Fix focus-trap crash when a bottom-docked modal's initialFocus ref is null (Sentry APP-C26) by trasnake87 · Pull Request #92747 · Expensify/App

trasnake87 · 2026-06-05T02:38:15Z

Explanation of Change

On web (narrow / mobile viewport), opening a bottom-docked popover or menu on /home — the FAB "create" menu, a report-action context menu, the emoji picker, attachment options (all render as bottom sheets on small screens) — could crash the screen with:

`initialFocus` was specified but was not a node, or did not return a node

(Sentry APP-C26, ~153 users / 171 events, non-deterministic.)

Root cause. BaseModal passes () => bottomDockedDismissButtonRef.current as initialFocus for bottom-docked modals (src/components/Modal/BaseModal.tsx). focus-trap reads initialFocus on a deferred setTimeout, and by then the dismiss button can already be unmounted — the sheet was dismissed quickly, or a relayout (resize / rotate / keyboard) flipped shouldShowBottomDockedDismissButton — so the getter returns null. focus-trap's getNodeForOption treats undefined ("auto-focus first tabbable") and false ("skip initial focus") as valid, but throws on any other falsy value (null / '' / 0). So the transient null crashes the trap.

Fix (two layers).

BaseModal coerces the getter's null to false at the source: () => bottomDockedDismissButtonRef.current ?? false.
FocusTrapForModal (web) normalizes any throwing initialFocus value to false at the single choke point every modal trap flows through, so no current or future caller can crash the trap. false / undefined and real nodes pass through unchanged.

Fixed Issues

$ #92415
PROPOSAL: #92415 (comment)

Tests

Added regression coverage in tests/unit/FocusTrapForModalTest.tsx: an initialFocus getter that returns null (and a bare null) now resolves to false instead of crashing; a real node and false pass through unchanged. The new cases return null (and would crash focus-trap) without this change.
Manual (web, narrow / mobile viewport, on /home): repeatedly open and quickly dismiss the FAB create menu, a report-action context menu, and the emoji picker; rotate / resize the window while one is opening. The screen no longer throws the initialFocus error.

Verify that no errors appear in the JS console

Offline tests

N/A — this is focus/UI behavior with no network dependency.

QA Steps

On web with a narrow viewport (or mobile web), go to /home.
Open the FAB (+) create menu and dismiss it quickly; repeat several times.
Open a report-action context menu and the emoji picker and dismiss quickly; rotate or resize the window while one is opening.
Verify the app does not crash and no `initialFocus` was specified but was not a node error appears in the console.

Verify that no errors appear in the JS console

PR Author Checklist

Note: this is a web-only fix — focus-trap doesn't run on native — so the native platform items below are N/A (the change can't affect native); verified on web (recording in Screenshots/Videos).

Screenshots/Videos

Android: Native

Android: mWeb Chrome

iOS: Native

iOS: mWeb Safari

MacOS: Chrome / Safari

92415_web_fab_menu.mp4

melvin-bot · 2026-06-05T02:38:21Z

@bernhardoj Please copy/paste the Reviewer Checklist from here into a new comment on this PR and complete it. If you have the K2 extension, you can simply click: [this button]

bernhardoj · 2026-06-05T02:47:38Z

                trapStack: sharedTrapStack,
                clickOutsideDeactivates: true,
-                initialFocus,
+                initialFocus: getSafeInitialFocus,


Let's revert this one, I don't think we need this.

Feels like we are recreating the focus-trap getNodeForOption.

I think it's the consumer's responsibility to make sure they don't pass 0 or '' or invalid element. If in the future there is a bug where the initial focus is an invalid element, someone will patch getSafeInitialFocus to check the validity of the element, and we will end up with an "inferior getNodeForOption".

For null, though, we can actually prevent this by type-checking. However, the focus-trap codebase accepts () => void as the initial focus, which accepts () => null.

Reverted the FocusTrapForModal layer — you're right, that's the consumer's responsibility and it would've drifted into an inferior getNodeForOption over time. Kept the fix at the source: BaseModal now coerces the bottom-docked getter's null to false. On the checklist + recordings next.

