Software generating system

Abstract

To automatically perform appropriate correction and generation such as modernization of an application program, while securing consistency.SOLUTION: A software generating system includes: a consistency ensuring information input part for obtaining consistency ensuring information which is information ensuring consistency between a plurality of screens or classes of a Web application program; a prompt template input part for obtaining a prompt template which is an instruction for generation or correction when the Web application program is generated or corrected; a prompt generating part for generating a prompt in which the consistency ensuring information is embedded in the prompt template on the basis of the consistency ensuring information input to the consistency ensuring information input part; and a source code generating part for generating source code of a generation portion or a correction portion of the Web application program on the basis of a prompt generated by the prompt generating part.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a software generation system. [Background technology]

[0002] Once created, web application programs are often used for a long period of time, and the programs need to be updated according to changes in the environment in which they are used, etc. For example, traditional web application programs were composed of multi-page applications (MPA), but in recent years single-page applications (SPA) have become mainstream, creating a need to convert web application programs into SPAs.

[0003] When a user requests a specific page on a website, an MPA sends all of the HTML data required to display that page, whereas an SPA sends only the data that is different from the page that was previously displayed. Updating a web application program from an MPA to an SPA is one example; web application programs may need to be updated for a variety of reasons.

[0004] Such updates to web application programs are called application modernization. Modernization requires experts who are familiar with the code of the relevant programs, but in many cases, the experts who were involved in the development of legacy systems have already retired, making it difficult to secure the necessary personnel. For this reason, there has been a demand for the development of technology for automatically modifying programs. Patent Document 1 describes a technology for generating a program by searching for source code developed in the past with similar specifications and modifying the source code using a genetic algorithm or the like. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent No. 737535 Summary of the Invention [Problem to be solved by the invention]

[0006] The technology described in Patent Document 1 makes it possible to generate source code. However, when the source code to be obtained is used in multiple files, it is necessary to ensure consistency by reflecting the mutual relationships among the multiple files in the source code, but it is difficult to ensure such consistency with the technology described in Patent Document 1.

[0007] For example, when modifying a web application program, suppose that the technology described in Patent Document 1 is applied to modify a certain web page so that it looks the same as before the modification. However, even if the display format of the web page is the same as before the modification, it may not be possible to modify the web page so that it transitions to another web page by clicking or entering information at a specific point on the web page.

[0008] In view of the above, an object of the present invention is to provide a software generation system that can automatically perform appropriate modifications and generation such as modernization of application programs while ensuring consistency. [Means for solving the problem]

[0009] In order to solve the above problems, for example, the configurations described in the claims are adopted. The present application includes multiple means for solving the above-mentioned problems, and one example thereof is a software generation system comprising: a consistency assurance information input unit that obtains consistency assurance information, which is information that ensures consistency between multiple screens or classes of a Web application program; a prompt template input unit that obtains a prompt template, which is an instruction for generation or modification when generating or modifying a Web application program; a prompt generation unit that generates a prompt in which the consistency assurance information is embedded in the prompt template based on the consistency assurance information input to the consistency assurance information input unit and the consistency assurance information input to the consistency assurance information input unit; and a source code generation unit that generates source code for the generation or modification portion of the Web application program based on the prompt generated by the prompt generation unit. [Effects of the Invention]

[0010] According to the present invention, it is possible to generate or modify a Web application program while ensuring the consistency between multiple files, and it is possible to generate or modify an appropriate Web application program. Problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a diagram showing an outline of an example of processing performed by a software generation system according to a first embodiment of the present invention. [Figure 2] FIG. 2 is a diagram showing an example of the flow of processing by the software generation system according to the first embodiment of the present invention. [Figure 3] 1 is a configuration diagram showing an example of a software generation system according to a first embodiment of the present invention. [Figure 4] 5 is a flowchart showing an example of processing (main routine) of a prompt generating unit according to the first embodiment of the present invention. [Figure 5]5 is a flowchart showing an example of processing (subroutine) of a prompt generating unit according to the first embodiment of the present invention. [Figure 6] 5 is a flowchart showing an example of processing (subroutine) of a prompt generating unit according to the first embodiment of the present invention. [Figure 7] 1A is a diagram showing an example of a case where there is no consistency assurance information and FIG. 1B is a diagram showing an example of a case where there is consistency assurance information according to a first embodiment of the present invention. [Figure 8] FIG. 10 is a diagram showing an example of the flow of processing by the software generation system according to the second embodiment of the present invention. [Figure 9] FIG. 10 is a configuration diagram showing an example of a software generation system according to a second embodiment of the present invention. [Figure 10] FIG. 10 is a diagram showing an example of analysis in the software generation system according to the second embodiment of the present invention. [Figure 11] FIG. 10 is a diagram showing an example of the flow of processing by the software generation system according to the third embodiment of the present invention. [Figure 12] FIG. 10 is a configuration diagram showing an example of a software generation system according to a third embodiment of the present invention. [Figure 13] FIG. 10 is a diagram illustrating an example of processing by a software generation system according to a third embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0012] <First embodiment> A software generation system according to a first embodiment of the present invention will be described below with reference to FIGS.

[0013] [Example of software updated by the software generation system] First, with reference to FIG. 1, an example of a Web application program that is updated by the software generation system of this embodiment will be described. In this embodiment, as shown on the left side of FIG. 1, it is assumed that a P layer (presentation layer) 2 of a legacy Web application program 1, which is an already created Web application, executes a multi-page application (MPA). At this time, the software generation system of this embodiment creates a modern Web application program 3 by updating the legacy Web application program 1, as shown on the right side of FIG.

