Abstract ::
Business Definition Language(BDL) The Business Definition Language is a simple procedural scripting language with some specialized features for defining Business Flows. The language supports variables, evaluation of conditional expressions, and execution of a few basic flow control statements. The primary purpose of the BDL is to enhance usability by: · Guiding users through complex Insurance Businesses flows. · Permitting the display of information to be tailored both to specific tasks and to the users role · Flowing data automatically from screen to screen to minimize re-keying of data To achieve these goals, the starts with a simple conceptual model: every task a user can perform on the system, from simple data retrieval to a complex business process, can be divided into a sequence of granular tasks, or "Steps". This sequence of Steps, called a "Business Flow", can be arbitrarily complex and can include conditional branching. A ?Business Process Definition" specifies both the sequence of Steps that make up a business and how data flows forward from Step to Step. Insurance application developers and business analysts can use a text editor to create or modify new or existing "Business Flow" source modules. The topics that follow provide the conceptual detail underlying the setup of business definitions along with information regarding how to program Business Flows. BDL incorporates the following Steps: · IO-Steps - interactions with the end user. IO-Steps define the set of data that is presented to, and/or returned from, the user. For each HTML page there is a corresponding IO-Step declaration identifying each field on the page. · Business Flow Steps - calls to other Business definitions. The called Business acts much like a subroutine (a "sub-business flow") with respect to the calling Business. BDL can also have input data and output data elements specified for them in the same way as IO-Steps. Every Step performs some discrete data processing task. For example, a typical business flow might be summarized as "Retrieve client record". For this IO-Step, two corresponding IO-Steps might be "Get client number from user", and "Display client name and accept updates". In the case of the example, most of the IO-Steps in the system are closely related to a business flow. IO-Steps are typically designed to get user input required for a business flow, or to display the output of a business flow. Modules: 1. Step Declarations 2. Process Definitions 3. Data Flow 4. Flow Control Step Declarations The language provides declarative support for presentation components (i.e data elements) and Business Functions The following is the syntax for Step declarations: IO-STEP A { ATTRIBUTES { Import = "B,C"; } } Process Definitions The following is an example of a Process definition that implements the example Process Flow: # Sample Process PROCESS id { STEP SampleStep1 { USES IO-STEP "GetClientNumber"; } STEP SampleStep2 { USES IO-STEP "ClientRetrieve"; } STEP SampleStep3 { USES IO-STEP "DisplayName"; } } Another sample process: PROCESS id { ATTRIBUTES { } VARIABLES { } STEP Sample1 { Uses IO-Step StepName } STEP Sample2 { USES IO-STEP StepName BRANCH(ACTION=="X") Sample1; } } PROCESS starts a Process definition where id is the unique identifier assigned to the process. The id name must be the same as the file name using a .f extension . The ATTRIBUTE block is optional. The VARIABLES block is optional. but must be there if there are any process parameters. The process parameter list defines the input and output fields for the process in a format similar to the Step statements used in IO-Step declaration. Data Flow Using the IO-Step declarations, which define the data requirements of each Step, the process has granular control over the flow of data from Step to Step. A key role of Step declarations is the assignment of an identifier, or field name to each data element of the Step. Field names are rationalized across the complete set of Step declarations so that, for example, a field "client-id" refers to the same type of client identifier in every Step. You can create variables in process scripts using simple assignment statements in the format: foo-variable="some string"; bar-variable=client-id; Flow Control The normal flow of a process is a linear progression through the Process definition: PROCESS sample { step1; step2; ... } You can use flow control statements to alter the flow based on transaction return codes, values of variables and host fields, or user-entered data: IF FOO = "X" BRANCH step3; A BRANCH statement target can be any Step defined in the Process definition or a label. Labels are defined using the syntax: : For example: CollectName: The name of a label cannot be the same as the name of a Step because the PathFinder will not know where to go. A looping construct is also provided: WHILE continue == "Yes" { step1; step2; ... } The BRANCH statement can be made conditional using the following syntax: BRANCH(X > Y) step3 The branch to step 3 will only take place if the variable X has a greater value than the variable Y. Any conditional statement that evaluates to true or false can be used in the parentheses. This statement is the equivalent of: IF X > Y BRANCH step3 Scripting Language Element: Element Description Example Comments A comment starts with the number sign # and ends at the end of the line. # # This is a comment line; it is ignored by PathFinder.# Identifiers An identifier is a sequence of letters, digits, underscores, and dashes starting with a letter. MIR-TABLEMIR_AGE01 Keywords All keywords are case-sensitive; they all must be in uppercase. Keywords have the same format as identifiers. ATTRIBUTES BRANCH CONST ELSE EMBEDS EXIT IF IN INCLUDE INOUT LOADSTRINGS MESSAGES-T OUT PROCESS STEP STRINGTABLE TERMINATE USES VALUEOF VARIABLES WHILE Functions Functions consist of the name of the function followed by parentheses "( )" that enclose the arguments to the function. X = LENGTH("Policy ID"); Literals - Numbers Other tokens in the script are numbers and strings. A number is an integer, that is, a sequence of digits. 1234 # a number Literals - Strings A string constant is defined as a quoted sequence of characters. You can use the escape character "" to define a non-printable character. "This is a string" Braces { } Used to surround statements or blocks of statements { #process statements} Comments: A comment starts with the number sign "#" and ends at the end of the line; for example: # # This is a comment line; it is ignored by PathFinder. # Identifiers: An identifier is a sequence of letters, digits, underscores, and dashes that start with a letter. For example, the following are valid identifiers: MIR-TABLE MIR_AGE01 Keywords: All keywords are case-sensitive; they must be typed in uppercase. Keywords have the same format as identifiers but are reserved for the PathFinder use; the following is the list of keywords: ATTRIBUTES BRANCH ELSE EMBEDS EXIT IF IN INCLUDE INOUT LENGTH MESSAGES-T OUT PROCESS STEP STRINGTABLE TERMINATE USES VALUEOF VARIABLES WHILE The following keywords are described: ATTRIBUTES Keyword BRANCH Keyword EXIT Keyword IF Keyword IN Keyword INCLUDE Keyword INOUT Keyword MESSAGES-T Keyword OUT Keyword STEP Keyword STRINGTABLE Keyword TERMINATE Keyword WHILE Keyword