KR20030010217A - Method for processing an application data in integration broker system - Google Patents

Abstract

PURPOSE: A method for processing application data in an integrated broker system is provided to effectively and easily process a data integration and a communication between applications and a data process between business logic by defining new XML standard for embodying an EAI(Enterprise Application Integration) solution adapted to an integration of different applications in a company or between companies. CONSTITUTION: Data are processed synchronously by at least one service selected in services of the least significant program module unit being used in common. The processed data are processed successively by at least one package selected in a group of packages being used in common. An application data processing method may be applied to an integrated broker system which mediates a data communication between different kinds of applications and databases. The service is a program module which processes the least significant company business and the service is a resource for integrating a database, a web, or an on-line application. The package is a program module which processes a synchronous company business having at least one service and has mapping information with respect to input/output data between each service as a script form.

Description

{Method for processing an application data in integration broker system}

The present invention relates to a method for processing application data in an integrated brokerage system, and more particularly, to application data in an integrated brokerage system for centrally integrating data communication between various applications such as different types of computers or databases. It relates to a treatment method.

First, FIG. 1 illustrates a general state in which a plurality of applications and databases are connected and connected. In particular, a plurality of applications, such as a computer system, are point-to-point for data communication with a database or different types of applications. Point to Point method can be connected and connected, the different types of applications connected by the point-to-point method as described above, the interface interface and communication unique for data communication with the counterpart application Each requires a protocol.

For example, as shown in FIG. 1, in a state in which the first and second databases 10 and 20 and the first, second and third applications 110, 111, and 112 are connected in a point-to-point manner, When the n-th application 120 is to be additionally connected and connected, the n-th application 120 requires unique interface logic and communication protocols for data communication with the counterpart application, respectively.

In addition, even in a counterpart application or a database connected to the n-th application 120, a unique interface logic and a communication protocol for data communication with the n-th application should be further provided. Due to the different data formats and platforms between applications, the interface logic is designed with complex business logic, and when adding or correcting the business logic, it is necessary to modify all the algorithms and modules associated with it. there was.

Therefore, as shown in FIG. 2, a new application, for example, n, is used using an integration broker system 200 that centrally integrates data communication between different types of applications. In order to more easily add the second application 120, a plan has been proposed, and such an integrated mediation system is expected to be commercialized.

On the other hand, as a solution for implementing the integrated mediation system 200, Enterprise Application Integration (EAI), eBusiness, Customer Relationship (CRM), Knowledge Management System (KMS), Enterprise Information Portal (EIP), Supply Chain (SCM) Various solutions such as management (DW) and data warehouse (DW) have been proposed. The EAI solution is evaluated as the most effective solution for integrating and brokering applications of different types of applications within an enterprise or between different companies. I am getting it.

In addition, an optimal programming language for the EAI solution is XML (eXtensible Markup Language), which is an international standard programming language related to electronic commerce between electronic documents and enterprises. It has the advantage of enabling communication.

Accordingly, many companies are looking for ways to express existing data in XML so that the data can be reused and integrated mediation between applications of different types.

However, in order to convert a large amount of existing data into XML, the work is not only enormous, but very complicated, and since data conversion itself is impossible without modification of the application, a new definition of XML for integrating various applications is required. There is an urgent need for a solution that can simplify and design complex business logic more efficiently.

Accordingly, the present invention was created in view of the above situation, and defines a new XML standard for implementing an enterprise application integration (EAI) solution that is optimal for integrating different applications within or between enterprises, and between various applications. Its purpose is to provide a method for processing application data in an integrated brokerage system to facilitate data integration, communication, and data processing between business logic.

1 illustrates a state in which a plurality of general applications and databases are connected and connected.

2 illustrates an integrated mediation system to which an application data processing method according to the present invention is applied, and a state in which a plurality of applications and databases are connected and connected.

3 illustrates a hierarchical structure for services, packages, and flows defined in an application data processing method according to the present invention;

4 illustrates a data processing process of a package defined by an application data processing method according to the present invention.

5 illustrates a data transformation process between services defined by an application data processing method according to the present invention.

6 illustrates a process of processing data divergence between services defined by an application data processing method according to the present invention.

7 illustrates a data merging process between services defined by an application data processing method according to the present invention.

8 illustrates a program module included in the integrated mediation system according to the present invention.

9 illustrates an embodiment of a raw data structure and a destination data structure according to the present invention,

10 to 12 illustrate input / output data in a lexical analysis module, a grammar error checking module, and a target data generation module according to the present invention.

※ Explanation of code for main part of drawing

10,20 Database 110-112,120 Application

200: integrated mediation system 210: data type analysis module

220: lexical analysis module 230: grammar error check module

240: purpose data generation module S: Service

M: Meta Service Q: Queue

F: Flow

Application data processing method in the integrated mediation system according to the present invention for achieving the above object, a first step of processing data synchronously by one or more services selected from the set of services of the lowest level module module used in common ; And two steps of sequentially processing the processed data by one or more packages selected from a set of commonly used packages that are sets of services.

Hereinafter, a preferred embodiment of an application data processing method in an integrated mediation system according to the present invention will be described in detail with reference to the accompanying drawings.

Application data processing method in the integrated mediation system according to the present invention, as shown in Figure 2, to be applied to the integrated mediation system 200 for centralized mediation of data communication between different types of applications and databases 3 is a hierarchical structure for a service, a package, and a flow defined by an application data processing method in an integrated mediation system according to the present invention.

The service, package, and flow represent a program module unit for describing an application in an enterprise. As shown in FIG. 3, the service or meta service is independently reusable at the package level, and the packages are Independent flow re-use is possible at the flow level, for example, continuous data processing between the packages at the flow level by a queue for first in first out (FIFO) of input / output data. Will be performed.

On the other hand, the service is a program module that processes the lowest level of corporate business, for example, is defined as a resource (Resource) for unifying the database, web (web) and online applications.

In addition, the package is a program module for processing a synchronous corporate task consisting of at least one or more services, for example, mapping of data flows, that is, input / output data, between services. Information is defined in the form of a script.

In addition, the flow is a program module for processing an asynchronous enterprise task consisting of at least one package and external applications, and for example, each element constituting the flow is defined as a task. .

As described above, the above services, packages, and flows are program modules implemented and defined using XML, which is optimal for implementing an EAI solution, and may be composed of a header part and a content part, respectively. place. In the header part, a transaction (node: transaction) type corresponding to the basic information of each module, catalog information including creation / modification date and description, and execution condition, result, and cache related information are included. Consist of included.

On the other hand, by defining the transaction type to enable package-level transaction management, when the performance of a particular service fails, it is possible to define that the entire package, which is a synchronous business unit, will be rolled back. It includes the creation / modification date of the module and its creator, and a keyword (node: keywordList) describing the characteristics of the module. This information includes information for allowing each module to be integrated and managed as a resource in the enterprise. do.

In addition, the cache information is information that exists only in the service, and the service is guaranteed to not change the result for a certain period of time, it is information that enables to recycle the performance result of the service already performed when the module is executed.

On the other hand, the header portion of the package and flow includes a list (node: pkService, fwTask) for the sub-modules constituting the package and flow, the content portion includes detailed information for execution, for example, input and output Information on and information on each module is included.

The service includes information on a data source accessed by the module (node: resData). In this case, characteristic data is described according to the data source type of each service, and in the case of access to a database, SQL (Structured Query Language) Or, in the case of the web, a script. The package includes XML for defining a data flow between services and a service constituting the package.

On the other hand, the XML defining the service may include the entire XML by information (node: isRef) included in the package header, it may refer to the service definition that exists outside the execution. The flow includes information (node: fwFMD) for defining a data flow between tasks and managing asynchronous process flow. In addition, configuration information (node: fwNotify) is described for notifying progress of work requester during asynchronous task processing.

In addition, mapping information (node: fieldMap) of input / output data of a module is conceptually defined in the same structure as the definition of a row and a column of a relational database. Different types of column structures and the attributes of the rows that make up those columns are defined, and these input / output structures form the package from the service and the internal data generated during all the flows of the flow from the package and external applications. This is a common expression for.

