Fixes https://github.com/payloadcms/payload/issues/6648
This PR introduces a new `useFormBackgroundProcessing` hook and a corresponding `setBackgroundProcessing` function in the `useForm` hook.
Unlike `useFormProcessing` / `setProcessing`, which mark the entire form as read-only, this new approach only disables the Publish button during autosaving, keeping form fields editable for a better user experience.
I named it `backgroundProcessing` because it should run behind the scenes without disrupting the user. You could argue that it is a bit more generic than something like `isAutosaving`, but it signals intent: Background = do not disrupt the user.
### What?
For the join field query adds ability to specify `count: true`, example:
```ts
const result = await payload.find({
joins: {
'group.relatedPosts': {
sort: '-title',
count: true,
},
},
collection: "categories",
})
result.group?.relatedPosts?.totalDocs // available
```
### Why?
Can be useful to implement full pagination / show total related
documents count in the UI.
### How?
Implements the logic in database adapters. In MongoDB it's additional
`$lookup` that has `$count` in the pipeline. In SQL, it's additional
subquery with `COUNT(*)`. Preserves the current behavior by default,
since counting introduces overhead.
Additionally, fixes a typescript generation error for join fields.
Before, `docs` and `hasNextPage` were marked as nullable, which is not
true, these fields cannot be `null`.
Additionally, fixes threading of `joinQuery` in
`transform/read/traverseFields` for group / tab fields recursive calls.
Added a new Reserved Field Names section to the migration guide.
Clarified that certain field names (`__v`, `salt`, `hash`, `file`, etc.)
are reserved for internal use and will be sanitized from the config if
used.
Included additional reserved names specific to `MongoDB`, `auth`-enabled
collections, and `upload`-enabled collections.
Added a note recommending against using field names with an underscore
(`_`) prefix, as they are reserved for internal columns and may cause
conflicts in `SQL` and other contexts.
Fixes#11159
This reverts commit 69c0d09 in #11390.
In order to future proof column prefs, it probably is best to continue
to use the current shape. This change was intended to ensure that as
little transformation to URL params was made as possible for #11387, but
we will likely transform them after all.
This will ensure that we can add support for additional properties over
time, as needed. For example, if we hypothetically wanted to add a
custom `label` or similar feature to columns prefs, it would make more
sense to use explicit properties to identity `accessor` and `active`.
For example:
```ts
[
{
accessor: "title",
active: true,
label: 'Custom Label' // hypothetical
}
]
```
This bumps next.js to 15.2.0 in our monorepo, as well as all @types/react and @types/react-dom versions. Additionally, it removes the obsolete `peerDependencies` property from our root package.json.
This PR also fixes 2 bugs introduced by Next.js 15.2.0. This highlights why running our test suite against the latest Next.js, to make sure Payload is compatible, version is important.
## 1. handleWhereChange running endlessly
Upgrading to Next.js 15.2.0 caused `handleWhereChange` to be continuously called by a `useEffect` when the list view filters were opened, leading to a React error - I did not investigate why upgrading the Next.js version caused that, but this PR fixes it by making use of the more predictable `useEffectEvent`.
## 2. Custom Block and Array label React key errors
Upgrading to Next.js 15.2.0 caused react key errors when rendering custom block and array row labels on the server. This has been fixed by rendering those with a key
## 3. Table React key errors
When rendering a `Table`, a React key error is thrown since Next.js 15.2.0
This PR improves existing JSX converter docs and adds 2 new sections:
- **converting internal links** - addresses why a `"found internal link, but internalDocToHref is not provided"` error is thrown, and how to get around it
- **Overriding default JSX Converters**
Lexical server features are able to add components to the import map through the `componentImports` property. As of now, the client feature did not have access to those. This is usually not necessary, as those import map entries are used internally to render custom components server-side, e.g. when a request to the form state endpoint is made.
However, in some cases, these import map entries need to be accessed by the client feature (see "Why" section below).
This PR ensures that keyed `componentImports` entries are made available to the client feature via the new `featureClientImportMap` property.
## Why?
This is a prerequisite of the lexical [wrapper blocks PR](https://github.com/payloadcms/payload/pull/9289), where wrapper block custom components need to be made to the ClientFeature. The ClientFeature is where the wrapper block node is registered - in order to generate the wrapper block node, we need access to the component
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
Previously, behavior with custom IDs and `select` query was incorrect.
By default, the `id` field is guaranteed to be selected, even if it
doesn't exist in the `select` query, this wasn't true for custom IDs.
Previously, `updateOne` was using `buildFindManyArgs` and `findFirst` just to retrieve the ID of the document to update, which is a huge function that's not necessary to run just to get the document ID.
This PR refactors it to use a simple `db.select` query to retrieve the ID
This PR fixes an issue where bulk upload attempts to generate thumbnails
for non-image files, causing errors on the page.
