Skip to content

Update redux: 4.0.5 → 4.2.1 (minor)#98

Open
depfu[bot] wants to merge 1 commit intomasterfrom
depfu/update/likor-web-app/yarn/redux-4.2.1
Open

Update redux: 4.0.5 → 4.2.1 (minor)#98
depfu[bot] wants to merge 1 commit intomasterfrom
depfu/update/likor-web-app/yarn/redux-4.2.1

Conversation

@depfu
Copy link

@depfu depfu bot commented Jan 29, 2023

Here is everything you need to know about this update. Please take a good look at what changed and the test results before merging this pull request.

What changed?

✳️ redux (4.0.5 → 4.2.1) · Repo · Changelog

Release Notes

4.2.1

This bugfix release removes the isMinified internal check to fix a compat issue with Expo. That check has added in early 2016, soon after Redux 3.0 was released, at a time when it was still less common to use bundlers with proper production build settings. Today that check is irrelevant, so we've removed it.

What's Changed

Full Changelog: v4.2.0...v4.2.1

4.2.0

This release marks the original createStore API as @deprecated to encourage users to migrate to Redux Toolkit, and adds a new legacy_createStore API as an alias without the deprecation warning.

Goal

Redux Toolkit (the @reduxjs/toolkit package) is the right way for Redux users to write Redux code today:

https://redux.js.org/introduction/why-rtk-is-redux-today

Unfortunately, many tutorials are still showing legacy "hand-written" Redux patterns, which result in a much worse experience for users. New learners going through a bootcamp or an outdated Udemy course just follow the examples they're being shown, don't know that RTK is the better and recommended approach, and don't even think to look at our docs.

Given that, the goal is to provide them with a visual indicator in their editor, like createStore . When users hover over the createStore import or function call, the doc tooltip recommends using configureStore from RTK instead, and points them to that docs page. We hope that new learners will see the strikethrough, read the tooltip, read the docs page, learn about RTK, and begin using it.

To be extremely clear:

WE ARE NOT GOING TO ACTUALLY REMOVE THE createStore API, AND ALL YOUR EXISTING CODE WILL STILL CONTINUE TO WORK AS-IS!

We are just marking createStore as "deprecated":

"the discouragement of use of some feature or practice, typically because it has been superseded or is no longer considered efficient or safe, without completely removing it or prohibiting its use"

For additional details, see the extensive discussion in #4325 .

Rationale

  • RTK provides a vastly improved Redux usage experience, with APIs that simplify standard usage patterns and eliminate common bugs like accidental mutations
  • We've had suggestions to merge all of RTK into the redux core package, or fully deprecate the entire redux package and rename it to @reduxjs/core. Unfortunately, those bring up too many complexities:
    • We already had a package rename from redux-starter-kit to @reduxjs/toolkit, and all of our docs and tutorials have pointed to it for the last three years. I don't want to put users through another whiplash package transition for no real benefit
    • Merging or rearranging our packages would effectively require merging all of the Redux repos into a single monorepo. That would require hundreds of hours of effort from us maintainers, including needing to somehow merge all of our docs sites together. We don't have the time to do that.
  • I don't want to add runtime warnings that would be really annoying

So, this is the minimum possible approach we can take to reach out to users who otherwise would never know that they are following outdated patterns, while avoiding breaking running user code or having to completely rewrite our package and repo structure.

Results

When a user imports createStore in their editor, they will see a visual strikethrough. Hovering over it will show a doc tooltip that encourages them to use configureStore from RTK, and points to an explanatory docs page:

image

Again, no broken code, and no runtime warnings.

If users do not want to see that strikethrough, they have three options:

  • Follow our suggestion to switch over to Redux Toolkit and configureStore
  • Do nothing. It's just a visual strikethrough, and it doesn't affect how your code behaves. Ignore it.
  • Switch to using the legacy_createStore API that is now exported, which is the exact same function but with no @deprecation tag. The simplest option is to do an aliased import rename:

image

What's Changed

  • Mark createStore as deprecated, and add legacy_createStore alias by @markerikson in #4336

Full Changelog: v4.1.2...v4.2.0

4.1.2

This release fixes a small specific TS types issue where state types that had a nested unknown field inside would cause compilation failures when used as the preloadedState argument.

What's Changed

Full Changelog: v4.1.1...v4.1.2

4.1.1

Just a small fix for Safari users in development mode.

