Version 9.5 SP1
Quick Reference

 —  Software AG IDL Editor  —

Grammar for IDL Template Files

An IDL template file contains the rules that the Software AG IDL Compiler uses - together with the IDL file - to generate interface objects, skeletons and wrappers for a programming language. The Developer's Kit provides several templates for various programming languages.

Warning:
The information in this section is intended for users who wish to write their own template files. Do not change the delivered template files.

This document explains the syntax of the template files in a formal notation. For an introduction on how to write template files, see Writing Template Files for Software AG IDL Compiler.


Software AG Template File Grammar

Syntax

{ statement }

Description

A template contains the rules which the IDL Compiler uses with the IDL file. The template is the lexical entity to start with.

Example

See under the lexical entity statement.

Top of page

assign_statement

Syntax

assign_string_statement | assign_integer_statement

Description

These statements are used

Example

See the lexical entities assign_string_statement or assign_integer_statement.

Top of page

assign_integer_statement

Syntax

%compute variable_of_type_integer string_with_expression_contents

Description

Compute the expression in string_with_expression_contents and assign the result to variable_of_type_integer.

Example

%compute a "&b * (%before + %after) / %eLength"

Top of page

assign_string_statement

Syntax

%assign variable_of_type_string string |
%assign variable_of_type_indexed_string string

Description

Assign the string contents to variable_of_type_string or variable_of_type_indexed_string

Example

%assign A "Assign this string to variable A" %assign A[5] "Assign this
        string to occurrence 5 of variable A"

Top of page

block

Syntax

'{' statement [ block ] '}'

Description

A block is a sequence of statements.

Example

See the lexical entity statement.

Top of page

compare_expression

Syntax

compare_strings [logical_compare_operator compare_strings]

Description

The logical_compare_operator performs a logical operation of two compare_strings. See the description of logical_compare_operator and compare_strings for specific information.

Example

%if "&i" > "3" && "&k" < "20"
{
"the variable i is greaterer than three AND the variable k is less than twenty"
}
%if "&i" = "1" || "&i" > "3"
{
"the variable i is one OR is greater than three"
}

Top of page

compare_strings

Syntax

string [ compare_operator ] string

Description

Compare two strings for a logical condition. The condition can be TRUE or FALSE.

Example

See lexical entity compare_operator.

Top of page

compare_operator

Syntax

Operator Meaning
  equal to(default)
= equal to
<> not equal to
< less than
<= less than or equal to
> greater than
>= greater than or equal to

Description

The logical operators are used in the lexical entity compare_strings.

Example

equal to (default) "A" "?A"
equal to "A" = "?A"
not equal to "A" <> "?A"
less than "A" < "?A"
less than or equal to "A" <= "?A"
greater than "A" > "?A"
greater than or equal to "A" >= "?A"

Top of page

control_statement

Syntax

assign_statement | definition-statement | file_handling_statement |  if_statement  | loop_statement |
output_control_statement

Description

control_statements determine the processing logic of a template. They do not create output.

Example

See lexical entities:

Top of page

definition-statement

Syntax

definition-of-base-type-template | 
definition-of-direction-template | 
definition-of-group-template | 
definition-of-index-template | 
definition-of-line-number-format-template | 
definition-of-member-separator-template | 
definition-of-names-format-template | 
definition-of-nest-level-format-template | 
definition-of-parent-identifier-template | 
definition-of-parent-index-template | 
definition-of-structure-template | 
definition-of-UnboundedArray-template.

Description

definition_statements give directives to the IDL Compiler. They do not create output.

Example

See lexical entities:

Top of page

definition-of-base-type-template

Syntax

%using  definition-of-base-type output-statement

Description

All references to the output_substitution_sequence %type in a loop_over_parameters are in the output as specified in output_statement. Default output-statement is "" (empty).

Example

A %using A "unsigned char %name%index[%eLength]"
AV %using AV "ERX_HVDATA %name%index"
B %using B "unsigned char %name%index[%eLength]"
BV %using BV "ERX_HVDATA %name%index"
D %using D "unsigned char %name%index[ERX_GET_PACKED_LEN(7)]"
F4 %using F4 "float %name%index"
F8 %using F8 "double %name%index"
I1 %using I1 "signed char %name%index"
I2 %using I2 "short %name%index;"
I4 %using I4 "long %name%index;"
K %using K "unsigned char %name%index[%eLength]"
KV %usingKV "ERX_HVDATA %name%index"
L %using L "unsigned char %name%index"
N %using N "unsigned char %name%index[%before+%after]"
NU %using NU "unsigned char %name%index[%before+%after]"
P %using P "unsigned char" %name%index[ERX_GET_PACKED_LEN(%before+%after)]"
PU %using PU "unsigned char %name%index[ERX_GET_PACKED_LEN(%before+%after)]"
T %using T "unsigned char
%name%index[ERX_GET_PACKED_LEN(13)]"

Top of page

definition-of-base-type

Syntax

A | AV | B | BV | D | F4 | F8 | I1 | I2 | I4 | K | KV | L | N | NU | P | PU | T

Description

For a description of the definition-of-base-type see IDL Data Types.