bernhardoj · 2026-06-05T06:19:48Z

Reviewer Checklist

I have verified the author checklist is complete (all boxes are checked off).
I verified the correct issue is linked in the ### Fixed Issues section above
I verified testing steps are clear and they cover the changes made in this PR
- I verified the steps for local testing are in the Tests section
- I verified the steps for Staging and/or Production testing are in the QA steps section
- I verified the steps cover any possible failure scenarios (i.e. verify an input displays the correct error message if the entered data is not correct)
- I turned off my network connection and tested it while offline to ensure it matches the expected behavior (i.e. verify the default avatar icon is displayed if app is offline)
I checked that screenshots or videos are included for tests on all platforms
I included screenshots or videos for tests on all platforms
I verified that the composer does not automatically focus or open the keyboard on mobile unless explicitly intended. This includes checking that returning the app from the background does not unexpectedly open the keyboard.
I verified tests pass on all platforms & I tested again on:
- Android: HybridApp
- Android: mWeb Chrome
- iOS: HybridApp
- iOS: mWeb Safari
- MacOS: Chrome / Safari
If there are any errors in the console that are unrelated to this PR, I either fixed them (preferred) or linked to where I reported them in Slack
I verified proper code patterns were followed (see Reviewing the code)
- I verified that any callback methods that were added or modified are named for what the method does and never what callback they handle (i.e. toggleReport and not onIconClick).
- I verified that comments were added to code that is not self explanatory
- I verified that any new or modified comments were clear, correct English, and explained "why" the code was doing something instead of only explaining "what" the code was doing.
- I verified any copy / text shown in the product is localized by adding it to src/languages/* files and using the translation method
- I verified all numbers, amounts, dates and phone numbers shown in the product are using the localization methods
- I verified any copy / text that was added to the app is grammatically correct in English. It adheres to proper capitalization guidelines (note: only the first word of header/labels should be capitalized), and is either coming verbatim from figma or has been approved by marketing (in order to get marketing approval, ask the Bug Zero team member to add the Waiting for copy label to the issue)
- I verified proper file naming conventions were followed for any new files or renamed files. All non-platform specific files are named after what they export and are not named "index.js". All platform-specific files are named for the platform the code supports as outlined in the README.
- I verified the JSDocs style guidelines (in STYLE.md) were followed
If a new code pattern is added I verified it was agreed to be used by multiple Expensify engineers
I verified that this PR follows the guidelines as stated in the Review Guidelines
I verified other components that can be impacted by these changes have been tested, and I retested again (i.e. if the PR modifies a shared library or component like Avatar, I verified the components using Avatar have been tested & I retested again)
I verified all code is DRY (the PR doesn't include any logic written more than once, with the exception of tests)
I verified any variables that can be defined as constants (ie. in CONST.ts or at the top of the file that uses the constant) are defined as such
If a new component is created I verified that:
- A similar component doesn't exist in the codebase
- All props are defined accurately and each prop has a /** comment above it */
- The file is named correctly
- The component has a clear name that is non-ambiguous and the purpose of the component can be inferred from the name alone
- The only data being stored in the state is data necessary for rendering and nothing else
- For Class Components, any internal methods passed to components event handlers are bound to this properly so there are no scoping issues (i.e. for onClick={this.submit} the method this.submit should be bound to this in the constructor)
- Any internal methods bound to this are necessary to be bound (i.e. avoid this.submit = this.submit.bind(this); if this.submit is never passed to a component event handler like onClick)
- All JSX used for rendering exists in the render method
- The component has the minimum amount of code necessary for its purpose, and it is broken down into smaller components in order to separate concerns and functions
If any new file was added I verified that:
- The file has a description of what it does and/or why is needed at the top of the file if the code is not self explanatory
If a new CSS style is added I verified that:
- A similar style doesn't already exist
- The style can't be created with an existing StyleUtils function (i.e. StyleUtils.getBackgroundAndBorderStyle(theme.componentBG)
If the PR modifies code that runs when editing or sending messages, I tested and verified there is no unexpected behavior for all supported markdown - URLs, single line code, code blocks, quotes, headings, bold, strikethrough, and italic.
If the PR modifies a generic component, I tested and verified that those changes do not break usages of that component in the rest of the App (i.e. if a shared library or component like Avatar is modified, I verified that Avatar is working as expected in all cases)
If the PR modifies a component related to any of the existing Storybook stories, I tested and verified all stories for that component are still working as expected.
If the PR modifies a component or page that can be accessed by a direct deeplink, I verified that the code functions as expected when the deeplink is used - from a logged in and logged out account.
If the PR modifies the UI (e.g. new buttons, new UI components, changing the padding/spacing/sizing, moving components, etc) or modifies the form input styles:
- I verified that all the inputs inside a form are aligned with each other.
- I added Design label and/or tagged @Expensify/design so the design team can review the changes.
If a new page is added, I verified it's using the ScrollView component to make it scrollable when more elements are added to the page.
For any bug fix or new feature in this PR, I verified that sufficient unit tests are included to prevent regressions in this flow.
If the main branch was merged into this PR after a review, I tested again and verified the outcome was still expected according to the Test steps.
I have checked off every checkbox in the PR reviewer checklist, including those that don't apply to this PR.