The fix ensures that thumbnail generation is skipped for non-image
files, preventing unnecessary errors.
Fixes#10428
Previously, `hasNextPage` was working incorrectly with polymorphic joins
(that have an array of `collection`) in MongoDB.
This PR fixes it and adds extra assertions to the polymorphic joins
test.
---------
Co-authored-by: Jarrod Flesch <jarrodmflesch@gmail.com>
Transforms how column prefs are stored in the database. This change
reduces the complexity of the `columns` property by removing the
unnecessary `accessor` and `active` keys.
This change is necessary in order to [maintain column state in the
URL](https://github.com/payloadcms/payload/pull/11387), where the state
itself needs to be as concise as possible. Does so in a non-breaking
way, where the old column shape is transformed as needed.
Here's an example:
Before:
```ts
[
{
accessor: "title",
active: true
}
]
```
After:
```ts
[
{
title: true
}
]
```
Since codeowner approvals are not currently required, the codeowners
file is only serving to add reviewers to PRs.
Removing the codeowners file for now as this is not desired. Can be
re-introduced at a later date if required approvers are implemented.
### What?
Unable to update json fields externally. For example, calling `setValue`
on a json field would not be reflected in the admin panel UI.
### Why?
JSON fields use the monaco editor to manage state internally, so
programmatically updating the value in state does not change the
internal value.
### How?
Set a ref when the user updates the value and then unset the ref after
the change is complete.
Inside the hook that watches `value`, if the value changed and the
change came from the system (i.e. a programmatic change) refresh the
editor by adjusting its key prop. If the change was made by the user,
there is no need to refresh the editor.
Fixes https://github.com/payloadcms/payload/issues/10819
Continuation of https://github.com/payloadcms/payload/pull/11372 but for our collection, global and auth operations
Previously, we were quite frequently using `.reduce()` to run hooks. This PR replaces them with simple `for` loops, which is less overhead, less code, less confusing and simpler to understand.
Continuation of #11008. When `filterOptions` are set on a relationship
field that is _nested within another field_, those filter options are
not applied to `Filter` component in the list view. This is because we
were only shallowly resolving filter options on top-level fields, as
opposed to recursively traversing fields to resolve them even when
deeply nested.
Updated `formatBreadcrumb` to fall back to an empty string if the
`useAsTitle` field for the document is undefined.
This handles cases where the field is optional or not filled out,
ensuring the label is never `undefined`.
Fixes#10377
This PR ensures that when `titleField.label` is provided as an object,
it is correctly translated and displayed in the search filter's
placeholder.
Previously, the implementation only supported string values, which could
lead to issues with object type labels. With these changes, object type
labels will now properly show as intended in the search filter.
Fixes#11348
Previously, we were quite frequently using `.reduce()` to sequentially run field hooks. This PR replaces them with simple `for` loops, which is less overhead, less code, less confusing and simpler to understand.
Additionally, it refactors `mergeLocaleActions` which previously was unnecessarily complex. They no longer entail async code, thus we no longer have to juggle with promises
When blocks have custom row labels, those row labels become stale when
reordering blocks. After moving a block, for example, the row label will
jump back the original block until form state returns with the proper
rendering order. This is especially evident on slow networks.
### What?
Two things:
1. Users unassigned to a tenant could not access their own account
2. Custom `tenantsArrayFieldName` and `tenantsArrayTenantFieldName`
configurations were not being used in all cases
### Why?
1. The access constraint provided by the plugin would not allow them to
make changes to their own account
2. `getUserTenantIDs` and `afterTenantDelete` were not using the custom
field names properly
### How?
1. Adds constraint for users allowing them to manage their own account
by default. Externally nothing has changed. If you need to lock your
users access control down you should do that just as you would without
this plugin.
2. Threads the field names through for usage.
Fixes https://github.com/payloadcms/payload/issues/11317
### What?
The `plugin-nested-docs` returns an array of breadcrumbs - the
`resaveChildren` file accidentally processed the breadcrumbs twice, once
where the data is updated and once within the `populateBreadcrumbs`
function which was causing the objects to be double nested.
### How?
Removes the extra nesting from `resaveChildren` file and allows the
`populateBreadcrumbs` to return the final data.
Fixes#10855
### What?
Information that locale fields in database are changing to a simpler
data structure in v3.
### Why?
Simple data migration is not enough to get it working, I had to spend
quite some time to figure out migration files and still it required some
manual input. Maybe others will find it useful when starting a v3
migration.
### How?
I had to do some custom migration scripts on my own to get v3 to work
with existing pages data.
---------
Co-authored-by: Sasha <64744993+r1tsuu@users.noreply.github.com>
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`
Exports the `useTenantSelection` hook from the multi-tenant plugin, this
way other users can import and use the hook along with it's methods.
