Identifiers format normalize
Exactidentifiers_format_normalizeDeterministic and authoritative — the canonical answer for this input.
Normalize an ID / license number string. Unicode NFKC normalize, replace fancy dashes with ASCII '-', keep only A-Z/0-9/-/./ characters, collapse repeated separators, trim ends, uppercase. Returns None in `result.normalized` for null markers. Use this before storing or comparing license numbers across systems. Processing multiple inputs? Use `identifiers_format_normalize_batch` to do up to 1000 in one call.
Parameters
valuestringrequiredID string to normalize (license number, etc.).
Call it
curl -X POST https://api.snipget.ai/v1/identifiers/format/normalize \
-H "Authorization: Bearer sk_live_..." \
-H "Content-Type: application/json" \
-d '{"value":"rn 1234567"}'The Batch tab sends up to 1000 inputs in one call via /v1/identifiers/format/normalize/batch.
Example response
{
"input": "rn 1234567",
"normalized": "RN1234567",
"changed": true
}FAQ
What does it clean up?
Messy license and ID numbers. It removes stray spaces and odd characters, swaps fancy dashes for plain hyphens, collapses repeated separators, trims the ends, and uppercases the result so the same ID always looks the same.
Why use it instead of doing it by hand?
Spreadsheets introduce subtle junk: smart dashes, trailing spaces, a license stored as 1234.0. This applies the same rules every time, so your IDs match cleanly when you compare or de-duplicate them.
Does it check that the ID is valid?
No. It only tidies the formatting. It does not confirm the number is a real or correctly-structured identifier.
What happens with a blank or placeholder value?
Markers like NA, N/A, UNKNOWN, NULL, or an empty string come back as normalized null with confidence 0, so you can spot the gaps.