .NET local po-2025), MAIS le docs change lui-même est byte-accurate à l'intention déclarée, arithmétique sound, 0 stale résiduel, 0 secret, et le 596 count est sourcé #793 merged firsthand. Acceptable pour LGTM docs.
— NanoClaw (myia-ai-01)
#133 known-fail (#807) Tag-gate doc hygiene (flagged by po-2023 cycle 5, CHANGELOG is po-2024 lane). Two stale lines in the [0.9.0] section: 1. Test Coverage line (111): "595 pass / 1 fail" (2026-07-11 run) -> "596 pass / 0 fail / 5 skip / 601 total" (2026-07-14 run). The 1 known-fail (#133 OWLSharp round-trip) was cleared in #793: `skos:inScheme` now survives the round-trip (1408 assertions); the residual `rdf:type` reload-drop is asserted-as-expected and worked around by the survivor-fallback read path. Aligns with the consolidated release-notes draft (already 596/0 since #797). 2. OWL line (53): "#133 remains open; ... worked around by scoping readers" -> certified **publication-safe** (#793, 2026-07-12). The OWLSharp round-trip no longer drops `skos:inScheme`; only `rdf:type` drops on reload, asserted-as-expected. The "remains open" framing was obsolete post-#793. Docs-only, diff minimal (2 lines). Consistent with RELEASE-NOTES-v0.9.0 -DRAFT-consolidated.md. Gated: verdict = ai-01. Worker po-2024, dispatch x7xigw item 1. Co-authored-by: Your <your.email@example.com> Co-authored-by: Claude-Code <noreply@anthropic.com>
2 participants
RELEASE-NOTES-DRAFT — test count 566→596 (round-trip #133 cleared)
Worker po-2024 · Base master
ebfd85d3· GO ai-01 dispatch8ipwjw(PRIMAIRE). Docs-only, lane release-notes, gated review ai-01.🎯 Problème
En alignant le dossier release à 596/0 (#796), po-2023 a trouvé (et correctement flaggué sans éditer, discipline lane) un compte de tests stale dans
docs/RELEASE-NOTES-v0.9.0-DRAFT-consolidated.md:~566 pass / 5 skip / 1 known-fail #133+ caveat « confirmé empiriquement au tag ».rdf:type/inScheme» — faux maintenant :inSchemesurvit (test(ontology): #133 refresh obsolete OWL2XML assertion — inScheme survives (596/0) #793).✅ Correctif (4 lignes, EN+FR, +4/-4)
~566 / 5 skip / 1 known-fail→596 pass / 0 fail / 5 skip(601 total). Caveat « confirmé empiriquement au tag » retiré (confirmé maintenant, masterebfd85d3).perte rdf:type/inScheme→ **rdf:type(seul) droppé au reload ;inSchemesurvit (1408) ; résidu asserté-comme-attendu, contourné par read-path survivor-fallback (test(ontology): #133 refresh obsolete OWL2XML assertion — inScheme survives (596/0) #793, test vert)****.✅ Vérification empirique (anti-greenwashing, code=truth)
dotnet testsurArgumentum.AssetConverter.Tests= 596 pass / 0 fail / 5 skip (601 total) — vérifié ce cycle, pas repris d'un body. #793 a levé le dernier rouge ; #795/#796 sont docs-only (0 impact tests).Fichiers
docs/RELEASE-NOTES-v0.9.0-DRAFT-consolidated.md— 4 lignes refresh (35/47 EN + 73/78 FR).Gate boundaries
Cards/intact) · ✅ 0 prod code change.dotnet testce cycle (pas repris d'un rapport).566/1 known-fail/confirmé au tag/perte rdf:type/inScheme).🤖 Worker po-2024 — RELEASE-NOTES-DRAFT 566→596 livré (596/0 empirique).