The input / output data defined as above occurs in the module processing order within the definition of the package and the flow. The mapping operation is defined by the map in the task flow in the case of the package, and the external application in the flow. In the case of integrating them, the data structure may be different from that of the filed map structure, which is a representation method of internal data.

On the other hand, Figure 4 shows an XML configuration for the package defined by the application data processing method according to the present invention, for example, one of the first, second and third services (S1, S2, S3) The first package P1 includes a unique package ID (package id = "P1") for package identification; A package header (pkHead) in which basic information of a package such as an author and a version is recorded; The package contents (pkBody) include information on package input (pkInput), information on package output (pkOutput), information on business logic between each service (mapInf), and information on services included in the package (service). ) Is configured to include.

On the other hand, any one service, for example the third service (S3), has a unique service ID (service id = "S3") for service identification; A service header (svHead) in which basic information of a service such as an author and a version is recorded; The service contents (svBody) including information on service input (svInput), information on service output (svOutput), and information on internal processing of each service (svData) are included.

The information on the service output (svOutput) includes information on a field map name (filedMap name = "S3Out"), and the field map name information includes an output field name (filed name = "e" or "). f ") and information indicating the type thereof (type =" int "or" string "), wherein the data differs between services by business logic in a package consisting of multiple services as described above. A data processing method of transforming, devering, and merging structures is described in detail as follows.

First, FIG. 5 illustrates a data transformation process between services defined by an application data processing method according to the present invention. For example, the first service S1 may have three a, b, and c outputs. The type of data output through a and b outputs has integer and zero length, respectively, and the type of data output through the remaining c outputs is String and It can be 16 strings long.

Meanwhile, the second service S2 that receives and inputs data of the first service S1 may have two x and y inputs, and the types of data received through the x input may be integers and zeros. (Zero) length, the type of data received through the remaining b input may be an integer and 16 string length.

As illustrated in FIG. 5, the first script Script 1 of the business logic for transforming the input / output data between the first and second services may have zero outputs a and b respectively. Is greater than zero and the c output is not null, the x input is the data output through the a output, and the a and b outputs are respectively greater than zero and the c output is null. In this case, the x input has logic to be data output through the b output.

Also, in the second script (Script 2) of the business logic, when the average ((a + b) / 2) of the a output and the b output is 100, the y input has “$” added to the beginning. c output data, and if the average ((a + b) / 2) of the a output and the b output is not 100, the y input is a logic such that "$" is the c output data appended to the beginning. By doing so, the data between the first service S1 and the second service S2 included in one package can be transformed.

6 illustrates a process of data branching between services defined by an application data processing method according to the present invention. For example, the first service S1 may have four outputs a, b, c, and d. The data output through the a and b outputs may be a and b inputs of the second service S2, and may be branched to the c and d inputs of the third service S3.

Meanwhile, in the first business logic for branching the output data of the first service S1 to the second service S2, as illustrated in FIG. 6, the output a and the output b of the first service are shown. This includes first and second scripts for outputting the a and b inputs of the second service, respectively, and branching the remaining output data of the first service S1 to the third service S3. The second business logic to include first and second scripts for outputting the c output and the d output of the first service to the c input and the d input of the second service, respectively.

Therefore, the a, b, c, d output data of the first service S1 is branched into a, b inputs of the second service S2 and c, d inputs of the third service S3. You can do it.

FIG. 7 illustrates a process of merging data between services defined by the application data processing method according to the present invention. For example, two a and b outputs of the first service S1 and Data output through two c and d outputs of the second service S2 may be merged into two a1 and c1 inputs of the third service S3, respectively, as shown in FIG. 7. The business logic for the merge process includes a first script for merging a output and b output of the first service and outputting the output as a1 input of a third service, and c output and d output of the first service. A second script for merging and outputting to the c1 input of the third service is included.

Accordingly, the a, b output data and the c and d output data of the first service S1 may be merged, respectively, and the merge process may be performed so that the a1 and c1 inputs of the third service S3 are received respectively. .

