Skip to content
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

[BUGFIX beta] fix for lastObject/firstObject update issue #15510

Merged
merged 1 commit into from
Jul 18, 2017

Conversation

bekzod
Copy link
Contributor

@bekzod bekzod commented Jul 16, 2017

enhanced #15110 PR, @stefanpenner

@locks locks requested review from stefanpenner and krisselden July 16, 2017 11:41
propertyWillChange(array, 'lastObject', meta);
propertyDidChange(array, 'lastObject', meta);
if (cache !== undefined) {
if (cache.firstObject !== undefined && startIdx === 0 && objectAt(array, 0) !== cacheFor.get(cache, 'firstObject')) {
Copy link
Member

Choose a reason for hiding this comment

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

i think one of the goals @hjdivad had, was to avoid the objectAt(x) lookups, as they would force the CP to be accessed.

I'll ping @hjdivad to chime in.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

yeah but you need to invalidate cp when it changes no ?

Copy link
Member

Choose a reason for hiding this comment

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

If the index for firstObject / lastObject changed, then we should broadcast. Even if the object is equal. That means we wont need both the objectAt stuff and the range based detection.

@bekzod bekzod force-pushed the array-first-last-object branch from 2220610 to 45ebc8f Compare July 17, 2017 14:27
@bekzod
Copy link
Contributor Author

bekzod commented Jul 17, 2017

I updated my PR, so it does not call objectAt but now behavior is changed of firstObject/lastObject change event, so now it notifies change even if same object as firstObject/lastObject is shifted/pushed to array , I think it is still better than how was before.
adjusted/added test accordingly @stefanpenner

@bekzod bekzod force-pushed the array-first-last-object branch 2 times, most recently from 2754363 to b036634 Compare July 17, 2017 14:40
@stefanpenner
Copy link
Member

@bekzod ya, that was the approach @hjdivad was taking. It is likely a trade-off we are good with.

}
if (cache !== undefined) {
if (cache.firstObject !== undefined && startIdx === 0) {
propertyWillChange(array, 'firstObject');
Copy link
Member

Choose a reason for hiding this comment

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

This PR drops && (addAmt > 0 || removeAmt > 0) because we can safely assume one of those must be > 0?

Copy link
Member

Choose a reason for hiding this comment

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

(which is fine, just want to confirm)


let lastIndex = length - delta - 1;
let lastAffectedIndex = startIdx + removedAmount;
if (lastIndex < lastAffectedIndex) {
Copy link
Member

Choose a reason for hiding this comment

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

I think this will miss cases where lastObject changes from negative indices, eg

[0, 1, 2, 3].replace(-2, 2);
// => [0, 1]

Copy link
Contributor Author

Choose a reason for hiding this comment

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

thanks, was not sure if negative startIndex could be passed will handle that

Copy link
Member

Choose a reason for hiding this comment

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

awesome thanks @bekzod 👍

@@ -54,4 +54,26 @@ suite.test('[A,B,C].pushObject(X) => [A,B,C,X] + notify', function() {
equal(observer.validate('firstObject'), false, 'should NOT have notified firstObject');
});

suite.test('[A,B,C,C].pushObject(A) => [A,B,C,C] + notify', function() {
Copy link
Member

Choose a reason for hiding this comment

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

👍

We'll need a test for removing the tail from a negative index

Copy link
Contributor Author

Choose a reason for hiding this comment

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

added that, you could take another shot when you have time, thanks

@bekzod bekzod force-pushed the array-first-last-object branch from 8c953db to f3e2ffd Compare July 17, 2017 20:49
Copy link
Member

@hjdivad hjdivad left a comment

Choose a reason for hiding this comment

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

LGTM

@@ -126,6 +126,27 @@ suite.test('[A,B,C,D].replace(2,2) => [A,B] + notify', function() {
equal(observer.validate('firstObject'), false, 'should NOT have notified firstObject once');
});

suite.test('[A,B,C,D].replace(-1,1) => [A,B,C] + notify', function() {
Copy link
Member

Choose a reason for hiding this comment

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

👍

let delta = addedAmount - removedAmount;
let previousLength = length - delta;

let normalStartIdx = startIdx < 0 ? previousLength + startIdx : startIdx;
Copy link
Member

Choose a reason for hiding this comment

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

👍

@stefanpenner
Copy link
Member

@krisselden bugfix beta or let it ride the canary train?

@stefanpenner
Copy link
Member

chatted with @krisselden hes ok with beta

@stefanpenner
Copy link
Member

@bekzod can you squash, and prefix with [BUGFIX beta] ?

@bekzod bekzod changed the title fix for lastObject/firstObject update issue [BUGFIX beta] fix for lastObject/firstObject update issue Jul 18, 2017
@bekzod bekzod force-pushed the array-first-last-object branch from f3e2ffd to 5752d95 Compare July 18, 2017 05:55
@bekzod
Copy link
Contributor Author

bekzod commented Jul 18, 2017

@stefanpenner done

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.

5 participants