import - Add the first row of csv as the attributes of a node type - Neo4J

Question

I have csv with a lot of columns, which is why it's not a productive and automatic way to add all columns as attributes by hand when writing a query. The csv has the first row with the names of the columns, and the remaining rows represent the nodes with the values of the attributes.

I tried to add the attributes using SET += but doesn't work.

LOAD CSV FROM 'file:///stations.csv' AS row
WITH row LIMIT 1
UNWIND row as att
WITH att
CREATE(n:Station)
SET n += att

question from:https://stackoverflow.com/questions/66066668/add-the-first-row-of-csv-as-the-attributes-of-a-node-type-neo4j

Answer 1

It would be best to choose only the few properties that are really needed "in the graph", and only load those. Generally speaking it is not best practice to replicate a relational data store table structure (with large column counts) inside Neo4j, first sketch out a new graph meta-model, then load it, this usually requires some ETL steps, and often data cleaning. One can sketch out the meta-model on paper with a pencil, or using the arrows tool, which provides a more polished diagram if needed for the documentation.

Note, I can however envision a need to dynamically handle the property list, perhaps the property list is short but always changing? I imagine there are several ways to do it, but off the top of my head one way would be to dynamically build the cypher statement in a string and then execute it using this APOC call

CALL apoc.cypher.run()