A Reference to IDL data type Alphanumeric.
AV Reference to IDL data type Alphanumeric variable length
B Reference to IDL data type Binary.
BV Reference to IDL data type Binary variable length.
D Reference to IDL data type Date.
F4 Reference to IDL data type Floating point (small).
F8 Reference to IDL data type Floating point (large).
I1 Reference to IDL data type Integer (small).
I2 Reference to IDL data type Integer (medium).
I4 Reference to IDL data type Integer (large).
K Reference to IDL data type Kanji.
KV Reference to IDL data type Kanji variable length.
L Reference to IDL data type Logical.
N Reference to IDL data type Unpacked decimal.
NU Reference to IDL data type Unpacked decimal unsigned.
P Reference to IDL data type Packed decimal.
PU Reference to IDL data type Packed decimal unsigned.
T Reference to IDL data type Time.

Top of page

definition-of-direction-template

Syntax

%using  %direction output-statement output-statement output-statement

Description

All references to the output_substitution_sequence %direction in a loop_over_parameters are in the output as specified in output-statement. If a parameter is of direction IN, the first output-statement will be used, the second output-statement will be used for direction OUT and the third for INOUT (see attribute-list). Default output-statement is "" "" "" (empty).

Example

%using %direction "In" "Out" "In Out"

Top of page

definition-of-group-template

Syntax

%using G output-statement output-statement

Description

If the parameter is a group (see group-parameter-definition) in a loop_over_parameters, all references to the output_substitution_sequence %type will be written into the output as specified in the output-statements. The first output-statement is the group prefix, typically the data type of the target programming language. The second output-statement is the group suffix which usually indicates the end of the group. Default output-statement is "" (empty).

Example

%using G "struct {" "%outBlank} %name;"

Top of page

definition-of-index-template

Syntax

%using  %index output-statement output-statement output-statement output-statement

Description

All references to the output_substitution_sequence %index in a loop_over_parameters will be written into the output as specified in the output-statements. According to the IDL (see array-definition), up to 3 dimensions are supported. The first output-statement is for scalar parameters, the second output-statement for 1-dimensional arrays, the third output-statement for 2-dimensional arrays and the fourth output-statement for 3-dimensional arrays. Default output-statement is "" (empty).

Example

%using %index "" "[%1_index]" "[%2_index][%1_index]"
        "[%3_index][%2_index][%1_index]"

Top of page

definition-of-line-number-format-template

Syntax

%using %NumberLine output-statement

Description

All references to the output_substitution_sequence %LibCount, %ProgCount and %NameCount are written into the output as specified in the output-statements. The output-statement uses the C printf format notation. Default output-statement is %u.

Example

%using %NumberLine "%.4u"

Top of page

definition-of-member-separator-template

Syntax

%using %member output-statement

Description

Specify a template for a fully qualified name of parameters. All references to the output_substitution_sequence %member in a loop_over_parameters are written into the output as specified in the output-statements. The IDL Compiler builds an internal tree hierarchy of parameters, structures and groups. If a parameter has a parent, it will be inserted before the fully qualified name. If a parameter has no parent, the output-statement will not be used. Default output-statement is "" (empty).

Example

%using %member "%name%Index."

Top of page

definition-of-names-format-template

Syntax

%using  %Format output-statement

Description

Specify a template for library, program or parameter name strings. All references to the output_substitution_sequence %OutputLevel in a loop_over_parameters are written into the output as specified in the output-statements. The output-statement uses the C printf format notation. Default output-statement is %s.

Example

%using %Format "%s.ext"

Top of page

definition-of-OutBlank-template

Syntax

%using %outBlank output-statement

Description

The output-statement definition replaces the blank (default), which will be used with the %outBlank statement. The statement will not be interpreted and will be used as it is. You cannot write an expression (string_with_expression_contents) or a variable (variable_of_type_string string) in this output statement. See the table entry definition-of-OutBlank-template in the section output_substitution_sequence for futher information.

Example

%using %outBlank "\t"

Top of page

definition-of-nest-level-format-template

Syntax

%using  %OutputLevel output-statement

Description

Specify a template for nesting level strings. The nesting level is the depth where a parameter is specified. All references to the output_substitution_sequence %OutputLevel in a loop_over_parameters are written into the output as specified in the output-statements. The output-statement uses the C printf format notation. Default output-statement is %using.

Example

%using %OutputLevel "%u"

Top of page

definition-of-parent-identifier-template

Syntax

%using %Xparent output-statement output-statement

Description

Specify a template for the parent. All references to the output_substitution_sequence %Xparent in a loop_over_parameters are written into the output as specified in the output-statements. If a parameter has no parent and the second output-statement is not empty, the second output-statement is used, otherwise, the first. The first output-statement uses the C printf format notation. Default is "%u" "" (second statement is empty).

Example

%using %Xparent "%d" "ERX_NO_PARENT_V2"

Top of page

definition-of-parent-index-template

Syntax

%using %Index output-statement output-statement output-statement output-statement

Description

Specify a template for a parameter's parent index. The IDL Compiler builds an internal tree hierarchy of parameters, structures and groups. A parameter's immediate parent in this hierarchy can be an array. All references to the output_substitution_sequence %Index in a loop_over_parameters are written into the output source code as specified in the output-statements. The output-statement describes the syntax for defining arrays of up to 3 dimensions as defined by the IDL (see array-definition). The Output_substitution_sequence %Index (uppercase I) is very similar to the output_substitution_sequence %index (defined with definition-of-index-template), but is useful only for building member names using output_substitution_sequence %member. Default output-statement is "" (empty).