Can be imported:
```ts
import { useTenantSelection } from '@payloadcms/plugin-multi-tenant/client'
```
The context returned:
```ts
type ContextType = {
/**
* Array of options to select from
*/
options: OptionObject[]
/**
* The currently selected tenant ID
*/
selectedTenantID: number | string | undefined
/**
* Prevents a refresh when the tenant is changed
*
* If not switching tenants while viewing a "global", set to true
*/
setPreventRefreshOnChange: React.Dispatch<React.SetStateAction<boolean>>
/**
* Sets the selected tenant ID
*
* @param args.id - The ID of the tenant to select
* @param args.refresh - Whether to refresh the page after changing the tenant
*/
setTenant: (args: { id: number | string | undefined; refresh?: boolean }) => void
}
```
Small performance improvement for types generation and anywhere else
`configToJSONSchema` can be used.
We did a deep copy of the whole sanitized entity config, which can be
expensive. Now, to do the needed mutation of `flattenedFields`, we just
create a new reference of `flattenedFields` for that.
When rendering a list drawer, you can pass a custom `onSelect` callback
to execute when the user clicks on the linked cell within the table. The
underlying handler, however, only passes the `docID` and
`collectionSlug` args through the callback, rather than the document
itself. This makes it impossible to perform side-effects that require
the data of the row that was selected.
Instances of this callback were also largely untyped.
Needed for #11330.
Bulk edit controls are currently displayed within the search bar of the
list view. This doesn't make sense from a UX perspective, as the current
selection is displayed somewhere else entirely. These controls also take
up a lot of visual real estate which is beginning to get overused
especially after the introduction of "list menu items" in #11230, and
the potential introduction of "saved filters" controls in #11330.
Now, they are rendered contextually _alongside_ the selection count. To
make room for these new controls, they are displayed in plain text and
the entity labels have been removed from the selection count.
### What?
Updated `docker-compose.yml` to use `Postgres` by default, with
`MongoDB` commented out in the templates that are by default using the
postgres adapter.
Fixes#11322
`payload.find` queries can be made faster by specifying `limit: 1` and `pagination: false` when only the first document is needed. This PR applies those options to various queries to improve performance.
Previously, the `bin` configuration wasn't working at all.
Possibly because in an ESM environment this cannot work, because
`import` always returns an object with a default export under the
`module` key.
```ts
const script: BinScript = await import(pathToFileURL(userBinScript.scriptPath).toString())
await script(config)
```
Now, this works, but you must define a `script` export from your file.
Attached an integration test that asserts that it actually works. Added
documentation on how to use it, as previously it was missing.
This can be also helpful for plugins.
### Documentation
Using the `bin` configuration property, you can inject your own scripts
to `npx payload`.
Example for `pnpm payload seed`:
Step 1: create `seed.ts` file in the same folder with
`payload.config.ts` with:
```ts
import type { SanitizedConfig } from 'payload'
import payload from 'payload'
// Script must define a "script" function export that accepts the sanitized config
export const script = async (config: SanitizedConfig) => {
await payload.init({ config })
await payload.create({ collection: 'pages', data: { title: 'my title' } })
payload.logger.info('Succesffully seeded!')
process.exit(0)
}
```
Step 2: add the `seed` script to `bin`:
```ts
export default buildConfig({
bin: [
{
scriptPath: path.resolve(dirname, 'seed.ts'),
key: 'seed',
},
],
})
```
Now you can run the script using:
```sh
pnpm payload seed
```
Improves performance and optimizes memory usage for mongodb adapter by
cutting down copying of results via `JSON.parse(JSON.stringify())`.
Instead, `transform` does necessary transformations (`ObjectID` ->
`string,` `Date` -> `string`) without any copying
Previously, we were calling `getNavPrefs` (a payload.find call) three times for every single page load.
This PR:
1. Ensures that `getNavPrefs` is called only once per page load, reducing two unnecessary `payload.find` calls every time a page is loaded or navigated to.
2. Adds `pagination: false` to the `payload.find` call, making it more efficient and improving performance.
## How?
We were using React's cache to ensure that navigation preferences (`getNavPrefs`) were fetched only once per request. However, this wasn't working as expected because the first argument of `getNavPrefs` was an object. Each time it was called, a new object reference was passed, preventing React from caching it properly.
To fix this, this PR ensures that only primitive values are used as arguments for caching, following best practices and making the cache function work as intended.
Just like https://github.com/payloadcms/payload/pull/11269, we stop passing through `req` to db operations in `checkDocumentLockStatus`.
After extensive testing, this seems to get rid of all transaction errors that occurred when I was testing autosave against a remote mongo DB.
We keep the `req` for postgres, as it mysteriously breaks in CI - this cannot be reproduced locally