Changes

  • Move miniKindOf out of if scope to fix ES5 compatibility issue (#4090 by @embeddedt)

4.1.0

This release shrinks our bundle size via error message extraction, updates several error messages for clarity, and optimizes our list of runtime dependencies.

Overall, version 4.1 shrinks from 2.6K min+gz to 1.6K min+gz thanks to these changes.

Be sure to check out the Redux Toolkit 1.6 alpha containing our new "RTK Query" data fetching APIs! It also includes Redux 4.1 as a dependency.

Changelog

Error Message Extraction and Improvements

We now extract all of our error messages from production builds in order to save on bundle size, using a technique inspired from React's error code extraction. The error messages will still show as normal in development, but in production they will reference a specific numeric error code and provide a link to a Redux docs page that has the full error message.

An example of this is: https://redux.js.org/errors?code=5 , which shows the "can't subscribe while reducers are executing" error.

The error code extraction saves about 800 bytes out of a production build.

Thanks to @andrewmcgivery for doing all the hard work on implementing the error extraction!

We've also updated many of our error messages to provide additional details at runtime about what happened, especially runtime type checks such as "actions must be plain objects". They now provide a more specific type for the unexpected value, such as indicating promise or function:

    expect(() => store.dispatch(() => {})).toThrow(
      /the actual type was: 'function'/
    )
<span class="pl-en">expect</span><span class="pl-kos">(</span><span class="pl-kos">(</span><span class="pl-kos">)</span> <span class="pl-c1">=&gt;</span> <span class="pl-s1">store</span><span class="pl-kos">.</span><span class="pl-en">dispatch</span><span class="pl-kos">(</span><span class="pl-k">new</span> <span class="pl-v">Date</span><span class="pl-kos">(</span><span class="pl-kos">)</span><span class="pl-kos">)</span><span class="pl-kos">)</span><span class="pl-kos">.</span><span class="pl-en">toThrow</span><span class="pl-kos">(</span>
  <span class="pl-pds"><span class="pl-c1">/</span>the actual type was: 'date'<span class="pl-c1">/</span></span>
<span class="pl-kos">)</span></pre></div>

Dependency Updates

We've updated the list of runtime dependencies for Redux:

  • We inlined the symbol-observable polyfill. This shrinks bundle size by a few bytes,
  • We've removed the legacy loose-envify dependency, which was only ever needed by Browserify users. If you still happen to be using Browserify, please review your build settings and see if you need to make any updates.
  • We now explicitly depend on @babel/runtime to extract some additional helpers out of our bundle. It's likely that your app already is pulling in those helpers anyway, so that removes some potential duplication.

Typing Tweaks

We've merged fixes for a couple edge cases in the 4.x TS typings related to state types.

Changes

v4.0.5...v4.1.0

Does any of this look wrong? Please let us know.

Commits

See the full diff on Github. The new version differs by 45 commits:


👉 No CI detected

You don't seem to have any Continuous Integration service set up!

Without a service that will test the Depfu branches and pull requests, we can't inform you if incoming updates actually work with your app. We think that this degrades the service we're trying to provide down to a point where it is more or less meaningless.

This is fine if you just want to give Depfu a quick try. If you want to really let Depfu help you keep your app up-to-date, we recommend setting up a CI system:

  • Circle CI, Semaphore and Travis-CI are all excellent options.
  • If you use something like Jenkins, make sure that you're using the Github integration correctly so that it reports status data back to Github.
  • If you have already set up a CI for this repository, you might need to check your configuration. Make sure it will run on all new branches. If you don’t want it to run on every branch, you can whitelist branches starting with depfu/.

Depfu Status

Depfu will automatically keep this PR conflict-free, as long as you don't add any commits to this branch yourself. You can also trigger a rebase manually by commenting with @depfu rebase.

All Depfu comment commands
@​depfu rebase
Rebases against your default branch and redoes this update
@​depfu recreate
Recreates this PR, overwriting any edits that you've made to it
@​depfu merge
Merges this PR once your tests are passing and conflicts are resolved
@​depfu close
Closes this PR and deletes the branch
@​depfu reopen
Restores the branch and reopens this PR (if it's closed)
@​depfu pause
Ignores all future updates for this dependency and closes this PR
@​depfu pause [minor|major]
Ignores all future minor/major updates for this dependency and closes this PR
@​depfu resume
Future versions of this dependency will create PRs again (leaves this PR as is)

@depfu depfu bot added the depfu label Jan 29, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

0 participants