Full Text index search query is not working

[amanvbhojak23]

This is my data

{
"identity": 15,
"labels": [
"Book"
],
"properties": {
"description": "This is the book description",
"id": "2"
}
}
{
"identity": 17,
"labels": [
"Book"
],
"properties": {
"description": "This is the description",
"id": "1"
}
}
{
"identity": 19,
"labels": [
"Book"
],
"properties": {
"description": "This is description",
"id": "3"
}
}
{
"identity": 20,
"labels": [
"Book"
],
"properties": {
"description": "description is This is description",
"id": "3"
}
}
{
"identity": 25,
"labels": [
"Book"
],
"properties": {
"description": "This is description",
"id": "4"
}
}
{
"identity": 29,
"labels": [
"Book"
],
"properties": {
"description": "This is description",
"id": "10"
}
}
{
"identity": 30,
"labels": [
"Book"
],
"properties": {
"description": "This is invalid description",
"id": "10"
}
}

This is how i created full text index
-- create full text index
CALL db.index.fulltext.createNodeIndex("book_index_1",["Book"],["description"], { analyzer: "keyword", eventually_consistent: "true" });

This is working.
CALL db.index.fulltext.queryNodes('book_index_1', 'This is description') yield node as n, score MATCH (n) return n, score;

This is not working – similar data with space is present.

CALL db.index.fulltext.queryNodes('book_index_1', 'This is SPACE_HERE_AS_IN_DATA invalid description') yield node as n, score MATCH (n) return n, score;
@Ari_Neo4j @dana_canzano

[dana_canzano]

@amanvbhojak23

Any details on what Neo4j version?

Also, per Full-text indexes - Cypher Manual

Full-text indexes store individual words in a STRING property. 
This is achieved by tokenizer, which breaks up a stream of 
characters into individual tokens (usually individual words).

so i would expect whitespace characters are not included in the index

[hakan.lofqvist1 (Hakan Lofqvist)]

See lucene doc for keyword analyzer:

KeywordAnalyzer "Tokenizes" the entire stream as a single token. This is useful for data like zip codes, ids, and some product names.

So your "invalid description" will yield ONE token:

keyword=[This is invalid description]

It really goes crazy if you have tab or other characters in there (it becomes difficult to search and find the term).

I am not an expert, but I would say they nailed the usecases for KeywordAnalyzer in the documentation. Maybe you should pick a different analyzer for book descriptions?