Read HL7 delimiters and field positions

Start with the header

Open the HL7 Message Analyzer and paste the synthetic message below. It is an inspection fixture, not a complete clinical message or proof of conformance.

MSH-1 is the field separator itself. MSH-2 contains the encoding characters; MSH-3 is the sending application. Counting every pipe-separated item as an ordinary field shifts the header. The HL7 MSH reference identifies those positions.

MSH|^~\&|SYNTHETIC|LOCAL|RECEIVER|LOCAL|202609140900||ADT^A01|GUIDE-001|T|2.5.1
PID|1||SYN-001^^^LOCAL^MR~SYN-002^^^OTHER^MR||EXAMPLE^SYNTHETIC
NTE|1||Encoded separator: \F\

Inspect a repeated field

Choose PID, then PID-3. With these delimiters, ~ separates repetitions, ^ separates components and & separates subcomponents. Empty components still occupy positions: the second identifier has OTHER in component 4. Expected observations:

MSH-1: |
MSH-2: ^~\&
MSH-3: SYNTHETIC
PID-3 repetitions: 2
PID-3[2].4: OTHER
NTE-3: Encoded separator: \F\

In module code, call inspectField(segment, 3) with the original segment returned by analyzeHL7. Its one-based paths describe repetitions, components and subcomponents. Re-analyze a serialized or cloned result before inspecting it.

Read separators from the message

The analyzer accepts distinct visible ASCII punctuation separators. This equivalent inspection fixture declares a field separator of *; MSH-2 declares component, repetition, escape and subcomponent separators in that order.

MSH*$%!@*SYNTHETIC*LOCAL*RECEIVER*LOCAL*202609140900**ADT$A01*GUIDE-002*T*2.5.1
PID*1**SYN-001$$$LOCAL$MR%SYN-002$$$OTHER$MR**EXAMPLE$SYNTHETIC

Delimiters: field=* component=$ repetition=% escape=! subcomponent=@
PID-3 repetitions: 2
PID-5[1].1: EXAMPLE

Keep encoding and transport limits in view

• Escape text remains encoded: \F\ stays \F\. Paired escapes protect embedded delimiters from splitting; the analyzer does not decode escapes or check their full grammar. Unmatched escapes produce a warning.
• A fifth MSH-2 truncation character is retained with a warning; truncation semantics are not interpreted.
• One message with CR, LF or CRLF breaks is accepted, including a single MLLP frame. Multiple messages, batch wrappers and prefixed log text are rejected.
• Copy and Download use normalized message text: supported framing and blank lines are removed, segments use CR, and one final CR is added. This does not preserve original transport bytes.
• Limits are 1 MiB of UTF-8 input, 2,000 segments and 50,000 structural parts. Header checks and common field labels do not establish message structure, datatype, terminology or profile conformance.