Specification reconstruction system

Abstract

To reconstruct specifications of an existing system in a form which can be effectively used by a developer who performs development based on the existing system.SOLUTION: The system includes an asset analyzing unit (11) that analyzes and grasps a relation between specifications, inputs a list of documents related to the specifications and a prompt specifying a predetermined condition to a generation AI3, and causes the generation AI3 to create detailed specifications (21) related to the documents according to the predetermined condition, a DB processing unit (12) that creates a specifications DB22 based on the detailed specifications (21), and a retrieval processing unit (13) that causes the generation AI3 to retrieve the specifications AI3 based on an interaction between a developer (2) and the generation DB22.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a software program development technology, and more particularly to a technology that is effective when applied to a specification reconstruction system that reconstructs specifications for an existing system. [Background technology]

[0002] When developing software based on an existing system, such as for maintenance or additional development of an information processing system or migration to a new system, specifications, related documents, source code, etc. (hereinafter collectively referred to as "specifications") are extremely important for developers to understand the system. In particular, with older systems such as legacy systems, there are cases where specifications were not prepared in the first place, or some have been scattered and lost, and the preparation of specifications is insufficient. In addition, there are cases where the experts from that time are no longer around and it is not possible to consult them about the contents, which can be a major obstacle when developing based on an existing system.

[0003] In response to this, for example, Japanese Patent Application Laid-Open No. 5-81004 (Patent Document 1) describes an automatic document generation system that automatically generates document files for process flow specifications, table definition specifications, and module definition specifications from source files written in a programming language, and when the source file is composed of multiple modules, the process flow specification file is divided for each module to create each document file.

[0004] Furthermore, Patent Publication No. 6932046 (Patent Document 2) describes that by converting source code into a process specification that represents the input, output, and processing of variables in a table, it is possible to generate a functional specification from the source code without any omissions, thereby ensuring traceability between the source code and the specification. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 5-81004 [Patent Document 2] Patent No. 6932046 Summary of the Invention [Problem to be solved by the invention]

[0006] According to the prior art, it is possible to create predetermined specifications etc. from source code etc.

[0007] However, simply explaining the behavior that can be understood from the source code makes it difficult to fully understand the significance of each behavior and operation in relation to the business functions and specifications that the target program or system achieves. Furthermore, even if a specification document can be created, it is meaningless if developers cannot use it during development, but simply creating and documenting a specification document makes it difficult for developers to use it.

[0008] SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide a specification reconstruction system that reconstructs the specifications of an existing system in a form that can be effectively used by developers who develop based on the existing system.

[0009] The above and other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings. [Means for solving the problem]

[0010] Among the inventions disclosed in this application, the outline of representative inventions will be briefly explained as follows.

[0011] A specification reconstruction system that is a representative embodiment of the present invention is a specification reconstruction system that reconstructs specifications of an existing system, and includes an asset analysis unit that analyzes and understands the relationships between each of the specifications, inputs a list of related documents in each of the specifications and a prompt specifying predetermined conditions into a generation AI (Artificial Intelligence), and causes the generation AI to create detailed specifications for the documents in accordance with the predetermined conditions; a database processing unit that creates a specification database based on the detailed specifications; and a search processing unit that causes the generation AI to search the specification database based on a dialogue between a developer and the generation AI. [Effects of the Invention]

[0012] The effects obtained by the representative inventions disclosed in this application can be briefly explained as follows.

[0013] That is, according to the representative embodiment of the present invention, it becomes possible to reconstruct the specifications of an existing system in a form that can be effectively used by a developer who develops a system based on the existing system. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a diagram illustrating an overview of a configuration example of a specification reconstruction system according to an embodiment of the present invention. [Figure 2] FIG. 10 is a diagram outlining an example of a prompt for causing a generation AI to create a detailed specification in one embodiment of the present invention. [Figure 3] FIG. 1 is a diagram showing an outline of an example of a detailed specification according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0015] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In all drawings used to explain the embodiments, the same parts are generally designated by the same reference numerals, and repeated explanations will be omitted. However, parts that have been designated and explained in one drawing may be referred to by the same reference numerals in the explanation of other drawings, although they will not be shown again.

[0016] <Summary> When reconstructing specifications for an existing system for which specifications have not been prepared, as mentioned above, it is difficult to fully understand the meaning of each behavior and operation in relation to the business functions and specifications realized by the target program or system by simply explaining the behavior that can be understood from the source code. For example, even if you understand the process of "setting the value of variable XX to 8," it is difficult to understand the meaning of the process if you do not know what "variable XX" represents or what the value "8" means.

