The articles listed in Prerequisite Concepts#Immutable Data Management give a number of good examples for how to perform basic update operations immutably, such as updating a field in an object or adding an item to the end of an array. However, reducers will often need to use those basic operations in combination to perform more complicated tasks. Here are some examples for some of the more common tasks you might have to implement.
The key to updating nested data is that every level of nesting must be copied and updated appropriately. This is often a difficult concept for those learning Redux, and there are some specific problems that frequently occur when trying to update nested objects. These lead to accidental direct mutation, and should be avoided.
Unfortunately, the process of correctly applying immutable updates to deeply nested state can easily become verbose and hard to read. Here's what an example of updating
state.first.second[someId].fourth might look like:
Obviously, each layer of nesting makes this harder to read, and gives more chances to make mistakes. This is one of several reasons why you are encouraged to keep your state flattened, and compose reducers as much as possible.
Defining a new variable does not create a new actual object - it only creates another reference to the same object. An example of this error would be:
This function does correctly return a shallow copy of the top-level state object, but because the
nestedState variable was still pointing at the existing object, the state was directly mutated.
Another common version of this error looks like this:
Doing a shallow copy of the top level is not sufficient - the
nestedState object should be copied as well.
splice. Since we don't want to mutate state directly in reducers, those should normally be avoided. Because of that, you might see "insert" or "remove" behavior written like this:
However, remember that the key is that the original in-memory reference is not modified. As long as we make a copy first, we can safely mutate the copy. Note that this is true for both arrays and objects, but nested values still must be updated using the same rules.
This means that we could also write the insert and remove functions like this:
The remove function could also be implemented as:
Updating one item in an array can be accomplished by using
Array.map, returning a new value for the item we want to update, and returning the existing values for all other items:
Some, like dot-prop-immutable, take string paths for commands:
Others, like immutability-helper (a fork of the now-deprecated React Immutability Helpers addon), use nested values and helper functions:
They can provide a useful alternative to writing manual immutable update logic.
Our Redux Toolkit package includes a
createReducer utility that uses Immer internally.
Because of this, you can write reducers that appear to "mutate" state, but the updates are actually applied immutably.
This allows immutable update logic to be written in a much simpler way. Here's what the nested data example
might look like using
This is clearly much shorter and easier to read. However, this only works correctly if you are using the "magic"
createReducer function from Redux Toolkit that wraps this reducer in Immer's
If this reducer is used without Immer, it will actually mutate the state!. It's also not obvious just by
looking at the code that this function is actually safe and updates the state immutably. Please make sure you understand
the concepts of immutable updates fully. If you do use this, it may help to add some comments to your code that explain
your reducers are using Redux Toolkit and Immer.
In addition, Redux Toolkit's
createSlice utility will auto-generate action creators
and action types based on the reducer functions you provide, with the same Immer-powered update capabilities inside.