Example

%using %Index "" "[0]" "[0][0]" "[0][0][0]"
%using %member "%Index."

Top of page

definition-of-structure-template

Syntax

%using S output-statement

Description

If the parameter is a structure (see structure-parameter-definition (IDL)) in a loop_over_parameters, all references to the output_substitution_sequence %type will be written into the output as specified in the output-statements. "INCLUDE AS GROUP" specified as the output-statement will embed all parameters (see parameter-data-definition) of the structure as if they were a group.

Example

%using S "INCLUDE AS GROUP"

Top of page

definition-of-UnboundedArray-template

Syntax

%using UnboundedArray output-statement
%using UnboundedArray ""

Description

If the parameter is an unbounded array (see array-definition) in a loop_over_parameters, all references to the output_substitution_sequence %type will be written into the output as specified in the output-statements statement, i.e. the settings of the definition-of-base-type are overwritten. The first form overwrites the settings of the definition-of-base-type. The second form switches back to the definition-of-base-type settings.

Example

%using UnboundedArray "ERX_HARRAY"

Top of page

error_statement

Syntax

%error output_character_sequence

Description

Use the error_statement to exit your template with an error message. The execution of the template will be stopped at this stetement and the error message will be given to the caller. The %error statement can be used in the main template and in subtemplates as well. The execution of the whole template compiling process will be stopped regardless of the type of template it is used in.

Example

%if "$(TARGET)" <> "COBOL" && "$(TARGET)" <> "BATCH"
{
   %error "TARGET not supported."
}

Top of page

execute_statement

Syntax

%execute output-statement [(parameter_list)] [return (return_list)]

Description

Use the %execute statement to include another template file in the current template file like a subprogram. You can outsource often-used code to an external template file. The executed template file uses a NEW CLEAN environment context. Only the %library, %program, %x_struct, %name, %LibCount, %ProgCount and %NameCount are known in the executed template file. No other variables or idl parameters (output_substitution_sequence) are defined in the executed template file. You must for example have a %name loop (loop_over_parameters) to have access to an idl parameter (output_substitution_sequence). All changes in the executed files will be lost after calling the include file and will have no effect in the calling template.

The %execute statement can be called with a list of parameters (parameter_list). The parameter_list needs to be set in brackets. This list of parameters will be copied into the variables in the executed file in the following order: ?A, ?B, ?C, ..., for example a list of parameters in the %execute statement ("AA" "BB" "CC" "DD"). These parameters will be available in the following variables: "?A" => "AA", "?B" => "BB", "?C" => "CC", "?D" => "DD".

The %execute statement can have a "return" statement to return a list of parameters (return_list), as well. This return_list has to be returned with the return_statement. The count of the expected and the returned parameter must be the same and the type of the parameters must match. For example, if the expected return parameter is a variable_of_type_string, the return parameter type must be a variable_of_type_string and if the expected return parameter is a variable_of_type_integer, the return parameter type must be a variable_of_type_integer.

Example 1

%execute "subprog.tpl" ("?A" "&i" "10")

%execute "subprog.tpl" ("?Z" "%OutputLevel" "subprog" "10") return ("?C" "&f")

Example 2

File main.tpl
%assign A "Test variable A"
%assign B "Test variable B"
%assign C "Length of A and B is"
%compute i "#A"  ; length of the variable A
%compute j "#B"  ; length of the variable B
%execute "calc.tpl" ("?C" "&i" "&j") return ("?Z")
"?Z\n" ; print the returned variable
; the ouput should be "Length of A and B is 30"
File calc.tpl
; the execute parameters are in the following variables
; ?A => the ?C of the main template = "Length of A and B is"
; ?B => the &i of the main template = "15"
; ?C => the &j of the main template = "15"
%compute a "?B + ?C"
%assign T "?A &a"
%return ("?T")

Top of page

file_handling_statement

Syntax

%file [output-statement]

Description

Use the %file statement to direct the output to a specific file. Only one file can be open at a time. If no file is open, all output goes to STDOUT by default. If output-statement is left blank, the file currently open is closed. All open files are implicitly closed if either a new open file statement is encountered or the IDL Compiler terminates.

Example

%file "%library.MAK" ; open a file
....
%file "" ; close the file

Top of page

if_statement

Syntax

%if compare_expression statement [if-elif-extension] [%else statement]

Description

If compare_expression is TRUE, then interpret statement. If compare_expression is FALSE and if there is an if-elif-extension, interpret the if-elif-extension.

If all compare_expressions in all if-elif-extensions are FALSE and if there is an else block, interpret the statement in the else block.

Example

%if "%type%index" ""
     "\n"
%elif "%index" ""
     "(%type)\n"
%else 
     "(%type%index)\n"

Top of page

if_elif_extension

Syntax

%elif compare_expression statement [if-elif-extension]

Description

If compare_expression is TRUE, then interpret the statement. If compare_expression is FALSE and if there is an if-elif-extension, interpret the if-elif-extension.

Example

See lexical entity if_statement.

