-
-
Notifications
You must be signed in to change notification settings - Fork 4.3k
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
Fix #708 #709
Fix #708 #709
Conversation
When I test in my repro repo with 2d2c68c, I get this error on build:
|
Ignore that last message, I tried installing straight from Github and forgot about the build step. When I check out this branch and build myself, and then use |
@TehShrike yep, was just looking into your previous comment, seeing the same thing. Thought I had a test that covered that. Investigating... |
Gah, I'm close. Every time I fix one test, another fails. I thought I'd figured it out for a bit back there, but not quite. Incidentally, that repro won't work even when this bug is fixed — Once the bug is fixed, to get the behaviour you want you would need to use a keyed each block. |
I had a suspicion that was related, which was why I through the |
I think this is ready to go, if anyone is interested in taking a look. The basic gist is this: callbacks that need to happen in a particular order (setting up observers for bindings, This way, we can enforce the order (ensuring that a Should also fix #714. |
Testing in my repro repo, everything is cool as long as I don't reverse the array or rearrange elements, which I understand is a different issue, so 👍 |
Where are you init options._root object? |
Code always looks simpler when you wrote it 😀 This PR is more comprehensive, which is why it looks more complex — it makes the order of operations more explicit and predictable, and avoids passing down additional information on each update. The test in #715 passes in this branch, whereas the test in this PR doesn't pass in the other branch |
now, I understand. It's really similar approach, I mean _block = true. But maybe better use _domBlock or _blockDom name? Because we have already "block" in another sense.
yep, I think you should merge this PR and reject mine. |
Maybe |
This definitely isn't the answer, but I think it's heading in the right direction, so I figured I'd leave it here for tomorrow.