Skip to content

Value classifications

Value classifications check the value of one column in a row and decide whether a condition is true or false.

They are used to describe simple conditions such as:

  • Is the email field empty?
  • Is the price greater than 0?
  • Is the booking date in the future?
  • Is the status confirmed?

The result is always true or false.

Other parts of the system can then use that result to control behavior, for example:

  • hiding rows from users
  • blocking invalid edits
  • filtering lists
  • triggering notifications

What a value classification is

A value classification is a named condition attached to a table and a column.

For each row in the table, the system checks the value in the selected column and evaluates a condition.

Example:

Table: Person
Column: email

Classification name: Has email address

Condition:

email IS NOT EMPTY

Result:

Person Email Has email address
Anna anna@example.com true
Erik (empty) false

The classification does not modify the data.
It simply describes something about the row.

How value classifications work

A value classification is defined using four main parts.

Table
The table where the classification applies.

Column
The column whose value will be checked.

Predicate
The condition applied to the value.

Parameters (optional)
Some predicates require additional values such as numbers or dates.

Example:

Classification: High price

price > 1000

Result:

Product Price High price
Chair 200 false
Sofa 1500 true

Predicates

Predicates define how the value is evaluated.

Different column types support different predicates.

Examples:

Text columns

  • is empty
  • is not empty
  • equals

Numeric columns

  • greater than
  • less than
  • equals

Date columns

  • before
  • after
  • within range

The available predicates depend on the type of column.

When to use value classifications

Value classifications are best used for simple checks on a single column.

Examples:

  • checking whether a required value exists
  • validating numeric ranges
  • identifying rows with specific status values
  • flagging rows that need attention

If the condition depends on related rows, a relational classification should be used instead.

A simple way to think about value classifications

A value classification answers a simple question about a row:

Does this column satisfy a specific condition?

The system evaluates that condition automatically and other parts of the system can reuse the result.

Reference

How-to

Related concepts