## Summary
Postgres columns can be
- [generated as
identity](https://www.postgresqltutorial.com/postgresql-tutorial/postgresql-identity-column/)
- [generated by a custom
expression](https://www.postgresql.org/docs/current/ddl-generated-columns.html)
In these 2 cases, the column is not required when inserting a new row.
This PR makes sure these types of column are not marked required in n8n.
### How to test
1. Create a Postgres table with all types of generated columns:
for version >= 10
```sql
CREATE TABLE "public"."test_table" (
"id" int8 NOT NULL DEFAULT nextval('test_table_id_seq'::regclass),
"identity_id" bigint GENERATED ALWAYS AS IDENTITY,
"id_plus" numeric GENERATED ALWAYS AS (id + 5) STORED,
"title" varchar NOT NULL,
"created_at" timestamp DEFAULT now(),
PRIMARY KEY ("id")
)
```
Before 10 you have to use serial or bigserial types:
```sql
CREATE TABLE distributors (
did serial not null primary key,
name varchar(40) NOT NULL CHECK (name <> '')
);
```
2. Add a postgres node to canvas and try to insert data without the
generated columns
3. Should successfully insert
More info in Linear/Github issue ⬇️
## Related tickets and issues
- fixes#7084
-
https://linear.app/n8n/issue/NODE-816/rmc-not-all-id-fields-should-be-required
-
https://linear.app/n8n/issue/NODE-681/postgres-cant-map-automatically-if-database-requires-a-field
## Review / Merge checklist
- [ ] PR title and summary are descriptive. **Remember, the title
automatically goes into the changelog. Use `(no-changelog)` otherwise.**
([conventions](https://github.com/n8n-io/n8n/blob/master/.github/pull_request_title_conventions.md))
- [ ] [Docs updated](https://github.com/n8n-io/n8n-docs) or follow-up
ticket created.
- [ ] Tests included.
> A bug is not considered fixed, unless a test is added to prevent it
from happening again.
> A feature is not complete without tests.
---------
Co-authored-by: Michael Kret <michael.k@radency.com>
Github issue / Community forum post (link here to close automatically):
https://github.com/n8n-io/n8n/pull/6348
---------
Co-authored-by: Giulio Andreini <g.andreini@gmail.com>
Co-authored-by: Marcus <marcus@n8n.io>
* fix(Postgres Node): For select queries, empty result should be be replaced with `{"success":true}`
* ⚡ less checks
---------
Co-authored-by: Michael Kret <michael.k@radency.com>
* Draft setup
* ⚡ Implemented expression evaluation in Postgres node, minor SQL editor UI improvements, minor refacring
* ⚡ Added initial version of expression preview for SQL editor
* ⚡ Linking npm package for codemirror sql grammar instead of a local file
* ⚡ Moving expression editor wrapper elements to the component
* ⚡ Using expression preview in SQL editor
* Use SQL parser skipping whitespace
* ✨ Added support for custom skipped segments specification
* ✨ Fixing highlight problems with dots and expressions that resolve to zero
* 👕 Fixing linting error
* ✨ Added current item support
* ⚡ Added expression support to more nodes with sql editor
* ✨ Added expression support for other nodes
* ✨ Implemented different SQL dialect support
* 🐛 Fixing hard-coded parameter names for editors
* ✨ Fixing preview for nested queries, updating query when input data changes, adding keyboard shortcut to toggle comments
* ✨ Adding a custom automcomplete notice for different editors
* ⚡ Updating SQL autocomplete notice
* ✅ Added unit tests for SQL editor
* ⚡ Using latest grammar
* 🐛 Fixing code node editor rendering
* 💄 SQL preview dropdown matches editor width. Removing unnecessary css
* ⚡ Addressing PR review feedback
* 👌 Addressing PR review feedback pt2
* 👌 Added path alias for utils in nodes-base package
* 👌 Addressing more PR review feedback
* ✅ Adding tests for `getResolvables` utility function
* ⚡Fixing lodash imports
* 👌 Better focus handling, adding more plugins to the editor, other minor imrovements
* ⚡ Not showing SQL autocomplete suggestions inside expressions
* ⚡ Using npm package for sql grammar
* ⚡ Removing autocomplete notice, adding line highlight on syntax error
* 👌 Addressing code review feedback
---------
Co-authored-by: Milorad Filipovic <milorad@n8n.io>
* Boilerplate for PostgresTrigger
* Create trigger function as a query
* Add additional fields to customize trigger query
* Add customizable channel name && operation name
* Add concat () for function name
* Add hints and placeholders
* Add resource Locator to trigger postgres
* Add the ability for knowing trigger event
* Throw error for same function name
* Remove console.logs
* Remove schema from Chanel notifcation mode
* Add UUID and save trigger in workflow static data drop function
* Fix bug where wrongfully casted result in pgl
* Correctly drops the resources when manually executing the trigger
* Remove manual execution with special interaction
* Remove console.logs
* ♻️ Move related trigger functions to new file
* fix target using 'schema."tableName"' in quotes
To support targets with Uppercase table names
* Remove static Data and use node id for uuid
* Update deleting of the trigger and function
* Fix regex expression for channel name
* Change to drop cascade the trigger function
* Replace functions on restart if no name has been defined
* Parse payload result
* Improve handling with hyphens in names
* Remove duplicate code and clean up
* Add payload on delete
* Fix rlc
* fixing uppercase tableName
* fix multiple triggers/connections issues
* fixing rlc pgp.end() issues
* unify pgp init db method
* drop trigger only in createTrigger mode
---------
Co-authored-by: Marcus <marcus@n8n.io>
