Skip to content

Matrix: renderer user input state #2581

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 11 commits into
base: renderer-user-input-core
Choose a base branch
from

Conversation

handeyeco
Copy link
Contributor

@handeyeco handeyeco commented Jun 11, 2025

Summary:

See the parent PR for more context. Part of LEMS-3208.

In #2566 I add additional APIs to let widgets consume userInput from a parent and to update that parent state using handleUserInput, while also supporting the legacy way of storing user input (which was a combination of internal widget state and stashing state in a random blob in Renderer). This PR is part of a series of PRs that go widget-by-widget to use the new API.

This is a generic summary I'm putting on each PR and not every point will be applicable to every widget, but the common themes are:

  • I added a test for serialization to make sure we're still backwards compatible (see LEMS-3185), then implemented the helpers getSerializedState and getUserInputFromSerializedState to keep supporting the expected results.
  • I moved user input for the component into the new userInput state in Renderer. This means consuming userInput in the component and calling handleUserInput on change.
  • When transform did something to initialize user input state, I moved the logic to getStartUserInput.
  • When staticTransform did something to get correct state in static widgets, I moved the logic to getCorrectUserInput.
  • Editors that use a widget to collect answer data are changed to support the new API.

Please see the parent PR for more information.


MatrixEditor does use Matrix, so this will affect the editing experience

@handeyeco handeyeco self-assigned this Jun 11, 2025
Copy link
Contributor

github-actions bot commented Jun 11, 2025

Size Change: -28 B (-0.01%)

Total Size: 473 kB

Filename Size Change
packages/perseus-editor/dist/es/index.js 91.2 kB +17 B (+0.02%)
packages/perseus/dist/es/index.js 201 kB -45 B (-0.02%)
ℹ️ View Unchanged
Filename Size
packages/kas/dist/es/index.js 20.7 kB
packages/keypad-context/dist/es/index.js 1 kB
packages/kmath/dist/es/index.js 5.98 kB
packages/math-input/dist/es/index.js 98.6 kB
packages/math-input/dist/es/strings.js 1.61 kB
packages/perseus-core/dist/es/index.js 20.4 kB
packages/perseus-linter/dist/es/index.js 7.14 kB
packages/perseus-score/dist/es/index.js 9.23 kB
packages/perseus-utils/dist/es/index.js 403 B
packages/perseus/dist/es/strings.js 7.56 kB
packages/pure-markdown/dist/es/index.js 1.22 kB
packages/simple-markdown/dist/es/index.js 6.72 kB

compressed-size-action

Copy link
Contributor

github-actions bot commented Jun 11, 2025

npm Snapshot: Published

Good news!! We've packaged up the latest commit from this PR (a4c8cc4) and published it to npm. You
can install it using the tag PR2581.

Example:

pnpm add @khanacademy/perseus@PR2581

If you are working in Khan Academy's frontend, you can run the below command.

./tools/bump_perseus_version.js -t PR2581

If you are working in Khan Academy's webapp, you can run the below command.

./dev/tools/bump_perseus_version.js -s n -t PR2581

@handeyeco handeyeco marked this pull request as ready for review June 16, 2025 16:49
@handeyeco handeyeco requested review from jeremywiebe, Myranae and a team June 16, 2025 16:49
Copy link
Contributor

@SonicScrewdriver SonicScrewdriver left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good to me. emptyMatrix gave me some pause, but I think it all makes sense.

@@ -37,7 +38,7 @@ class MatrixEditor extends React.Component<Props> {
static defaultProps: MatrixDefaultWidgetOptions =
matrixLogic.defaultWidgetOptions;

change: (arg1: any, arg2: any, arg3: any) => any = (...args) => {
change: (arg1: any, arg2?: any, arg3?: any) => any = (...args) => {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Again, it'd be nice to remove changeable here too, but it's not required.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I can't get rid of it everywhere right now because there are some uses that are beyond the scope of my current work. In this case the widget isn't wrapped in a Renderer so the editor is handling its own state. Maybe something that should change, but not related to my current work.

onBlur: () => {},
onFocus: () => {},
trackInteraction: () => {},
const matrixProps: Partial<PropsFor<typeof Matrix>> = {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I've already mentioned this separately in other PR's, but I wonder if we couldn't simplify how we're doing these types since we're making a partial and then casting it as the full props later.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

From the other PR:

I don't know what to do about the cast on 317. We're expecting props for an Expression widget but the editor doesn't provide all of those props. This happens in a couple of places in the editors because the widgets in the editors don't need all of the stuff they need in the learner experience. For the most part this worked because of poor type coverage or use of ts-expect-error which is much worse than type casting.

I think this is still an improvement than what was here before and I think further work on this is beyond the scope of my current work.

type RenderProps = MatrixPublicWidgetOptions & {
emptyMatrix: ReadonlyArray<ReadonlyArray<string>>;
};
// type RenderProps = MatrixPublicWidgetOptions;
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Should you clear this line out, given we have the proper declaration right after? If not, we might want to add a comment explaining why we're keeping this here.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good catch!

@@ -104,29 +83,32 @@ type ExternalProps = WidgetProps<
PerseusMatrixUserInput
> satisfies PropsFor<typeof Matrix>;

type RenderProps = MatrixPublicWidgetOptions & {
emptyMatrix: ReadonlyArray<ReadonlyArray<string>>;
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I see we're losing this emptyMatrix prop - was it unused?

const {matrixBoardSize, prefix, suffix} = widgetOptions;
return {
matrixBoardSize,
prefix,
suffix,
emptyMatrix: blankInput,
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Oh I see, here it is! Seems like a decent cleanup considering the tests are proving the inputs are properly empty. What a strange thing that doesn't seem to have been helping much at all.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants