Jurisdictions state normalize
Heuristicjurisdictions_state_normalizeBest-effort — fuzzy, approximate, or inferred; check the confidence score.
Normalize a US state/territory/DC/AFPO reference. Accepts 2-letter USPS codes (case-insensitive) or full names. Returns the canonical record with `code`, `name`, `iso_code` (the ISO 3166-2 code), and `type` (state / district / territory / armed_forces_postal). Confidence 1.0 on exact match, lower on fuzzy. Processing multiple inputs? Use `jurisdictions_state_normalize_batch` to do up to 1000 in one call.
Parameters
valuestringrequiredUS state code or full name.
Call it
curl -X POST https://api.snipget.ai/v1/jurisdictions/state/normalize \
-H "Authorization: Bearer sk_live_..." \
-H "Content-Type: application/json" \
-d '{"value":"ny"}'The Batch tab sends up to 1000 inputs in one call via /v1/jurisdictions/state/normalize/batch.
Example response
{
"code": "NY",
"name": "New York",
"iso_code": "US-NY",
"type": "state",
"parent_code": "US",
"matched_by": "code"
}FAQ
What does this do?
It cleans a single US state value into its official record. Type ny, NY, or New York and you get back the USPS code NY, the full name, the ISO 3166-2 code (US-NY), and the type. Great for tidying a State column.
How is this different from the general jurisdiction tools?
This one is US-only and takes no filters, so you never have to worry about CA meaning Canada. It is a focused shortcut for the common case; the Normalize a jurisdiction tool is the choice when you also need countries.
Does it cover more than the 50 states?
Yes. It also recognizes DC, the US territories, and the armed-forces postal codes (AA, AE, AP), and the type field tells you which one you got.
Is the match exact or fuzzy?
Exact first: a 2-letter code or full name (case-insensitive) matches at confidence 1.0. When that misses, a conservative fuzzy fallback catches a small typo ('Californa', confidence 0.7) or a unique prefix ('calif', confidence 0.8) and sets matched_by to 'fuzzy'. Heavier mangling like Caleeforniya still returns no match rather than a guess.
What if the value is not a US state?
You get confidence 0 with all fields null and a reason, not an error. A foreign country name simply returns no match, since this tool is scoped to the US.