generate RDB DDL/DML from GraphDB (Tinkerpop).

• this script generate 4 SQL files.
  1. [DDL] CREATE TABLE "vertex_${vertex label}"
     • the columns are as follows.
       • id
       • property_${all property keys}
  2. [DDL] CREATE TABLE "edge_${edge label}from${out vertex label}to${in vertex label}"
     • the columns are as follows.
       • id
       • id_in_v (= vertex_{vertex label}.id)
       • id_out_v (= vertex_{vertex label}.id)
       • property_${all property keys}
  3. [DML] INSERT INTO "vertex_${vertex label}"
  4. [DML] INSERT INTO "edge_${edge label}from${out vertex label}to${in vertex label}"

• Java 11
• Scala 2.13
• sbt 1.9.8

1. checkout this repository

   gh repo clone kazumatsudo/GraphDB2RDB

1. launch your GraphDB Server

   docker compose up -d

2. select analysis method
   choose analysis method to search Vertex/Edge algorithm.

   • "by_exhaustive_search" (default)
   • "using_specific_key_list"

   please see following passage if you want to change method.

   % export ANALYSIS_METHOD=by_exhaustive_search

3. execute the script by sbt

   sbt run

4. generate SQL files

   • sql/ddl_edge.sql
   • sql/ddl_vertex.sql
   • sql/insert_edge.sql
   • sql/insert_vertex.sql

no advance preparation required because it's selected by default.

1. set environment variable "ANALYSIS_METHOD" to "using_specific_key_list"

   % export ANALYSIS_METHOD=using_specific_key_list

2. set target vertex label, property key, and its values in using_key_list_file.json
   • json schema: using_key_list_file_schema.json

You can define following settings as you like.

• GraphDB connection settings
• output SQL directory

The settings you can be changed are summarized in application.conf.

• CODE_OF_CONDUCT

• CONTRIBUTING

• LICENSE

• kazumatsudo

• xxx