Top of page

logical_compare_operator

Syntax

&& logical-AND operator
|| logical-OR operator

Description

The logical operators perform logical-AND (&&) and logical-OR ( || ) operations. The logical-AND operator has a higher piority than the logical-OR operator.

Example

See the lexical entity compare_expression.

Top of page

loop_statement

Syntax

loop_over_libraries | loop_over_parameters | loop_over_programs | loop_over_structures | loop_of_while

Description

Loop sequences instruct the IDL Compiler to loop through all occurrences of libraries (see library-definition), programs (see program-definition), structures (see structure-definition) and parameters (see parameter-data-definition).

Example

See lexical entities:

Top of page

loop_over_libraries

Syntax

%library statement

Description

This loop statement instructs the IDL Compiler to loop through all occurrences of libraries (see library-definition).

Example

%library
{
   ....
}

Top of page

loop_over_parameters

Syntax

%name statement

Description

This loop statement instructs the IDL Compiler to loop through all occurrences of parameters (see parameter-data-definition). A loop over parameters must be placed in a loop_over_programs or loop_over_structures.

Example

%name 
{
   ....
}

Top of page

loop_over_programs

Syntax

%program statement

Description

This loop statement instructs the IDL Compiler to loop through all occurrences of programs (see program-definition). A loop over programs must be placed in a loop_over_libraries.

Example

%program
{
   ....
}

Top of page

loop_over_structures

Syntax

%x_struct statement

Description

This loop statement instructs the IDL Compiler to loop through all occurrences of structures (see structure-definition). A loop over structures must be placed in a loop_over_libraries or loop_over_programs. In a loop_over_libraries all structures in the library (see library-definition) are accessed. In a loop_over_programs all structures in the current program (see program-definition) are accessed.

Example

%structure
{
   ....
}

Top of page

loop_of_while

Syntax

%while compare_expression statement

Description

Loop while compare_strings is TRUE.

Example

%compute i "0"
%while "&i" < "10"
{
   ....
  %compute i "&i + 1"
}

Top of page

message_statement

Syntax

%message output_character_sequence

Description

Use the message_statement to notify the template user with a message. The output_character_sequence will report as the message on the console.

Example

%if "%eLength" > "32766"
{
  %message "Maximum length for %type usually 32766"
}

Top of page

output

Syntax

output_character_sequence | output_escape_sequence  | output_formatting_sequence|  output_of_variable |
output_substitution_sequence

Description

See lexical entities:

Example

See lexical entities:

Top of page

output_character_sequence

Description

Simple sequences of characters not matching other output such as output_escape_sequence, output_formatting_sequence, output_of_variable or output_substitution_sequence form a character sequence.

Example

"This is a character sequences in an output statement."

Top of page

output_control_ims

Syntax

%IMS+ | %IMS- | %IMS

Description

The IMS flag. If this flag is set, the parameter in the loop_over_parameters, that are marked with the IMS attribute in the IDL file will also be taken into consideration. If this flag is off, all parameters that are marked with the IMS attribute will be ignored. If + or - are not specified, the flag will be toggled. The default is off.

Example

%IMS+ 
%library
{
   %program
   {
      %name
      {
         "%name"
      }
   }
}

Top of page

output_control_imsonly

Syntax

%IMSONLY+ | % IMSONLY - | % IMSONLY

Description

The IMSONLY flag. If this flag is set, parameters in the loop_over_parameters that are not marked with the IMS attribute in the IDL file will be ignored. If this flag is off, the loop_over_parameters will work as usual. If + or - are not specified, the flag will be toggled. The default is off.

Example

%IMSONLY+ 
%library
{
   %program
   {
      %name
      {
         "%name - this parameter has an ims attribute"
      }
   }
}

Top of page

output_control_lower_upper

Syntax

(Defaults are underlined.)

%UpperCase+    | %UpperCase-    | %UpperCase
%UpperCasePgm+ | %UpperCasePgm- | %UpperCasePgm
%LowerCase+    | %LowerCase-    | %LowerCase

Description

UpperCase Name Uppercase flag. If this flag is set, %name substitution_sequences written to the output will be converted to uppercase. If no + or - is specified, the flag will be toggled. The default is off.
UpperCasePgm Program Uppercase flag. If this flag is set, %program substitution_sequences written to the output will be converted to uppercase. If no + or - is specified, the flag will be toggled. The default is off.
LowerCase Name Lowercase flag. If this flag is set, %name substitution_sequences written to the output will be converted to lowercase. If no + or - is specified, the flag will be toggled. The default is on.

Example

UpperCase

%UpperCase+ ; set name uppercase flag on
%UpperCase- ; set name uppercase flag off
%UpperCase ; toggle name uppercase flag

UpperCasePgm

%UpperCasePgm+ ; set program uppercase flag on
%UpperCase- ; set program uppercase flag off
%UpperCase ; toggle program uppercase flag

LowerCase

%LowerCase+ ; set name lowercase flag on
%LowerCase- ; set name lowercase flag off
%LowerCase ; toggle name lowercase flag

Top of page

output_control_sanitize

Syntax

(Defaults are underlined.)

