Skip to content

Comments

Remove macro/micro tasks during subscriber update#724

Open
VickyStash wants to merge 22 commits intoExpensify:mainfrom
callstack-internal:VickyStash/poc/get-rid-of-macro-micro-tasks
Open

Remove macro/micro tasks during subscriber update#724
VickyStash wants to merge 22 commits intoExpensify:mainfrom
callstack-internal:VickyStash/poc/get-rid-of-macro-micro-tasks

Conversation

@VickyStash
Copy link
Contributor

@VickyStash VickyStash commented Jan 23, 2026

Details

Check the issue description for details.

Related Issues

$ Expensify/App#82871

Automated Tests

Should be covered by existing tests

Manual Tests

The E/App should work the same way as before. Let's verify basic test steps:

  1. Login with a account.
  2. Navigate to a chat, send a message. React to that message.
  3. Reply to that message in a new thread.
  4. Create a new task, navigate to its details and mark it as completed.
  5. Create a new expense to someone else, navigate to its details.
  6. Go to Reports tab, Expenses, make sure you can find your expense and see its details.
  7. Create a new workspace, make sure the rooms were created.

Author Checklist

  • I linked the correct issue in the ### Related Issues section above
  • I wrote clear testing steps that cover the changes made in this PR
    • I added steps for local testing in the Tests section
    • I tested this PR with a High Traffic account against the staging or production API to ensure there are no regressions (e.g. long loading states that impact usability).
  • I included screenshots or videos for tests on all platforms
  • I ran the tests on all platforms & verified they passed on:
    • Android / native
    • Android / Chrome
    • iOS / native
    • iOS / Safari
    • MacOS / Chrome / Safari
    • MacOS / Desktop
  • I verified there are no console errors (if there's a console error not related to the PR, report it or open an issue for it to be fixed)
  • I followed proper code patterns (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 the left part of a conditional rendering a React component is a boolean and NOT a string, e.g. myBool && <MyComponent />.
    • 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 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 followed the guidelines as stated in the Review Guidelines
  • I tested other components that can be impacted by my changes (i.e. if the PR modifies a shared library or component like Avatar, I verified the components using Avatar are working as expected)
  • 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.js or at the top of the file that uses the constant) are defined as such
  • I verified that if a function's arguments changed that all usages have also been updated correctly
  • 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
    • If we are not using the full Onyx data that we loaded, I've added the proper selector in order to ensure the component only re-renders when the data it is using changes
    • 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 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 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 author checklist, including those that don't apply to this PR.

Screenshots/Videos

Android: Native
android.mp4
Android: mWeb Chrome
android_web.mp4
iOS: Native
ios.mov
iOS: mWeb Safari
ios_web.mov
MacOS: Chrome / Safari
web.mp4

@koko57
Copy link
Contributor

koko57 commented Feb 16, 2026

LGTM

Copy link
Contributor

@kubabutkiewicz kubabutkiewicz left a comment

Choose a reason for hiding this comment

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

lgtm!

@VickyStash VickyStash changed the title [POC] Remove macro/micro tasks during subscriber update Remove macro/micro tasks during subscriber update Feb 20, 2026
@VickyStash VickyStash marked this pull request as ready for review February 20, 2026 08:55
@VickyStash VickyStash requested a review from a team as a code owner February 20, 2026 08:55
@melvin-bot melvin-bot bot requested review from grgia and removed request for a team February 20, 2026 08:56
Copy link

@chatgpt-codex-connector chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

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

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 6318f5ef65

ℹ️ About Codex in GitHub

Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

if (lastConnectionCallbackData.has(mapping.subscriptionID) && valueToPass === lastValue) {
// If the subscriber was already notified (e.g. by a synchronous keyChanged call),
// skip the initial data delivery to prevent duplicate callbacks.
if (lastConnectionCallbackData.has(mapping.subscriptionID)) {

Choose a reason for hiding this comment

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

P1 Badge Preserve initial hydration after synchronous updates

Keep sendDataToConnection() from returning solely on lastConnectionCallbackData.has(...); this now drops the initial hydration callback whenever any synchronous keyChanged/keysChanged ran first. In the common race where Onyx.connect() is followed by an immediate Onyx.set() in the same tick, the subscription gets marked as "already notified" and the later storage-backed init payload is skipped even if it contains additional data (especially for collection subscribers), leaving subscribers with a partial state until a future update arrives.

