Field validations currently run very often, such as within form state on
type. This can lead to serious performance implications within the admin
panel if those validation functions are async, especially if they
perform expensive database queries. One glaring example of this is how
all relationship and upload fields perform a database lookup in order to
evaluate that the given value(s) satisfy the defined filter options. If
the field is polymorphic, this can happen multiple times over, one for
each collection. Similarly, custom validation functions might also
perform expensive tasks, something that Payload has no control over.
The fix here is two-fold. First, we now provide a new `event` arg to all
`validate` functions that allow you to opt-in to performing expensive
operations _only when documents are submitted_, and fallback to
significantly more performant validations as form state is generated.
This new pattern will be the new default for relationship and upload
fields, however, any custom validation functions will need to be
implemented in this way in order to take advantage of it. Here's what
that might look like:
```
[
// ...
{
name: 'text'
type: 'text',
validate: async (value, { event }) => {
if (event === 'onChange') {
// Do something highly performant here
return true
}
// Do something more expensive here
return true
}
}
]
```
The second part of this is to only run validations _after the form as
been submitted_, and then every change event thereafter. This work is
being done in #10580.
### What
This PR adds a filtering mechanism to exclude certain reserved fields
from being displayed in the `Edit Many` drawer for bulk uploads. This
ensures that only relevant fields are available for bulk editing.
### Why
Fields like `filename`, `mimeType`, and `filesize` are not intended to
be edited in bulk. Filtering these fields streamlines the interface and
focuses on fields that are meaningful for bulk operations.
### How
- Introduced a `filterOutUploadFields` utility to exclude reserved
fields from the field selection in bulk uploads.
- Applied this filter to the `Edit Many` drawer, ensuring a more
relevant and user-friendly experience.
- Reserved fields include properties like `file`, `mimeType`, `url`,
`width`, `height`, and others that are not applicable for bulk editing.
### What?
This PR introduces the ability to bulk edit multiple uploads
simultaneously within the `Edit all` option for bulk uploads. Users can
now select fields to update across all selected uploads in a single
operation.
### Why?
Managing multiple uploads individually can be time-consuming and
inefficient, especially when updating common fields. This feature
streamlines the process, improving user experience and productivity when
handling bulk uploads.
### How?
* Added an `Edit Many` drawer component specific to bulk uploads that
allows users to select fields for bulk editing.
* Enhanced the FormsManager and related logic to ensure updates are
applied consistently across all selected uploads.
- Fixes collection and tab descriptions not having a bottom padding:
- Deprecates `description` on tabs config in favour of
`admin.description` but supports both
- Adds test to make sure `admin.description` renders with static string
and function too for tabs
Previously, we were unnecessarily passing the `ClientCollectionConfig`
down from the Table to the Client, even though the client cell
components could simply access it via the `useConfig` hook. This
resulted in redundant data being sent to the client for every single
table cell. Additionally, we were performing a deep copy of the
`ClientCollectionConfig`, which wasted both memory and CPU resources on
the server.
---------
Co-authored-by: Jacob Fletcher <jacobsfletch@gmail.com>
Custom block row labels defined on `admin.components.Label` were not
rendering despite existing in the config. Instead, if a custom label
component was defined on the _top-level_ blocks field itself, it was
incorrectly replacing each blocks label _in addition_ to the field's
label. Now, custom labels defined at the field-level now only replace
the field's label as expected, and custom labels defined at the
block-level are now supported as the types suggest.
Following https://github.com/payloadcms/payload/pull/10551, I found and
fixed a handful more bugs:
- When writing to the input, the results that were already there were
not cleaned, causing incorrect results to appear.
- the scroll was causing an infinite loop that showed repeated elements
- optimization: only the required field is selected (not required)
- refs are set to the initial value to avoid a state where nothing can
be searched
### What?
The `pasteURL` feature for Upload fields has been updated to support
both **client-side** and **server-side** URL fetching. Previously, users
could only paste URLs from the same domain as their Payload instance
(internal) or public domains, which led to **CORS** errors when trying
to fetch files from external URLs.
Now, users can choose between **client-side fetching** (default) and
**server-side fetching** using the new `pasteURL` option in the Upload
collection config.
### How?
- By default, Payload will attempt to fetch the file client-side
directly in the browser.
- To enable server-side fetching, you can configure the new `pasteURL`
option with an `allowList` of trusted domains.
- The new `/api/:collectionSlug/paste-url` endpoint is used to fetch
files server-side and stream them back to the browser.
#### Example
```
import type { CollectionConfig } from 'payload'
export const Media: CollectionConfig = {
slug: 'media',
upload: {
// pasteURL: false, // Can now disable the pasteURL option entirely by passing "false".
pasteURL: {
allowList: [
{
hostname: 'payloadcms.com', // required
pathname: '',
port: '',
protocol: 'https', // defaults to https - options: "https" | "http"
search: ''
},
{
hostname: 'example.com',
pathname: '/images/*',
},
],
},
},
}
```
### Why
This update provides more flexibility for users to paste URLs into
Upload fields without running into **CORS errors** and allows Payload to
securely fetch files from trusted domains.
