### What
When using the `Restore as draft` action from the version view, the
restored document incorrectly appears as `published` in the API. This
issue was reported by a client.
### Why
In the `restoreVersion` operation, the document is updated via
`db.updateOne` (line 265). The update passes along the status stored in
the version being restored but does not respect the `draft` query
parameter.
### How
Ensures that the result status is explicitly set to `draft` when the
`draft` argument is `true`.
Fixes#13574.
When editing an autosave-enabled document within a document drawer, any
changes made will while the autosave is processing are ultimately
discarded from local form state. This makes is difficult or even
impossible to edit fields.
This is because we server-render, then replace, the entire document on
every save. This includes form state, which is stale because it was
rendered while new changes were still being made. We don't need to
re-render the entire view on every save, though, only on create. We
don't do this on the top-level edit view, for example. Instead, we only
need to replace form state.
This change is also a performance improvement because we are no longer
rendering all components unnecessarily, especially on every autosave
interval.
Before:
https://github.com/user-attachments/assets/e9c221bf-4800-4153-af55-8b82e93b3c26
After:
https://github.com/user-attachments/assets/d77ef2f3-b98b-41d6-ba6c-b502b9bb99cc
Note: ignore the flashing autosave status and doc controls. This is
horrible and we're actively fixing it, but is outside the scope of this
PR.
---
- To see the specific tasks where the Asana app for GitHub is being
used, see below:
- https://app.asana.com/0/0/1211139422639700
Fixes#10526
### What?
`updateVersion` throws error at deleting before version of draft.
### Why?
When triggers `db.updateVersion` with non-mongodb environment,
`saveVersion` in `payload/src/versions/saveVersions.ts` doesn't remove
`id` in `versionData`.
### How?
Add `delete versionData.id` for non-mongodb environments.
---------
Co-authored-by: German Jablonski <43938777+GermanJablo@users.noreply.github.com>
This PR fixes some incorrect field paths handling (=> should not pass
path and schema paths that contain index paths down to sub-fields
outside of the indexPath property) when building the version fields.
Fixes https://github.com/payloadcms/payload/issues/12376
---
- To see the specific tasks where the Asana app for GitHub is being
used, see below:
- https://app.asana.com/0/0/1210932060696925
Follow-up to #13416. Supersedes #13434.
When autosave is triggered and the user continues to modify fields,
their changes are overridden by the server's value, i.e. the value at
the time the form state request was made. This makes it almost
impossible to edit fields when using a small autosave interval and/or a
slow network.
This is because autosave is now merged into form state, which by default
uses `acceptValues: true`. This does exactly what it sounds like,
accepts all the values from the server—which may be stale if underlying
changes have been made. We ignore these values for onChange events,
because the user is actively making changes. But during form
submissions, we can accept them because the form is disabled while
processing anyway.
This pattern allows us to render "computed values" from the server, i.e.
a field with an `beforeChange` hook that modifies its value.
Autosave, on the other hand, happens in the background _while the form
is still active_. This means changes may have been made since sending
the request. We still need to accept computed values from the server,
but we need to avoid doing this if the user has active changes since the
time of the request.
---
- To see the specific tasks where the Asana app for GitHub is being
used, see below:
- https://app.asana.com/0/0/1211027929253429
Fixes#10515. Needed for #12956.
Hooks run within autosave are not reflected in form state.
Similar to #10268, but for autosave events.
For example, if you are using a computed value, like this:
```ts
[
// ...
{
name: 'title',
type: 'text',
},
{
name: 'computedTitle',
type: 'text',
hooks: {
beforeChange: [({ data }) => data?.title],
},
},
]
```
In the example above, when an autosave event is triggered after changing
the `title` field, we expect the `computedTitle` field to match. But
although this takes place on the database level, the UI does not reflect
this change unless you refresh the page or navigate back and forth.
Here's an example:
Before:
https://github.com/user-attachments/assets/c8c68a78-9957-45a8-a710-84d954d15bcc
After:
https://github.com/user-attachments/assets/16cb87a5-83ca-4891-b01f-f5c4b0a34362
---
- To see the specific tasks where the Asana app for GitHub is being
used, see below:
- https://app.asana.com/0/0/1210561273449855
Field-level permissions were not handled correctly at all. If you had a
field set with access control, this would mean that nested fields would
incorrectly be omitted from the version view.
