Delete from Database · Watchflows Docs

At a glance

Takes: Any payload — its variables feed the conditions
Emits: rowsAffected
Good for: Retention rules and cleanup jobs

Set it up

Say which rows with the where-builder, and they go. This is the node behind every retention rule: a nightly Schedule into a Delete whose only condition is created_at older than (days) 30 keeps a table at a month, forever.

The inspector for Delete from Database: a table and the conditions that pick the rows.

Configure

Field	Default	What it does
Database	—	Which database to write to. The picker fills in the id.
Table	—	The table to delete from. It must already exist.
Where	—	Conditions, each a column, an operator, and usually a value.
Match	All of the conditions	All (AND) or Any (OR).
No conditions — reach EVERY row	Off	The deliberate opt-in for emptying the table.
Match capitalisation exactly	Off	Off means `equals`, `contains` and friends ignore case.

The where-builder offers the same eighteen operators as Find in Database. Two of them read a date column against a window of time — older than (days) and within last (days) — and their value is a count of days, never a date. created_at works out of the box; a column of your own works too as long as it holds an ISO-8601 stamp, because both sides go through SQLite's date parser rather than being compared as text.

Output

Every key is available downstream as {{key}}.

Key	Type	Value
`rowsAffected`	Number	How many rows were removed. `0` is a normal answer — it means the conditions matched nothing

Example: prune a clipping by id

Input supplies the row id. Delete from Database removes the row where id equals {{text}}. Log reports the count.

Three nodes: name a row, remove it, see the count.

Paste this into the Flow Builder to get the same flow:

Take a row id from an Input trigger, delete that row from the clips table in Clipboard History, and log how many rows were removed.

Good to know

A database has no Trash. Deleted rows are gone. This node is always badged destructive.

No conditions means the node refuses to run. A blank condition list would empty the table, so it stops rather than guessing. Turn on No conditions — reach EVERY row if that really is what you want.

A window that isn't a positive number of days stops the run. last week, 30 days, 0 and a missing variable are all refused rather than turned into some other window — the same rule every condition follows, for the same reason.

A value that isn't a date simply never matches. Point older than (days) at a column of names and it selects no rows, in either direction. Anything more particular than "N days" belongs in Run SQL.