Screenshots/Videos

Android: HybridApp

Android: mWeb Chrome

android.mweb.mp4

iOS: HybridApp

iOS: mWeb Safari

ios.mweb.mp4

MacOS: Chrome / Safari

web.mp4

bernhardoj · 2026-06-05T06:21:11Z

@trasnake87 please complete the checklist (including recordings) and address #92747 (comment)

JmillsExpensify

No product review required.

trasnake87 · 2026-06-06T00:11:06Z

@bernhardoj done on all three:

Reverted the FocusTrapForModal layer — kept the fix at the source (BaseModal coerces the null getter to false).
Completed the checklist. This is a web-only fix (focus-trap doesn't run on native), so native items are N/A; left the unit-test box unchecked with a note — the fix is a one-line source guard and the crash is a non-deterministic deferred-setTimeout race, so a unit test would only assert ?? false returns false. Happy to add a render test if you'd prefer.
Attached the web recording (Screenshots/Videos → MacOS: Chrome / Safari): the FAB create menu opening + dismissing on /home, including rapid cycles to stress the path, no crash.

Ready for re-review.

trasnake87 · 2026-06-06T00:16:09Z

Update: added the regression test (tests/unit/BaseModalInitialFocusTest.tsx) and ticked the box, so the checklist is now fully complete. The test asserts the bottom-docked dismiss-button getter resolves to false (not null) when its ref is unmounted — the exact state that crashed focus-trap.

bernhardoj · 2026-06-06T04:16:03Z

+// Capture the props BaseModal passes to the underlying modal (specifically `initialFocus`).
+// Render null so the bottom-docked dismiss button never mounts — i.e. its ref stays `null`,
+// which is exactly the state that makes focus-trap throw "`initialFocus` was specified but was
+// not a node" (Sentry APP-C26). The fix must coerce that `null` to `false`.


I don't like that we need to put the sentry number reference or the issue number reference (line 28)

bernhardoj · 2026-06-06T04:18:34Z

+    isInLandscapeMode: false,
+}));
+
+describe('BaseModal — bottom-docked initialFocus null guard (#92415)', () => {


Btw, can we create a BaseModalTest file that will contain all BaseModal test cases instead of creating 1 file for 1 test case?

Then we need to mock the ReanimatedModal for this specific test case only instead for the whole test suite.

bernhardoj · 2026-06-06T04:18:53Z

Also, lint fails

trasnake87 · 2026-06-06T05:25:47Z

@bernhardoj addressed both — renamed to a general tests/unit/BaseModalTest.tsx, removed the Sentry/issue-number references, and scoped the ReanimatedModal + useResponsiveLayout mocks to just this test case via jest.doMock (so they don't leak into future BaseModal tests). Also fixed the lint failure (the unused prop-type on the mock + the restricted Text import) — green locally now. Ready for another look.

bernhardoj · 2026-06-06T07:31:47Z

+describe('BaseModal', () => {
+    afterEach(() => {
+        jest.dontMock('@components/Modal/ReanimatedModal');
+        jest.dontMock('@hooks/useResponsiveLayout');


I think jest.resetModules(); is simpler here instead of dontMock.

bernhardoj · 2026-06-06T07:32:21Z

+        // modal, the dismiss-button ref can be `null` by the time focus-trap reads it (the read is deferred), so
+        // the getter must coerce that `null` to `false`. The mocks are scoped to this test (via jest.doMock) so
+        // they don't leak into other BaseModal cases.
+        let captured: Record<string, unknown> = {};


Let's use ReanimatedModalProps as the props type and for this specifically, add | undefined (ReanimatedModalProps | undefined).

bernhardoj · 2026-06-06T07:32:59Z

+        // the getter must coerce that `null` to `false`. The mocks are scoped to this test (via jest.doMock) so
+        // they don't leak into other BaseModal cases.
+        let captured: Record<string, unknown> = {};
+        jest.doMock('@hooks/useResponsiveLayout', () => ({


We don't need this because by default, test runs on small screen

App/__mocks__/react-native.ts

Lines 12 to 21 in 380da60

// Tests will run with the app in a typical small screen size by default. We do this since the react-native test renderer

// runs against index.native.js source and so anything that is testing a component reliant on withWindowDimensions()

