[NoQA] [Audit][Implementation] - Increase max cache keys count and use waitForCollectionCallback #38464

hurali97 · 2024-03-18T11:49:42Z

Details

This PR is part of the implementation phase of App Startup Audit by Callstack.

Originally the proposal is here but after thinking it thoroughly, we decided to follow a better solution.

We have lots of connections to Reports data in Onyx in our utils files and some of them uses waitForCollectionCallback: true and the rest of them doesn't. Which adds up to the app startup time. The reason is that when we don't set waitForCollectionCallback to true in Onyx.js the following code is executed:

if (mapping.waitForCollectionCallback) {
    getCollectionDataAndSendAsObject(matchingKeys, mapping);
    return;
}

 // We did not opt into using waitForCollectionCallback mode so the callback is called for every matching key.
for (let i = 0; i < matchingKeys.length; i++) {
    get(matchingKeys[i]).then((val) => sendDataToConnection(mapping, val, matchingKeys[i], true));
}

This loop will be executed for all the keys and will individually inform the subscriber of updates. The get function first checks whether the value for a key is present in cache and if it doesn't it then gets it from the storage.

We have 6 connections to Reports without the waitForCollectionCallback so this loop will be executed 6 times. Bearing in mind that Reports can be a huge collection, for eg on a heavy account it can get to 15k reports.

The solution is to refactor our current Onyx.connect calls to Reports to include waitForCollectionCallback and adjust the code accordingly. There's one place where we won't be changing it and that's in ReportUtils.ts where we have to call handleReportChanged on each report.

One more thing to notice is that currently our cache limit is 10k which is easily hit by a heavy account of 15k reports, so this cache limit isn't helping the app in case of a heavy account. We can increase it to 20k so that it can hold more reports.

Below are Hermes traces that have been gathered on:

a real android device
with release variant of Expensify App
on a heavy account

Baseline

With waitForCollectionCallback

`

With increased cache and waitForCollectionCallback

To summarise, When launching the app on Android:

We get 32 seconds on baseline, which is what we currently have 🔴
25 seconds with improvements by adding waitForCollectionCallback 🟢
18 seconds with increased cache to 20k and improvements by adding waitForCollectionCallback 🟢 🟢

Fixed Issues

$ #38586
PROPOSAL:

Tests

Verify that no errors appear in the JS console

Offline tests

QA Steps

Verify that no errors appear in the JS console

PR Author Checklist

Screenshots/Videos

Android: Native

android.mov

Android: mWeb Chrome

android-web.mov

iOS: Native

ios.mov

iOS: mWeb Safari

ios-web.mov

MacOS: Chrome / Safari

web.mov

MacOS: Desktop

desktop.mov

kacper-mikolajczak

LGTM 👍

Question @hurali97, what's stopping us from having single connection to ONYXKEYS.COLLECTION.REPORT, which would result in single instance of reports object that then later could be shared across the app?

hurali97 · 2024-03-19T08:36:43Z

what's stopping us from having single connection to ONYXKEYS.COLLECTION.REPORT, which would result in single instance of reports object that then later could be shared across the app?

This is more complicated then it looks like. A straight forward solution is to implement something like this. This will keep only two connections to Onyx, one for waitForCollectionCallback and second without it. It uses a singleton based approach with event driven updates.

However, this will get more complicated over the time as we'll have to do the same for each of the collection, for eg, ReportAction.

I also evaluated whether we can update some parts of Onyx to have only one connection per key but that's easier said than done. The reason is that we have callbacks with each connection and even if we manage to keep only one connection per key, we will have to store the callbacks like below:

{
key: 'report',
callbacks: [
  {
   callback: () => {},
   waitForConnectionCallback: false,
   ...otherProps,
  },
  {
    callback: () => {},
    waitForConnectionCallback: true,
    ...otherProps,
  }
 ]
}

Even with above structure, we will have to loop over the callbacks to fire them whenever Report changes.

Now, given all of the above I think it's better to avoid the individual loop within the Onyx for the larger keys. Which means reduce the chances of iterating over the keys and getting them from cache, then from storage by using waitForCollectionCallback: true.

kacper-mikolajczak · 2024-03-19T09:29:57Z

What I had in mind is starting small. The passive usages of Report collection, the ones where we connect and callback only serves us to dump fresh data into local variable. An example of it:

App/src/libs/OptionsListUtils.ts

Lines 251 to 259 in bbf4bee

    
           let policyExpenseReports: OnyxCollection<Report> = {}; 
        
           Onyx.connect({ 
        
               key: ONYXKEYS.COLLECTION.REPORT, 
        
               waitForCollectionCallback: true, 
        
               callback: (report) => { 
        
                   policyExpenseReports = report; 
        
               }, 
        
           });

Could be replaced with single instance which we could import in many places. This would reduce code complexity as all passive connection would be explicitly handled in one place, e.g. OnyxStore.ts file.

For interactive usages (where live callback is needed), then we could implement it with Observer pattern like so (presented in simple form for brevity):

const store = {};
const listeners = new Set();

Onyx.connect({
   key: 'some_key',
   callback: (data) => {
       store = data;
       listeners.forEach(l => l(store));
   },
});

const addListener = (l) => { 
   listeners.add(l);
   return () => listeners.remove(l)
}

// Rest of the module

export { addListener, removeListener, data }

This is more complex topic but the main take-away is that, as you pointed out in your analysis, we have many redundant connections scattered across the app which builds up both mental and perf overheads.

Question is, the Onyx.connect itself is kind of an observer, so I am concerned such solution adds another layer of abstraction where it is not needed. In this case, is it possible to implement it in Onyx itself?

hurali97 · 2024-03-19T09:50:10Z

@kacper-mikolajczak I see. I implemented something similar when working in analysis phase here. What you’ve proposed is more or less the same thing.

The only issue is the complexity that this solution brings. Just see that we will have to do this for each key, if we want to use it. Also, in my profiling with the changes in the PR, I realise that this single connection approach doesn’t bring much of an improvement. So I guess with these improvements, this doesn’t make sense to introduce more complexity just for the sake of single connection because it’s not hurting the performance, courtesy of the changes added in the PR now :thinking_face:

Previously, what you’ve mentioned was my go to approach but I guess we have something better now, so I will keep it until in the future we encounter it again then we can continue the discussion.

kacper-mikolajczak · 2024-03-19T09:51:11Z

Great, thanks for the explanation! We can definitely continue the discussion :) GJ on the PR 🚀

melvin-bot · 2024-03-19T13:01:37Z

@grgia 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]

src/libs/OptionsListUtils.ts

src/libs/actions/Report.ts

src/libs/OptionsListUtils.ts

…y-App into audit/app-startup/increase-cache-and-add-waitforcollectioncallback

hurali97 · 2024-03-22T11:26:14Z

@shubham1206agra any updates on the review? 🙂

shubham1206agra · 2024-03-22T11:30:11Z

@hurali97 I thought you were going to implement https://github.com/Expensify/App/pull/38464/files#r1531973454. Let me ping @roryabraham here to confirm if we can proceed with this solution.

hurali97 · 2024-03-22T12:52:25Z

@hurali97 I thought you were going to implement https://github.com/Expensify/App/pull/38464/files#r1531973454. Let me ping @roryabraham here to confirm if we can proceed with this solution.

@shubham1206agra Yeah I am willing to implement this just waiting for a green light from the reviewers 👍

shubham1206agra · 2024-03-25T12:54:33Z

@roryabraham Friendly bump here.

src/libs/OptionsListUtils.ts

shubham1206agra · 2024-03-25T17:00:10Z

@grgia Can you comment on #38464 (comment) instead if possible?

shubham1206agra · 2024-03-27T09:53:11Z

@hurali97 Shouldn't the same thing applies to getReportOption?

hurali97 · 2024-03-27T10:07:40Z

@hurali97 Shouldn't the same thing applies to getReportOption?

@shubham1206agra I think no, because the getReportOption is using participant to access the report from allReports. If getReportOption was being passed the report as parameter then we could have done the same here as well.

Edit: Yeah sure thing 👍 I didn't see that allReports is not used now, so we can just remove the connect altogether.

shubham1206agra · 2024-03-27T10:27:30Z

@hurali97 You may have been misguided by the type. You may want to recheck the usage of getPolicyExpenseReportOption.

hurali97 · 2024-03-27T10:44:31Z

@hurali97 You may have been misguided by the type. You may want to recheck the usage of getPolicyExpenseReportOption.

@shubham1206agra Yeah I was because I checked before merging the main. Anyways, things are resolved now 👍

shubham1206agra · 2024-03-27T11:01:00Z

@hurali97 You may have been misguided by the type. You may want to recheck the usage of getPolicyExpenseReportOption.

@shubham1206agra Yeah I was because I checked before merging the main. Anyways, things are resolved now 👍

@hurali97 No, you are misguided now. The correct type of input is Participant only :).

hurali97 · 2024-03-27T11:06:11Z

@shubham1206agra Oh.... so you mean I should rename report to participant in getPolicyExpenseReportOption ?

shubham1206agra · 2024-03-27T11:20:57Z

Yes, and update the function accordingly.

hurali97 · 2024-03-27T11:28:19Z

Yes, and update the function accordingly.

I am understanding that we can rename the parameter report to participant in getPolicyExpenseReportOption but what are you referring to update besides it?

Are you referring to add ReportUtils.getReport(participant.reportID) in getPolicyExpenseReportOption to get the report using this helper method? Something like below:

function getPolicyExpenseReportOption(participant: Participant): ReportUtils.OptionData {
    const expenseReport = ReportUtils.isPolicyExpenseChat(participant) ? ReportUtils.getReport(participant.reportID) : null;

....

shubham1206agra · 2024-03-27T11:56:15Z

@hurali97 Yes

src/libs/OptionsListUtils.ts

shubham1206agra · 2024-03-27T16:05:44Z

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 tests pass on all platforms & I tested again on:
- Android: Native
- Android: mWeb Chrome
- iOS: Native
- iOS: mWeb Safari
- MacOS: Chrome / Safari
- MacOS: Desktop
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 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 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.js 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.
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: Native

Screen.Recording.2024-03-27.at.9.26.31.PM.mov

Android: mWeb Chrome

Screen.Recording.2024-03-27.at.9.06.25.PM.mov

iOS: Native

Screen.Recording.2024-03-27.at.9.14.18.PM.mov

iOS: mWeb Safari

Screen.Recording.2024-03-27.at.9.00.42.PM.mov

MacOS: Chrome / Safari

Screen.Recording.2024-03-27.at.8.11.53.PM.mov

MacOS: Desktop

Screen.Recording.2024-03-27.at.9.09.48.PM.mov

Co-authored-by: Shubham Agrawal <58412969+shubham1206agra@users.noreply.github.com>

mountiny

Thank you!

mountiny · 2024-03-28T11:28:53Z

Going to go ahead since C+ reviewed and Georgia approved before too

melvin-bot · 2024-03-28T11:29:08Z

@mountiny looks like this was merged without a test passing. Please add a note explaining why this was done and remove the Emergency label if this is not an emergency.

mountiny · 2024-03-28T12:59:51Z

Melvin the gaslighter the tests were passing

OSBotify · 2024-03-29T02:27:11Z

🚀 Deployed to staging by https://github.com/mountiny in version: 1.4.58-0 🚀

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

OSBotify · 2024-04-02T11:03:27Z

🚀 Deployed to production by https://github.com/Beamanator in version: 1.4.58-8 🚀

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

getusha · 2024-05-10T04:23:40Z

src/libs/TaskUtils.ts

@@ -42,11 +38,11 @@ function getTaskReportActionMessage(action: OnyxEntry<ReportAction>): Pick<Messa
 }

 function getTaskTitle(taskReportID: string, fallbackTitle = ''): string {
-    const taskReport = allReports[taskReportID] ?? {};
+    const taskReport = allReports?.[taskReportID] ?? {};


This should've been updated to const taskReport = allReports?.[${ONYXKEYS.COLLECTION.REPORT}${taskReportID}] ?? {}; after updating how we key the reports. it caused a regression #40121 More information: #40121 (comment)

hurali97 added 5 commits March 18, 2024 16:45

perf: use waitForCollectionCallback

c826b13

perf: remove redundant report connection

decdb46

perf: increase max cached keys count

a9f7d46

fix: linting

8078034

fix: add disable eslint for handleReportChanged

bbf4bee

kacper-mikolajczak approved these changes Mar 19, 2024

View reviewed changes

hurali97 marked this pull request as ready for review March 19, 2024 13:01

hurali97 requested a review from a team as a code owner March 19, 2024 13:01

melvin-bot bot requested review from grgia and removed request for a team March 19, 2024 13:01

muttmuure mentioned this pull request Mar 19, 2024

[HOLD for payment 2024-04-09] [$500] [Performance] Refactor our current Onyx.connect calls to Reports to include waitForCollectionCallback #38586

Closed

mountiny assigned hurali97 Mar 19, 2024

shubham1206agra reviewed Mar 19, 2024

View reviewed changes

src/libs/OptionsListUtils.ts Outdated Show resolved Hide resolved

roryabraham reviewed Mar 19, 2024

View reviewed changes

src/libs/OptionsListUtils.ts Outdated Show resolved Hide resolved

src/libs/OptionsListUtils.ts Outdated Show resolved Hide resolved

src/libs/actions/Report.ts Show resolved Hide resolved

src/libs/OptionsListUtils.ts Outdated Show resolved Hide resolved

hurali97 added 3 commits March 20, 2024 15:12

Merge branch 'main' of https://github.com/callstack-internal/Expensif…

2fe89d4

…y-App into audit/app-startup/increase-cache-and-add-waitforcollectioncallback

refactor: update report to reports

c9cb60a

Merge branch 'main' of https://github.com/callstack-internal/Expensif…

3086cfc

…y-App into audit/app-startup/increase-cache-and-add-waitforcollectioncallback

shubham1206agra reviewed Mar 25, 2024

View reviewed changes

src/libs/OptionsListUtils.ts Outdated Show resolved Hide resolved

refactor: remove unnecessary variables for reports

95ae11e

grgia previously approved these changes Mar 25, 2024

View reviewed changes

refactor: remove not needed Onyx.connect for Report collection

c4b6d85

refactor: use correct types

11917c4

hungvu193 mentioned this pull request Mar 27, 2024

perf: Cache search options #38207

Merged

50 tasks

shubham1206agra reviewed Mar 27, 2024

View reviewed changes

src/libs/OptionsListUtils.ts Outdated Show resolved Hide resolved

refactor: use participant instead of expenseReport

acc3455

Co-authored-by: Shubham Agrawal <58412969+shubham1206agra@users.noreply.github.com>

shubham1206agra approved these changes Mar 28, 2024

View reviewed changes

melvin-bot bot requested a review from mountiny March 28, 2024 08:48

mountiny approved these changes Mar 28, 2024

View reviewed changes

mountiny merged commit 5955e99 into Expensify:main Mar 28, 2024
16 checks passed

melvin-bot bot added the Emergency label Mar 28, 2024

mountiny changed the title ~~[Audit][Implementation] - Increase max cache keys count and use waitForCollectionCallback~~ [NoQA] [Audit][Implementation] - Increase max cache keys count and use waitForCollectionCallback Mar 28, 2024

mountiny removed the Emergency label Mar 28, 2024

github-actions bot mentioned this pull request Mar 29, 2024

Deploy Checklist: New Expensify 2024-03-29 #39222

Closed

bernhardoj mentioned this pull request Apr 12, 2024

[HOLD for payment 2024-05-08] [$250] Task – When task name is changed offline, the task's chat report in the LHN doesn't update after going back online #40121

