Trying to make the event a bit more performant for events.
Feel free to reject this because the API inevitably isn't great. It's good for perf though, and since we're only using `accumulate` in very restrained places, I think we're fine.
`accumulateInto` is `accumulate` that mutates more and allocates less. I kept `accumulate` in case we want that in the future.
This moves all logic around legacy descriptors to ReactLegacyDescriptor. This
is responsible for the layer that knows that createClass exports a legacy
factory. When passed one of these classes, it unwraps it to be a real class.
If it is passed a non legacy factory, it is assumed to be a non-react component
that needs to be invoked as a plain function.
The semantic change is that a descriptor is now always returned if passed a
legacy factory. Even if that factory is a mock. A mock would previously return
undefined.
For mocks, I treat the factory as the authoritative function. I call it to extract
the instance or fill it with an empty component placeholder.
Additionally, I make the classes take props as the first argument to the
constructor. This is what the new class system will do.
We currently need to set up some internals by calling the internal construct
method. Instead of doing that automatically in the constructor, I now move that
to a second pass so that mocks can get the plain props.
This means that we can assert that a mock has been called once it's mounted
with it's final props. Instead of the descriptor factory being called.
Relax the argument type checks. Currently we throw for non-objects and terminals
but Object.assign does a coercion to Object instead. It also allows merging
Arrays as if they are objects.
This also relaxes the check for dependents such as ImmutableObject. This sucks
but it will allow us to use a fast code path to native Object.assign.
We always have the option of adding warnings to Object.assign or static type
checks.
I'm keeping the null check. Object.assign throws for null checks.
We'll also start returning the result of coercions just like Object.assign.
This detail is going to become more important once the idiom
`className={joinClasses(this.props.className, newClass)}` becomes more
common, as it will when we move away from `this.transferPropsTo`.
Breaking changes
- key/ref are no longer accessible on props but they are accessible on the
descriptors. This means that parents/owners can access it but not the
component itself.
- Descriptor factories are now plain functions and you can't rely on the
prototype or constructors of descriptors to identify the component type.
Existing descriptor factories are now wrapped in a legacy factory. Currently it
does nothing but it will give us a hook to track callers to factories that are
not using JSX but just invoking the function directly. It also proxies static
methods/properties to the underlying class. The newer factories don't have this
feature.
ReactTextComponent has it's own little factory because it's props is not an
object. This is a detail and will go away once ReactTextComponent no longer
needs descriptors.
The component that gets passed into renderComponent isn't guaranteed to be the
instance that gets mounted. We want to clone the instance.
Unit tests need to reason about the mounted instance. The first code mod changes:
ReactTestUtils.renderIntoDocument(<identifier>)
into
<identifier> = ReactTestUtils.renderIntoDocument(<identifier>)
Using this scripts:
scripts/bin/codemod -m -d ~/www --extensions js \
'^(\s*)ReactTestUtils\.renderIntoDocument\(\s*([$a-zA-Z0-9_]+)\s*\)' \
'\1\2 = ReactTestUtils.renderIntoDocument(\2)'
In the second case I do the same for React.renderComponent. However, there are
alot more unnecessary matches so I only codemod if the same identifier occurs
later in the file.
scripts/bin/codemod -m -d ~/www --extensions js \
'^(\s*)React.renderComponent\(\s*([$a-zA-Z0-9_]+)\s*?,(.*?\n?.*?\s\2\b)' \
'\1\2 = React.renderComponent(\2,\3'
And one more for ReactMount.renderComponent used by internals.
scripts/bin/codemod -m -d ~/www --extensions js \
'^(\s*)ReactMount.renderComponent\(\s*([$a-zA-Z0-9_]+)\s*?,(.*?\n?.*?\s\2\b)' \
'\1\2 = ReactMount.renderComponent(\2,\3'
This still matches many unnecessary cases where the second occurance of the
identifier is a redeclaration or comment. But this code mod doesn't hurt in
those cases.
Finally I have to do the same for:
this.<identifier> = React.renderComponent(this.<identifier>,
This is a common pattern for production code but not tests. Some of these call
sites will likely break when we move to true descriptors.
scripts/bin/codemod -m -d ~/www --extensions js \
'^(\s*)React.renderComponent\((\s*)this\.([$a-zA-Z0-9\_\.]+)\s*?,' \
'\1this.\3 = React.renderComponent(\2this.\3,'
This is the first step towards descriptors. This will start cloning the
component when it's mounted instead of mounting the first instance.
This avoids an issue where a reference to the first instance can hang around
in props. Since a mounted component gets mutated, the descriptor changes.
We don't need to clone the props object itself. Mutating the shallow props
object of a child that's passed into you is already flawed. Those cases need to
use cloneWithProps. A props object is considered shallow frozen after it leaves
the render it was created in.
After we run `require('mock-modules').dumpCache()`, the object exported by
the `emptyObject` module will no longer be identical to previously
exported objects, so tests like `expect(component.refs).toBe(emptyObject)`
will fail.
Note that this behavior only manifests itself in tests, because of course
we do not call `dumpCache` in production code.
We could consider storing the `emptyObject` globally to thwart the effects
of `dumpCache`, but it's more idiomatic simply to re-`require` the latest
version of `emptyObject`.
If no refs are rendered, `this.refs` is undefined. This is bad since it deopts & is hard to look for. Instead we should make `this.refs` an immutable empty object.
This adds an instrumentation hook for logging so that we can monitor invalid API
usage before we're ready to issue a warning.
I took the opportunity to update some console.warns to use the warning module
instead. The remaining console.warns
will be replaced by the warning module after we've cleaned up the callsites.
grep -rl 'Copyright 2013 Facebook' static_upstream | xargs perl -pi -w -e s/Copyright 2013 Facebook/Copyright 2013-2014 Facebook/g;'
Not going to check in a script to do this since it will just change every year.
Closes#1006
Also added tests for PooledClass.
Noticed that some places use `ReactReconcileTransaction.release(transaction)`, when `transaction` might be of another class, say `ReactServerRenderingTransaction` (still a Github PR). This catches that.
If a transaction wrapper's closer throws (such as flushBatchedUpdates in ReactDefaultBatchingStrategy) then we still want to properly deinitialize the transaction by setting isInTransaction to false. Without this, we can't reuse the transaction object (in this case, all future batched updates would fail because nested transactions aren't allowed).
Rethrowing errors makes debugging harder. This makes it so that an exception in a render method can be caught using the purple stop sign (or the blue one, of course) in Chrome.
`cloneWithProps` uses `ReactPropTransferer`, which ignores the `key`
prop. See https://github.com/facebook/react/pull/713
However, this is not the case with `cloneWithProps` because when someone
is cloning a component and provides a key, they mean for the clone to
take it.
Andres Suarez
Miorel Palii
Cheng Lou
Sebastian Markbage
Ben Newman
Ben Alpert
Paul O'Shannessy
Guangqiang Dong
Jan Kassens
Paul O’Shannessy
Christoph Pojer
Kunal Mehta
Ben Newman
Pete Hunt
Cheng Lou
plievone
Paul Shen
Andreas Svensson
