Conditions

---

Purpose

The conditions section allows you to filter source data before any transformation is applied.
It is equivalent to a SQL WHERE clause: only rows matching the condition are processed by the stream.

Conditions are defined at the root level of a .stream.json file.

---

Basic Structure

Simple Condition

{
  "condition": {
    "left": "a.storeId",
    "op": "==",
    "right": "$storeId$"
  }
}

• left: Column reference (source or lookup)
• op: Comparison operator
• right: Comparison value (literal, variable, list, or null)

---

Condition Types

1. Simple Condition

Used when a single comparison is sufficient.

{
  "condition": {
    "left": "a.status",
    "op": "==",
    "right": "'ACTIVE'"
  }
}

2. Indexed Condition

Indexed conditions are required when building complex logical expressions.

{
  "left": "a.status",
  "op": "==",
  "right": "'ACTIVE'",
  "index": 1
}

• index uniquely identifies the condition
• Indexes are referenced later in the logic tree

3. Complex Condition (Logical Combination)

Multiple conditions can be combined using logical operators.

{
  "condition": {
    "conditions": [
      { "left": "a.country", "op": "==", "right": "'FR'", "index": 1 },
      { "left": "a.type", "op": "IN", "right": ["STORE", "WAREHOUSE"], "index": 2 }
    ],
    "logic": {
      "left": 1,
      "op": "&&",
      "right": 2
    }
  }
}

Resulting logic:

(country == 'FR') AND (type IN ['STORE', 'WAREHOUSE'])

---

Join Conditions with Lookups

Conditions can reference lookup columns using their alias.

{
  "lookups": [
    { "table": "stores", "alias": "b", "columns": [...] }
  ],
  "condition": {
    "left": "b.storeType",
    "op": "==",
    "right": "'RETAIL'"
  }
}

---

Supported Operators

Comparison Operators

Operator	Description	Example
==	Equality comparison	"== 'ACTIVE'"
IN	Value membership in a list	IN ["A", "B", "C"]
is not	Not null / existence check	is not null

---

Logical Operators

Operator	Description
&&	Logical AND
`

---

Value Syntax

SQL Databases (PostgreSQL, MySQL, SQL Server)

Syntax	Usage	Example
"a.column"	Source column reference	"a.storeId"
"b.column"	Lookup column reference	"b.storeType"
"$var$"	Runtime variable	"$storeId$"
"'VALUE'"	String literal	"'ACTIVE'"
100	Numeric literal	100
"_raw(...)"	Raw SQL expression	"_raw(DATE('$startDate$'))"

Example:

{
  "right": "_raw(DATE('$startDate$'))"
}

---

Couchbase

Syntax	Usage	Example
#column.X#	Column reference by index	#column.2#
alias(a).#column.X#	Column reference with explicit alias	alias(a).#column.3#
fieldName	Field existence check	recipe
$var$	Runtime variable (no quotes)	$travelId$

Example:

{
  "condition": {
    "left": "#column.2#",
    "op": "==",
    "right": "$travelId$"
  }
}

---

Nested Logical Trees

Complex logical trees can be nested arbitrarily.

{
  "conditions": [
    { "left": "a.country", "op": "==", "right": "'FR'", "index": 1 },
    { "left": "a.city", "op": "==", "right": "'Paris'", "index": 2 },
    { "left": "a.status", "op": "==", "right": "'ACTIVE'", "index": 3 }
  ],
  "logic": {
    "left": {
      "left": 1,
      "op": "&&",
      "right": 2
    },
    "op": "&&",
    "right": 3
  }
}

Logical result:

(country == 'FR' AND city == 'Paris') AND status == 'ACTIVE'

---

Checklist

Simple Condition

• Define left
• Define op
• Define right

Complex Condition

• Define conditions[]
• Assign unique index values
• Build the logic tree
• Validate final logical expression

---