toneScreensTests.testSuite
README.md
toneScreens tests
Checks the package against styleKit's block rules and against what its docs promise, partly on the presented frame of a small set the suite builds itself (a light wall, a dark wall and a camera of its own), since the test runner empties the scene.
- Rules: all eighteen screens are installed from toneScreens, pass
StyleKit.checkBlockand sit in the Tone slot. - Two forms, one set of settings: every page screen (holds still on the screen) has a screen on surfaces (stays on things) with exactly the same settings, so switching between them is one swap.
- One change: any screen swaps for any other in one
StyleKit.swap("screen", id)that keeps the entry's id. - Dot size follows tone: with the screen's paper lift and print fade off, the round, square, line and newspaper screens each take a larger share of the dark wall's brightness than of the light wall's.
- Still screen: on a still camera two settled frames a second apart match under the round, chain, cross-line and four-colour screens, misregistration included. Nothing crawls, shimmers or jitters on the clock.
- Clean removal: clearing a stack of screens puts the frame back within 0.5 of 255.
The scene's shared stack, the viewport camera and the test set are put back after each test. Run with zero tests run toneScreensTests or asset.resolve("toneScreensTests", "testSuite"):run(). It takes about a minute, most of it waiting for frames to settle.
Known limits
The tests judge the frame by mean brightness over regions, which is what tone means for a screen; they do not count individual dots. Motion of the camera (surface screens staying on things) is covered by the Lookbook's motion numbers in styleKit, not here.