Useful? React with 👍 / 👎.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

In this case, by the time the hydration promise resolves, it would only have equal or older data, making the skip correct.

Copy link
Contributor

Choose a reason for hiding this comment

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

@VickyStash I asked Claude about this comment and according to it it's valid, here's a unit test it designed for me

diff --git a/tests/unit/onyxTest.ts b/tests/unit/onyxTest.ts
index 64caec5..94aff05 100644
--- a/tests/unit/onyxTest.ts
+++ b/tests/unit/onyxTest.ts
@@ -140,6 +140,51 @@ describe('Onyx', () => {
         });
     });
 
+    it('should deliver full collection data when connect() is followed by immediate set() of a single member in the same tick', () => {
+        const mockCallback = jest.fn();
+        const collectionKey = ONYX_KEYS.COLLECTION.TEST_CONNECT_COLLECTION;
+
+        // Write collection members directly to storage, bypassing Onyx cache values.
+        // This simulates data that exists in persistent storage but hasn't been loaded into cache yet
+        // (e.g. from a previous session).
+        return StorageMock.setItem(`${collectionKey}1`, {ID: 1, value: 'one'})
+            .then(() => StorageMock.setItem(`${collectionKey}2`, {ID: 2, value: 'two'}))
+            .then(() => StorageMock.setItem(`${collectionKey}3`, {ID: 3, value: 'three'}))
+            .then(() => {
+                // Register the keys in Onyx's key cache so getAllKeys() can discover them.
+                // We intentionally do NOT add values to the cache — only keys — to simulate
+                // data that is known to exist but whose values haven't been hydrated yet.
+                cache.addKey(`${collectionKey}1`);
+                cache.addKey(`${collectionKey}2`);
+                cache.addKey(`${collectionKey}3`);
+
+                // Connect to the collection — this starts an async storage read (microtask)
+                connection = Onyx.connect({
+                    key: collectionKey,
+                    waitForCollectionCallback: true,
+                    callback: mockCallback,
+                });
+
+                // Immediately set a single member in the same synchronous tick.
+                // This triggers synchronous keyChanged() which calls the subscriber with a partial
+                // collection (just _1 from the cache). This sets lastConnectionCallbackData for this
+                // subscriber. The async hydration from subscribeToKey should still deliver the full
+                // collection afterwards, since the data is different.
+                Onyx.set(`${collectionKey}1`, {ID: 1, value: 'updated'});
+
+                // Wait for all async operations (storage reads from subscribeToKey) to complete
+                return waitForPromisesToResolve();
+            })
+            .then(() => {
+                // The subscriber's final state should contain ALL collection members,
+                // including _2 and _3 that were only in storage (not cache) at the time of the synchronous keyChanged call.
+                const lastCall = mockCallback.mock.calls[mockCallback.mock.calls.length - 1][0];
+                expect(lastCall).toHaveProperty(`${collectionKey}1`);
+                expect(lastCall).toHaveProperty(`${collectionKey}2`);
+                expect(lastCall).toHaveProperty(`${collectionKey}3`);
+            });
+    });
+
     it('should merge an object with another object', () => {
         let testKeyValue: unknown;
 

Could you validate if it makes sense?

@Krishna2323
Copy link

Reviewing...

@Krishna2323
Copy link

Krishna2323 commented Feb 20, 2026

Encountered a console error — checking whether it’s related to the PR changes. Also repro on main.

@Krishna2323
Copy link

@VickyStash I believe this might be related to the changes. Could you please take a look? Thanks!

image

@VickyStash
Copy link
Contributor Author

@VickyStash I believe this might be related to the changes. Could you please take a look? Thanks!

