2

Can anyone help me figure out how we can load data into a Tuple created inside TigerGraph DB? I have created the following Tuple and I am trying to load data from a file using Spark.

Graph Name: MyGraph Tuple Name: MyTuple

Attributes of MyTuple: ID String(50),Datee Datetime

The tuple MyTuple is being referred into an edge (ed_newEdge) of MyGraph as an edge attribute. Seeking help to figure out how we can load data into the tuple present in MyGraph using Spark.

Stanislav Kralin
  • 11,070
  • 4
  • 35
  • 58
Surya
  • 45
  • 5

1 Answers1

4

You can insert data to the tuple using tigergraph jdbc driver. You can find an example here https://github.com/tigergraph/ecosys/blob/master/tools/etl/tg-jdbc-driver/tg-jdbc-examples/src/main/java/com/tigergraph/jdbc/UpsertQuery.java

After following above code for creating connection , you can use below code snippet

try (java.sql.Statement stmt = con.createStatement()) {
      query = "INSERT INTO edge ed_newEdge(Vertex1, Vertex2) VALUES(MyTuple("1d123","2011-02-03 01:02:03"))";
      stmt.addBatch(query);}

Alternatively you can process your data on SPARK and serialize your data on a Parquet or csv file. And then you can create a GSQL loading job to read that file and insert the value to your tuple. https://docs.tigergraph.com/dev/gsql-ref/ddl-and-loading/creating-a-loading-job#loading-a-user-defined-type-udt-attribute

TYPEDEF TUPLE <f1 INT (1), f2 UINT, f3 STRING (10), f4 DOUBLE > myTuple   # define a UDT
CREATE VERTEX v_udt  (id STRING PRIMARY KEY, att_udt myTuple)
CREATE GRAPH test_graph (v_udt)
CREATE LOADING JOB load_udt FOR GRAPH test_graph {
    DEFINE FILENAME f;
    LOAD f TO VERTEX v_udt VALUES ($0, myTuple($1, $2, $3, $4) ); 
    # $1 is loaded as f1, $2 is loaded as f2, and so on
}
RUN LOADING JOB load_udt USING f="./udt.csv"

Now you can invoke above created loading job using rest request https://docs.tigergraph.com/dev/gsql-ref/ddl-and-loading/running-a-loading-job

M.S.
  • 66
  • 2