---
- To see the specific tasks where the Asana app for GitHub is being
used, see below:
- https://app.asana.com/0/0/1210932060696919
Fixes#12975.
When editing autosave-enabled documents through the join field, the
document drawer closes unexpectedly on every autosave interval, making
it nearly impossible to use.
This is because as of #12842, the underlying relationship table
re-renders on every autosave event, remounting the drawer each time. The
fix is to lift the drawer out of table's rendering tree and into the
join field itself. This way all rows share the same drawer, whose
rendering lifecycle has been completely decoupled from the table's
state.
Note: this is very similar to how relationship fields achieve similar
functionality.
This PR also adds jsdocs to the `useDocumentDrawer` hook and strengthens
its types.
---
- To see the specific tasks where the Asana app for GitHub is being
used, see below:
- https://app.asana.com/0/0/1210906078627353
Required for #13005.
Opening an autosave-enabled document within a drawer triggers an
infinite loop when the root document is also autosave-enabled.
This was for two reasons:
1. Autosave would run and change the `updatedAt` timestamp. This would
trigger another run of autosave, and so on. The timestamp is now removed
before comparison to ensure that sequential autosave runs are skipped.
2. The `dequal()` call was not being given the `.current` property off
the ref object. This meant that is was never evaluate to `true` and
therefore never skip unnecessary autosaves to begin with.
---
- To see the specific tasks where the Asana app for GitHub is being
used, see below:
- https://app.asana.com/0/0/1210697235723932
### What?
This PR fixes a runtime error that occurs when opening the "More
versions..." drawer while browsing the versions for a global. It also
fixes a minor runtime error when navigating to a global version view
where an optional chaining operator was missing as the collection
variable would be undefined as we are viewing a global.
This PR also adds an e2e test to ensure the versions drawer is
accessible and renders the appropriate number of versions for globals.
### Why?
To properly render global version views without errors.
### How?
By threading the global slug to the versions drawer and adjusting some
properties of the `renderDocument` server function call there. This PR
also adds an optional chaining operator the `versionUseAsTitle` in the
original view to prevent an error in globals.
Notes:
- This was brought to my attention in Discord by a handful of users
Before: (Missing optional chaining error)
[error1-verions-Editing---Menu---Payload.webm](https://github.com/user-attachments/assets/3dc4dbe4-ee5a-43df-8d25-05128b05e063)
Before: (Versions drawer error)
[error2-versions-Editing---Menu---Payload.webm](https://github.com/user-attachments/assets/98c3e1da-cb0b-4a36-bafd-240f641e8814)
After:
[versions-globals-Dashboard---Payload.webm](https://github.com/user-attachments/assets/c778d3f0-a8fe-4e31-92cb-62da8e6d8cb4)
### What?
After creating a new document from a relationship field, this doc should
automatically become the selected document for that relationship field.
This is the expected and current behavior. However, when the
relationship ties to a collection with autosave enabled, this does not
happen.
### Why?
This is expected behavior and should still happen when the relationship
is using an autosave enabled collection.
### How?
1. The logic in `addNewRelation` contained an `if` statement that
checked for `operation === 'create'` - however when autosave is enabled,
the `create` operation runs on the first data update and subsequently it
is a `update` operation.
2. The `onSave` from the document drawer provider was not being run as
part of the autosave workflow.
#### Reported by client.
When saving a global with versioning enabled as draft, and then
publishing it, the following error may appear: `[16:50:35] ERROR: Could
not find createdAt or updatedAt in latestVersion.version`
This is due to an incorrect check to appease typescript strict mode. We
shouldn't throw if `version.updatedAt` doesn't exist - the purpose of
this logic is to add that property if it doesn't exist
This clarifies that jobs.autoRun only *runs* already-queued jobs. It does not queue the jobs for you.
Also adds an e2e test as this functionality had no e2e coverage
I noticed a few issues when running e2e tests that will be resolved by
this PR:
- Most important: for some test suites (fields, fields-relationship,
versions, queues, lexical), the database was cleared and seeded
**twice** in between each test run. This is because the onInit function
was running the clear and seed script, when it should only have been
running the seed script. Clearing the database / the snapshot workflow
is being done by the reInit endpoint, which then calls onInit to seed
the actual data.
