This introduction covers the following topics:
This section defines basic terms used throughout this documentation. For your convenience, the definitions are repeated in a glossary, along with other terms introduced in later documents.
An object is one of the following:
A unit of programming code (for example, a program or subroutine);
A data-dictionary definition (for example, a view of a database file);
A user-written error message.
Natural views (which can be generated from Predict views) are called data definition modules (DDMs). To accommodate the terminology's used at different sites; they are called views/DDMs in this documentation.
An object version is an object that has been put under the control of PAC and assigned a version number. An object version is sometimes called a PAC object or versioned object.
Source code means a unit of code written in a high-level language, such as Natural. Source code is equivalent to Natural "saved" code.
Executable code or executable object means a unit of code that has been compiled and can be translated by Natural. Executable code is equivalent to Natural "cataloged" code. A PAC entity is any of the structures that PAC uses to control or facilitate the movement of objects through the life-cycle. These entities are discussed in Using PAC.
An application is a set of objects that work together to perform a task. Thus, an application object is a component of the application.
A Natural library is a set of Natural objects that are stored together; you can add objects to a library or delete objects from it.
A foreign location is the physical location of the dataset that contains the foreign objects.
A lock prohibits access to a library, application, or object.
System files are Adabas files that contain the following (the name of each system file is in parentheses):
Software resources (FNAT)
Application libraries (FUSER)
Predict information (FDIC)
Security information (FSEC)
In this documentation, the FUSER file, which contains application objects, is called the user system file. The FDIC file is called the Predict file. In addition to Predict data definitions and cross-reference information, Natural rules and views/DDMs generated from Predict entries are stored in the Predict file. User system files and Predict files are identified in PAC by the database number (DBnr) and file number (Fnr).
In addition, there are three system files reserved to PAC and PAA:
The PAC ACF systems file stores the saved and cataloged code for every version of every object under PAC control. The ACF also stores extended information about object versions and PAC entities.
The PAC PCF system file stores cross-reference data, keywords, and the latest version of Natural and Predict objects in the PAC-controlled environment. Natural objects are cataloged and Predict objects are generated in the PCF.
The PAA system file stores audit data for production objects.
The user environment is the mix of hardware, operating systems, teleprocessing (TP) systems, and database managers (DBMSs) at a site. An application environment or operating environment includes the physical locations where application objects are stored, and the operating system, TP system, and DBMS with which they are developed, tested, or used. The PAC-controlled environment refers to the PAC ACF and PCF system files and the operating environments that are governed by site-specific procedures defined in PAC. The PAA-controlled environment refers to production environments that are protected and audited by PAA. A Natural directory provides information about the environment in which an object was developed, saved, or cataloged.
PAC is fully integrated with other Software AG products and at various levels does have certain version requirements. These version requirements are described in detail in the PAC Installation documentation and PAC Release Notes that are issued with each version. Other Software AG products that may be required by PAC are:
Natural for Mainframes
Natural Security for Mainframe
Natural for UNIX / OpenVMS
Predict
Adabas
Entire System Server
Predict Case