> ## Documentation Index
> Fetch the complete documentation index at: https://docs.turingdb.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Migrate from Neo4j

> Export a Neo4j graph and import it into TuringDB

This guide walks you through importing a Neo4j graph into TuringDB. The process uses **APOC** to export the Neo4j graph as JSONL, which TuringDB can load directly.

## Prerequisites

* A running or restorable Neo4j instance (version 4.x or 5.x) with the **APOC plugin** installed
* A running TuringDB instance

<Tip>
  APOC comes bundled with Neo4j Desktop. For server installs, follow the [APOC installation guide](https://neo4j.com/docs/apoc/current/installation/).
</Tip>

## Migrating from Neo4j 4.x

If your database was created with Neo4j 4.x, you need to migrate it to the 5.x format before exporting. Using `neo4j-admin`:

```bash theme={null}
# Load your v4 dump
neo4j-admin database load --from-stdin neo4j --overwrite-destination=true < mydb.dump

# Migrate to v5 format
neo4j-admin database migrate neo4j --force-btree-indexes-to-range
```

Then start Neo4j 5.x and proceed with the export below.

<Steps>
  <Step title="Export from Neo4j with APOC">
    Connect to your Neo4j instance (via Neo4j Browser, `cypher-shell`, or any client) and run:

    ```jsx theme={null}
    CALL apoc.export.json.all("output.json", {useTypes: true});
    ```

    This writes the entire graph to `output.json` in Neo4j's import directory (typically `$NEO4J_HOME/import/`). The file contains one JSON object per line — one for each node and one for each relationship.

    <Warning>
      Make sure Neo4j has write access to its import directory and that APOC's configuration allows file export (`apoc.export.file.enabled=true` in `apoc.conf`).
    </Warning>
  </Step>

  <Step title="Copy the file to TuringDB's data directory">
    TuringDB only reads external files from the `data` subdirectory of its working directory (set by `--turing-dir`, default `$HOME/.turing`).

    ```bash theme={null}
    cp $NEO4J_HOME/import/output.json ~/.turing/data/
    ```
  </Step>

  <Step title="Load the file into a new graph">
    <Tabs>
      <Tab title="Cypher">
        ```jsx theme={null}
        LOAD JSONL 'output.json' AS mygraph
        ```
      </Tab>

      <Tab title="Python SDK">
        ```python theme={null}
        from turingdb import TuringDB

        client = TuringDB(host="http://localhost:6666")
        client.query("LOAD JSONL 'output.json' AS mygraph")
        ```
      </Tab>
    </Tabs>

    TuringDB creates a new graph named `mygraph` and populates it with all nodes and relationships from the file.
  </Step>

  <Step title="Set the graph as active and query it">
    <Tabs>
      <Tab title="Cypher">
        ```jsx theme={null}
        cd mygraph

        MATCH (n) RETURN n
        ```
      </Tab>

      <Tab title="Python SDK">
        ```python theme={null}
        client.set_graph("mygraph")

        df = client.query("MATCH (n) RETURN n")
        print(df)
        ```
      </Tab>
    </Tabs>
  </Step>
</Steps>

<Check>
  Your Neo4j graph is now available in TuringDB as `mygraph`.
</Check>

## Verify the import

Use meta-queries to inspect what was imported:

<Tabs>
  <Tab title="Cypher">
    ```jsx theme={null}
    CALL db.labels()
    CALL db.edgeTypes()
    CALL db.propertyTypes()
    ```
  </Tab>

  <Tab title="Python SDK">
    ```python theme={null}
    print(client.query("CALL db.labels()"))
    print(client.query("CALL db.edgeTypes()"))
    print(client.query("CALL db.propertyTypes()"))
    ```
  </Tab>
</Tabs>