                     %Sanitize+ | %Sanitize- | %Sanitize %SanitizeCamelCased+ | %SanitizeCamelCased- | 
%SanitizeCamelCased %SanitizeCobol+ | %SanitizeCobol- | %SanitizeCobol %SanitizePascalCased+ | 
%SanitizePascalCased- | %SanitizePascalCased

Description

Sanitize Sanitize flag for C programming language. If this flag is set, %x_struct, %u_struct, %name, %program and %library substitution sequences written to the output will be forced to follow C conventions. The special characters '#', '$', '&', '+', '-', '.', '/' and '@' in parameter names permitted in the IDL file will be converted to underscores '_' to produce valid C names. If + or - are not specified, the flag will be toggled. The default is on.
SanitizeCamelCased Sanitize flag for C# programming language. If this flag is set, %x_struct, %u_struct,%name, %program and %library substitution sequences written to the output will be forced to follow camel cased naming conventions as they are used in C#. The special characters '#', '$', '&', '+', '-', '.', '/', '@' and '_' in parameter names permitted in the IDL file will be removed. The character following the special character will be converted to uppercase and all other characters to lowercase. The very first character within %name, %program, and %library substitution sequences will be converted to lowercase. %UpperCase- and %LowerCase- must be set also to have CamelCased names. If + or - are not specified, the flag will be toggled. The default is off.
SanitizeCobol Sanitize flag for COBOL programming language. If this flag is set, x_struct, %u_struct, %name, %program and %library substitution sequences written to the output will be forced to follow COBOL conventions. The special characters '#', '$', '&', '+', '.', '/', '@' and '_' in parameter names permitted in the IDL file will be converted to hyphen '-' to produce valid COBOL names. . If a parameter name starts with a digit, e.g. '1', it is prefixed with the character 'P'. If + or - are not specified, the flag will be toggled. The default is off.
SanitizeDCOMWrapper Sanitize flag for DCOM Wrapper. If this flag is set, %x_struct, %u_struct, %name, %program and %library substitution sequences written to the output will be forced to follow DCOM conventions. The special characters '#', '$', '&', '+', '-', '.', '/' and '@' in parameter names permitted in the IDL file will be converted to underscores '_' to produce valid DCOM names. All preceding underscores in parameter names are deleted. If a parameter name starts with a digit, e.g. '1', it is prefixed with the character 'P'. If + or - are not specified, the flag will be toggled. The default is on.
SanitizePascalCased Sanitize flag for C# programming language. If this flag is set, %x_struct, %u_struct,%name, %program and %library substitution sequences written to the output will be forced to follow Pascal-cased naming conventions as they are used in C#. The special characters '#', '$', '&', '+', '-', '.', '/', '@' and '_' in parameter names permitted in the IDL file will be removed. The character following the special character will be converted to uppercase and all other characters to lowercase. The very first character in the %name, %program, and %library substitution sequences will be converted to uppercase. %UpperCase- and %LowerCase- must be set also to have PascalCased names. If + or - are not specified, the flag will be toggled. The default is off.
SanitizePLI Sanitize flag for the PL/I programming language. If this flag is set, %x_struct, %u_struct, %name, %program and %library substitution sequences written to the output will be forced to follow PL/I conventions. The special character '&', '+', '-', '.' and '/' in parameter names permitted in the IDL file will be converted to underscores '_' to produce valid PL/I names. If + or - are not specified, the flag will be toggled. The default is off.

Example

Sanitize
%Sanitize+ set Sanitize flag for C variable names on
%Sanitize- set Sanitize flag for C variable names off
%Sanitize toggle Sanitize flag for C variable names
SanitizeCamelCased
%SanitizeCamelCased+ set Sanitize flag for C# parameter names on
%SanitizeCamelCased- set Sanitize flag for C# parameter names off
%SanitizeCamelCased toggle Sanitize flag for C# parameter names
SanitizeCobol
%SanitizeCobol+ set Sanitize flag for COBOL variable names on
%SanitizeCobol- set Sanitize flag for COBOL variable names off
%SanitizeCobol toggle Sanitize flag for COBOL variable names
SanitizeDCOMWrapper
%SanitizeDCOMWrapper+ set Sanitize flag for DCOM names on
%SanitizeDCOMWrapper- set Sanitize flag for DCOM names off
%SanitizeDCOMWrapper toggle Sanitize flag for DCOM names
SanitizePascalCased
%SanitizePascalCased+ set Sanitize flag for C# member names on
%SanitizePascalCased- set Sanitize flag for C# member names off
%SanitizePascalCased toggle Sanitize flag for C# member names
SanitizePLI
%SanitizePLI+ set Sanitize flag for PL/I member names on
%SanitizePLI- set Sanitize flag for PL/I member names off
%SanitizePLI toggle Sanitize flag for PL/I member names

Top of page

output_control_statement

Syntax

output_control_lower_upper | output_control_sanitize |
        output_control_verbose

Description

Use the flags to force upper/lowercase conversions, C, C# or COBOL language conventions or, for example, for comments to be written into the output. The default setting when you do not code any output_control_statements are forced to follow the C programming language convention.

Example

See under the lexical entities:

Top of page

output_control_verbose

Syntax

%verbose+ | %verbose- | %verbose

Description

Verbose flag. If this flag is set, template file output-statements will be written to the output without being interpreted, e.g. the substitution_sequences are output as is and not replaced by their meaning. If + or - are not specified, the flag will be toggled. The default is off.

