I am using Data Importer to upload csv-files.
A lot of property cells in the csv-file are empty (really, so no spaces).
I would expect them to appear as null, but they don't.
See top of attachment: I checked with coalesce and that returns a ""-value.
After removing the property, coalesce does recognize the null, see bottom of attachment.
My conclusion is that Data Importer doesn't import empty cells as null.
It is my understanding that this can be solved while importing manually, but I can't find any setting in Data Importer.
Is there any solution?
Any help greatly appreciated!
We are aware of this behavior and it will be eventually solved. Meanwhile, fastest way to clean this properties is:
call apoc.periodic.iterate("MATCH (n) return id(n) as id","MATCH (n) where id(n) = id set n = apoc.map.clean(properties(n), [],['',' ','-'])",{batchSize:50000, parallel:true})
I don’t think so, but can’t you get along without doings so? In queries, you can query with condition ‘property is null’ if you want to identify these nodes after import.
Hi @bennu_neo ,
let's hope so!
But in the meantime: the problem is that I can remove all the empty-but-not-null properties in Neo4j Browser, but every time I want to add stuff with Data Importer, it wants me to re-upload all the csv's and when I hit "Run Import", it puts the original stuff back in (which is good because it allows me to make corrections in the original csvs, but in this respect it is killing)...
Hi @glilienfield, thanks for thinking along!
The problem is that the properties are not recognised as null...
Even if they were, I am using NeoDash and I can't add code wherever I need it.
An additional problem is that I can remove all the empty-but-not-null properties in Neo4j Browser, but every time I want to add stuff with Data Importer, it wants me to re-upload all the csv's and when I hit "Run Import", it puts the original stuff back in...