How do I parse a data with column that needs to be splitted while using neo4j admin-import?

[shonesum]

Hello all,
I have a node and relationship csv. In relationship, I have a column where I need to split the column on a pipe. I tried using --array-delimiter = '|' but It doesn't split the data. My data looks like

node.csv

identifier:ID,name:LABEL
1,apple
2,ball
3,cat

rel.csv

source_ids:START_ID,target_ids:END_ID,type:TYPE,data
1,2,connection,
2,3,relation,test1|test2
3,1,connection,test1
1,3,relation,test4|test3|tet6

I used the code
neo4j-admin import --verbose --ignore-extra-columns=true --array-delimiter "|" --nodes C:/Users/Sam/Documents/node.csv --relationships C:/Users/Sam/Documents/rel.csv.

The result I get is :

{
 "data": "test4|test3|tet6"
}

what I want is :

{
  "data": ["test4","test3","tet6"]
}

I am using Neo4j 4.0.0 community.

[mike_r_black]

I've never tried the array delimiter option. If you can't get it to work my other suggestion would be that after you import the data you could write a cypher command to refactor the data into correct model after it's loaded. The APOC library can help iterate the same command over and over again until all the data is found to be clean.

[ameyasoft]

Try this:

with split("test4|test3|tet6", "|")  as t1 
with [t1[0], t1[1], t1[2]] as t2 return t2

Replace "test4|test3|tet6" with data per your csv.

Result: ["test4", "test3", "tet6"]

[shonesum]

shonesum:

source_ids:START_ID,target_ids:END_ID,type:TYPE,data

Instead of data, if we use data:STRING

it worked. Thank you