Integration Soup Paths (Concise)

Paths are how you locate values inside messages so you can use them in variables, transformers, filters, bindings, database parameters, HTTP arguments, and other workflow settings.

Each MessageType has its own path syntax. The most important rule is: always be explicit about where the data comes from.

---

HL7 Paths (HL7 v2.x)

Syntax

Format: Segment[index]-FieldNumber[repetition].ComponentNumber.SubComponentNumber

• Segment: PID, OBX, PV1, etc.
• index is 1-based and optional (first occurrence assumed if omitted)
• Field repetition [repetition] is optional (first repetition assumed if omitted)
• Component/subcomponent are strongly preferred

Examples

• PID-5.1
  First component of field 5 in PID

• OBX[2]-5.2
  Second OBX segment, field 5, component 2

• Prefer: PID-8.1 over PID-8
  Even if there’s only one component, including it is preferred.

Receiver vs Sender access

• Receiver value:
  PID-3.1
• Sender value:
  'Sender Name' PID-3.1
'Sender ID' PID-3.1

---

XML Paths

Syntax

• Nodes separated by /
• Attributes use @
• Always include the full path from the root

Examples

• Element path:
  Patient/Demographics/Name/First

• Attribute path:
  Patient/Demographics/Name/@type

Receiver vs Sender access

• Receiver value:
  Patient/ID
• Sender value:
  'Sender Name' Patient/ID
'Sender ID' Patient/ID

---

JSON Paths

Syntax

Integration Soup uses the same style as XML here: - Nodes separated by / - Attributes use @ - Always include the full path from the root

Examples

• Value path:
  Patient/Demographics/Name/First

• Attribute-style path:
  Patient/Demographics/Name/@type

Receiver vs Sender access

• Receiver value:
  Patient/ID
• Sender value:
  'Sender Name' Patient/ID
'Sender ID' Patient/ID

---

CSV Paths

Syntax

• Use column ordinals (0-based) in brackets
• Row selection is not supported
• If the column order isn’t obvious, instructions must list the presumed column order

Examples

• Valid:
  [5]
• Invalid:
  [Name]

Receiver vs Sender access

• Receiver value:
  [2]
• Sender value:
  'Sender Name' [1]
'Sender ID' [1]

---

SQL Paths (Query Results)

Syntax

For query results, paths behave like CSV, but: - Use column positions (1-based) in brackets - Multiple rows / multiple result sets are not supported - If needed, the instructions should explicitly say the action must be coded.

Examples

• Valid:
  [5]
• Invalid:
  [Name]

Receiver vs Sender access

• Receiver value:
  [2]
• Sender value:
  'Sender Name' [1]
'Sender ID' [1]

---

Practical Patterns You’ll Use Often

Pattern A — Receiver → Variable → Message

1) Set variable from receiver path
Set ${PatientID} from PID-3.1 2) Insert ${PatientID} into the current activity’s message via mapping transformer

Pattern B — DB Sender result → TCP Sender message

1) DB sender returns columns (SQL paths like [1], [2], …)
2) TCP sender transformer references DB sender explicitly:
Map the value from 'DB Insert Patient Encounter' [1] to replace PID-3.1

---