- The slowest part of `clearAndSeedEverything` is recreating indexes on
mongodb. This PR slightly improves performance here by:
- Skipping this process for the built-in `['payload-migrations',
'payload-preferences', 'payload-locked-documents']` collections
- Previously we were calling both `createIndexes` and `ensureIndexes`.
This was unnecessary - `ensureIndexes` is a deprecated alias of
`createIndexes`. This PR changes it to only call `createIndexes`
- Makes the reinit endpoint accept GET requests instead of POST requests
- this makes it easier to debug right in the browser
- Some typescript fixes
- Adds a `dev:memorydb` script to the package.json. For some reason,
`dev` is super unreliable on mongodb locally when running e2e tests - it
frequently fails during index creation. Using the memorydb fixes this
issue, with the bonus of more closely resembling the CI environment
- Previously, you were unable to run test suites using turbopack +
postgres. This fixes it, by explicitly installing `pg` as devDependency
in our monorepo
- Fixes jest open handles warning
I think it's easier to review this PR commit by commit, so I'll explain
it this way:
## Commits
1. [parallelize eslint script (still showing logs results in
serial)](c9ac49c12d):
Previously, `--concurrency 1` was added to the script to make the logs
more readable. However, turborepo has an option specifically for these
use cases: `--log-order=grouped` runs the tasks in parallel but outputs
them serially. As a result, the lint script is now significantly faster.
2. [run pnpm
lint:fix](9c128c276a)
The auto-fix was run, which resolved some eslint errors that were
slipped in due to the use of `no-verify`. Most of these were
`perfectionist` fixes (property ordering) and the removal of unnecessary
assertions. Starting with this PR, this won't happen again in the
future, as we'll be verifying the linter in every PR across the entire
codebase (see commit 7).
3. [fix eslint non-autofixable
errors](700f412a33)
All manual errors have been resolved except for the configuration errors
addressed in commit 5. Most were React compiler violations, which have
been disabled and commented out "TODO" for now. There's also an unused
`use no memo` and a couple of `require` errors.
4. [move react-compiler linter to eslint-config
package](4f7cb4d63a)
To simplify the eslint configuration. My concern was that there would be
a performance regression when used in non-react related packages, but
none was experienced. This is probably because it only runs on .tsx
files.
5. [remove redundant eslint config files and fix
allowDefaultProject](a94347995a)
The main feature introduced by `typescript-eslint` v8 was
`projectService`, which automatically searches each file for the closest
`tsconfig`, greatly simplifying configuration in monorepos
([source](https://typescript-eslint.io/blog/announcing-typescript-eslint-v8#project-service)).
Once I moved `projectService` to `packages/eslint-config`, all the other
configuration files could be easily removed.
I confirmed that pnpm lint still works on individual packages.
The other important change was that the pending eslint errors from
commits 2 and 3 were resolved. That is, some files were giving the
error: "[File] was not found by the project service. Consider either
including it in the tsconfig.json or including it in
allowDefaultProject." Below I copy the explanatory comment I left in the
code:
```ts
// This is necessary because `tsconfig.base.json` defines `"rootDir": "${configDir}/src"`,
// And the following files aren't in src because they aren't transpiled.
// This is typescript-eslint's way of adding files that aren't included in tsconfig.
// See: https://typescript-eslint.io/troubleshooting/typed-linting/#i-get-errors-telling-me--was-not-found-by-the-project-service-consider-either-including-it-in-the-tsconfigjson-or-including-it-in-allowdefaultproject
// The best practice is to have a tsconfig.json that covers ALL files and is used for
// typechecking (with noEmit), and a `tsconfig.build.json` that is used for the build
// (or alternatively, swc, tsup or tsdown). That's what we should ideally do, in which case
// this hardcoded list wouldn't be necessary. Note that these files don't currently go
// through ts, only through eslint.
```
6. [Differentiate errors from warnings in VScode ESLint
Rules](5914d2f48d)
There's no reason to do that. If an eslint rule isn't an error, it
should be disabled or converted to a warning.
