- new status `ENQUEUED` added. With a command to backfill
- counter in cache per workspace, managed by a new service
[workflow-run-queue.workspace-service.ts](https://github.com/twentyhq/twenty/compare/tt-improve-workflow-run-queueing?expand=1#diff-1e2de2a48cd482a3bd7e8dedf1150a19d0b200afbd9282181a24ecddddb56927)
- cron added that will run every minute to look for not started
workflows
Here is the new flow:
- When executing a workflow, we check if the queue is not full. If not,
run is created as `ENQUEUED` and the run workflow job is triggered as
usual. If full, create the run as NOT_STARTED and do not trigger the job
- Cron will look for NOT_STARTED workflows and queue some if there is
some place again
- Only MANUAL and Form submit skip the queue limit
- id field should only be available for search records action
- create record action does not work for relations. Requires to send
`accountOwner: { id: string }` instead of `accountOwner: string`
- hidding `runs` for version views as we did for workflows
Workflow views and versions are seed being opened by default into record
page. Issue is that:
- new views are set by default to side panel. Updated by copying the
current opensIn value to the new view
- users can still select side panel into their options. Disabling the
button.
<img width="650" alt="Capture d’écran 2025-06-19 à 16 15 34"
src="https://github.com/user-attachments/assets/0ddc3284-0fed-404f-9c1d-225c65549fd1"
/>
- Variables can now be handled for select/multiselect/relations
- Hide field not supported in forms (source, rating)
- Add tests for schemas
Remaning issues:
- country/currency pickers not working
- stories for components
- variable picker hidden for dates
First PR to add filters to send records. Lot of work left, but I want to
split. I mainly want to validate the architecture there.
https://github.com/user-attachments/assets/63375a75-ba88-49df-8c12-5e3e58de5342
TODO in next PRs:
- fix design
- make filters reliable. Some composite fields are not implemented and
some fields like datetime do not work well
- improve typing
Fixes https://github.com/twentyhq/twenty/issues/12337
When importing emails, matched companies are added, but no event is
triggered. Which means that workflows are not triggered. Adding the
event.
To test:
- create a workflow that listens to company creation
- import emails
- make sure workflow has been triggered
Backend part of https://github.com/twentyhq/core-team-issues/issues/928
- Add fields to database event settings
- If not set, match all automated triggers with the right event name
- If set, event needs at least one updated field listened to be treated
For database event triggers, we remove the before / after logic. We go
directly with the properties
<img width="211" alt="Capture d’écran 2025-05-27 à 11 40 36"
src="https://github.com/user-attachments/assets/a05bd3c1-104b-477b-be52-d56846ce7e63"
/>
To achieve this without changing the shape of events, we need to handle
keys using dots, such:
```
'properties.after.name': {
icon: 'IconBuildingSkyscraper',
type: FieldMetadataType.TEXT,
label: 'Name',
value: 'My text',
isLeaf: true,
},
```
This PR:
- adds logic to handle the case where the key has dot included
- adds tests
Workflow statuses are often broken. I did not figured out why yet. But I
see two causes that can be fixed:
- statuses calculation are really complicated today, just to spare a
call to the database
- job is not indempotent, it is using the combination of the previous
statuses + the update to calculate the new statuses. Which means that
once broken, next updates will be broken as well
Instead, we now:
- fetch workflow versions
- get the statuses from these.
It simplifies the code and make the job indempotent.
Fixes https://github.com/twentyhq/twenty/issues/12125
The root cause of the infinite loop was the calendar cursor. In some
cases, it was not properly displayed and was causing the loop because of
its animation that was always restarting.
We agreed with @FelixMalfait and @Bonapara that given the current
importance of the feature and the amount of issues associated, we remove
the cursor for now.
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.
- enrich response so the record is available in the step output. Today
this is available in the schema but only the id is set
- make the full record picker clickable instead of the arrow only
<img width="467" alt="Capture d’écran 2025-04-30 à 16 08 04"
src="https://github.com/user-attachments/assets/db74b9a6-7f1d-4e54-bf06-9be3d67ee398"
/>
Twenty prod DB has been exported for testing.
Main updates:
- do not process workflow runs with less than 2 steps. Nothing to do.
- update runs by batches of 500. Will avoid java heap space issue
- add more logs
When inserting a new step between step 1 et step 2, then step 1 should
have the new step as next step id, add stop having step 2.
