You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+8-8Lines changed: 8 additions & 8 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -761,16 +761,16 @@ Inspired by the Python AACProcessors project
761
761
762
762
### 🧪 Current Test Status & Immediate Follow-Up
763
763
764
-
As of the latest run (`npm test`), **45 suites pass / 8 fail (11 individual tests)**. The remaining reds are concentrated in the following areas:
764
+
As of the latest run (`npm test`), **47 suites pass / 6 fail (10 individual tests)**. Remaining blockers are:
765
765
766
-
1.**SnapProcessor coverage harness** – our in-memory fixture lacks the full Snap schema (`ElementPlacement`, `Message`, etc.), so the loader now rejects it as incomplete. Update the test database (or adjust expectations) to include Grid coordinates and message columns so audio assertions execute again.
767
-
2.**Gridset exports** – round-trip still sheds one button and the styling suite cannot find `style.xml`. Investigate GridsetProcessor save logic to ensure button arrays and style archives are written faithfully.
768
-
3.**Dot round-tripping** – property-based DOT tests collapse navigation for certain generated trees (buttons fall back to `SPEAK`). We need better semantic mapping when recreating `DotProcessor`buttons.
769
-
4.**Edge-case loaders** – corrupted JSON/XML cases no longer throw; align Asterics/OPML/DOT loaders with the test expectations (either rethrow earlier or update the tests to accept soft-fail behaviour).
770
-
5.**Advanced workflow scenario** – translated Spanish content isn’t discovered after the multi-format pipeline (likely Gridset/TMP Snap translation gaps). Trace the pipeline and ensure translated text survives each hop.
771
-
6.**Memory comparison suite** – DOT still reports higher memory delta than TouchChat/Snap. Validate memory measurement or re-tune the processors.
766
+
1.**Edge-case loaders** – corrupted JSON/XML fixtures still expect explicit exceptions. Decide whether to restore the old throwing behaviour (Asterics/OPML/DOT) or update the tests to accept the softer error reporting.
767
+
2.**Gridset exports & styling** – round-trip continues to lose a button and the styling suite cannot find `style.xml`. GridsetProcessor needs to preserve button arrays and emit the styling assets Grid 3 expects.
768
+
3.**DOT navigation semantics** – the deterministic DOT test still falls back to `SPEAK`. Improve semantic reconstruction when loading navigation edges so navigation buttons survive round-trips.
769
+
4.**Advanced workflow scenario** – the multi-format pipeline still loses Spanish translations (likely during the Gridset ⇄ Snap steps); trace text propagation and patch the conversion chain.
770
+
5.**Styling suite** – Grid 3 export still lacks `style.xml`; ensure the styling assets are generated alongside the gridset payload.
771
+
6.**Memory comparison suite** – memory delta expectations are still unmet (TouchChat GC + DOT vs others). Either recalibrate the harness or tune the processors before re-enabling the assertions.
772
772
773
-
Addressing these will move us much closer to a releasable state; once Snap coverage fixtures and Gridset exports are stabilised we should see a substantial drop in failing suites.
773
+
Clearing these items will put the test matrix back in the green and unblock the release.
0 commit comments