Yeah, it looks so.
This Onyx PR removes macro/micro task deferral during subscriber updates, making notifications synchronous.
The updateLastAccessedWorkspace inside WithPolicy HOC is called right during the render and it makes it trigger updates in other components within the same render cycle, which React doesn't allow.
The fix can be just wrapping the call into useEffect. But I'm thinking if there are any other places in the app like this.
cc @fabioh8010

@VickyStash
Copy link
Contributor Author

VickyStash commented Feb 20, 2026

I've run checks with Claude and I haven't found other places like that, though it's not 100% guarantee

Summary

Problem

updateLastAccessedWorkspace(policyID) in src/pages/workspace/withPolicy.tsx:94 is called directly in the render body of the WithPolicy HOC. This function calls Onyx.set(), which is a side effect that violates React's rule that rendering must be pure.

Why it's a problem

• React may re-render components multiple times, discard renders, or pause them (Strict Mode, concurrent features). Side effects in the render body can execute unpredictably.
• It triggers the error: "Cannot update a component while rendering a different component" — because Onyx.set() synchronously notifies subscribers, causing state updates in other components mid-render.

Why it wasn't caught before

The code was always technically wrong, but previously Onyx.set() likely deferred or batched its subscriber notifications in a way that didn't trigger a synchronous re-render during the calling component's render phase. The Onyx update changed the notification path to be more synchronous, exposing this pre-existing bug.

Scope of the issue

After searching the entire codebase across HOCs, custom hooks, providers, and all components importing from @userActions/— this is the only instance of this anti-pattern. All other action/Onyx calls are properly wrapped in useEffect, useCallback, or event handlers.

Fix

Wrap the call in useEffect:

useEffect(() => {
    if (policyID && policyID.length > 0) {
        updateLastAccessedWorkspace(policyID);
    }
}, [policyID]);

@VickyStash VickyStash changed the title Remove macro/micro tasks during subscriber update [HOLD on Fabio review]Remove macro/micro tasks during subscriber update Feb 20, 2026
@fabioh8010
Copy link
Contributor

I asked @VickyStash to hold this PR because of this plan to handle the canBeMissing situation -> Expensify/App#80095 (comment)

@VickyStash VickyStash changed the title [HOLD on Fabio review]Remove macro/micro tasks during subscriber update Remove macro/micro tasks during subscriber update Feb 23, 2026
@VickyStash
Copy link
Contributor Author

Not holded anymore!

@mountiny
Copy link
Contributor

@Krishna2323 can you please test this pr with App?

@Krishna2323
Copy link

Will be on it in an hour or two.

@Krishna2323
Copy link

Reviewing...

@VickyStash
Copy link
Contributor Author

@Krishna2323 Please, take another look!

@Krishna2323
Copy link

Krishna2323 commented Feb 24, 2026

@VickyStash could you please check this comment? Sorry about the wrong suggestion. 🙏🏻

@Krishna2323
Copy link

I’ll try to finish the review by tomorrow — I had to step out for a doctor’s appointment. Thanks for the patience.

Comment on lines 688 to 707
if (isSubscribedToCollectionKey) {
if (subscriber.waitForCollectionCallback) {
lastConnectionCallbackData.set(subscriber.subscriptionID, cachedCollection);
subscriber.callback(cachedCollection, subscriber.key, partialCollection);
continue;
}

// If they are not using waitForCollectionCallback then we notify the subscriber with
// the new merged data but only for any keys in the partial collection.
const dataKeys = Object.keys(partialCollection ?? {});
for (const dataKey of dataKeys) {
if (deepEqual(cachedCollection[dataKey], previousCollection[dataKey])) {
continue;
}

subscriber.callback(cachedCollection[dataKey], dataKey);
}
lastConnectionCallbackData.set(subscriber.subscriptionID, cachedCollection);
continue;
}
Copy link
Contributor

Choose a reason for hiding this comment

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

diff --git a/lib/OnyxUtils.ts b/lib/OnyxUtils.ts
index dd08e6b..06d17c1 100644
--- a/lib/OnyxUtils.ts
+++ b/lib/OnyxUtils.ts
@@ -686,8 +686,9 @@ function keysChanged<TKey extends CollectionKeyBase>(
             // If they are subscribed to the collection key and using waitForCollectionCallback then we'll
             // send the whole cached collection.
             if (isSubscribedToCollectionKey) {
+                lastConnectionCallbackData.set(subscriber.subscriptionID, cachedCollection);
+
                 if (subscriber.waitForCollectionCallback) {
-                    lastConnectionCallbackData.set(subscriber.subscriptionID, cachedCollection);
                     subscriber.callback(cachedCollection, subscriber.key, partialCollection);
                     continue;
                 }
@@ -702,7 +703,6 @@ function keysChanged<TKey extends CollectionKeyBase>(
 
                     subscriber.callback(cachedCollection[dataKey], dataKey);
                 }
-                lastConnectionCallbackData.set(subscriber.subscriptionID, cachedCollection);
                 continue;
             }
 