7. [Disable skip lint, and lint over the entire repo now that it's
faster](e4b28f1360)
The GitHub action linted only the files that had changed in the PR.
While this seems like a good idea, once exceptions were introduced with
[skip lint], they opened the door to propagating more and more errors.
Often, the linter was skipped, not because someone introduced new
errors, but because they were trying to avoid those that had already
crept in, sometimes accidentally introducing new ones.
On the other hand, `pnpm lint` now runs in parallel (commit 1), so it's
not that slow. Additionally, it runs in parallel with other GitHub
actions like e2e tests, which take much longer, so it can't represent a
bottleneck in CI.
8. [fix lint in next
package](4506595f91)
Small fix missing from commit 5
9. [Merge remote-tracking branch 'origin/main' into
fix-eslint](563d4909c1)
10. [add again eslint.config.js in payload
package](78f6ffcae7)
The comment in the code explains it. Basically, after the merge from
main, the payload package runs out of memory when linting, probably
because it grew in recent PRs. That package will sooner or later
collapse for our tooling, so we may have to split it. It's already too
big.
## Future Actions
- Resolve React compiler violations, as mentioned in commit 3.
- Decouple the `tsconfig` used for typechecking and build across the
entire monorepo (as explained in point 5) to ensure ts coverage even for
files that aren't transpiled (such as scripts).
- Remove the few remaining `eslint.config.js`. I had to leave the
`richtext-lexical` and `next` ones for now. They could be moved to the
root config and scoped to their packages, as we do for example with
`templates/vercel-postgres/**`. However, I couldn't get it to work, I
don't know why.
- Make eslint in the test folder usable. Not only are we not linting
`test` in CI, but now the `pnpm eslint .` command is so large that my
computer freezes. If each suite were its own package, this would be
solved, and dynamic codegen + git hooks to modify tsconfig.base.json
wouldn't be necessary
([related](https://github.com/payloadcms/payload/pull/11984)).
### What?
If an error occurs while unpublishing a document in the edit view UI,
the toast which shows the error message now displays the actual message
which is sent from the server, if available.
### Why?
Only a generic error message was shown if an unpublish operation failed.
Some errors might be solvable by the user, so that there is value in
showing the actual, actionable error message instead of a generic one.
### How?
The server response is parsed for error message if an unpublish
operation fails and displayed in the toast, instead of the generic error
message.
### What?
Fixes the label for documents which were the current published document
but got unpublished in the version view.
### Why?
If the most recent published document was unpublished, it remained
displayed as "Currently published version" in the version list.
### How?
Checks whether the document has a currently published version instead of
only looking at the latest published version when determining the label
in the versions view.
Fixes https://github.com/payloadcms/payload/issues/10838
---------
Co-authored-by: Alessio Gravili <alessio@gravili.de>
Co-authored-by: Dan Ribbens <dan.ribbens@gmail.com>
This adds a new `showSaveDraftButton` option to the
`versions.drafts.autosave` config for collections and globals.
By default, the "Save as draft" button is hidden when autosave is
enabled. This new option allows the button to remain visible for manual
saves while autosave is active.
Also updates the admin UI logic to conditionally render the button when
this flag is set, and updates the documentation with an example usage.
Fixes an issue where an autosave being triggered would turn off the
ability to schedule a publish. This happened because we check against
`modified` on the form but with autosave modified is always true.
Now we make an exception for autosave enabled collections when checking
the modified state.
This PR adds a new `SchedulePublish` config type on our schedulePublish
configuration in versions from being just boolean.
Two new options are supported:
- `timeFormat` which controls the formatting of the time slots, allowing
users to change from a 12-hour clock to a 24-hour clock (default to 12
hour)
- `timeIntervals` which controls the generated time slots (default 5)
Example configuration:
```
versions: {
drafts: {
schedulePublish: {
timeFormat: 'HH:mm',
timeIntervals: 5,
},
},
},
```
<!--
Thank you for the PR! Please go through the checklist below and make
sure you've completed all the steps.
Please review the
[CONTRIBUTING.md](https://github.com/payloadcms/payload/blob/main/CONTRIBUTING.md)
document in this repository if you haven't already.
