Micro Focus Enterprise Developer 3.0 for Eclipse > Programming > Mainframe Programming > SQL Option for DB2 > How to... > Declaration Generator > Creating a .DGE File

Creating a .DGE File

Restriction: This topic applies to Windows environments only.

To run the Declaration Generator in batch mode, you must create a .DGE file specifying the tables for which you want declaration files generated, and the language(s) in which you want the declaration files written.

A .DGE file can be created in two ways:

You can enter your specifications in the Declaration Generator window and save them with the SAVE AS button. See Running Interactively.
You can create a .DGE file using any ASCII text editor.

The .DGE file must be stored in the DEFAULT APPLICATION PATH as specified in Options, PATHS tab, and it must have a .DGE extension.

A .DGE file contains the instructions needed to generate one or more declaration files. Each line in the file specifies a single "transaction," which creates a single declaration file in a single language (COBOL, C or PL/I). A transaction is specified as follows:

			tablename
			decFileName [option1
			option2
			option3...]

Where:

tablename specifies the name of the table for which a declaration file is generated. You can specify location and AuthID information using location.authid.tablename notation.

decFileName specifies the name of the file to which the declaration statements are written.

Do not include path information in decFileName. The Declaration Generator automatically writes the file to the application path (the application path is specified in the PATHS tab of Options).
Do not include an extension in decFileName. An extension is automatically assigned based on the language you've specified.

The option parameters specify the programming language, structure name and other parameters, if you don't want to use the defaults. See

DGE File Transaction Options

The following table describes the optional parameters you can include in a transaction statement. If an option is not included in a transaction statement, a default value is used.

Option	Description	Example
a=`Case`	Where `Case` determines whether C element names are written in upper or lower case. If `Case` is set to upper, element names are written in uppercase. If `Case` is set to lower, or if this option is not set, element names are written in lowercase.	`a=upper` generates uppercase elements.
c=`CompValue`	`CompValue` specifies the COMP type assigned to binary integers (for COBOL only; this option is ignored if language is set to C). `CompValue` must be one of the following: 0 - uses COMP 4 - uses COMP-4 5 - uses COMP-5 0 (COMP) is the default.	`c=4` assigns COMP-4 to binary integers.
d=`DStuc`	Where `DStruc` determines whether a C structure is written as a variable declaration. If `DStruc` is set to decl, the structure is declared as a variable. If `DStruc` is set to nodecl, or if this option is not set, the C structure is defined as a data type, but is not declared as a variable.	`d=decl` creates a variable declaration from the structure.
g=`PicType`	Where `Pictype` specifies the COBOL data type that will be used for graphic, vargraphic and long vargraphic fields. `Pictype` must be one of the following: PICN declares graphic fields as PIC N(x). PICG declares graphic fields as PIC G(x) DISPLAY-1 PICX declares graphic fields as PIC X(x) PICN is the default.	`g=PICG` declares graphic fields as follows: `PIC G(x) DISPLAY-1,`
k=`PageSize`	Where `PageSize` specifies the page size that will be assigned to COBOL declarations for long varchar, long vargraphic, and long varbinary fields. `PageSize` must be one of the following: 4K 32K 4K is the default.	`k=32K` assigns a 32k pages size to long variables.
l=`Language`	`Language` specifies the language in which the declaration will be written. `Language` must be one of the following: C COBOL COBOL is the default.	`l=C` generates a C header file. `l=COBOL` generates a COBOL copybook.
p=`Prefix`	`Prefix` specifies the prefix that will be affixed to each of the element names. Depending upon the state of the u=`Names` option, this prefix will either be concatenated with a field name or a number to form an element name. See u=`Names`, below.	`p=sales- u=real` combines the prefix sales- with the field names to form element names such as sales-dept, sales-mgr, and so on. `p=sales- u=n/a` combines the prefix sales- with a three-digit number to form element names such as sales-001, sales-002, and so on.
s=`Structure`	`Structure` specifies the name of the record definition (COBOL) or structure (C). This name must be valid for the chosen language. If `Structure` is not specified, the following defaults are used: For C: c_tablename For COBOL: dcltablename	`s=c_SalesReps` assigns the name c_SalesReps to the generated structure
u=`Names`	Where `Names` specifies whether field names or numbers are concatenated to `Prefix` to form element names (see p=`Prefix`, above). If `Names` is set to real, field names are concatenated to `Prefix`. If `Names` is set to n/a, or if this option is not set, numbers are concatenated to `Prefix`.	`p=sales- u=real` combines prefix sales- with field names to form element names such as sales-dept,sales-mgr, and so on. `p=sales- u=n/a` combines the prefix sales- with a three-digit number to form element names such as sales-001, sales-002, and so on.
v=`Level`	Where `Level` is an integer specifying the number to be assigned to the field-level definitions in a COBOL record description.	`v=3` writes level-3 field definitions, for example: 3 ST PIC X(2). 3 ZIP PIC X(5)