|Basic Topics||Advanced Topics||Reference Topics|
|Documentation Home Page
Creating an InstantDB Database
The ScriptTool Sample Program
Viewing the Database
Apple MAC Users
Visual Age for Java users
Note: JTA is not required if you are using InstantDB in conjunction with Enhydra Enterprise.
You should now have the following sub-directories:
|Classes||holds the jar files containing the database classes|
|Examples||Holds various example files|
|Functions||Holds the source code for InstantDB's SQL functions|
java -ms16m -mx32m org.enhydra.instantdb.ScriptTool
or, if you are running JDK 1.2 or above:
java -Xms16m -Xmx32m org.enhydra.instantdb.ScriptTool
If all goes well, ScriptTool will create further sub-directories and create database tables in those sub-directories. You will also see some output on the screen. This is due to the fact that the supplied properties file enables a certain amount of logging. This can be easily switched off by editing the sample.prp file and setting traceLevel to zero.
Note: The Classes/idb.jar file is the only .jar that you need to keep in order to use InstantDB. idbexmpl.jar contains only pre-compiled versions of the example programs. It can be deleted once the example programs are no longer required.
If you get a ClassNotFoundException, or for some reason the example program doesn't load, then try the following:
If you were paying very close attention to the sample output, you may have spotted the URL "jdbc:idb:sample.prp". InstantDB refers to a database via a Java properties file. The properties file defines everything that InstantDB needs to know in order to create or open the database. Even the name of the database is derived from the properties file.
There is no separate database creation utility. If the properties file directs InstantDB to a database that does not exist, InstantDB creates the database itself.
The sample.prp file contains comments on the available configuration options. Those that are not self-explanatory are discussed later.
The ScriptTool program by default reads its SQL from the file called sql1.txt, although any other file can be supplied as a parameter. The particular sql1.txt supplied runs threads to interpret sql2.txt and sql3.txt in the sample sub-directory. This includes examples of some (but certainly not all) of the SQL syntax supported by InstantDB. Full details of the sample program are given in the examples section.
The test program also includes a test import file: "import1.txt". The format of this file is found in import_schema.txt. More information on importing tables is described later.
You can view the contents of the newly created database using the command:
java -ms16m -mx32m org.enhydra.instantdb.DBBrowser
This starts a small database browser. It initially displays the system properties such as the Java Virtual Machine being used. Click Browse. This displays the contents of the current directory. Click on sample.prp and then click on Open. This constructs the URL for the database. Click on Connect and the program connects to the database. The left hand panel displays the tables held in the database, including system tables and indexes. Click on the import1 table - the program creates the SQL: SELECT * FROM import1. click on Submit and this gets submitted to the database. The results are displayed in the right hand panel.
The properties file which defines the database, contains information about the location of various database files. These are specified using the following properties.
|tablePath||Y||Ordinary database tables|
|tmpPath||N||Temporary tables such as Results sets|
|systemPath||N||System tables, such as the table of all columns|
If tmpPath, systemPath and indexPath are not specified then they default to the same value as tablePath.
Paths can be specified as either absolute paths, or as relative paths. If relative paths are used, then the paths are relative to the current user directory as specified in the system property user.dir. This is usually the directory that the Java Virtual Machine is being run in (i.e., the directory you were in when you typed the "java" command). Note that changing user.dir does not change the user's current directory, it simply changes the property. There is currently no way to change the user directory in Java.
Paths relative to the properties file itself, rather than to the current user directory, are also supported. To enable this option, include the line:
in the properties file. Note that both absolute and relative paths can be specified using either forward slashes / or double backslashes \\ on Windows 95 and NT. Using relative paths with forward slashes provides the highest degree of portability.
The sample.prp file provided in the sample directory uses paths that are relative to itself.
Paths can also be specified using system properties. This is done by prefixing the path value with a '$' dollar sign. For example, setting tablePath=$user.dir will cause tablePath to be set to the runtime value of the System property "user.dir".
When table paths are specified relative to the properties file, make sure you pass a url that contains the full path of the properties file. e.g., jdbc:idb:c:\idb\sample.prp (double back-slashes if in Java code).
A url such as jdbc:idb:sample.prp will work fine when you are actually running your program from the directory containing sample.prp, but you must use the full path of the properties file if you are running the program from a different directory.
The current example text files are saved with CR/LF line terminators. You may have to convert some of these files to use standard MAC line terminators (CR only) to get the sample programs to work.
Thanks to David Bowser-Chao for pointing this out.
InstantDB automatically generates certain file names. Because of filename length limitations on the MAC, users are advised to restrict table and column name lengths. In particular, if binary columns are being used then the table and column names should be restricted to no more than 4 or 5 characters.
For some reason, Visual Age for Java cannot import the current InstantDB JAR files. To import into VAJ, simply unarchive the JAR files into an empty directory and import the directory instead.