Oracle8i SQLJ Developer's Guide and Reference
Release 3 (8.1.7)

Part Number A83723-01


Solution Area



Go to previous page Go to beginning of chapter Go to next page

Translating SQLJ Source on a Client and Loading Components

One approach to developing SQLJ code for the server is to first run the SQLJ translator on a client machine to take care of translation, compilation, and profile customization. Then load the resulting class and resource files (including your SQLJ profiles) into the server, typically using a Java archive (.jar) file.

If you are developing your source on a client machine, as is usually the case, and have a SQLJ translator available there, this approach is advisable. It allows the most flexibility in running the translator, because option-setting and error-processing are not as convenient in the server.

It might also be advisable to use the SQLJ -ser2class option during translation when you intend to load an application into the server. This results in SQLJ profiles being converted from .ser serialized resource files to .class files and simplifies their naming. Be aware, however, that profiles converted to .class files cannot be further customized. To further customize, you would have to rerun the translator and regenerate the profiles. For information about the -ser2class option, see "Conversion of .ser File to .class File (-ser2class)".

When you load .class files and .ser resource files into the server, either directly or using a .jar file, the resulting database library units are referred to as Java class schema objects (for Java classes) and Java resource schema objects (for Java resources). Your SQLJ profiles will be in resource schema objects if you load them as .ser files, or in class schema objects if you enabled -ser2class during translation and load them as .class files.

Loading Classes and Resources into the Server

Once you run the translator on the client, use the Oracle loadjava client-side utility to load class and resource files into schema objects in the server. This utility is discussed in detail in the Oracle8i Java Developer's Guide.

Either specify the class and resource files individually on the loadjava command line, or put them into a .jar file and specify the .jar file on the command line. A separate schema object is created for each .class or .ser file in the .jar file or on the command line.


As of release 8.1.6, the loadjava utility does support compressed files.  

Consider an example where you: 1) translate and compile Foo.sqlj, which includes an iterator declaration for MyIter; 2) enable the -ser2class option when you translate Foo.sqlj; and 3) archive the resulting files (Foo.class, MyIter.class, Foo_SJProfileKeys.class, and Foo_SJProfile0.class) into Foo.jar. Then run loadjava with the following command line (plus any options you want to specify). This examples uses the default OCI driver:

loadjava -user scott/tiger Foo.jar

Or, alternatively, use the original files:

loadjava -user scott/tiger Foo.class MyIter.class Foo_SJProfileKeys.class Foo_SJProfile0.class


loadjava -user scott/tiger Foo*.class MyIter.class

Or, to use the Thin driver for loading (specifying the -thin option and an appropriate URL):

loadjava -thin -user scott/tiger@localhost:1521:ORCL Foo.jar

For information about files generated by the SQLJ translator, see "Code Generation" and "Java Compilation".


  • When you load a profile into the server as a .ser file, it is first customized if it was not already customized on the client. If it was already customized, it is loaded as is.

  • You can access the USER_OBJECTS view in your database schema to verify that your classes and resources are loaded properly. This is discussed in the Oracle8i Java Developer's Guide.


Although the loadjava utility is recommended for loading your SQLJ and Java applications into the server, you can also use Oracle SQL CREATE JAVA commands such as the following:



See the Oracle8i SQL Reference for more information about the CREATE JAVA commands.

Loaded Class and Resource Schema Objects

This section discusses how schema objects for classes and profiles are named when you load classes and profiles into the server.

If the SQLJ -ser2class option was enabled when you translated your application on the client, then profiles were converted to .class files and will be loaded into class schema objects in the server. If -ser2class was not enabled, then profiles were generated as .ser serialized resource files and will be loaded into resource schema objects in the server.

In the following discussion, it is assumed that you use only the default connection context class for any application that will run in the server; therefore, there will be only one profile.


There are two forms of schema object names in the server: full names and short names.

Full names are fully qualified and are used as the schema object names wherever possible. If any full name is longer than 31 characters, however, or contains characters that are illegal or cannot be converted to characters in the database character set, then the Oracle8i server converts the full name to a short name to employ as the name of the schema object, keeping track of both names and how to convert between them. If the full name is 31 characters or less and has no illegal or inconvertible characters, then the full name is used as the schema object name.

For more information about these and about other file naming considerations, including DBMS_JAVA procedures to retrieve a full name from a short name and the converse, see the Oracle8i Java Developer's Guide.  

