A village is contained within a prefecture, and a district has many villages or towns that not overlap an no uncovered area.

By importing the table from wikipedia, we can get the name, en (english name), prefecture, district and area details of the village.

So we can use the following code to get the data and run the cypher query.

const createVillages = async (ctx) => {
  const url = `https://www.wikitable2json.com/api/List_of_villages_in_Japan?table=0&keyRows=1`;
  const { data } = await axios.get(url);

  const villages = data[0].map((village) => {
    return {
      area: village["Area (in km²)"],
      district: village["District"],
      name: village["Japanese"],
      prefecture: village["Prefecture"],
      en: village["Village"],
    };
  });

  const result = await ctx.session.run(geography.createVillages, { villages });

  logger.info(`Villages created`);

  return result;
};

And the cypher is pretty simple, for each village, create Village node and set area, en property. Create a district node and bind the village, district and prefecture relationships.

UNWIND $villages as village
MERGE (v:Village:Municipal{name: village.name})
SET v.area = village.area,
    v.en = village.en

WITH v, village
MERGE (d:District{name: village.district})
MERGE (v)-[:IN]->(d)

WITH v, village, village.prefecture as prefecture, d
MATCH (p:Prefecture)
WHERE p.en = toLower(prefecture)
  OR p.en = toLower(
    replace(
      replace(prefecture, 'ō', 'o'),
      'Ō',
      'O'
    )
  )
MERGE (v)-[:IN]->(p)
MERGE (d)-[:IN]->(p)

Thoughts

By building the fundamentals of data in the database, I know more than yesterday.

Reference

1. List of villages in Japan - Wikipedia