The following items will ensure that your PR is handled as smoothly as
possible:
- PR Title must follow conventional commits format. For example, `feat:
my new feature`, `fix(plugin-seo): my fix`.
- Minimal description explained as if explained to someone not
immediately familiar with the code.
- Provide before/after screenshots or code diffs if applicable.
- Link any related issues/discussions from GitHub or Discord.
- Add review comments if necessary to explain to the reviewer the logic
behind a change
### What?
### Why?
### How?
Fixes #
-->
### What?
This PR aims to deflake the `test/versions/e2e.spec.ts:925:5 › Versions
› Collections with draft validation › - with autosave - shows a prevent
leave alert when form is submitted but invalid` e2e test.
The issue seems to be that the `fill` call followed by a `page.reload`
sometimes conflicts with autosave which may cause the test to flake.
### Why?
To deflake this test in ci.
### How?
Adds a single `waitForAutoSaveToRunAndComplete` function call prior to
the last call to `page.reload`. In my testing, on my local machine,
adding the `waitForAutoSaveToRunAndComplete` function allows the test to
pass every time. Without this, the tests fails on my machine
consistently.
Fixes#11458
Some complex, nested fields were receiving incorrect field paths and
schema paths, leading to a `"Error: No client field found"` error.
This PR ensures field paths are calculated correctly, by matching it to
how they're calculated in payload hooks.
Consolidates all bulk edit related tests into a single, dedicated suite.
Currently, bulk edit tests are dispersed throughout the Admin > General
and the Versions test suites, which are considerably bloated for their
own purposes. This made them very hard to locate, mentally digest, and
add on new tests. Going forward, many more tests specifically for bulk
edit will need to be written. This gives us a simple, isolated place for
that.
With this change are also a few improvements to the tests themselves to
make them more predictable and efficient.
Implements a form state task queue. This will prevent onChange handlers
within the form component from processing unnecessarily often, sometimes
long after the user has stopped making changes. This leads to a
potentially huge number of network requests if those changes were made
slower than the debounce rate. This is especially noticeable on slow
networks.
Does so through a new `useQueue` hook. This hook maintains a stack of
events that need processing but only processes the final event to
arrive. Every time a new event is pushed to the stack, the currently
running process is aborted (if any), and that event becomes the next in
the queue. This results in a shocking reduction in the time it takes
between final change to form state and the final network response, from
~1.5 minutes to ~3 seconds (depending on the scenario, see below).
This likely fixes a number of existing open issues. I will link those
issues here once they are identified and verifiably fixed.
Before:
I'm typing slowly here to ensure my changes aren't debounce by the form.
There are a total of 60 characters typed, triggering 58 network requests
and taking around 1.5 minutes to complete after the final change was
made.
https://github.com/user-attachments/assets/49ba0790-a8f8-4390-8421-87453ff8b650
After:
Here there are a total of 69 characters typed, triggering 11 network
requests and taking only about 3 seconds to complete after the final
change was made.
https://github.com/user-attachments/assets/447f8303-0957-41bd-bb2d-9e1151ed9ec3
Previously, `db.updateVersion` had a mistake with using `transform({
operation: 'write' })` instead of `transform({ operation: 'read' })`
which led to improper DB data sanitization (like ObjectID -> string,
Date -> string) when calling `payload.update` with `autosave: true` when
some other autosave draft already exists. This fixes
https://github.com/payloadcms/payload/issues/11542 additionally for this
case.
Maintains column state in the URL. This makes it possible to share
direct links to the list view in a specific column order or active
column state, similar to the behavior of filters. This also makes it
possible to change both the filters and columns in the same rendering
cycle, a requirement of the "list presets" feature being worked on here:
#11330.
For example:
```
?columns=%5B"title"%2C"content"%2C"-updatedAt"%2C"createdAt"%2C"id"%5D
```
The `-` prefix denotes that the column is inactive.
This strategy performs a single round trip to the server, ultimately
simplifying the table columns provider as it no longer needs to request
a newly rendered table for itself. Without this change, column state
would need to be replaced first, followed by a change to the filters.
This would make an unnecessary number of requests to the server and
briefly render the UI in a stale state.
This all happens behind an optimistic update, where the state of the
columns is immediately reflected in the UI while the request takes place
in the background.