Example

%verbose+ 
"/* This is file %program.c */"
"/* Please do not modify this file */"
%verbose

Top of page

output_escape_sequence

Syntax

\\

Description

The escape character is used to change the meaning of the special characters &, ? and # back to their normal meaning. Special characters access variables using the output_of_variable lexical entity. With escape characters it is possible to insert a plain & by typing : \\&.

Example

"This string contains an ampersand \\&."

Top of page

output_formatting_sequence

Syntax

Sequence Meaning
\n Newline
\r Carriage return
\t Horizontal tab
\ddd ASCII character in octal notation, e.g. \012 for the new line
\xdd ASCII character in hex notation, e.g. \x09 for the horizontal tab

Description

Formatting sequences are output control characters such as newline, backspace, etc. For characters in hexadecimal notation, the IDL Compiler ignores all leading zeros. It establishes the end of the hex-specified escape character when it encounters either the first non-hex character or more than two hex characters not including leading zeros. In the latter case, it reports an error and ignores all characters beyond the second one.

Example

"This string ends on this line.\n"

Top of page

output_of_variable

Syntax

??variable_of_type_indexed_string | ## variable_of_type_indexed_string
        | &variable_of_type_integer | ?variable_of_type_string | #variable_of_type_string | $(...)

Description

This form of accessing variables can be used in output_statements. If it is used, the variable content is written to the output.

Example

&a substitutes the integer variable with its number.
?A substitutes the string variable with its contents.
??A[0] substitutes the indexed string variable with its contents.
#A substitutes the string variable with an integer of the length of its contents.
##B[%OutputLevel] substitutes the indexed string variable with an integer of the length of its contents.
$(TEMP) substitutes the option variable with its contents.

Top of page

output_statement

Syntax

'"' { output } '"'

Description

output is a string consisting of an output_character_sequence, output_escape_sequence, output_formatting_sequence, output_of_variable or an output_substitution_sequence in any order.

Example

"This is simple output."

Top of page

output_substitution_sequence

Syntax

Sequence Meaning
%after Inserts the digits after the decimal point of the current parameter (see simple-parameter-definition) into the output.

This substitution sequence can only be used in an active loop_over_parameters, and if the current parameter is of data type N, NU, P, PU. Using it with other data types will lead to an error

%Alias If a library alias name is used, inserts the library alias name of the current library (see library-definition) into the output. If no alias is provided, the library name (contents of %library) is provided in the %Alias substitution sequence. This substitution sequence can only be used in an active loop_over_libraries.
%before Inserts the digits before decimal point of the current parameter (see simple-parameter-definition) into the output.

This substitution sequence can be used only in an active loop_over_parameters, and if the current parameter is of data type N, NU, P, PU. Using it with other data types will lead to an error.

%Count Inserts the output's current line number. The format is controlled by the definition-of-names-format-template.
%direction Inserts the direction of the current parameter (see parameter-data-definition) into the output as specified with definition-of-direction-template. This substitution sequence can only be used in an active loop_over_parameters.
%eLength
Inserts the logical length of the current parameter (see simple-parameter-definition) into the output. This substitution sequence can only be used in active loop_over_parameters. The logical length according to the Software AG IDL data type is as follows:
for type A,K,B,I,F,U logical length as set in the IDL file
for type AV,KV,BV,UV maximum logical length as set in the IDL file
for type L Type L has no explicit logical length in the Software AG IDL file. However, the length is always set to 1.
for type N,NU,P,PU Digits before and after decimal point in encoded form as given in the Software AG IDL file. Use the macro ERX_GET_DIGITS (see erx.h) to get the digits before the decimal point. Use the macro ERX_GET_DECIMALS (see erx.h) to get the number of digits after the decimal point.
for type T Type T has no explicit logical length in the Software AG IDL file. However, the length is always set to 12.
for type D Type D has no explicit logical length in the Software AG IDL file. However, the length is always set to 6.
%file Inserts the current filename into the output.
%Format Inserts into the output the base name as given by the -F parameter on start of the The Software AG IDL Compiler. If no -F parameter is provided during start of the IDL Compiler, the base name of the idl-file without path and extension is inserted into the output.
%index Inserts the index of the current parameter (see parameter-data-definition) into the output as specified with definition-of-parent-index-template. This substitution sequence can only be used in an active loop_over_parameters.
%LibCount Inserts the total number of libraries (see library-definition) as given in the IDL file into the output. The format is controlled by the definition-of-line-number-format-template.
%library Inserts the current library name (see library-definition) into the output. The inserted library name is controlled by output_control_statement. This substitution sequence can only be used in an active loop_over_libraries.
%member Inserts the fully qualified member name: (parent...parent...child) into the output. The member name is controlled by definition-of-member-separator-template. This substitution sequence can only be used in an active loop_over_parameters.
%Method If a program alias is used, inserts the program alias name of the current program (see program-definition) into the output. If no alias is provided, the program name (contents of %program) is provided. This substitution sequence can only be used in an active loop_over_programs.
%name Inserts the current parameter name (see parameter-data-definition) into the output. The parameter name is controlled by output_control_statement. This substitution sequence can only be used in an active loop_over_parameters.
%NameCount Inserts the total number of parameters (see parameter-data-definition) of the current program (see program-definition) into the output. The format is controlled by the definition-of-line-number-format-template. This substitution sequence can only be used in an active loop_over_programs.
%outBlank Writes n blanks (n = nesting level of parameter) into the output according to the level of the current parameter (see parameter-data-definition). This substitution sequence can only be used in an active loop_over_parameters. You can replace the blank with another output statement by using (%using %outBlank output-statement).
%OutputLevel Inserts the level of the parameter (see parameter-data-definition) into the output. This substitution sequence can only be used in an active loop_over_parameters.
%ProgCount Inserts the total number of programs (see program-definition) in the current library into the output. The format is controlled by the definition-of-line-number-format-template. This substitution sequence can only be used in an active loop_over_libraries.
%program Inserts the current program (see program-definition) name into the output. The output is controlled by output_control_statement. This substitution sequence can only be used in an active loop_over_programs.
%type Inserts the parameter type (see parameter-data-definition) into the output. The parameter type is controlled by definition-of-base-type-template, definition-of-group-template, definition-of-UnboundedArray-template and definition-of-structure-template. This substitution sequence can only be used in an active loop_over_parameters.
%size (deprecated, should no longer be used!) The %size substitution sequence is deprecated and should no longer be used. Use %eLength substitution sequence instead. Using %size will lead to an error.
%TypeAttributes The substitution sequence %TypeAttributes produces a 2-byte bitmask indicating

