This PR removes the effect component that was synchronizing the record
store recoil state with the context selector equivalent state that is
used for performance on the tables.
Now we only set the context selector in parallel with the recoil state,
thus avoiding any synchronization side-effect between those two states.
Fix the following error:
Cannot use a pool after calling end on a pool
<img width="917" alt="Screenshot 2025-05-17 at 14 56 18"
src="https://github.com/user-attachments/assets/63081831-9a7e-4633-8274-de9f8a48dbae"
/>
The problem was that the datasource manager was destroying the
connections when a datasource cache expired.
This PR implements a global PostgreSQL connection pool sharing
mechanism.
- Patches pg.Pool to reuse connection pools across the application when
connection parameters match, reducing resource overhead.
- New environment variables allow enabling/disabling sharing and
configuring pool size, idle timeout, and client exit behavior.
WorkspaceDatasourceFactory will now use shared pools if enabled, this
will avoid recreating 10 connections for each pods for each workspace.
---------
Co-authored-by: Charles Bochet <charlesBochet@users.noreply.github.com>
Fixes#11762
The `copy-button` classname from scalar was overwritting the style of
our copy button. It only happened in production build and not with `npx
nx run twenty-front:start` so it was quite hard to catch. Thanks
@charlesBochet for finding the root cause.
- Removed `copy-button` classname since it was unused
- Added `?inline` when importing scalar css to have inline css
Before:
<img width="210" alt="Capture d’écran 2025-05-16 à 17 43 50"
src="https://github.com/user-attachments/assets/7256b5e6-0b61-4590-a4de-d6b28ab2b4ed"
/>
After:
<img width="230" alt="Capture d’écran 2025-05-16 à 17 43 32"
src="https://github.com/user-attachments/assets/a763172b-0566-413d-bbdd-470f28097ae4"
/>
Yoga graphql error were not correctly interpreted by the exception
handler. Mostly validations on the scalars such as bad enum options,
wrong format for uuid and such.
This PR adds a new convertGraphQLErrorToBaseGraphQLError utility
function in graphql-errors.util.ts that converts those errors to our
custom BaseGraphQLError by using the extension.http.code from the error
when possible so they can be handled the same way we treat the graphql
errors we throw ourselves.
Before
<img width="799" alt="Screenshot 2025-05-16 at 11 04 08"
src="https://github.com/user-attachments/assets/08b0a908-34d8-45a6-b315-8e211d1104ce"
/>
After
<img width="797" alt="Screenshot 2025-05-16 at 11 16 37"
src="https://github.com/user-attachments/assets/3fff0a70-6c3f-413a-b458-56030377fec9"
/>
# Modal API Refactoring
This PR refactors the modal system to use an imperative approach for
setting hotkey scopes, addressing race conditions that occurred with the
previous effect-based implementation.
Fixes#11986Closes#12087
## Key Changes:
- **New Modal API**: Introduced a `useModal` hook with `openModal`,
`closeModal`, and `toggleModal` functions, similar to the existing
dropdown API
- **Modal Identification**: Added a `modalId` prop to uniquely identify
modals
- **State Management**: Introduced `isModalOpenedComponentState` and
removed individual boolean state atoms (like
`isRemoveSortingModalOpenState`)
- **Modal Constants**: Added consistent modal ID constants (e.g.,
`FavoriteFolderDeleteModalId`, `RecordIndexRemoveSortingModalId`) for
better maintainability
- **Mount Effects**: Created mount effect components (like
`AuthModalMountEffect`) to handle initial modal opening where needed
## Implementation Details:
- Modified `Modal` and `ConfirmationModal` components to accept the new
`modalId` prop
- Added a component-state-based approach using
`ModalComponentInstanceContext` to track modal state
- Introduced imperative modal handlers that properly manage hotkey
scopes
- Components like `ActionModal` and `AttachmentList` now use the new
`useModal` hook for better control over modal state
## Benefits:
- **Race Condition Prevention**: Hotkey scopes are now set imperatively,
eliminating race conditions
- **Consistent API**: Modal and dropdown now share similar patterns,
improving developer experience
## Tests to do before merging:
1. Action Modals (Modal triggered by an action, for example the delete
action)
2. Auth Modal (`AuthModal.tsx` and `AuthModalMountEffect.tsx`)
- Test that auth modal opens automatically on mount
- Verify authentication flow works properly
3. Email Verification Sent Modal (in `SignInUp.tsx`)
- Verify this modal displays correctly
4. Attachment Preview Modal (in `AttachmentList.tsx`)
- Test opening preview modal by clicking on attachments
- Verify close, download functionality works
- Test modal navigation and interactions
5. Favorite Folder Delete Modal (`CurrentWorkspaceMemberFavorites.tsx`)
- Test deletion confirmation flow
- Check that modal opens when attempting to delete folders with
favorites
6. Record Board Remove Sorting Modal (`RecordBoard.tsx` using
`RecordIndexRemoveSortingModalId`)
- Test that modal appears when trying to drag records with sorting
enabled
- Verify sorting removal works correctly
7. Record Group Reorder Confirmation Modal
(`RecordGroupReorderConfirmationModal.tsx`)
- Test group reordering with sorting enabled
- Verify confirmation modal properly handles sorting removal
8. Confirmation Modal (base component used by several modals)
- Test all variants with different confirmation options
For each modal, verify:
- Opening/closing behavior
- Hotkey support (Esc to close, Enter to confirm where applicable)
- Click outside behavior
- Proper z-index stacking
- Any modal-specific functionality
Follow-up on https://github.com/twentyhq/twenty/pull/12007
In this PR
- adding a filter on HttpExceptionHandlerService to filter out 4xx
errors from driver handling (as we do for graphQL errors: see
useGraphQLErrorHandler hook - only filteredIssues are sent to`
exceptionHandlerService.captureExceptions()`.)
- grouping together more missing metadata issues
- attempting to use error codes as issues names in sentry to improve UI;
for now it says "Error" all the time
As discussed with @Weiko
Even though we cache the datasource, the connection expire after
10minutes in TypeORM, that might be the reason why our app is spamming
the proxy asking for connections. Also lowering the pool size.
Updated URL reference from getFrontUrl to getBaseUrl to ensure correct
hostname handling. Adjusted record filtering logic to exclude successful
records, preventing unnecessary rendering in the UI.
This PR fixes the infinite loop that was happening in `RecordShowEffect`
component due to a useEffect on `recordStoreFamilyState` which was
creating a non-deterministic open loop.
The fix was to use a recoilCallback to avoid reading a stale state from
Recoil with `useRecoilValue`, useRecoilCallback always gets the most
fresh data and commits everything before React goes on with rendering,
thus avoiding any stale value in a useEffect.
Fixes https://github.com/twentyhq/twenty/issues/11079
Fixes https://github.com/twentyhq/core-team-issues/issues/957
# Introduction
Added a no-explicit-any rule to the twenty-server, not applicable to
tests and integration tests folder
Related to https://github.com/twentyhq/core-team-issues/issues/975
Discussed with Charles
## In case of conflicts
Until this is approved I won't rebased and handle conflict, just need to
drop two latest commits and re run the scripts etc
## Legacy
We decided not to handle the existing lint error occurrences and
programmatically ignored them through a disable next line rule comment
## Open question
We might wanna activate the
[no-explicit-any](https://typescript-eslint.io/rules/no-explicit-any/)
`ignoreRestArgs` for our use case ?
```
ignoreRestArgs?: boolean;
```
---------
Co-authored-by: etiennejouan <jouan.etienne@gmail.com>
# Handling Google error
This bug is hard to reproduce so the resolution is based only on the
Sentry logs. There is not definite error code
for this error message. in the documentation.
https://developers.google.com/workspace/gmail/api/guides
This is why i added more 5xx code types in order to be sure we catch
this. In order to refine this later, i added the error code to the
message.
Fixes https://github.com/twentyhq/twenty/issues/12025
Fixes https://github.com/twentyhq/twenty/issues/11864 and
https://github.com/twentyhq/core-team-issues/issues/908
We should not send `createManyXXX` mutations with FE-forged ids in the
payload if we want to do an upsert, because that 1) prevents records
from being merged 2) triggers optimistic rendering while we can't know
before-hand which records will actually be created and which records
will only be updated
Also noticed createdBy was being overriden even for records we are
updating and not creating, which did not seem right, so fixed that too
## Description
When a calendar channel fails, its status is not reset during the
reconnect step.
In some cases, resetting is necessary—especially when we’ve deployed a
fix and users try to reconnect their accounts after the patch.
## Why reseting to initial state
Our processes are idempotent, so we can safely set the status to null to
restart the flow. This approach covers all cases and avoids edge
conditions caused by inconsistent statuses.
Fixes : https://github.com/twentyhq/twenty/issues/12026
closes#11996
- Switched the “Object destination” select in
SettingsDataModelFieldRelationForm to use
activeObjectMetadataItems.filter(...) so workflows, system, and remote
objects are excluded
- Updated useRelationSettingsFormInitialValues to fall back on the same
filtered activeObjectMetadataItems list for its initial value
This ensures workflows (and any unwanted system/remote objects) no
longer show up in the dropdown or as the default.
Fixes https://github.com/twentyhq/twenty/issues/12040
When fields are deleted but still used in workflows we do not update
create record action settings.
It breaks all following workflow execution and the user cannot update
the settings anymore.
This PR fixes the bug by filtering on existing fields.
Next step will be to clean settings on field deletion. Adding it to fast
follows.
Also lowering throttle limit because some infinite loops are not
catched.
Fixes https://github.com/twentyhq/core-team-issues/issues/956
This PR fixes a bug that appeared when switching between two kanban
views multiple times.
Steps to reproduce :
- Go to kanban view A
- Go to kanban view B
- Go back to kanban view A
Video before :
https://github.com/user-attachments/assets/4fa789ae-7187-498e-82b4-ee7896cd95d1
Video after :
https://github.com/user-attachments/assets/2b323a2d-2f76-405d-9abd-38fe72ee2214
The problem was that we allowed a hook to take a nullable parameter that
can be nullable between page switch, and threw when it was undefined.
In order to be more cautious in the future, let's be sure that we don't
throw for undefined props of a hook, when it is expected that those
props can be in an undefined state for several React render loops, while
fetching new data.
Here I identified the parent effect : `<RecordIndexBoardDataLoader />`
that loads all states for a board when we switch between views, for each
column, by calling `<RecordIndexBoardColumnLoaderEffect />` in a
`.map()`.
Each `RecordIndexBoardColumnLoaderEffect` was calling
`useLoadRecordIndexBoardColumn` with a potentially undefined
`boardFieldMetadataId`.
So to fix this, I cut the render flow higher than the throw in
`useLoadRecordIndexBoardColumn`, and by doing that I was able to ensure
that `recordIndexKanbanFieldMetadataItem` in
`RecordIndexBoardDataLoader` was already defined when using it inside
`useLoadRecordIndexBoardColumn`.
`recordIndexKanbanFieldMetadataItem` was unnecessarily fetched two
times, one time in `RecordIndexBoardDataLoader` and another time in its
child `useLoadRecordIndexBoardColumn` hook.
By implementing this flow-cut higher up, I could then remove the `|
null` in TypeScript in children components, and expect a defined value
in all the children of `RecordIndexBoardDataLoader`, thus removing the
need to ask ourselves if we should throw or not in
`useLoadRecordIndexBoardColumn`.
closes#11849
The Logo component’s fallback URL was pointing to
`/icons/android/android-launchericon-192-192.png`, but the asset lives
under `/images/icons/....`. This updates defaultPrimaryLogoUrl to use
the correct `/images/icons/android/android-launchericon-192-192.png`
path, restoring the default logo display when no primaryLogo prop is
provided.
This PR fixes an infinite loop than happens due to a useEffect in a non
deterministic manner.
The fix is to put a `isDeeplyEqual()` to avoid re-rendering the
useEffect if the value is the same.
Fixes : https://github.com/twentyhq/core-team-issues/issues/957
Context
workspaceMemberId is not always present in AuthContext. Solution
implemented here is to fetch workspaceMemberId via userId.
QRQC coming ...
Tested
- FindManyCalendarEvents / FindOneCalendarEvent
- FindManyMessages / FindOneMessage
closes https://github.com/twentyhq/twenty/issues/12027
This PR attemps at improving sentry grouping and filtering by
- Using the exceptionCode as the fingerprint when the error is a
customException. For this to work in this PR we are now throwing
customExceptions instead of internalServerError deprived of their code.
They will still be converted to Internal server errors when sent back as
response
- Filtering 4xx issues where it was missing (for emailVerification
because errors were not handled, for invalid captcha and billing errors
because they are httpErrors and not graphqlErrors)
---------
Co-authored-by: Félix Malfait <felix@twenty.com>
## Changes
- Updated the property name in the upgrade guide to reflect the
permission split in Admin Panel:
- `canImpersonate`: only provides access to impersonate functionality
- `canAccessFullAdminPanel`: provides access to all other admin panel
features
Several users have complained about not being able to read their emails
anymore.
This is because the find-messages post query hook is expecting
ObjectRecord[] as an input but is actually getting a graphql Connection
Typing was wrong. This PR fixes the typing and make sure the post query
hook always get an ObjectRecord[]
Using useEffect triggered at ActivityRichTextEditor unmount, to delete
attachments only when note is closed (and not when file block is deleted
during note update to keep command + z shortcut)
closes : https://github.com/twentyhq/twenty/issues/11229
In this PR:
- Set the default position for the DONE option of the task's status
option to `2` instead of `1`, which was the same as `IN_PROGRESS`
option's position.
- Write a command to prevent position duplicates in the database for the
task's status field.
What I've checked before setting this PR as ready to be reviewed:
- De-duplicating the position solves the issue and it's possible to edit
the field (solves the related issue)
- The upgrade command de-duplicates the position for each workspace.
There are no more DONE options with `position=2`. I ran the upgrade
command on the `database-snapshot-manager` dataset.
- Suspended workspaces aren't fixed
---
To test the script:
```ts
const scannedPositions = new Set();
let biggestPosition = -1;
// Sort options by position for consistent processing
const sortedOptions = [
{ name: 'a', position: 2 },
{ name: 'b', position: 1 },
{ name: 'c', position: 1 },
{ name: 'd', position: 2 },
].sort((a, b) => a.position - b.position);
for (const option of sortedOptions) {
if (scannedPositions.has(option.position)) {
option.position = biggestPosition + 1;
}
biggestPosition = Math.max(biggestPosition, option.position);
scannedPositions.add(option.position);
}
console.log('Sorted options:', sortedOptions);
```
Closes https://github.com/twentyhq/twenty/issues/11790