// would be most commonly assumed to be on a mobile phone vs. a tablet or desktop style view. This behavior can be

// overridden by explicitly setting the dimensions inside a test via Dimensions.set()

let dimensions: Record<string, number> = {

width: 300,

height: 700,

scale: 1,

fontScale: 1,

};

bernhardoj · 2026-06-06T07:34:19Z

+            </BaseModal>,
+        );
+
+        const initialFocus = captured.initialFocus as (() => unknown) | undefined;


Suggested change

const initialFocus = captured.initialFocus as (() => unknown) | undefined;

const initialFocus = captured?.initialFocus;

expect(typeof initialFocus).toBe('function');

// dismiss button never mounted -> ref.current is null -> the getter resolves to false (no crash)

expect((initialFocus as () => unknown)()).toBe(false);

bernhardoj · 2026-06-06T07:34:54Z

Also, please move the file under /tests/ui folder

…dals For a bottom-docked modal, BaseModal passes () => bottomDockedDismissButtonRef.current as initialFocus. focus-trap reads it on a deferred setTimeout, by which point the dismiss button can be unmounted, so the getter returns null — and focus-trap throws on a null (vs false/undefined). Coerce the null to false at the source, with a regression test.

trasnake87 · 2026-06-06T17:45:35Z

@bernhardoj all five addressed — moved to tests/ui/BaseModalTest.tsx, switched to jest.resetModules(), typed the mock + captured value as ReanimatedModalProps | undefined, dropped the redundant small-screen mock (confirmed it still passes on the default 300px viewport), and applied your assertion phrasing. Green locally. Ready for another look.

bernhardoj

LGTM!

OSBotify · 2026-06-08T09:05:03Z

✋ This PR was not deployed to staging yet because QA is ongoing. It will be automatically deployed to staging after the next production release.

github-actions · 2026-06-08T09:06:32Z

🚧 @mountiny has triggered a test Expensify/App build. You can view the workflow run here.

github-actions · 2026-06-08T09:11:51Z

🧪🧪 Use the links below to test this adhoc build on Android, iOS, and Web. Happy testing! 🧪🧪
Built from App PR #92747.

Android 🤖	iOS 🍎
https://ad-hoc-expensify-cash.s3.us-east-1.amazonaws.com/rock-artifacts/ad-hoc/rock-android-Adhoc-c72d62d7d4e-3ea5527-34dc8bf8a8634d646c8e08399c08bbad5d4ce074/index.html	https://ad-hoc-expensify-cash.s3.us-east-1.amazonaws.com/rock-artifacts/ad-hoc/rock-ios-device-AdHoc-c72d62d-3ea5527-43be1f9f0b3f2ac24c705286b1959a30f1097a48/index.html

Web 🕸️
https://92747.pr-testing.expensify.com

👀 View the workflow run that generated this build 👀

OSBotify · 2026-06-09T04:35:27Z

🚀 Deployed to staging by https://github.com/mountiny in version: 9.4.1-0 🚀

platform	result
🕸 web 🕸	success ✅
🤖 android 🤖	success ✅
🍎 iOS 🍎	success ✅

Bundle Size Analysis (Sentry):

Android
iOS

MelvinBot · 2026-06-09T04:36:48Z

Help site review — no changes required

I reviewed the changes in this PR against Expensify's help site files under App/docs/articles, applying HELPSITE_NAMING_CONVENTIONS.md, HELP_AUTHORING_GUIDELINES.md, and TEMPLATE.md.

Conclusion: No help site documentation changes are required, so I did not create a draft PR.

Why

This is an internal, web-only bug fix with no user-facing behavior change:

It coerces a null initialFocus value to false in src/components/Modal/BaseModal.tsx so focus-trap no longer crashes when a bottom-docked modal's dismiss-button ref is unmounted before focus-trap reads it (Sentry APP-C26).
The only other change is added regression test coverage in tests/ui/BaseModalTest.tsx.

There are no new or renamed features, tabs, settings, buttons, or workflows — nothing that the customer-facing help site describes. The help articles document product features and how-to flows, none of which are affected by this crash fix.

