Reset the canvas transform instead of compounding scale on resize

A canvas can grow its contents on every resize when a density transform is multiplied repeatedly. This fixture isolates the matrix-state mistake behind that cumulative scaling.

This edition is a laboratory fixture. Node and the companion HTML page execute the authored functions against synthetic cases. No physical device panel was run, no ad network reviewed the build, and no campaign result is inferred from a passing test.

What the documentation actually supplies

MDN: canvas transform is the governing reference for the browser primitive. MDN documents setTransform as setting the canvas transformation matrix, in contrast to composing an additional transform. The fixture models the scale component of that state rather than rendering through a real canvas context. The fixture policy around that primitive is ours.

The failure the fixture isolates

The input is a sequence of desired density scales. The corrected branch replaces the previous scale with the latest value. The broken branch multiplies every value into the accumulated transform. Repeating the same resize therefore changes the broken result even though the desired final layout is unchanged.

The broken branch is retained so a producer can see the exact mistake, not only the repaired output.

All 3 fixed-branch cases matched the spec when executed in Node on 19 September 2026.

What a pass does not prove

The test proves idempotent scale assignment in the simplified model. It does not verify the actual canvas state, drawing order, clipping region or image quality. A production renderer should explicitly establish its transform at the start of a drawing pass rather than depend on whichever state a previous pass happened to leave.

Untested cells in this edition: physical devices, host SDKs, network packaging limits, store-opening callbacks, and campaign performance. Those remain empty on purpose.

Next check on a real build

Trigger repeated same-size layout events in the real build and compare known object coordinates before and after. Restore or reset the transform before applying the desired logical-to-backing mapping. Test this together with the density policy, because a correct buffer size cannot compensate for an incorrectly accumulated drawing transform.

Open the runnable fixture to re-execute the cases in a browser. Download the case record to keep inputs, expectations and limits with a QA ticket.

Background in the archive: Playable ad device testing matrix and Playable spec sheet before you build. Those guides describe practice; they are not a substitute for this fixture and this check does not recertify their every claim.