The bitmask displays the following:

  • If an array and the 1st dimension is unbounded then bit 0 is ON rightmost.

  • If an array and the 2nd dimension is unbounded then bit 1 is ON.

  • If an array and the 3rd dimension is unbounded then bit 2 is ON.

  • If the aligned attribute is set then bit 3 is ON.

For users of the EntireX RPC C Runtime the bitmask corresponds directly to the ERXeAttributes defined for the ERX_PARAMETER_DEFINITION_V3. (see erx.h) This substitution sequence can only be used in an active loop_over_parameters.

%u_struct Inserts the name of the referenced structure (see structure-definition) into the output. The inserted name is controlled by output_control_statement. This substitution sequence can only be used in an active loop_over_parameters and only when the current parameter uses a structure as its type definition (see structure-parameter-definition (IDL)).
%x_struct Inserts the name of the structure (see structure-definition) into the output. The inserted name is controlled by output_control_statement. This substitution sequence can only be used in an active loop_over_parameters.
%Xparent Inserts the parameter's parent into the output. This substitution sequence can only be used in an active loop_over_parameters.
%0_index Inserts the number of indices of the current parameter (see parameter-data-definition) into the output. This substitution sequence can only be used in an active loop_over_parameters.
%1_index Inserts the count of elements in dimension 1 of the current parameter (see parameter-data-definition) into the output. The substitution sequence should be used for 1-dimensional parameters only (this can be checked with the %0_index substituion sequence). This substitution sequence can only be used in an active loop_over_parameters.
%2_index Inserts the count of elements in dimension 2 of the current parameter (see parameter-data-definition) into the output. The substitution sequence should be used for 2-dimensional parameters only (this can be checked with the %0_index substituion sequence).This substitution sequence can only be used in an active loop_over_parameters.
%3_index Inserts the count of elements in dimension 3 of the current parameter (see parameter-data-definition) into the output. The substitution sequence should be used for 3-dimensional parameters only (this can be checked with the %0_index substituion sequence). This substitution sequence can only be used in an active loop_over_parameters.
%SameLineComment Inserts the text of the comment line of the current parameter from the IDL file. Use the parameter properties of the IDL Editor to set this comment line in the IDL file.
%SVMMetaData Inserts the metadata part contained in a related client-side server mapping file (see CVM File) of the current IDL program into the output. This substitution sequence can only be used in an active loop_over_programs. If there is no related CVM file, an empty string is inserted
%SVMFormatArea Inserts the format area contained in a related CVM file of the current IDL program into the output. This substitution sequence can only be used in an active loop_over_programs. If there is no related CVM file, an empty string is inserted
%SVMValueArea Inserts the value area contained in a related CVM file of the current IDL program into the output. This substitution sequence can only be used in an active loop_over_programs. If there is no related CVM file, an empty string is inserted
%SVMStringArea Inserts the string area contained in a related CVM file of the current IDL program into the output. This substitution sequence can only be used in an active loop_over_programs. If there is no related CVM file, an empty string is inserted
%SVMRpcProtocol Inserts the RPC protocol version contained in a related CVM file of the current IDL program into the output. This substitution sequence can only be used in an active loop_over_programs. If there is no related CVM file, an empty string is inserted

Description

Substitution sequences are substituted by their actual contents during generation.

Example

"This is a substitution sequence containing the library: %library."

Top of page

parameter_list

Syntax

(parameter_list)

Description

The parameter_list is an unnumbered count of parameters. This list of parameters needs to be set in brackets. Each parameter needs to be set in quotation marks and will be interpreted before using. Parameters will be separated by blanks. The parameter_list can be defined as an empty list, in which case the return_list needs to be defined only with the brackets "()".

Example

("param" "10" "?A" "&i" "%0_index" "%OutputLevel")

Top of page

return_list

Syntax

(return_list)

Description

The return_list is an unnumbered count of parameters. This list of return parameters needs to be set in brackets. Each return parameter needs to be set in quotation marks. Parameters will be separated by blanks. Only variable_of_type_string and variable_of_type_integer are allowed in this list. variable_of_type_indexed_string is not allowed. It is also not allowed to use any constant string. The return_list can be defined as an empty list, in which case the return_list needs to be defined only with the brackets "()".