@trasnake87, since no help site changes are required, there is no linked help site PR to review. If you believe a docs update is warranted, reply with details and I'll create one.

trasnake87 requested review from a team as code owners June 5, 2026 02:38

melvin-bot Bot assigned trasnake87 Jun 5, 2026

melvin-bot Bot requested review from bernhardoj and removed request for a team June 5, 2026 02:38

melvin-bot Bot requested review from JmillsExpensify and removed request for a team June 5, 2026 02:38

bernhardoj reviewed Jun 5, 2026

View reviewed changes

trasnake87 force-pushed the fix-initialfocus-crash-92415 branch from b33cdc5 to 6773e36 Compare June 5, 2026 03:34

JmillsExpensify reviewed Jun 5, 2026

View reviewed changes

trasnake87 force-pushed the fix-initialfocus-crash-92415 branch from 6773e36 to 866e1b7 Compare June 5, 2026 13:29

trasnake87 force-pushed the fix-initialfocus-crash-92415 branch from 866e1b7 to abf69d2 Compare June 6, 2026 00:15

bernhardoj reviewed Jun 6, 2026

View reviewed changes

trasnake87 force-pushed the fix-initialfocus-crash-92415 branch from abf69d2 to 286212f Compare June 6, 2026 05:25

bernhardoj reviewed Jun 6, 2026

View reviewed changes

trasnake87 force-pushed the fix-initialfocus-crash-92415 branch from 286212f to 1cb4a97 Compare June 6, 2026 17:45

bernhardoj approved these changes Jun 7, 2026

View reviewed changes

melvin-bot Bot requested a review from mountiny June 7, 2026 06:27

mountiny approved these changes Jun 8, 2026

View reviewed changes

mountiny merged commit c72d62d into Expensify:main Jun 8, 2026
33 checks passed

OSBotify mentioned this pull request Jun 9, 2026

Deploy Checklist: New Expensify 2026-06-09 #92988

Open

	// Tests will run with the app in a typical small screen size by default. We do this since the react-native test renderer
	// runs against index.native.js source and so anything that is testing a component reliant on withWindowDimensions()
	// would be most commonly assumed to be on a mobile phone vs. a tablet or desktop style view. This behavior can be
	// overridden by explicitly setting the dimensions inside a test via Dimensions.set()
	let dimensions: Record<string, number> = {
	width: 300,
	height: 700,
	scale: 1,
	fontScale: 1,
	};

-        const initialFocus = captured.initialFocus as (() => unknown) | undefined;
+        const initialFocus = captured?.initialFocus;
+expect(typeof initialFocus).toBe('function');
+// dismiss button never mounted -> ref.current is null -> the getter resolves to false (no crash)
+expect((initialFocus as () => unknown)()).toBe(false);

Conversation

trasnake87 commented Jun 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Explanation of Change

Fixed Issues

Tests

Offline tests

QA Steps

PR Author Checklist

Screenshots/Videos

Uh oh!

melvin-bot Bot commented Jun 5, 2026

Uh oh!

bernhardoj Jun 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

trasnake87 Jun 5, 2026

Choose a reason for hiding this comment

Uh oh!

bernhardoj commented Jun 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Reviewer Checklist

Screenshots/Videos

Uh oh!

bernhardoj commented Jun 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

JmillsExpensify left a comment

Choose a reason for hiding this comment

Uh oh!

trasnake87 commented Jun 6, 2026

Uh oh!

trasnake87 commented Jun 6, 2026

Uh oh!

bernhardoj Jun 6, 2026

Choose a reason for hiding this comment

Uh oh!

bernhardoj Jun 6, 2026

Choose a reason for hiding this comment

Uh oh!

bernhardoj commented Jun 6, 2026

Uh oh!

trasnake87 commented Jun 6, 2026

Uh oh!

bernhardoj Jun 6, 2026

Choose a reason for hiding this comment

Uh oh!

bernhardoj Jun 6, 2026

Choose a reason for hiding this comment

Uh oh!

bernhardoj Jun 6, 2026

Choose a reason for hiding this comment

Uh oh!

bernhardoj Jun 6, 2026

Choose a reason for hiding this comment

Uh oh!

bernhardoj commented Jun 6, 2026

Uh oh!

trasnake87 commented Jun 6, 2026

Uh oh!

bernhardoj left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

OSBotify commented Jun 8, 2026

Uh oh!

github-actions Bot commented Jun 8, 2026

Uh oh!

github-actions Bot commented Jun 8, 2026

Uh oh!

OSBotify commented Jun 9, 2026

Uh oh!

MelvinBot commented Jun 9, 2026

Help site review — no changes required

Uh oh!

Reviewers

Assignees

Labels

Projects

trasnake87 commented Jun 5, 2026 •

edited

Loading

bernhardoj Jun 5, 2026 •

edited

Loading

bernhardoj commented Jun 5, 2026 •

edited

Loading

bernhardoj commented Jun 5, 2026 •

edited

Loading