Full Names of Loaded Classes (including profiles if -ser2class enabled)

The full name of the class schema object produced when you load a .class file into the server is determined by the package and class name in the original source code. Any path information you supply on the command line (so that loadjava can find it, for example) or in the .jar file is irrelevant in determining the name of the schema object. For example, if Foo.class consists of a class Foo which was specified in the source code as being in package x.y, then the full name of the resulting class schema object is as follows:


Note that ".class" is dropped.

If Foo.sqlj declares an iterator MyIter, then the full name of its class schema object is:


(Unless it is a nested class, in which case it will not have its own schema object.)

The related profile-keys class file, generated by SQLJ when you translate Foo.sqlj, is Foo_SJProfileKeys.class; therefore, the full name of its class schema object is:


If the -ser2class option was enabled when you translated your application, then the resulting profile was generated in file Foo_SJProfile0.class; therefore, the full name of the class schema object is:


Full Names of Loaded Resources (including profiles if -ser2class not enabled)

This discussion is relevant only if you did not enable the -ser2class option when you translated your application, or if you use other Java serialized resource (.ser) files in your application.

The naming of resource schema objects is handled differently from class schema objects--their names are not determined from the contents of the resources. Instead, their full names are identical to the names that appear in a .jar file or on the loadjava command line, including path information. Note also that the .ser extension is not dropped.

It is important to note that because resource names are used to locate the resources at runtime, their names must include the correct path information. In the server, the correct full name of a resource is identical to the relative path and file name that Java would use to look it up on a client.

In the case of a SQLJ profile, this is a subdirectory under the directory specified by the translator -d option, according to the package name. If the -d option, used to specify the top-level output directory for generated .class and .ser files, is set to /mydir and the application is in package abc.def, then .class and .ser files generated during translation will be placed in the /mydir/abc/def directory. (For more information about the SQLJ -d option, including the default value, see "Output Directory for Generated .ser and .class Files (-d)".)

At runtime, /mydir would presumably be in your CLASSPATH, and Java will look for your application components in the abc/def directory underneath it.

Therefore, when you load this application into the server, you must run loadjava or jar from the -d directory so that the path you specify on the command line to find the files also indicates the package name, as follows:

cd /mydir
loadjava <...options...> abc/def/*.class abc/def/*.ser

Or, if you use a .jar file:

cd /mydir
jar -cvf myjar.jar abc/def/*.class abc/def/*.ser
loadjava <...options...> myjar.jar

If your application is App and your profile is App_SJProfile0.ser, then either of the above examples will correctly result in the following full name of the created resource schema object:


Note that ".ser" is retained.

Note also that if you set -d to a directory whose hierarchy has no other contents (which is advisable), you can simply run jar as follows to recursively get your application components:

cd /mydir
jar -cvf myjar.jar *
loadjava <...options...> myjar.jar

Publishing the Application After Loading Class and Resource Files

Before using your SQLJ code in the server, you must publish the top-level methods, as is true of any Java code you use in the server. Publishing includes writing call descriptors, mapping datatypes, and setting parameter modes. For information, see the Oracle8i Java Stored Procedures Developer's Guide.

Summary: Running a Client Application in the Server

This section summarizes the typical steps of running a client application in the server. As an example, it uses the NamedIterDemo sample application provided in "Named Iterator--NamedIterDemo.sqlj".

  1. Create a .jar file for your application components. For NamedIterDemo, the components include SalesRec.class as well as the application class and profile.

    You can create a .jar file niter-server.jar as follows:

    jar cvf niter-server.jar Named*.class Named*.ser SalesRec.class
  2. Load the .jar file into the server.

    Use loadjava as follows. This example instructs loadjava to use the OCI8 driver in loading the files. The -resolve option results in the class files being resolved.

    loadjava -oci8 -resolve -force -user scott/tiger niter-server.jar
  3. Create a SQL wrapper in the server for your application.

    For example, run a SQL*Plus script that executes the following:

    set echo on
    set serveroutput on
    set termout on
    set flush on
    execute dbms_java.set_output(10000);
    create or replace procedure SQLJ_NAMED_ITER_DEMO as language java 
    name 'NamedIterDemo.main (java.lang.String[])';

    The DBMS_JAVA.SET_OUTPUT() routine reroutes default output to your screen, instead of to a trace file; the input parameter is the buffer size in bytes.

  4. Execute the wrapper.

    For example:

    sqlplus> call SQLJ_NAMED_ITER_DEMO();

Go to previous page
Go to beginning of chapter
Go to next page
Copyright © 1996-2000, Oracle Corporation.

All Rights Reserved.


Solution Area