[performance] Inline ReactChildReconciler.instantiateChildren #3922
Conversation
| if (__DEV__) { | ||
| warning( | ||
| true, | ||
| 'mountChildren(...): Encountered two children with the same key, ' + |
|
Code looks fine. Will try to replicate your perf-test results and if it looks positive I'll merge. |
mridgway
force-pushed
the
inlineInstantiateChildren
branch
from
861b870
to
b4d2d1f
Compare
facebook-github-bot
added
GH Review: review-needed
and removed
GH Review: needs-revision
labels
|
Updated and rebased. |
|
@jimfb Assigning to you. |
|
Yep, this diff is freakishly similar to my diff. We decided to merge #4408, since I was already iterating on it with Sebastian. For this reason, I'm closing out this diff. @mridgway Your diff was spot on. I truly wish we had seen it earlier. If we had merged this back in May, it would have saved us a whole bunch of time/energy :P. Ah well, such is life. We didn't know then what we know now (didn't have the perf tests on our end). If you have any other ideas for improving perf and/or otherwise improving React, please do ping us; we can do a video chat or grab lunch or whatever. |
|
@mridgway Sorry for dropping the ball on this one. Got caught up in preparing for the conference and then I was out on vacation for a couple weeks. I could've passed this to @jimfb when he started focusing on perf work but didn't think of it at the time. Please do let us know if you find more improvements and hopefully we can get them merged sooner next time. |
|
No worries. I'm just glad someone is looking into performance 👍. @jimfb I may take you up on lunch sometime. I'm eager to work with you guys on optimizing the server rendering flow but need some more information on what the plan is before I can contribute. Looks like #4012 has some more details that I'll look into. |
Follow up from #3717