To migrate the data, you have to convert SQL data to graph data. This is a long-lasting activity of the process and probably awkward as well. Because you have to migrate data from SQL, you must read data using ad hoc SQL queries and write data using Cypher queries.
Note that another strategy would be importing data from a CSV file, with the LOAD CSV FROM statement, which is available from Cypher 2.1 at http://docs.neo4j.org/chunked/milestone/import-importing-data-from-a-single-csv-file.html.
Because we have to create a node with its attributes, we have to read data using a SQL statement that returns all the data we need to create a node. Of course, you can migrate an entity in more steps, but we would usually prefer to perform only one CREATE query for performance reasons because otherwise we would have to migrate a huge database.
For example, let's start by migrating the
Authorsentity. This entity is easy to deal with; in fact, all attributes are only in one table...