javadbchem Wiki

If you have decided to use javadbchem, you need to check out the project or download the release. You also need to have ant installed (version 1.9 was used for development). In both cases, you should find the following file in your directory: src/schema/example-schema.xml - this file describes the structure of the database you are going to create (the file name can be anything ending with -schema.xml, so you do not need to change the name, but you could). All tables and fields currently contained are needed for the structure database to work properly, so you should not remove them. You can add further columns as you like to any table (e. g. to MOLECULE for molecule properties or descriptors) and you can add new tables. These can link to the existing tables using foreign keys (e. g. the ATOM_ID of the ATOM table for a complex atomic property you don't want to put directly into the ATOM table). So you are free to modify this as long as you leave the current structure in place (if you change something you risk the generated Java code not to work). You can also change the name attribute of the database tag, this is the name which will be used for your database (or schema in Oracle).

There is also a build.properties file. It contains the settings which are used for code generation by Torque. The following settings are important (consult the http://db.apache.org/torque/torque-3.3/releases/torque-3.3/generator/index.html for further information):

• torque.project: The name which will be used for a couple of filenames, including the jar generated for inclusion in your final project
• torque.database: The type of database to use. Only mysql is currently tested.
• torque.targetPackage: The Java package in which the OR mapping Java classes will be created.

The settings in torque.database are not used in the process described here.

test/net/sf/javadbchem/test/Torque.properties: Such a file is supplying run time configuration including database credentials to Torque. You do not need the file now, but you can change it now. The settings torque.dsfactory.molecules.connection are the important ones, for the rest, defaults should work.

Once done with these files, you can change to the root directory of the checkout (i. e. the one containing build.xml). If you run ant here without further parameters, Torque generates the sql scripts and the Java mapping classes. The Java code is then wrapped in a file dist/<yourprojectname>.jar. This is the file you can use in your project to read and write chemical structures from and to the database. See Howto on using the generated code how to do this.