This adds support for calculating and displaying file sizes for JPEG XL
images.
Image resizing is not supported by sharp out-of-the-box yet:
https://github.com/lovell/sharp/issues/2731
- reduces unnecessary shallow copying within operations by removing
unnecessary spreads or .map()'s
- removes unnecessary `deleteMany` call in `deleteUserPreferences` for
auth-enabled collections
- replaces all instances of `validOperators.includes` with
`validOperatorMap[]`. O(n) => O(1)
- optimizes the `sanitizeInternalFields` function. Previously, it was
doing a **lot** of shallow copying
Data for the EditMany view was fetched even though the EditMany Drawer
was not open. This, in combination with the router.replace call to add
the default limit query param, caused the root layout to re-render
This PR modifies `tsconfig.base.json` by setting the following
strictness properties to true: `strict`, `noUncheckedIndexedAccess` and
`noImplicitOverride`.
In packages where compilation errors were observed, these settings were
opted out, and TODO comments were added to make it easier to track the
roadmap for converting everything to strict mode.
The following packages now have increased strictness, which prevents new
errors from being accidentally introduced:
- storage-vercel-blob
- storage-s3*
- storage-gcs
- plugin-sentry
- payload-cloud*
- email-resend*
- email-nodemailer*
*These packages already had `strict: true`, but now have
`noUncheckedIndexedAccess` and `noImplicitOverride`.
Note that this only affects the `/packages` folder, but not
`/templates`, `/test` or `/examples` which have a different `tsconfig`.
The `access.read` function executed within form state was missing the
`id` arg, and was also incorrectly setting `data` as `doc`. When
building form state, there is no concept of a "doc" because it is
possible to build form state using only a subset of fields. There is
"data", however, which represents the schema path at the entry point of
the function. Similarly, when building form state on within an
`onChange` function, for example, we do not send the original doc
through the request, which is what "doc" would represent. Instead, we
send either `data` or `formState`, both of which could represent a
_modified_ doc. This particular invocation of read access does not
effect the visibility of fields themselves, but rather their return
values from the form state endpoint. Field visibility is determined at
the request level.
This significantly optimizes the form state, reducing its size by up to
more than 3x and improving overall response times. This change also has
rolling effects on initial page size as well, where the initial state
for the entire form is sent through the request. To achieve this, we do
the following:
- Remove `$undefined` strings that are potentially attached to
properties like `value`, `initialValue`, `fieldSchema`, etc.
- Remove unnecessary properties like empty `errorPaths` arrays and empty
`customComponents` objects, which only need to exist if used
- Remove unnecessary properties like `valid`, `passesCondition`, etc.
which only need to be returned if explicitly `false`
- Remove unused properties like `isSidebar`, which simply don't need to
exist at all, as they can be easily calculated during render
## Results
The following results were gathered by booting up each test suite listed
below using the existing seed data, navigating to a document in the
relevant collection, then typing a single letter into the noted field in
order to invoke new form-state. The result is then saved to the file
system for comparison.
| Test Suite | Collection | Field | Before | After | Percentage Change |
|------|------|---------|--------|--------|--------|
| `field-perf` | `blocks-collection` | `layout.0.field1` | 227kB | 110
kB | ~52% smaller |
| `fields` | `array-fields` | `items.0.text` | 14 kB | 4 kB | ~72%
smaller |
| `fields` | `block-fields` | `blocks.0.richText` | 25 kB | 14 kB | ~44%
smaller |
After working on this I found a more accurate way to reproduce the bug:
- in the issue repro, type a letter in the select menu.
- delete the letter and wait for the debounce (300ms)
- type another letter.
- in devtools, you should see that the query increases the pagination by
+1. With this change, the pagination is reset when the input changes.
Fixes#10496
I'd like to do integration testing. But since there is no isolated `/ui`
test yet, this requires some planning. I have it pending.
### What?
Allows a user to delete a scheduled publish event after it has been
added:
### Why?
Previously a user had no control over making changes once scheduled.
### How?
Extends the `scheduledPublishHandler` server action to accept a
`deleteID` for the event that should be removed and exposes this to the
user via the admin UI in a new column in the Upcoming Events table.
Continuation of #10540. Passes server props to custom label components
rendered within table columns, such as the list view. This way custom
server components can have access to `payload`, `i18n`, etc. as
expected.
Since postgres uses number IDs by default, when we were storing the
relationship field value with postgres we weren't able to query it
This fixes that problem by casting the ID to always a string making it
safe for querying inside the JSON field
### What?
The list view was throwing a hydration error for date fields.
### Why?
The issue really stems from the fact that cells are client rendered. We
dynamically load the dateFNS Locale object at runtime to keep the bundle
size small — which makes sense. But on the first render that means we do
not have the Locale object from the known locale so the server/client
determines what to render it as. This causes a mismatch when hydrating.
In the future I think cells could be server rendered and that would
solve the need for this fix which adds "Loading..." while the dateFNS
Locale is loaded.
I think server rendering the cells would allow us to import the dateFNS
Locale inline (blocking) and then pass the rendered string down to the
list view. This should work because we **know** the locale on the
server.
### How?
In this PR, it adds a "Loading..." fallback for the date cell if the
dateFNS Locale has not loaded yet.
Fixes#10044
This PR adds `cacheTags: boolean` (default `true`) to allow users to
disable the appended document updatedAt value in the case of hosting
with third party CDNs which may not allow additional search params and
throw an error.
It also fixes how we append this value to consider the case where the
URL already contains parameters and appends it with `&` instead.
In the future `cacheTags` can be made an object to allow granularity for
disabling `eTag` headers used for caching as well.
The cache tag control should help with these two issues:
- Fixes https://github.com/payloadcms/payload/issues/9880
- Fixes https://github.com/payloadcms/payload/issues/9993
The appending of the value correctly addresses this:
- Fixes https://github.com/payloadcms/payload/issues/10139
Fixes https://github.com/payloadcms/payload/issues/10436
Fixes an issue where drafts' updatedAt timestamp is not being updated.
We weren't updating the `versionData` to have the right timestamp in the
saveVersion operation when drafts were being updated.
Added e2e tests to make sure 'Last Modified' is always different in both
autosave and non-autosave drafts.
Editing fields during a locale change on slow networks can lead to
changes being reset when the new form state is returned. This is because
the fields receive new values from context when the new locale loads in,
which may have occurred _after_ changes were made to the fields. The fix
is to subscribe to a new `localeIsLoading` context which is set
immediately after changing locales, and then reset once the new locale
loads in. This also removes the misleading `@deprecated` flag from the
`useLocale` hook itself.
The `ListPreferences` and `ColumnPreferences` types were defined
multiple times in different places, making it difficult to make changes
across the board. Now, the `ListPreferences` type is exported directly
from `payload` alongside the other preferences types, and
`ColumnPreferences` has been merged directly into this type to simplify
usage as this is not a standalone preference.
Currently, unless a locale is present in the URL search params, the
locale context is instantiated using the default locale until prefs load
in client-side. This causes the locale selector to briefly render in
with the incorrect (default) locale before being replaced by the proper
locale of the request. For example, if the default locale is `en`, and
the page is requested in `es`, the locale selector will flash with
English before changing to the correct locale, even though the page data
itself is properly loaded in Spanish. This is especially evident within
slow networks.
The fix is to query the user's locale preference server-side and thread
it into the locale provider to initialize state. Because search params
are not available within server layouts, we cannot pass the locale param
in the same way, so we rely on the provider itself to read them from the
`useSearchParams` hook. If present, this takes precedence over the
user's preference if it exists.
Since the root page also queries the user's locale preference to
determine the proper locale across navigation, we use React's cache
function to dedupe these function calls and ensure only a single query
is made to the db for each request.
### What?
In the scheduled publish feature, it is possible to open the modal to
schedule publishing a document even before you have saved a draft. This
change removes the option to open the drawer to even attempt this.
This also fixes an issue if you do not have permission to publish, you
cannot schedule publish either.
### Why?
There were numerous problems:
1. The schedule publish events would show all other publish events for
the collection without an ID in the query
2. Scheduling a publish would not work without an ID for a document to
publish
### How?
Removes the Schedule Publish menu item and drawer if you are on a
collection without an ID or you do not have permission to publish.
Without an ID:
With an ID:
Perviously, you would always have the option to Schedule Publish.
### What?
The status indicator was not updating properly when users were clicking
"unpublish" and "revert changes"
### Why?
hasPublishedDoc, unpublishedVersionCount and
mostRecentVersionIsAutosaved states were not being updated properly.
### How?
This PR updates the variables when interacting with the status actions
and sets mostRecentVersionIsAutosaved to false when the publish button
is clicked.
Fixes https://github.com/payloadcms/payload/issues/9531
The localization e2e test is notorious for flaking, consuming a lot of
time and resources continually retrying. This was because the test was
attempting to click DOM elements using selectors that never resolve, or
attempting to click inaccessible DOM nodes such as those behind a modal.
The fix is to ensure that the dot nav, for example, is disabled while
form state loads, and that modals are properly closed prior to executing
subsequent tests, etc. Tests also needed to explicitly check for
_enabled_ states before performing click actions, rather than simply
awaiting their visibility.
Fixes#10018. When toggling columns, then sorting them, the table is
reset to the collection's default columns instead of the user's
preferred columns. This is because when sorting columns, a stale
client-side cache of the user's preferences is used to update their sort
preference. This is because when column state is constructed
server-side, it completely bypasses the client-side cache. To fix this,
sort preferences are now also set on the server right alongside column
preferences, which performs an upsert-like operation to ensure that no
existing preferences are lost.
