Importing & Exporting Data-2

 

📤 Exporting Data from Neo4j

1. Export to CSV

You can export query results as CSV.

Example

MATCH (u:User)-[:WORKS_ON]->(p:Project)
RETURN u.name AS user, p.title AS project

👉 In Neo4j Browser, you can click “Download CSV”.

Or in Cypher Shell:

cypher-shell -u neo4j -p password \
"CALL apoc.export.csv.query(
  'MATCH (u:User)-[:WORKS_ON]->(p:Project) RETURN u.name, p.title',
  'export.csv',
  {}
)"

📄 export.csv

u.name,p.title
Alice,Graph DB
Alice,AI System
Bob,Graph DB

---

2. Export to JSON

JSON is handy for APIs or app integration.

Using APOC plugin:

CALL apoc.export.json.query(
  "MATCH (u:User)-[:WORKS_ON]->(p:Project)
   RETURN u.name, p.title",
  "export.json",
  {}
)

📄 export.json

[
  {"u.name":"Alice", "p.title":"Graph DB"},
  {"u.name":"Alice", "p.title":"AI System"},
  {"u.name":"Bob", "p.title":"Graph DB"}
]

---

3. Exporting Whole Database

• CSV:

  CALL apoc.export.csv.all("alldata.csv", {})
  

• JSON:

  CALL apoc.export.json.all("alldata.json", {})
  

---

4. Other Options

• apoc.export.cypher.* → export Cypher scripts (for re-import).

• apoc.export.graphml.* → export GraphML (XML format).

• In Neo4j Bloom or Browser, you can visually download query results as CSV.

---

✅ Summary

• For small queries → just RETURN and download.

• For full database or structured exports → use APOC export functions.

• Choose CSV for spreadsheets/SQL imports, JSON for APIs/web apps.

APOC (Awesome Procedures On Cypher) is like a Swiss army knife for Neo4j 🛠️.
When it comes to ETL (Extract, Transform, Load), APOC is the go-to extension for importing, transforming, and exporting data.

---

⚡ APOC for ETL in Neo4j

1. Extract (Getting data in)

APOC provides procedures to read data from external sources (CSV, JSON, XML, databases, APIs).

Examples

• From CSV

CALL apoc.load.csv("file:///users.csv") YIELD map
RETURN map;

• From JSON (API or file)

CALL apoc.load.json("https://api.example.com/data") YIELD value
RETURN value;

• From JDBC (RDBMS)

CALL apoc.load.jdbc("jdbc:mysql://localhost:3306/mydb","SELECT * FROM users")
YIELD row
RETURN row;

---

2. Transform (Clean / reshape data)

APOC helps in data transformation before saving into graph.

Examples

• String functions

RETURN apoc.text.upper("neo4j");    // "NEO4J"

• Splitting and trimming

RETURN apoc.text.split("Alice,Bob,Charlie", ","); 
// ["Alice","Bob","Charlie"]

• Mapping JSON fields to nodes

CALL apoc.load.json("file:///users.json") YIELD value AS u
MERGE (p:Person {id: u.id})
SET p.name = u.name,
    p.email = apoc.text.lower(u.email);

---

3. Load (Insert into Neo4j)

After transforming, load data into nodes and relationships.

Example – Import CSV into graph

users.csv

id,name,role
1,Alice,Developer
2,Bob,Manager

Cypher:

CALL apoc.load.csv("file:///users.csv") YIELD map AS row
MERGE (u:User {id: row.id})
SET u.name = row.name,
    u.role = row.role;

Example – Import JSON with relationships

[
  {"name": "Alice", "projects": ["GraphDB", "AI System"]},
  {"name": "Bob", "projects": ["GraphDB"]}
]

Cypher:

CALL apoc.load.json("file:///projects.json") YIELD value AS row
MERGE (u:User {name: row.name})
FOREACH (proj IN row.projects |
  MERGE (p:Project {title: proj})
  MERGE (u)-[:WORKS_ON]->(p)
);

---

4. Export (Send data out)

• CSV:

CALL apoc.export.csv.all("all_data.csv", {})

• JSON:

CALL apoc.export.json.all("all_data.json", {})

• To relational DB:

CALL apoc.export.jdbc.query(
  "MATCH (u:User)-[:WORKS_ON]->(p:Project) RETURN u.name, p.title",
  "jdbc:mysql://localhost:3306/mydb?user=root&password=pass",
  "user_projects"
)

---

✅ Summary

• Extract → apoc.load.csv/json/xml/jdbc

• Transform → apoc.text.*, apoc.map.*, FOREACH, Cypher functions

• Load → MERGE nodes/relationships

• Export → apoc.export.csv/json/jdbc/cypher/graphml

👉 In short: APOC makes Neo4j ETL pipelines super flexible, connecting graphs to almost any system