[0017] Furthermore, even if specifications can be created, it is meaningless if developers cannot use them during development.In order for developers to effectively use the created specifications, they need to be organized and restructured so that the desired information described in the specifications can be searched and accessed smoothly.

[0018] Therefore, a specification reconstruction system, which is one embodiment of the present invention, creates detailed specifications from documents such as source code and related documents related to existing systems using so-called generative AI (artificial intelligence) such as ChatGPT (registered trademark), and then reconstructs the specifications by combining them with the original source code, etc., into a database that developers can search and refer to through dialogue with the generative AI, etc. This makes it possible to efficiently support developers in understanding the specifications of existing systems, for example, where specifications are not well-organized and have become black boxes, and where there are no experts and the system includes programs such as assemblers and JCL (Job Control Language) that are difficult to transcribe.

[0019] <System configuration> 1 is a diagram showing an overview of an example configuration of a specification reconstruction system 1 according to an embodiment of the present invention. The specification reconstruction system 1 is an information processing system that is configured, for example, by one or more server devices, a virtual server constructed on a cloud computing service, or an information processing terminal, and that realizes various functions related to the reconstruction of specifications for an existing system by executing middleware such as an OS (Operating System), a DBMS (DataBase Management System), and a Web server program, which are loaded onto memory from a recording device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive), and software running on the OS (Operating System), which are loaded onto memory from a recording device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive), and software running on the OS (Operating System), which is ... run on the memory.

[0020] This specification reconstruction system 1 has various units, such as an asset analysis unit 11, a database processing unit 12, and a search processing unit 13, which are implemented as software. It also has a data store, such as a specification database (DB) 22, which is implemented as a database, file table, or the like. The asset analysis unit 11, the database processing unit 12, and the search processing unit 13 can each access and use the generation AI 3 implemented in an external or local environment via an API (Application Programming Interface), or the like.

[0021] The asset analysis unit 11 has the function of taking documents such as source code 31 and specifications 32 of an existing system as input, analyzing them to understand their relationships, organizing the call structures between the source code 31 and the data call structures, and using a series of documents that are determined to be related as input to create a detailed specification 21 using the generation AI 3 (making the generation AI 3 create the detailed specification 21).Paper documents (handwritten or printed) for which no data exists may be input using so-called AI-OCR (Optical Character Recognition) by the generation AI 3.

[0022] To understand the relationship between the source code 31 and the specifications 32 and to organize the call structure, various well-known techniques in software engineering, such as source code and document analysis techniques, can be used as appropriate. If an expert is available, information can be obtained from that expert, or the results of the analysis and organization by the asset analysis unit 11 and the results of the detailed specifications 21 can be presented to that expert for confirmation.

[0023] 2 is a diagram outlining an example of instructions (prompts) for causing the generation AI 3 in one embodiment of the present invention to generate the detailed specification 21. In this embodiment, the input files that are the inputs required to generate the detailed specification 21 and the output request for the content to be output as the detailed specification 21 are specified together in a prompt and input to the generation AI 3.

[0024] In the example of Figure 2, the conditions for processing by the generation AI 3 are set in "#SituationSetting," which specifies the situation setting, such as the role and purpose of the processing, and in "#Constraints," which specifies the content and items to be output. Then, in "#Output," the creation of a specification is instructed, and in "#InputFile," a series of related documents (source code 31 and specification 32) are specified. The series of related documents here is, for example, a series of documents determined to be related by the analysis of the source code 31 and specification 32 by the asset analysis unit 11. In this way, by inputting a series of related documents together, rather than creating specifications for each individual document, it is possible to understand and explain the content and meaning of the processing in relation to business functions and specifications.

[0025] This prompt can be created by setting the required content, such as a list of input files, to a pre-prepared template. It is not necessarily required to complete the detailed specification 21 with a single instruction to the generation AI 3. The output content of the detailed specification 21 may be gradually fulfilled by issuing additional instructions or requests to the output from the generation AI 3 and repeating automatic and manual interactions.

[0026] Fig. 3 is a diagram outlining an example of a detailed specification 21 in one embodiment of the present invention. Fig. 3 shows an example of a detailed specification 21 output for each program by the generation AI 3 in accordance with the output conditions, based on the list of input files specified by the prompt such as the example in Fig. 2 above.

