Skip to main content

Record Types

Define structured data types and access their fields with compile-time validation.

Use Case​

You receive structured user data and need to extract and process individual fields.

The Pipeline​

# simple-pipeline.cst - Working with record types

# Define a record type for user data
type User = {
  id: Int,
  name: String,
  email: String,
  score: Int
}

# Input: a single user record
in user: User

# Extract and process fields
user_name = trim(user.name)
is_high_score = gt(user.score, 100)

# Output the results
out user_name
out is_high_score

Explanation​

StepExpressionPurpose
1type User = { ... }Defines a record type with four typed fields
2in user: UserDeclares input using the custom type
3user.nameAccesses the name field — compiler verifies it exists
4gt(user.score, 100)Compares score to a threshold

Field access is validated at compile time. Writing user.nme would produce a compile error: field 'nme' not found. Did you mean 'name'?

tip

The compiler catches field access typos at compile time, not runtime. This eliminates entire classes of bugs before your pipeline runs.

Running the Example​

Input​

{
  "user": {
    "id": 1,
    "name": "  Alice  ",
    "email": "alice@example.com",
    "score": 150
  }
}

Expected Output​

{
  "user_name": "Alice",
  "is_high_score": true
}

Variations​

Nested processing​

type User = { name: String, email: String, score: Int }

in user: User

upper_name = Uppercase(user.name)
name_length = TextLength(user.name)

out upper_name
out name_length

Multiple record types​

type User = { name: String, role: String }
type Config = { maxRetries: Int, timeout: Int }

in user: User
in config: Config

greeting = concat("Welcome, ", user.name)
totalWait = config.maxRetries * config.timeout

out greeting
out totalWait
warning

Field names are case-sensitive. user.Name and user.name are different fields. The compiler will report an error if the field doesn't exist.

Best Practices​

  1. Define types at the top — type definitions should appear before they are used
  2. Use descriptive field names — field names become part of the compile-time contract
  3. Prefer record types over loose inputs — grouping related fields into a record makes pipelines more maintainable