Continuation of #6644
Now chromium browser is used in workspaces tests instead of firefox and
screenshots after each test are properly saved in one folder when run
from IDE and from terminal using `yarn test:e2e` command
I have investigated the performance of our frontend vite build:
`npx nx run twenty:start` of `npx nx run twenty:build`
RAM usage:
- 160Mb: vite serve
- background typescript checker: 2.5GB
- background eslint checker: 3.5GB
I'm introducing two environment variables in FE .env to disable these
checkers on lower configuration (and to disable them from CD build):
```
# VITE_DISABLE_TYPESCRIPT_CHECKER=true
# VITE_DISABLE_ESLINT_CHECKER=true
```
This PR was created by [GitStart](https://gitstart.com/) to address the
requirements from this ticket:
[TWNTY-6692](https://clients.gitstart.com/twenty/5449/tickets/TWNTY-6692).
This ticket was imported from:
[TWNTY-6692](https://github.com/twentyhq/twenty/issues/6692)
---
### Description
The problem is not related to the API, what was happening was a failure
in the form validation, because in the changed file, the `currencyCode`
value, which should be a string with single quotes around it, was
receiving single quotes again, unnecessarily, and this affected field
validation in
`packages/twenty-front/src/modules/object-record/record-field/validation-schemas/currencyFieldDefaultValueSchema.ts`
please, make this change below before testing the PR(to fix a bug, the
slice is not updated yet):\
<aa4ae53fb4>
### Demo
<https://www.loom.com/share/2ce130f2e2fe46868e9b1e9119f65cde?sid=dbcb2da2-3641-423c-bdfc-01b0fc52162a>
### Refs
#6692Fixes#6692
---------
Co-authored-by: gitstart-twenty <gitstart-twenty@users.noreply.github.com>
Co-authored-by: Marie Stoppa <marie.stoppa@essec.edu>
The update of all fields was caused by `useContextSelector` not being
properly implemented.
As it is a memoization library the `useRecordFieldValue` hook wasn't
giving to the library the required things to allow memoization.
I just added recordId + fieldName to the memoization function so that
`useContextSelector` doesn't recompute itself whenever any record
changes.
## Features
- Fetch a workflow and display it in a tree with the React Flow library
- The nodes are positioned by an algorithm
- The feature is put behind a feature flag. The `/workflow/:id` route is
disabled if the flag is off.
- I started implementing a right drawer. That's a big WIP and it will be
finished in another PR.
## How to test this feature
1. Create a workflow instance in the database through a GraphQL query.
See below for instructions.
2. After enabling the feature flag, you should be able to see the
workflow you created in the workflows list. To visualize the workflow,
go to the `/workflow/:id` page where the id is the id of the workflow.
See the video for a quick way to do so.
```gql
// First
mutation createWorkflow($data: WorkflowCreateInput!) {
createWorkflow(data: $data) {
id
}
}
// Result
{
"data": {
"name": "test"
}
}
// Second
mutation createWorkflowVersion($data: WorkflowVersionCreateInput!) {
createWorkflowVersion (data: $data) {
id
}
}
// Result
{
"data": {
"name": "v1",
"trigger": {
"name": "trigger",
"displayName": "New or Updated Row",
"type": "DATABASE_EVENT",
"settings": {
"eventName": "company.created",
"triggerName": "Company Created"
},
"nextAction": {
"name": "step_1",
"displayName": "Code",
"type": "CODE",
"valid": true,
"settings": {
"serverlessFunctionId": "function_id",
"errorHandlingOptions": {
"retryOnFailure": {
"value": false
},
"continueOnFailure": {
"value": false
}
}
}
}
},
"workflowId": "workflow_id"
}
}
```
https://github.com/user-attachments/assets/42bbd98c-5e13-447c-9307-461a18ac2195
At field creation we are checking the availability of the name by
comparing it to the other fields' names' on the object; but for
composite fields the fields' names' as indicated in the repository do
not exactly match the column names' on the tables (e.g "createdBy" field
is actually represented by columns createdByName, createdBySource etc.).
In this PR we prevent the conflict with the standard composite fields'
names.
There is still room for errors with the custom composite fields: for
example a custom composite field "address" of type address on a custom
object "listing" will introduce the columns addressAddressStreet1,
addressAddressStreet2 etc. while we won't prevent the user from later
creating a custom field named "addressAddressStreet1".
For now I decided not to tackle this as this seem extremely edgy + would
impact performance on creation of all fields while never actually useful
(I think).
Fixes (#6643)
- According to `aws-sdk` v3, each service is packaged as a separate
module, and the codebase uses all of the required AWS services
accordingly. Therefore, removing the direct dependency on `aws-sdk:v2`
to resolve the warning shown in (#6643)
Co-authored-by: Charles Bochet <charles@twenty.com>
- improvements on serverless function behavior (autosave performances,
deploy on execution only)
- add versioning to serverless functions
- add a publish endpoint to create a new version of a serverless
function
- add deploy and reset to lastVersion button in the settings section:
<img width="736" alt="image"
src="https://github.com/user-attachments/assets/2001f8d2-07a4-4f79-84dd-ec74b6f301d3">
@lucasbordeau @charlesBochet
Issue #4826
Could u review this changes.
Let me know what do you think.
---------
Co-authored-by: Lucas Bordeau <bordeau.lucas@gmail.com>
We want to avoid the nested structure of active pieces. Steps to execute
will now be separated from the trigger. It will be an array executed
sequentially.
For now a step can only be an action. But at some point it will also be
a branch or a loop
### Description:
- we move all logic about the unmatchedOptions to a new component called
UnmatchColumn, because as it will be a full line in the table, it was
better to update where the component will be rendered
- In the latest changes to keep the columns when we change the step to
step 3 and go back to step 2, we added a fallback state
initialComputedColumnsState that saves the columns and only reverts the
updates when we go back to step 1 or close by clicking the X button
### Refs:
#6135
```
It was necessary to add references and floating styles to the generic component to fix the bug when the last option was open and the dropdown was being hidden in the next row of the spreadsheet table. We fixed the same problem that occurs in the companies table as well
```
we used this approach mentioned on this documentation to be able to use
the hook without calling it on each component, we are calling only once,
on the shared component
<https://floating-ui.com/docs/useFloating#elements>\
before:
now:
###
Demo: <https://jam.dev/c/e0e0b921-7551-4a94-ac1c-8a50c53fdb0c>
Fixes#6135
NOTES: the enter key are not working on main branch too
---------
Co-authored-by: gitstart-twenty <gitstart-twenty@users.noreply.github.com>
Co-authored-by: Lucas Bordeau <bordeau.lucas@gmail.com>
WorkspaceMemberId is mandatory in the jwt token generated for a given
user on a given workspace.
However, when a user signs up, it does not have a workspaceMemberId yet.
Fix https://github.com/twentyhq/twenty/issues/6669
- create a commun function `startWorkflowRun` that both create the run
object and the job for executing the workflow
- use it in both the `workflowEventJob` and the `runWorkflowVersion`
endpoint
Bonus:
- use filtering for exceptions instead of a util. It avoids doing a try
catch in all endpoint
Issue #6630
It seems this bug is caused by `targetableObjects` being assigned an
empty array, which then leads to an error due to it being undefined.
I've made some changes that should address the issue, but I would
appreciate any feedback or suggestions on alternative solutions.
Please let me know if there is a better approach to resolving this.
Thank you!
https://github.com/user-attachments/assets/d6409798-3320-49b3-834f-2b6888847ed8
Fix
[#web](https://github.com/orgs/twentyhq/projects/1/views/3?pane=issue&itemId=75329194)
This PR does 2 things:
- migrate webhooks to TwentyORM
- Fix inversion between objectNameSingular and operation in webhook
eventName. It is stored as {objectNameSingular}.{operation} and we were
querying {operation}.{objectNameSingular}
We have found the root cause of the issue:
- when using a datasource (including the cached ones), we are fetching
ObjectMetadataCollection from cache (700kB). Datasource usage is
happening any time we are using twentyORM, which is everywhere in the
jobs and in some resolvers (including the GetCurrentUser one). This is
leading to a high load on redis and leading to the performance issues we
are seeing.
- we actually don't need to fetch this objectMetadataCollection while
using a cached datasource, only when we instantiate a new one
This PR was created by [GitStart](https://gitstart.com/) to address the
requirements from this ticket:
[TWNTY-5370](https://clients.gitstart.com/twenty/5449/tickets/TWNTY-5370).
This ticket was imported from:
[TWNTY-5370](https://github.com/twentyhq/twenty/issues/5370)
---
### Description
- We updated the logic in
packages/twenty-server/src/engine/metadata-modules/object-metadata/object-metadata.service.ts
Test cases:
1. Ensure that when an object is disabled, all related relationships are
also disabled.
a. Example disable the people object
b. Check the company object and verify that the people field has been
disabled too
c. Check the opportunity object and check that the point of contact
field has been disabled too
2. Verify that when a previously disabled object is restored, the
relationships are also restored.
3. Ensure that previously disabled relationships remain disabled when
the object is disabled and later restored.
4. Verify that relationships of a disabled object are not visible in the
UI.
5. Ensure that relationships to a disabled object are marked as inactive
in the data models screen
### Refs
#5370
### Demo
<https://www.loom.com/share/2b0a91f463ca4e02a6963f9a8796a0d9?sid=1e9c4fb8-8fb9-4c6c-b43a-c50f3776e1d3>
Fixes#5370
---------
Co-authored-by: gitstart-twenty <gitstart-twenty@users.noreply.github.com>
Co-authored-by: Marie Stoppa <marie.stoppa@essec.edu>
## Context
As we grow, the messaging scripts are experiencing performance issues
forcing us to temporarily disable them on the cloud.
While investigating the performance, I have noticed that generating the
entity schema (for twentyORM) in the repository is taking ~500ms locally
on my Mac M2 so likely more on pods. Caching the entitySchema then!
I'm also clarifying naming around schemaVersion and cacheVersions ==>
both are renamed workspaceMetadataVersion and migrated to the workspace
table (the workspaceCacheVersion table is dropped).
Fixes: #6436
Changes made:
- Added typecheck step before twenty-ui build to check stories TS errors
- Added a tsconfig.dev.json to add stories and tests to typecheking when
in dev mode
- Added tsconfig.dev.json to storybook dev command of twenty-ui to
typecheck stories while developing
- Fixed twenty-ui stories that were broken
- Added a serve command to serve front build
- Fixed unit test from another PR
---------
Co-authored-by: Félix Malfait <felix.malfait@gmail.com>
Co-authored-by: Lucas Bordeau <bordeau.lucas@gmail.com>
Implement soft delete on standards and custom objects.
This is a temporary solution, when we drop `pg_graphql` we should rely
on the `softDelete` functions of TypeORM.
---------
Co-authored-by: Félix Malfait <felix.malfait@gmail.com>
Co-authored-by: Lucas Bordeau <bordeau.lucas@gmail.com>
## ISSUE (Warning)
- Fixes#6437
### Things needed to investigate :
- [x] Can we exclude @tabler-icons from Babel / wyw-in-js / the vite
config of Storybook ?
> Yes we can
- [ ] Is there a reproducible difference between non-Linaria components
loading time and Linaria components loading times ?
> Couldn't find anything related to this, yet. but changes fix the
problem.
- Remove raw queries and replace them by using `twentyORM`
- Refactor into services and utils
---------
Co-authored-by: Charles Bochet <charles@twenty.com>
Activity creation was duplicating the new activity on already visited
records of the same type.
The fix was to activate `shouldMatchRootQueryFilter` on createOne for
activity.
- create a workflow run every time a workflow is triggered in
not_started status. This status will be helpful later for once workflows
will be scheduled
- update run status once workflow starts running
- complete status once the workflow finished running
- add a failed status if an error occurs
Modal clickoutside was modified recently.
It was considering portaled component like dropdown outside, so
comparePixel mode for click outside is needed here.
Calling `getObjectMetadata` from `WorkspaceCacheStorageService` in every
query was causing big performance issues. The `objectMetadataCollection`
is now part of the `WorkspaceInternalContext` so we only instance it
once in the `WorkspaceDatasourceFactory`.
Queries are now much faster, for instance for TimelineCalendar, it went
from ~450ms to 80ms.