Technically speaking, an additional database query in performed compared
to the old strategy, whereas before we'd send the data through the
request to avoid this. But this is a necessary tradeoff and doesn't have
huge performance implications. One could argue that this is actually a
good thing, as the data might have changed in the background which would
not have been reflected in the result otherwise.
Removes all unnecessary `page.waitForURL` methods within e2e tests.
These are unneeded when following a `page.goto` call because the
subsequent page load is already being awaited.
It is only a requirement when:
- Clicking a link and expecting navigation
- Expecting a redirect after a route change
- Waiting for a change in search params
Fixes
https://github.com/payloadcms/payload/issues/11359#issuecomment-2678213414
The link element by using startTransitionRoute and manually calling
router.push would technically cause links to be clicked twice, by not
preventing default browser behaviour.
This caused a problem on clicking /create links as it hit the route
twice. Added a test making sure Create new doesn't lead to abnormally
increased document counts
Changes:
- Added `e.preventDefault()` in our Link element
- Added `preventDefault` as an optional prop to this element so that
people can handle it on their own if needed via a custom `onClick`
### What?
When you first edit a document and then open the Schedule publish
drawer, you can schedule publish changes but the current changes made to
the form won't be included.
### Why?
The UX does not make it clear that the changes you have in the form are
not actually going to be published.
### How?
Instead of allowing that we just disable the Schedule Publish drawer
toggler so that users are forced to save a draft first.
In addition to the above, this change also passes a defaultType so that
an already published document will default the radio type have
"Unpublish" selected.
There are nearly a dozen independent implementations of the same modal
spread throughout the admin panel and various plugins. These modals are
used to confirm or cancel an action, such as deleting a document, bulk
publishing, etc. Each of these instances is nearly identical, leading to
unnecessary development efforts when creating them, inconsistent UI, and
duplicative stylesheets.
Everything is now standardized behind a new `ConfirmationModal`
component. This modal comes with a standard API that is flexible enough
to replace nearly every instance. This component has also been exported
for reuse.
Here is a basic example of how to use it:
```tsx
'use client'
import { ConfirmationModal, useModal } from '@payloadcms/ui'
import React, { Fragment } from 'react'
const modalSlug = 'my-confirmation-modal'
export function MyComponent() {
const { openModal } = useModal()
return (
<Fragment>
<button
onClick={() => {
openModal(modalSlug)
}}
type="button"
>
Do something
</button>
<ConfirmationModal
heading="Are you sure?"
body="Confirm or cancel before proceeding."
modalSlug={modalSlug}
onConfirm={({ closeConfirmationModal, setConfirming }) => {
// do something
setConfirming(false)
closeConfirmationModal()
}}
/>
</Fragment>
)
}
```
Fixes https://github.com/payloadcms/payload/issues/11224
Fixes https://github.com/payloadcms/payload/issues/10492
This PR fixes a few weird behaviours when `validate: true` is set on drafts:
- when autosave is on and you submit an invalid form it would get stuck in an infinite loop
- PreventLeave would not trigger for submitted but invalid forms leading to potential data loss
Changes:
- Adds e2e tests for the above scenarios
- Adds a new `isValid` flag on the `Form` context provider to signal globally if the form is in a valid or invalid state
- Components like Autosave will manage this internally since it manages its own submission flow as well
- Adds PreventLeave to Autosave too for when form is invalid meaning data hasn't been actually saved so we want to prevent the user accidentally losing data by reloading or closing the page
The following tests have been added
This PR extends timezone support to scheduled publish UI and collection,
the timezone will be stored on the `input` JSON instead of the
`waitUntil` date field so that we avoid needing a schema migration for
SQL databases.
If a timezone is selected then the displayed date in the table will be
formatted for that timezone.
Timezones remain optional here as they can be deselected in which case
the date will behave as normal, rendering and formatting to the user's
local timezone.
For the backend logic that can be left untouched since the underlying
date values are stored in UTC the job runners will always handle this
relative time by default.
Todo:
- [x] add e2e to this drawer too to ensure that dates are rendered as
expected
### What?
If you had multiple operator constraints on a single field, the last one
defined would be the only one used.
