chore(Scope): short-circuit after dirty-checking last dirty watcher

[IgorMinar]

Stop dirty-checking during $digest after the last dirty watcher has been re-checked.
This prevents unneeded re-checking of the remaining watchers (They were already checked in the previous iteration), bringing a substantial performance improvement to the average case run time of $digest.

Closes #5272

[mary-poppins]

Thanks for the PR!

• Contributor signed CLA now or in the past
  • If you just signed, leave a comment here with your real name
• PR's commit messages follow the commit message format

If you need to make changes to your pull request, you can update the commit with git commit --amend.
Then, update the pull request with git push -f.

Thanks again for your help!

[IgorMinar]

@kseamon please review this PR. I've modified your commit slightly and added tests.

[petebacondarwin]

I guess this assumes that the watch functions themselves (as opposed to watch handlers) are in fact idempotent and have no side-effects?

[kseamon]

LGTM

[kseamon]

@pete - It does, but the old behavior did as well. If we couldn't assume that the watch value functions were without side-effects then we'd have to iterate forever.

[petebacondarwin]

I guess this was always an assumption but it could have been possible for a
watch function to have "stable" side effects that stabilized within 10
digest iterations
On 5 Dec 2013 13:10, "Karl Seamon" [email protected] wrote:

@pete https://github.com/pete - It does, but the old behavior did as
well. If we couldn't assume that the watch value functions were without
side-effects then we'd have to iterate forever.

—
Reply to this email directly or view it on GitHubhttps://github.com//pull/5287#issuecomment-29896009
.

[pete]

@kseamon Sure. I'm in favor of that.

[rino65]

Hi
regarding "idempotent" , I have a use case that has a problem with this change.
my watch methods can add and remove other watches , and I noticed that during the digest watcher array loop , if I remove a watch whose index is below the current variable length value of the loop , the watcher array is shifted to the left , and next iteration length-- will point to the same watch as the previous one because all elements are shifted and I get watch === lastDirtyWatch and break from traverseSceopesLoop
and all the other watches dont get called I need the others to get called as well.

[kseamon]

All righty. You should probably file a separate ticket for this, but the simple solution is to clear the lastDirtyWatch variable whenever an unwatch call is made.

Sent from my iPhone

On Dec 23, 2013, at 9:45 AM, rino65 [email protected] wrote:

Hi
regarding "idempotent" , I have a use case that has a problem with this change.
my watch methods can add and remove other watches , and I noticed that during the digest watcher array loop , if I remove a watch whose index is below the current variable length value of the loop , the watcher array is shifted to the left , and next iteration length-- will point to the same watch as the previous one because all elements are shifted and I get watch === lastDirtyWatch and break from traverseSceopesLoop
and all the other watches dont get called I need the others to get called as well.

—
Reply to this email directly or view it on GitHub.

[rino65]

opened issue on this
#5525