When deleting a step, we link the parent and next steps together. It may
change in the future
2025-04-16 15:30:05 +02:00
Thomas Trompetteandgreptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
- add next step id on step
- backfill next step id on step, except for the last one
- backfill flow for workflow run, when it exists
---------
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
We do not manage rich text properly in workflows. This is because API
has a layer called transformer service. Looks a bit as a duplicate of
format data, but this api layer was already there for position anyway.
Using it in workflow record actions.
I hope at some point we merged formatData util and transformer.
- requires a refacto so several fields can be updated at once
- updating object name on record picker will now update placeholder
- add a min-height to label so fields do not get moved when the label is
deleted
- Add position during workflow version / creation. It will allow to have
the versions and runs ordered
- Block the creation from generated api for versions. We use workflow
post hooks or create from draft
Currently, when filling the form, values are not saved in the action
settings. This is an issue because we do not see the response in the
node settings, only in the output of the step.
This PR:
- adds a new endpoint to update a step in the run flow output
- updates this flow when a step is updated
https://github.com/user-attachments/assets/2e74a010-a0d2-4b87-bd1f-1c91f7ca6b60
---------
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
- create a form filler component
- send the response on submit
- put back a field name. We need it for the step output
- validate a form is well set before activation
TODO:
- we need to refresh to see the form submitted. We need to discuss about
a strategy
- the response is not saved in the step settings. We need a new endpoint
to update workflow run step
https://github.com/user-attachments/assets/0f34a6cd-ed8c-4d9a-a1d4-51455cc83443
Workflows step details in workflows and versions should be different
from the node tab in run. For most cases, it was using the same
component. But for forms, it will be a different one.
This PR:
- renames form action into formBuilder. formFiller is coming
- put code into a separated folder
- creates a new component for node details
When creating an object from kanban, we are using
`labelIdentifier.toLowerCase()` for the labelIdentifier field.
Issue is that labelIdentifier is translated.
Using `labelIdentifierField.name` instead.
When a step is deleted in a draft version, its variable are still
available in the following steps. This is because step output schema was
not reset. We needed either to refresh or to change version so output
schema gets updated.
This PR:
- migrates to a family state global + context not linked to a component
- add a reset step output schema function
- reset when a step is removed
Steps are broken when a variable is set.
This is because component instance is not set for version and run
visualizers.
Each step viewer should be wrapped by the instance context.
Each diagram visualizer should be responsible for populating the output
schema.
Also fixing a billing error when running workflow.
- Create a workflow version component family state for each workflow
version : `stepId` => `StepOutputSchema`
- Populate this state when reaching the workflow visualizer of the
workflow version
- Wrap the right drawer when in edit mode with the context. It is the
only one who needs this schema
Next step:
- read this state from the variables
- add name to form field metadata
- extract field generation from object record schema
- use field generation to generate field from metadata
- add tests
We are starting to put too many services in common folder. Version and
step building should be separated into different services and go to the
builder folder. Today builder folder only manage schema.
We should:
- keep services responsible for only one action
- keep modules based on the actual action these provide rather than
having common module
This PR:
- creates a service for workflow version builder
- moves version and step builders to workflow builder folder rather than
commun
- creates separated folders for schema, version and steps
No logic has been added. Only modules created and functions moved.
- create form action
- add `pendingEvent` to executor output
- when receiving pendingEvent, exit workflow execution
- let the workflow in running status for now before taking a decision
Actions will now:
- receive the complete input
- get the step they want to execute by themself
- check that the type is the right one
- resolve variables
These all share a common executor interface.
It will allow for actions with a special execution process (forms, loop,
router) to have all required informations.
Main workflow executor should:
- find the right executor to call for current step
- store the output and context from step execution
- call next step index
Should be possible to end workflow, not matter what the current status
is.
On failure before the workflow was started, this error prevent the
workflow to be marked as failed with the right error message.
We need the version trigger and steps to be stored in the output. We
should not rely on the version itself because some run are made on draft
versions. Which means versions could be edited afterwards.
Close
https://discord.com/channels/1130383047699738754/1334441759484149793
Using refetch queries was not working for certain use cases.
To find manual active workflows in cmd+k, we use a query with specific
filters that was complicated to refetch.
Finally I will update the cache manually. It was not properly updated
before because the json value of the version trigger was stringified
without spaces. So the entity was not found in apollo cache.
Thomas Trompette