Bulk update workflow using `LOAD FROM` and `MERGE` with Polars DataFrames

update.py

import polars as pl
import kuzu
import shutil

shutil.rmtree("test_db", ignore_errors=True)
db = kuzu.Database("test_db")
conn = kuzu.Connection(db)

# Get a JSON object of persons and products purchased
data = [
    {
        "name": "Karissa",
        "products": ["iPhone", "MacBook"]
    },
    {
        "name": "Rhea",
        "products": ["Airpods", "MacBook"]
    }
]

df = pl.DataFrame(data)
print(df)

conn.execute("CREATE NODE TABLE Person(name STRING, PRIMARY KEY(name))")
conn.execute("CREATE NODE TABLE Product(name STRING, PRIMARY KEY(name))")
conn.execute("CREATE REL TABLE Purchased(FROM Person TO Product)")

conn.execute(
    """
    COPY Person FROM (
        LOAD FROM df 
        RETURN name
    );
    """
)
conn.execute(
    """
    COPY Product FROM (
        LOAD FROM df 
        WITH products 
        UNWIND products AS product 
        RETURN DISTINCT product
    );
    """
)
conn.execute(
    """
    COPY Purchased FROM (
        LOAD FROM df 
        WITH * 
        UNWIND products AS product 
        RETURN name, product
    );
    """
)

conn.execute(
    """
    LOAD FROM df
    WITH *
    UNWIND products AS prod
    MERGE (p:Person {name: name})
    MERGE (product:Product {name: prod})
    MERGE (p)-[:Purchased]->(product);
    """
)

# Query the database
res = conn.execute("MATCH (p:Person)-[:Purchased]->(product:Product) RETURN p.name, product.name;")
print(res.get_as_pl())