Closed

6 tasks

getusha reviewed May 10, 2024

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[NoQA] [Audit][Implementation] - Increase max cache keys count and use waitForCollectionCallback #38464

[NoQA] [Audit][Implementation] - Increase max cache keys count and use waitForCollectionCallback #38464

hurali97 commented Mar 18, 2024 •

edited by mountiny

Loading

kacper-mikolajczak left a comment

hurali97 commented Mar 19, 2024

kacper-mikolajczak commented Mar 19, 2024 •

edited

Loading

hurali97 commented Mar 19, 2024

kacper-mikolajczak commented Mar 19, 2024 •

edited

Loading

melvin-bot bot commented Mar 19, 2024

hurali97 commented Mar 22, 2024

shubham1206agra commented Mar 22, 2024

hurali97 commented Mar 22, 2024

shubham1206agra commented Mar 25, 2024

shubham1206agra commented Mar 25, 2024 •

edited

Loading

shubham1206agra commented Mar 27, 2024

hurali97 commented Mar 27, 2024 •

edited

Loading

shubham1206agra commented Mar 27, 2024

hurali97 commented Mar 27, 2024

shubham1206agra commented Mar 27, 2024

hurali97 commented Mar 27, 2024

shubham1206agra commented Mar 27, 2024

hurali97 commented Mar 27, 2024

shubham1206agra commented Mar 27, 2024

shubham1206agra commented Mar 27, 2024

mountiny left a comment

mountiny commented Mar 28, 2024

melvin-bot bot commented Mar 28, 2024

mountiny commented Mar 28, 2024

OSBotify commented Mar 29, 2024

OSBotify commented Apr 2, 2024

getusha May 10, 2024

[NoQA] [Audit][Implementation] - Increase max cache keys count and use waitForCollectionCallback #38464

[NoQA] [Audit][Implementation] - Increase max cache keys count and use waitForCollectionCallback #38464

Conversation

hurali97 commented Mar 18, 2024 • edited by mountiny Loading

Details

Fixed Issues

Tests

Offline tests

QA Steps

PR Author Checklist

Screenshots/Videos

kacper-mikolajczak left a comment

Choose a reason for hiding this comment

hurali97 commented Mar 19, 2024

kacper-mikolajczak commented Mar 19, 2024 • edited Loading

hurali97 commented Mar 19, 2024

kacper-mikolajczak commented Mar 19, 2024 • edited Loading

melvin-bot bot commented Mar 19, 2024

hurali97 commented Mar 22, 2024

shubham1206agra commented Mar 22, 2024

hurali97 commented Mar 22, 2024

shubham1206agra commented Mar 25, 2024

shubham1206agra commented Mar 25, 2024 • edited Loading

shubham1206agra commented Mar 27, 2024

hurali97 commented Mar 27, 2024 • edited Loading

shubham1206agra commented Mar 27, 2024

hurali97 commented Mar 27, 2024

shubham1206agra commented Mar 27, 2024

hurali97 commented Mar 27, 2024

shubham1206agra commented Mar 27, 2024

hurali97 commented Mar 27, 2024

shubham1206agra commented Mar 27, 2024

shubham1206agra commented Mar 27, 2024

Reviewer Checklist

Screenshots/Videos

mountiny left a comment

Choose a reason for hiding this comment

mountiny commented Mar 28, 2024

melvin-bot bot commented Mar 28, 2024

mountiny commented Mar 28, 2024

OSBotify commented Mar 29, 2024

OSBotify commented Apr 2, 2024

getusha May 10, 2024

Choose a reason for hiding this comment

hurali97 commented Mar 18, 2024 •

edited by mountiny

Loading

kacper-mikolajczak commented Mar 19, 2024 •

edited

Loading

kacper-mikolajczak commented Mar 19, 2024 •

edited

Loading

shubham1206agra commented Mar 25, 2024 •

edited

Loading

hurali97 commented Mar 27, 2024 •

edited

Loading