我们可以在 Java 中使用 JDBC 驱动程序使用 Select 查询来获取顶点数据吗

Question

I am trying to run select statement for tigerGraph using JDBC Driver for JAVA but getting java.sql.SQLException: Unsupported dbtable: InfectionCase Error. Is select statement supported by tigerGraph using JDBC or is there any other way to execute Select Statement.

try (Connection con = getConnection()) {
        String query = "select * From InfectionCase where id == \"Hansarang Convalescent Hospital\"";
        System.out.println("Running :" + query);
        Statement statement = con.createStatement();
        ResultSet rs = statement.executeQuery(query);
        System.out.println(rs.toString());
        do {
            java.sql.ResultSetMetaData metaData = rs.getMetaData();
            System.out.println("Table: " + metaData.getCatalogName(1));
            System.out.print(metaData.getColumnName(1));
            for (int i = 2; i <= metaData.getColumnCount(); ++i) {
                System.out.print("\t" + metaData.getColumnName(i));
            }

java.sql.SQLException: Unsupported dbtable: InfectionCase
at com.tigergraph.jdbc.restpp.driver.QueryParser.<init>(QueryParser.java:629)
at com.tigergraph.jdbc.restpp.RestppStatement.execute(RestppStatement.java:41)
at com.tigergraph.jdbc.restpp.RestppStatement.executeQuery(RestppStatement.java:35)
at com.prissoft.connection.GraphJSON.selectStatementUsingSQL(GraphJSON.java:145)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:498)

Answer 1

To fetch vertex instances from TigerGraph you need to use a different syntax as the SELECT statement is not supported in the JDBC driver:

try (Connection con = getConnection()) {
    String query = "get vertex(InfectionCase, \"Hansarang Convalescent Hospital\")";
    <more statements>
}

You can use this get vertex(vertex_type) syntax to get a single vertex by primary ID or get a set of vertices (optionally filtered, sorted and limited). See the documentation on this.

I personally think that it is better to access TigerGraph via the REST API than using the JDBC driver. JDBC was designed for relation databases and thus it expects tabular data. The graph is not tabular, it's a data network. So, what you normally get out of a TigerGraph statement or query execution is a JSON document describing a subgraph: vertices and edges. This output can be flattened into rows and columns, but then you get a lesser image of the graph: a 2D projection of a 3D (or rather: a multi-dimensional) data set.

Another note: fetching individual vertices and edges, or even sets of these via JDBC calls (like above) or via the built-in REST endpoints (like http://localhost:9000/graph/MyGrah/vertices/InfectionCase/Hansarang+Convalescent+Hospital ) is not the most efficient use of TigerGraph. Surely, sometimes you just need a set of vertices to list in an app or visualise a graph. The real power comes from developing queries , TigerGraph's "stored procedures" that can execute even very complex and sophisticated graph algorithms like PageRank or Louvain community detection inside the graph very fast, in a parallelised fashion.