Reputation: 692
neo4j-sh (?)$ USING PERIODIC COMMIT
> LOAD CSV WITH HEADERS FROM "file:/home/ubuntu/data/woka-data/woka-data.csv" AS row
> MERGE (woka:Woka {woka_id: row.woka_id, woka_title: row.woka_title});
IllegalMultilineFieldException: At file:/home/ubuntu/data/woka-data/woka-data.csv:834 - Multi-line fields are illegal in this context and so this might suggest that there's a field with a start quote, but a missing end quote.
My kernel version is 2.2.3 and the file is quite large:
-rwxr-xr-x 1 ubuntu ubuntu 2508218727 Jul 21 17:54 woka-data.csv
Any clue how to fix this?
I changed the order of the fields in the file and now the first error is reported for line 83. Here is this line:
97801322605720000000,eng,Addison Wesley Longman,,"Wri& Speakg at Work& Build Own CC& Mst
: Perils"
It is a new line before
: Perils
How to fix this?
This error seems to be specific to this kernel version. Before upgrading I used a similar file to import the same data having multiline fields and it worked.
What can I do now?
Upvotes: 3
Views: 1114
Reputation: 692
The bug is still persistent in Kernel version 2.2.3 despite the claims that was fixed. I had to clean all data at the extraction time. I removed all characters that caused LOAD CSV to fail with reported error message or similar.
Upvotes: 3
Reputation: 1099
Found this gem in the source code. See if any rows(lines) in your csv have illegal quoting. Escape these and try again
public class IllegalMultilineFieldException extends FormatException
{
public IllegalMultilineFieldException( SourceTraceability source )
{
super( source, "Multi-line fields are illegal in this context and so this might suggest that " + "there's a field with a start quote, but a missing end quote." );
}
}
This was an issue before Neo4j 2.2.2. Bug report can be found here.
If youre using any version previous to 2.2.2 update it and it should work
I am using Kernel version Neo4j - Graph Database Kernel (neo4j-kernel), version: 2.2.3
Upvotes: 1