🗾 Japanese Address Splitter
Paste one address per line and the tool automatically splits each into prefecture, city/ward, and remainder. Results shown as a table with CSV (tab / comma) download.
🔒 About Privacy
- ・All processing happens entirely in your browser
- ・Address data is never sent to any server
- ・No storage logs, no upload history, no database
📖 Where people get stuck
Split one address per line into prefecture, municipality and the remainder. Everything runs in the browser. Addresses are data where the assumption "this can be split correctly" simply does not hold: there is no consistent convention for variants, former municipality names or how a building is written, and some addresses have no postal code at all. Treat the result as preparation that a human still has to look over.
| Case | What happens | What to do |
|---|---|---|
| Some rows have no prefecture | When a form keeps the prefecture in a separate field it disappears from the exported text. It is also common for people to omit it when the city name makes it obvious. Split mechanically from the start of such a string and part of the city name gets taken for the prefecture. | Before splitting, pull out the rows that do not begin with a prefecture. The set of cities where people habitually drop the prefecture is small, so filling those in from a lookup first raises accuracy noticeably. If a postal code is present, deriving the prefecture from it is more reliable — a full seven-digit code resolves uniquely. |
| The boundary between the number and the building | The same place can be written many ways, with different separators and orderings — a hyphenated form, a spelled-out form, half-width or full-width hyphens, with or without spaces. No single regular expression can reduce all of them to one shape. | Do not try to separate the building name precisely. What you usually need is the prefecture and municipality; keeping everything after that as a single string breaks far less often. If the goal is only to normalise notation, running the text through half-width and full-width conversion to unify hyphens and digits already improves duplicate detection considerably. |
| The municipality has since been renamed | After municipal mergers, an address that was valid at the time may not exist in a current list. Old customer records keep the old names and simply fail to match. The reverse also happens — records written with the post-merger name — and which is right depends on when the row was created. | Do not discard rows that fail to match — set them aside as needs review. If there are many, build an old-to-new mapping. Where a postal code exists, making it the key and re-deriving the address from it is the most robust approach: codes do change on merger, but far less often than address strings, and the postal service publishes the history of changes. |
In any system that handles addresses, the standard practice is to store both the normalised form and exactly what the user typed. When normalisation goes wrong a human can still judge from the original, but if you overwrote it there is nothing to fall back on. This matters most for delivery: a form a machine cannot parse is often perfectly clear to the courier, and reformatting it can be what stops the parcel arriving. Keep the split apart as derived data for search and aggregation only.
📖 How to Use
-
1
Paste addressesPaste one address per line.
-
2
Run splitClick Split. All 47 prefectures are detected via regex.
-
3
Download CSVDownload as CSV or TSV for Excel / Sheets.
❓ Frequently Asked Questions
What if the address has no prefecture?
Does it work for non-Japanese addresses?
Are addresses uploaded to your server?
🔗 Related Tools
🐛 Found a bug or issue with this tool?
Free to use, no signup. Even just the steps to reproduce are helpful. Reports go directly to the operator and help us fix issues.
Thanks for your report!
Your report has been delivered to the operator and will be used to improve the tool.