Example

("?A" "?Z" "&i" "&n")

Top of page

return_statement

Syntax

%return (parameter_list)

Description

The return_statement will be used to return "return parameters" from an executed subtemplate. No statements after the return_statement will be executed. The subtemplate will return to the main template with this statement. If the return_statement has been placed in the main template, the execution of the template will be stopped as normal at this point.

Example

%return ("param" "10" "?A" "&i" "%0_index" "%OutputLevel")

Top of page

statement

Syntax

block | control_statement | output_statement

Description

These are the 3 basic types of statements used in a template. A block is a sequence of statements. Output_statements create the output. Control_statements determine the processing logic.

Example

See under the lexical entities control_statement and output_statement.

Top of page

string

Syntax

'"' { output } '"'

Description

Any kind of output can be used to form a string. A string is used to form the condition criteria in a compare_strings lexical entity used e.g. in if_statement and loop_of_while. A string is not written to the output.

Example

"String with contents of variable $A"

Top of page

string_with_expression_contents

Syntax

'"' { output } '"'

Description

Any kind of output can be used to form String_with_expression_contents. However, this kind of string must adhere to the rules of an expression. A string_with_expression_rules is not written to the output.

Supported mathematical operations are:
+ addition
- subtraction
* multiplication
/ division
mod modulo operation; computes the remainder after dividing its first operand by its second

Supported bit operations are:
and bitwise AND operation
or bitwise OR operation
xor bitwise XOR operation

Precedence of operators:
*, /, %
+, -
xor, and, or

You may control the precedence of the operation with brackets.

Example

%compute a "%OutputLevel + 1"
%compute b "%OutputLevel * 10"
%compute c "%TypeAttribute mod 3"
%compute d "(%TypeAttribute and 7) * 10"
%compute e "%TypeAttribute or 1"
%compute f "%TypeAttribute xor 3"

Top of page

substring_statement

Syntax

%substring  variable_of_type_string string from_position length |
%substring  variable_of_type_indexed_string string from_position length |

Description

Extract from the source variable string the substring from_position up to the length to variable_of_type_string or variable_of_type_indexed_string.

The parameters from_position and length are of variable_of_type_integer. It is not possible to use a string_with_expression_contents for from_positon and length. For length, the constant "all" or "ALL" can be used to extract the rest of the source string starting from from_position. The first position of the source string is 0.

If length is longer as the length of the substring to extract, the available substring from from_position to the end of the string will be assigned (same as using the constant "all" for length). If the from_position is been higher as the length of the string, an empty string will be assigned. If the value of from_position or length is lower than 0, an error will occur.

Example

%compute f "0"
%compute l "32 + 10"
%substring A "These are all characters before position 42 and these are all characters after position 42" "&f" "&l"
%compute f "&f + &l"
%compute l "100"
%substring A[1] "These are all characters before position 42 and these are all characters after position 42" "&f" "&l"

After execution, the variable A contains the string "These are all characters before position 42" and variable A[1] contains the string " and this are all characters after position 42".

Top of page

UnsupportedProgram_statement

Syntax

%UnsupportedProgram output_character_sequence

Description

Use the UnsupportedProgram_statement to notify the IDL Compiler that the current program in the current library is not supported and needs to be ignored in further processing. The template writer can inform the template user of the reason why this program is not supported with the output_character_sequence. Usually the template writer will mark a program as unsupported if the program contains unsupported data type in the IDL definition for the target programming language. The output_character_sequence will report as a message on the console.

This statement must be embedded in loop_over_libraries and loop_over_programs. Furthermore it cannot be used if a file was already open using file_handling_statement.

Example

%using K  "K"
%library
{
   %program
   {
      %name
      {
        %if "%type" = "K"
         {
           %compute z "%eLength /2"
           %compute z "&z *2"
           %if "&z" <> "%eLength"
           {
             %UnsupportedProgram "Length for %type fields must be even."
           }
         }
      }
   }
}

Top of page

variable_index

Syntax

string_with_expression_contents

Description

Any variable_index follows the rules of a string_with_expression_rules. Additionally, the result of the expression is restricted to the range 0 - 8, i.e. indices must be in the range of 0 - 8. The variable_index is not written to the output.

Example

"2"
"%OutputLevel"
"&A + 1"

Top of page

variable_name

Syntax

The variable name can be A - Z and a - z.

Description

Variable names are not case-sensitive. variable_of_type_integer, variable_of_type_string and variable_of_type_indexed_string are distinct variables.

Example

A 
B
a
b

Top of page

variable_of_type_indexed_string

Syntax

variable-name[variable_index]

Description

variable_of_type_indexed_string are always correctly initialized to blanks.

Example

A[0]
B[%OutputLevel]

Top of page

variable_of_type_integer

Syntax

variable-name

Description

variable_of_type_integer are initialized to zero.

Example

A
B
a
b

Top of page

variable_of_type_string

Syntax

variable-name

Description

variable_of_type_string are always initialized to blanks.

Example

A 
B  
a  
b

Top of page