Optimize Neo4j Cypher query on Neo4jSearchDataExtractor (#213)

* Optimize Neo4j Cypher query on Neo4jSearchDataExtractor

* flake8

databuilder/extractor/neo4j_search_data_extractor.py

@@ -21,21 +21,27 @@ class Neo4jSearchDataExtractor(Extractor):
         MATCH (db:Database)<-[:CLUSTER_OF]-(cluster:Cluster)<-[:SCHEMA_OF]-(schema:Schema)<-[:TABLE_OF]-(table:Table)
         {publish_tag_filter}
         OPTIONAL MATCH (table)-[:DESCRIPTION]->(table_description:Description)
-        OPTIONAL MATCH (table)-[read:READ_BY]->(user:User)
-        OPTIONAL MATCH (table)-[:COLUMN]->(cols:Column)
-        OPTIONAL MATCH (cols)-[:DESCRIPTION]->(col_description:Description)
         OPTIONAL MATCH (table)-[:TAGGED_BY]->(tags:Tag) WHERE tags.tag_type='default'
+        WITH db, cluster, schema, table, table_description, COLLECT(DISTINCT tags.key) as tags
         OPTIONAL MATCH (table)-[:TAGGED_BY]->(badges:Tag) WHERE badges.tag_type='badge'
+        WITH db, cluster, schema, table, table_description, tags, COLLECT(DISTINCT badges.key) as badges
+        OPTIONAL MATCH (table)-[read:READ_BY]->(user:User)
+        WITH db, cluster, schema, table, table_description, tags, badges, SUM(read.read_count) AS total_usage,
+        COUNT(DISTINCT user.email) as unique_usage
+        OPTIONAL MATCH (table)-[:COLUMN]->(col:Column)
+        OPTIONAL MATCH (col)-[:DESCRIPTION]->(col_description:Description)
+        WITH db, cluster, schema, table, table_description, tags, badges, total_usage, unique_usage,
+        COLLECT(col.name) AS column_names, COLLECT(col_description.description) AS column_descriptions
         OPTIONAL MATCH (table)-[:LAST_UPDATED_AT]->(time_stamp:Timestamp)
         RETURN db.name as database, cluster.name AS cluster, schema.name AS schema,
         table.name AS name, table.key AS key, table_description.description AS description,
         time_stamp.last_updated_timestamp AS last_updated_timestamp,
-        EXTRACT(c in COLLECT(DISTINCT cols)| c.name) AS column_names,
-        EXTRACT(cd IN COLLECT(DISTINCT col_description)| cd.description) AS column_descriptions,
-        REDUCE(sum_r = 0, r in COLLECT(DISTINCT read)| sum_r + r.read_count) AS total_usage,
-        COUNT(DISTINCT user.email) as unique_usage,
-        COLLECT(DISTINCT tags.key) as tags,
-        COLLECT(DISTINCT badges.key) as badges
+        column_names,
+        column_descriptions,
+        total_usage,
+        unique_usage,
+        tags,
+        badges
         ORDER BY table.name;
         """
     )

setup.py

@@ -2,7 +2,7 @@ import os
 from setuptools import setup, find_packages
 
 
-__version__ = '2.2.0'
+__version__ = '2.3.1'
 
 
 requirements_path = os.path.join(os.path.dirname(os.path.realpath(__file__)), 'requirements.txt')