[0027] Returning to FIG. 1 , the database processing unit 12 has a function of creating a database of the contents of the detailed specification 21 output by the asset analysis unit 11 as a specification DB 22. The detailed specification 21 shown in the example of FIG. 3 can actually be enormous in volume, making it difficult for the developer 2 to identify and refer to the necessary or relevant parts when designing and developing a new system. Therefore, in this embodiment, the contents of the detailed specification 21 are created as a database so that the developer 2 can efficiently refer to them. By creating a database that includes not only the detailed specification 21 but also the original source code 31 and specifications 32, the developer 2 can confirm the original information that served as the basis for the detailed specification 21, thereby ensuring the accuracy of the specifications and improving development efficiency.

[0028] The search processing unit 13 has a function for actually searching the specification DB 22 and a function for providing a search interface to the developer 2. The search function can be configured to use, for example, RAG (Retrieval-Augmented Generation) technology, provide the contents of the specification DB 22 to the generation AI 3 as an information source, and perform a search (make the generation AI 3 search) through a dialogue with the generation AI 3 (a question to the generation AI 3). For example, in response to a question from the developer 2 such as "Where in the specification did you refer to for the processing of this part?", the relevant parts of the existing source code 31 and specification 32 answered by the generation AI 3 can be output.

[0029] In addition to the specification search function, for example, as a development support function, it is also possible to implement a function to generate and output sample source code in other languages ​​using the generation AI3 based on the contents of the specification DB22 (detailed specification 21) (to have the generation AI3 generate sample source code in other languages).

[0030] As explained above, according to the specification reconstruction system 1 which is one embodiment of the present invention, a detailed specification 21 is created using the generation AI 3 from documents such as source code 31 and specifications 32 related to an existing system, and this is then combined with the original source code 31 and specifications 32 to create a database as a specification DB 22, which is then organized and reconstructed so that the developer 2 can search and refer to it through dialogue with the generation AI 3. This makes it possible to efficiently support the developer 2 in understanding the specifications when designing and developing new systems based on existing systems for which specifications have not been organized.

[0031] The invention made by the inventor has been specifically described above based on the embodiments, but it goes without saying that the present invention is not limited to the above embodiments and can be modified in various ways without departing from the spirit of the invention. Furthermore, the above embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those having all of the described configurations. Furthermore, it is possible to add, delete, or replace part of the configuration of the above embodiments with other configurations.

[0032] Furthermore, the above-described configurations, functions, processing units, processing means, etc. may be partially or entirely implemented in hardware, for example, by designing them as integrated circuits. The above-described configurations, functions, etc. may also be implemented in software, with a processor interpreting and executing a program that implements each function. Information such as the programs, tables, and files that implement each function can be stored in a storage device such as a memory, hard disk, or SSD, or in a storage medium such as an IC card, SD card, or DVD.

[0033] In addition, in the above figures, the control lines and information lines shown are those that are considered necessary for explanation, and do not necessarily show all the control lines and information lines that are actually implemented. In reality, it can be assumed that almost all components are interconnected. [Industrial Applicability]

[0034] The present invention can be used in a specification reconstruction system that reconstructs specifications for an existing system. [Explanation of symbols]

[0035] 1...Specification reconstruction system, 2...Developer, 3...Generative AI, 11...asset analysis unit, 12...database processing unit, 13...search processing unit, 21...Detailed specifications, 22...Specification DB, 31...Source code, 32...Specification

Claims

1. A specification reconstruction system for reconstructing specifications of an existing system, an asset analysis unit that analyzes and understands the relationships between the specifications, inputs a list of documents related to the specifications and a prompt specifying predetermined conditions to a generation AI (Artificial Intelligence), and causes the generation AI to create detailed specifications related to the documents in accordance with the predetermined conditions; a database processing unit that creates a specification database based on the detailed specifications; A specification reconstruction system having a search processing unit that causes the generation AI to search the specification database based on an interaction between a developer and the generation AI.

2. 2. The specification reconstruction system according to claim 1, The asset analysis unit is a specification reconstruction system that gradually fulfills the contents of the detailed specification by repeatedly executing dialogue between the developer and the generation AI.

3. 2. The specification reconstruction system according to claim 1, A specification reconstruction system, wherein the specified conditions include, in addition to the list of documents, instructions regarding the role and status of the generating AI, and instructions regarding the output content of the detailed specification.

Images