On the other hand, by supporting a script for implementing and executing a variety of business logic for transforming, branching, or merging different types of data as described above, the script processing method for reducing execution time, enabling efficient design, maintenance, etc. in detail The explanation is as follows.

First, FIG. 8 illustrates a program module included in an integrated mediation system according to the present invention. In particular, a program module for script processing includes a data type analysis module 210, a lexical analysis module 220, and a grammar error check. Module 230, and the object data generation module 240, the data type analysis module 210, the structure of the raw data and the target data, the raw actual data, and a script for generating the target data After receiving the input, each data structure is analyzed.

On the other hand, in the data type analysis module 210, for example, as shown in Figure 9, the structure and purpose of the raw data largely divided into a data name (Data_Name), data type (Type) and data length (Length) By analyzing the structure of the data, you can verify the data format errors of the business logic in the script and generate the target data.

In the lexical analysis module 20, as shown in FIG. 10, the input script is divided into meaningful tokens, and constants and variables are stored and managed in respective tables as in the previously stored data structures. By doing so, it can be used for grammar error checking and object data generation.

In addition, as shown in FIG. 11, the grammar error checking module 230 parses a parsed program from a received token list, that is, a program written in a high-level language, to grammar each sentence. In addition to generating a parse tree configured in a tree form according to the structure, it is checked whether there is a grammar error in the script. In this case, the parse tree may be generated and used to generate the target data.

In the object data generation module 240, as shown in FIG. 12, the object data is generated by referring to the structure of the raw and object data, the parse tree, the constant, and the variable table. The parse tree of is generated with one target real data, and each generated target real data is composed of one meaningful record unit.

On the other hand, the format of the data generated by the target data generation module 240 and the data format stored in the data format analysis module 210 does not match, or the parse tree processed by the target data generation module 240 If their constants and variables do not match the data types, an error will occur.

As mentioned above, preferred embodiments of the present invention are disclosed for the purpose of illustration, and those skilled in the art can improve and change various other embodiments within the spirit and technical scope of the present invention disclosed in the appended claims below. , Replacement or addition would be possible.

The application data processing method in the integrated mediation system according to the present invention as described above defines a new XML standard for implementing an EAI solution that is optimal for integrating different applications, and integrates and communicates data between various applications and business. For data processing between logics, data is processed to be synchronized by services of the lowest level business program module used in common, and the processed data are sequentially processed by packages that are sets of the services. In this way, the integration of different applications can be easily implemented, and the development and maintenance of program modules can be simplified, and the work flow is flowed into the package and the top program module unit. Each accessible This enables both online and batch environments, and makes it easier for many users who are familiar with scripting languages and simple programming, by implementing and processing business logic for data exchange between applications using scripts. It's a very useful invention that allows you to design business logic.

Claims (8)

A step of synchronously processing data by at least one service selected from a set of services of a unit of the lowest program module used in common; And And sequentially processing the processed data by one or more packages selected from a set of commonly used packages that are sets of services. The method of claim 1, The service is a program module for processing the lowest enterprise business, the application data processing method in the integrated mediation system, characterized in that the resource (Resource) for integrating a database, web or online applications. The method of claim 1, The package is a program module for processing a synchronous enterprise work consisting of at least one service, the application data in the integrated mediation system, characterized in that the mapping information for the input / output data between each service in the form of a script Treatment method. The method of claim 1, The flow in which at least one package is included is an application data processing method in an integrated brokerage system, characterized in that a program module for processing asynchronous enterprise work consisting of the package and an external application. The method according to any one of claims 1 to 4, The program module, the application data processing method in the integrated mediation system, characterized in that defined and implemented by XML (eXtensible Markup Language) that is optimal for enterprise application integration (EAI) solution. The method of claim 1, The package is an application data processing method in the integrated brokerage system, characterized in that for processing the data processing between the services by at least one of the method of transformation, branching or merging. The method of claim 6, The business logic for the transformation, branching or merging process, at least one or more scripts, characterized in that the application data processing method in the integrated mediation system. The method of claim 1, Data or control transfer between the packages, the application data processing method in the integrated mediation system, characterized in that via a queue (Queue).