Example:
```ts
where: {
id: {
in: [doc2.id],
not_in: [], // <-- only respected this operator constraint
},
}
```
and
```ts
where: {
id: {
not_in: [],
in: [doc2.id], // <-- only respected this operator constraint
},
}
```
They would yield different results.
### Why?
The results were not merged into an `$and` query inside parseParams.
### How?
Merges the results within an `$and` constraint.
Fixes https://github.com/payloadcms/payload/issues/10944
Supersedes https://github.com/payloadcms/payload/pull/11011
Fixes https://github.com/payloadcms/payload/issues/11002
`buildVersionFields` was adding `null` version fields to the version fields array. When RenderVersionFieldsToDiff tried to render those, it threw an error.
This PR ensures no `null` fields are added, as `RenderVersionFieldsToDiff` can't process them. That way, those fields are properly skipped, which is the intent of `modifiedOnly`
This PR moves the logic for rendering diff field components in the
version comparison view from the client to the server.
This allows us to expose more customization options to the server-side
Payload Config. For example, users can now pass their own diff
components for fields - even including RSCs.
This PR also cleans up the version view types
Implements the following from
https://github.com/payloadcms/payload/discussions/4197:
- allow for customization of diff components
- more control over versions screens in general
TODO:
- [x] Bring getFieldPaths fixes into core
- [x] Cleanup and test with scrutiny. Ensure all field types display
their diffs correctly
- [x] Review public API for overriding field types, add docs
- [x] Add e2e test for new public API
## Description
Allows some fields to be collapsed in the version diff view. The fields
that can be collapsed are the ones which can also be collapsed in the
edit view, or that have visual grouping:
- `collapsible`
- `group`
- `array` (and their rows)
- `blocks` (and their rows)
- `tabs`
It also
- Fixes incorrect indentation of some fields
- Fixes the rendering of localized tabs in the diff view
- Fixes locale labels for the group field
- Adds a field change count to each collapsible diff (could imagine this
being used in other places)
- Brings the indentation gutter back to help visualize multiple nesting
levels
## Future improvements
- Persist collapsed state across page reloads (sessionStorage vs
preferences)
## Screenshots
### Without locales
### With locales
--------------
- [x] I have read and understand the
[CONTRIBUTING.md](https://github.com/payloadcms/payload/blob/main/CONTRIBUTING.md)
document in this repository.
## Type of change
<!-- Please delete options that are not relevant. -->
- [x] New feature (non-breaking change which adds functionality)
## Checklist:
- [x] I have added tests that prove my fix is effective or that my
feature works
- [x] Existing test suite passes locally with my changes
- [ ] ~I have made corresponding changes to the documentation~
### What?
Previously, field error messages displayed in toast notifications used
the field path to reference fields that failed validation. This
path-based approach was necessary to distinguish between fields that
might share the same name when nested inside arrays, groups, rows, or
collapsible fields.
However, the human readability of these paths was lacking, especially
for unnamed fields like rows and collapsible fields. For example:
- A text field inside a row could display as: `_index-0.text`
- A text field nested within multiple arrays could display as:
`items.0.subArray.0.text`
These outputs are technically correct but not user-friendly.
### Why?
While the previous format was helpful for pinpointing the specific field
that caused the validation error, it could be more user-friendly and
clearer to read. The goal is to maintain the same level of accuracy
while improving the readability for both developers and content editors.
### How?
To improve readability, the following changes were made:
1. Use Field Labels Instead of Field Paths:
- The ValidationError component now uses the label prop from the field
config (if available) instead of the field’s name.
- If a label is provided, it will be used in the error message.
- If no label exists, it will fall back to the field’s name.
2. Remove _index from Paths for Unnamed Fields (In the validationError
component only):
- For unnamed fields like rows and collapsibles, the _index prefix is
now stripped from the output to make it cleaner.
- Instead of `_index-0.text`, it now outputs just `Text`.
3. Reformat the Error Path for Readability:
- The error message format has been improved to be more human-readable,
showing the field hierarchy in a structured way with array indices
converted to 1-based numbers.
#### Example transformation:
##### Before:
The following fields are invalid: `items.0.subArray.0.text`
##### After:
The following fields are invalid: `Items 1 > SubArray 1 > Text`