Maybe we can just do lastConnectionCallbackData.set(subscriber.subscriptionID, cachedCollection); right inside if (isSubscribedToCollectionKey) { since it will be executed in all situations inside this condition?

if (lastConnectionCallbackData.has(mapping.subscriptionID) && valueToPass === lastValue) {
// If the subscriber was already notified (e.g. by a synchronous keyChanged call),
// skip the initial data delivery to prevent duplicate callbacks.
if (lastConnectionCallbackData.has(mapping.subscriptionID)) {
Copy link
Contributor

Choose a reason for hiding this comment

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

@VickyStash I asked Claude about this comment and according to it it's valid, here's a unit test it designed for me

diff --git a/tests/unit/onyxTest.ts b/tests/unit/onyxTest.ts
index 64caec5..94aff05 100644
--- a/tests/unit/onyxTest.ts
+++ b/tests/unit/onyxTest.ts
@@ -140,6 +140,51 @@ describe('Onyx', () => {
         });
     });
 
+    it('should deliver full collection data when connect() is followed by immediate set() of a single member in the same tick', () => {
+        const mockCallback = jest.fn();
+        const collectionKey = ONYX_KEYS.COLLECTION.TEST_CONNECT_COLLECTION;
+
+        // Write collection members directly to storage, bypassing Onyx cache values.
+        // This simulates data that exists in persistent storage but hasn't been loaded into cache yet
+        // (e.g. from a previous session).
+        return StorageMock.setItem(`${collectionKey}1`, {ID: 1, value: 'one'})
+            .then(() => StorageMock.setItem(`${collectionKey}2`, {ID: 2, value: 'two'}))
+            .then(() => StorageMock.setItem(`${collectionKey}3`, {ID: 3, value: 'three'}))
+            .then(() => {
+                // Register the keys in Onyx's key cache so getAllKeys() can discover them.
+                // We intentionally do NOT add values to the cache — only keys — to simulate
+                // data that is known to exist but whose values haven't been hydrated yet.
+                cache.addKey(`${collectionKey}1`);
+                cache.addKey(`${collectionKey}2`);
+                cache.addKey(`${collectionKey}3`);
+
+                // Connect to the collection — this starts an async storage read (microtask)
+                connection = Onyx.connect({
+                    key: collectionKey,
+                    waitForCollectionCallback: true,
+                    callback: mockCallback,
+                });
+
+                // Immediately set a single member in the same synchronous tick.
+                // This triggers synchronous keyChanged() which calls the subscriber with a partial
+                // collection (just _1 from the cache). This sets lastConnectionCallbackData for this
+                // subscriber. The async hydration from subscribeToKey should still deliver the full
+                // collection afterwards, since the data is different.
+                Onyx.set(`${collectionKey}1`, {ID: 1, value: 'updated'});
+
+                // Wait for all async operations (storage reads from subscribeToKey) to complete
+                return waitForPromisesToResolve();
+            })
+            .then(() => {
+                // The subscriber's final state should contain ALL collection members,
+                // including _2 and _3 that were only in storage (not cache) at the time of the synchronous keyChanged call.
+                const lastCall = mockCallback.mock.calls[mockCallback.mock.calls.length - 1][0];
+                expect(lastCall).toHaveProperty(`${collectionKey}1`);
+                expect(lastCall).toHaveProperty(`${collectionKey}2`);
+                expect(lastCall).toHaveProperty(`${collectionKey}3`);
+            });
+    });
+
     it('should merge an object with another object', () => {
         let testKeyValue: unknown;
 

Could you validate if it makes sense?

@mountiny
Copy link
Contributor

Thanks @Krishna2323 feel free to take time off too

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants