In this PR, we are refactoring two things:
- leverage field.defaultValue for Select and MultiSelect settings form
(instead of option.isDefault)
- use quoted string (ex: "'USD'") for string default values to embrace
backend format
---------
Co-authored-by: Thaïs Guigon <guigon.thais@gmail.com>
A user has reported an issue with REST API.
We have recently migrated the graphql IDs from UUID to ID type. As Rest
API is leveraging the graphql API under the hood, the Rest API query
builder should be updated accordingly
In this PR, I'm refactoring the way we associate messageParticipant post
person/company creation. Instead of looking a all person without
participant, we are passing the one that were just created.
Also, I'm making sure the message and messageParticipant creation
transaction is commited before creating person/company creation (and
then messageParticipant association)
This PR fixes several issues:
- enum naming should be: {tableName}_{fieldName}_enum and respecting the
case
- defaultValue format handled in the FE should respect the one in the BE
In my opinion we should refactor the defaultValue:
- we should respect backend format: "'myDefault'" for constant default
and "0" for float, "now" for expressions, "true" for booleans. we can
rename it to defaultValueExpression if it is more clear but we should
not maintain a parallel system
- we should deprecate option: isDefaultValue which is confusing
- we should re-work backend to have a more unified approach between
fields and avoid having if everywhere about select, multiselect, and
currency cases. one unified "computeDefaultValue" function should do the
job
What is still broken:
- currency default Value on creation. I think we should do the refactor
first
- select default value edition.
These cases do not break the schema but are ignored currently
The whole viewBar component was re-rendered on view changes which was
introducing performance issue. The need was to compute page title, this
should be done in a lower level component
While using middleware (executed pre-graphql) for graphql endpoint, we
need to swallow exception and return errors with a 200. Otherwise it's
not a valid graphql response
## Query depth deprecation
I'm deprecating depth parameter in our graphql query / cache tooling.
They were obsolete since we introduce the possibility to provide
RecordGqlFields
## Refactor combinedFindManyRecordHook
The hook can now take an array of operationSignatures
## Fix tasks issues
Fix optimistic rendering issue. Note that we still haven't handle
optimisticEffect on creation properly
## Context
We recently enabled the option to bypass SSL certificate authority
validation when establishing a connection to PostgreSQL. Previously, if
this validation failed, the server would revert to unencrypted traffic.
Now, it maintains encryption even if the SSL certificate check fails. In
the process, we overlooked a few DataSource setups, prompting a review
of DataSource creation within our code.
## Current State
Our DataSource initialization is distributed as follows:
- **Database folder**: Contains 'core', 'metadata', and 'raw'
DataSources. The 'core' and 'metadata' DataSources manage migrations and
static resolver calls to the database. The 'raw' DataSource is utilized
in scripts and commands that require handling both aspects.
- **typeorm.service.ts script**: These DataSources facilitate
multi-schema connections.
## Vision for Discussion
- **SystemSchema (formerly core) DataSource**: Manages system schema
migrations and system resolvers/repos. The 'core' schema will be renamed
to 'system' as the Core API will include parts of the system and
workspace schemas.
- **MetadataSchema DataSource**: Handles metadata schema migrations and
metadata API resolvers/repos.
- **(Dynamic) WorkspaceSchema DataSource**: Will be used in the Twenty
ORM to access a specific workspace schema.
We currently do not support cross-schema joins, so maintaining these
DataSources separately should be feasible. Core API resolvers will
select the appropriate DataSource based on the field context.
- **To be discussed**: The potential need for an AdminDataSource (akin
to 'Raw'), which would be used in commands, setup scripts, and the admin
panel to connect to any database schema without loading any model. This
DataSource should be reserved for cases where utilizing metadata,
system, or workspace entities is impractical.
## In This PR
- Ensuring all existing DataSources are compliant with the SSL update.
- Introducing RawDataSource to eliminate the need for declaring new
DataSource() instances in commands.
In this PR:
- Follow up on #5170 as we did not take into account not logged in users
- only apply throttler on root fields to avoid performance overhead
An error has been recently introduced in the sync of fieldMetadata. This
PR fixes it
Additionnally, we are enabling email for trialing and past_due
workspaces. There is an ongoing work to introduce a more robust
activationStatus on workspace.
While trying to migrate a workspace from 0.3.3 to 0.10.0, we've faced an
issue with the script to migrate default-values format.
This PR fixes it.
We really need to add tests on this part ;)
## Context
For users with many records, only the first n*60 records were loaded on
board views (n being the number of visible columns). This was because of
the following behavior:
- watch for end of column visibility changes. If an end of column is
visible, try to fetch more. However, watching for visbility changes is
not reliable enough.
## What we want
If an end of column is visible, try to fetch more. If no more records is
availble in pagination, do not fetch more
## Context
We have recently added an event listener to create audit logs on objects
update. However, we have only created the structure (relations on event
standard objects) for Company, Person, Opportunity and custom objects.
There is a larger effort in #4936 to refactor this.
For now, we are disabling log auditing on all other objects
## How
Add @IsNotAuditLogged() annotation on all standard objects except
Company, Person, Opportunity
Unfortunately, it is not possible in CSS to have an overflow:visible
over x-axis while having an overflow:hidden over y-axis, leading to the
following issue:
<img width="1512" alt="image"
src="https://github.com/twentyhq/twenty/assets/12035771/9b84cbbb-c6c4-4fd6-a630-a24f01eccf73">
I'm refactoring the RecordInlineCell and RecordTableCell to use
useFloating + createPortal to open the cell.
We have recently discovered that we were using ID type in place of UUID
type in many place in the code.
We have merged #4895 but this introduced bugs as we forgot to replace it
everywhere
We have discovered that GraphQL inputs for fields of type ids in create
/ update input where using a more permissive ID type than the type used
in FilterInput in queries.
This PRs fixes it and make sure that all Input are using UUID graphql
scalar types
## Removing repetitive queries (impacting performance on page load)
We have recently introduced the capability to detect schema version
mismatch. To do that, we add a new header in all our queries. On page
load, this header is added once we know the currentUser and especially
its currentWorkspace and related schema version.
However, applying this header to apollo client will re-trigger all
queries that have been already performed (GetClientConfig and
GetCurrentUser). To avoid re-triggering them, I'm introducing two new
"isLoaded" states and skip the query if the query has already been
performed
## Fixing Relation Detail not displaying data on show page
Small bug introduced in a previous PR
In this PR:
- fix empty list placeholder positionning
- prevent user from erasing custom address field as composite types
removal is not supported yet @ijreilly FYI
- fix show page relation error
- Implement address filter
In this PR, I'm fixing two things on the ViewPicker in Create mode:
- if the Dropdown has no max height, it should not be scrollable (which
is causing issue with inner dropdowns being cut by overflow: hidden
- if the user has changed the icon, the type or the name of the view,
consider the create form as isDirty and prevent its value to be
overriden by re-renders (cache updates for example)
We used to not type properly the return of getRecordFromCache before the
work from last week.
The getViewFromCache function was patching this temporarily in a "dirty"
way.
Now that this is cleaner, I'm removing the typescript patch. This fixing
several behaviors on viewBar
We were missing `JsDom` dependencies in the package.json generated by nx
while running `twenty-server`: `yarn nx build:packageJson`
Detailed explanation:
- we are currently using nx paradigm which is to put dependencies of all
projets at root, which enables global package migrations for the whole
monorepo
- for production containers, we only want specific project dependency to
be added. This is done by running `yarn nx build:packageJson` on
`twenty-server`. Nx is statically analyzing twenty-server dependencies
and generating a tailored package.json that production containers can
later use.
- However, `nx` static analysis is not flawless and is missing some
packages. We are going to stop using it as the value is not there yet
but the burden for developers is high. The guideline is to put back
project dependencies into specific package `package.json`
- Therefore, I'm adding `jsdom` to twenty-server `package.json`
The tests were broken. It turns out that it is not easy to mock two
apolloClients as apollo only provides "MockedProvider" component to mock
apollo in tests.
MockedProvider Api does not allow us to mock on client level.
For now, I'm defaulting to the base ApolloClient in test mode to avoid
the issue
When writing to the normalized cache (record), it's crucial to use _refs
for relationships to avoid many problems. Essentially, we only deal with
level 0 and generate all fields to be comfortable with their defaults.
When writing in queries (which should be very rare, the only cases are
prefetch and the case of activities due to the nested query; I've
reduced this to a single file for activities
usePrepareFindManyActivitiesQuery 🙂), it's important to use queryFields
to avoid bugs. I've implemented them on the side of query generation and
record generation.
When doing an updateOne / createOne, etc., it's necessary to distinguish
between optimistic writing (which we actually want to do with _refs) and
the server response without refs. This allows for a clean write in the
optimistic cache without worrying about nesting (as the first point).
To simplify the whole activities part, write to the normalized cache
first. Then, base queries on it in an idempotent manner. This way,
there's no need to worry about the current page or action. The
normalized cache is up-to-date, so I update the queries. Same idea as
for optimisticEffects, actually.
Finally, I've triggered optimisticEffects rather than the manual update
of many queries.
---------
Co-authored-by: Lucas Bordeau <bordeau.lucas@gmail.com>
