Documentation

Address Tokenizer Documentation

Public documentation covers evaluation, country support, licensing, data-source transparency, and curated examples from the live showcase output.

Country and language examples for dedicated parsers, with Core and Pro outputs.

Selected real-world cases such as city typos, PO boxes, long street names, and multi-line inputs.

Send an anonymized address input, actual output if available, and the behavior you expected.

How best-effort parsing works for countries without a dedicated parser.

The public showcase data file used to generate supported-country, edge-case, and limitation examples.

Blank, punctuation-only, postcode-only, and ambiguous inputs handled without crashing.

Examples that are outside current dedicated parser coverage and should be treated as reviewable.

Clear expectations around geocoding, deliverability, unsupported scripts, and ambiguous country signals.

High-level notes about public data attribution and third-party notices.