[0014] In the modern Web application program 3, the P layer 4 executes a single page application (SPA). An API (Application Programming Interface) 5 is provided between the P layer 4 and the data layer of the modern Web application program 3. In the following description and drawings, a Web application program will be referred to as a Web app.

[0015] [Overview of the process performed by the software generation system] FIG. 2 shows an overview of the processing performed by the software generation system of this embodiment. First, the software generation system acquires the source code 1a, 1b, 1c, ··· of the legacy web application and creates a prompt 11 for understanding the source code 1a, 1b, 1c, ···.

[0016] This prompt 11 is an instruction statement when using generative AI (Artificial Intelligence) technology. Specifically, prompt 11 creates a sentence (prompt) such as "Please understand the source code of XXX." Here, each acquired source code 1a, 1b, 1c, ... is inserted into the "XXX" field. In other words, prompt 11 is a prompt to understand each source code 1a, 1b, 1c, ... The software generation system executes the processing of the generated AI (step S11) using the created prompt 11, and obtains a response 12. The response 12 is a response saying, "I understand the source code of XXX."

[0017] The software generation system then uses a prepared prompt template 13 and the response 12 to obtain consistency assurance information 14. As shown in Figure 2, this prompt template 13 lists information necessary to ensure the consistency of a web application, such as "parameters to be passed to the transition destination," "parameter passing method," "name of the service class method to be executed," and "DTO class for storing the execution results." However, the prompt template 13 handled by the software generation system of this embodiment has some blank spaces (holes) and serves as an information template for ensuring consistency, for example.

[0018] For the prompt template 13 having blanks, a process is performed in which the source code understood from the response 12 obtained in step S11 is filled in the blanks. However, there may be consistency assurance information 14 that does not require the inclusion of source code or the like. Based on the consistency assurance information 14 thus obtained and the response 12, a prompt 15 for generating a Component class is obtained. When the prompt 15 is obtained, the software generation system executes the processing of the generated AI according to the created prompt 15 (step S12) and obtains the response 16 thereof. Furthermore, the software generation system obtains a prompt 17 for modifying the Component class from the response 16 and the consistency assurance information.

[0019] When the prompt 17 is obtained, the software generation system executes the generation AI process (step S13) according to the prompt 17, and obtains the generated source code 18 as a response. The generated source code 18 is source code for converting a legacy Web app into a modern Web app.

[0020] [Configuration of software generation system] FIG. 3 shows an example of the configuration of the software generation system 100. The software generation system 100 includes a prompt generation unit 110, a consistency processing unit 120, a source code generation unit 130, and a display unit (generation result display unit) 104.

[0021] The consistency processing unit 120 includes a consistency assurance information input unit 121 and a prompt template input unit 122 . The consistency assurance information input unit 121 acquires the consistency assurance information 14 prepared in advance, and obtains the consistency assurance information 14' by extracting the information necessary for the current processing. The prompt template input unit 122 obtains a prompt template 13' corresponding to the consistency assurance information corresponding to the current web application from the prompt templates 13 prepared in advance.

[0022] The consistency assurance information 14 ′ obtained by the consistency assurance information input unit 121 and the prompt template 13 ′ obtained by the prompt template input unit 122 are supplied to the prompt generation unit 110 . The prompt generation unit 110 generates a prompt 111 for performing processing of the generated AI based on the consistency assurance information 14' and the prompt template 13'. The generated prompt 111 includes the prompts 11, 15, and 17 described in Fig. 2. In other words, the prompt generation unit 110 performs processing to generate the prompts 11, 15, and 17 described in Fig. 2 in that order.

[0023] The prompt 111 generated by the prompt generator 110 is input to the source code generator 130 . Based on the sentence of the prompt 111, the source code generation unit 130 obtains an appropriate source code generation result 131 using generation AI technology. The source code generation result 131 is displayed on the display unit 104 .

[0024] The software generation system 100 shown in Fig. 3 can be configured by a computer, as shown in the upper right corner of Fig. 3. That is, the software generation system 100 includes, as computer hardware configuration, a CPU (Central Processing Unit) 101, memory 102, storage 103, display unit 104, and communication interface 105. These components are connected via a bus line to enable mutual data transfer.

[0025] CPU 101 executes programs (software) implemented in memory 102, thereby configuring a calculation unit that performs various processes on memory 102. For example, memory 102 includes a prompt generation unit 110, a consistency processing unit 120, and a source code generation unit 130. The memory 102 includes a work area for executing arithmetic processing, and also includes an area for storing programs executed under the control of the CPU 101. The storage 103 stores programs and various data. The display unit 104 displays the calculation results and the like.

[0026] The communication interface 105 executes communication processing with other systems. For example, the communication interface 105 communicates with a generation AI execution server 191 that executes the calculation processing of the generation AI. The generation AI processing in steps S11, S12, and S13 in FIG. 2 is executed by this generation AI execution server 191, and the software generation system 100 obtains prompt instructions required for executing the generation AI and the execution results of the generation AI.

[0027] [Processing performed by the prompt generation unit] Next, the prompt generation process performed by prompt generator 110 will be described. FIG. 4 is a flowchart showing the main routine of the prompt generation process performed by the prompt generation unit 110. First, the prompt generator 110 performs a data read process (step S110). Next, the prompt generator 110 performs an embedding process for the analysis results (step S120).

[0028] FIG. 5 is a flowchart showing the flow of the data read process in step S110 of FIG. First, when the data reading process is started, the prompt generation unit 110 reads a prompt template from the prompt template input unit 122 (step S111). Next, the prompt generator 110 reads the analysis results from the consistency processor 120 (step S112), and ends the data read process.

[0029] FIG. 6 is a flowchart showing the process of embedding the analysis results in step S120. When the process of embedding the analysis results begins, the prompt generation unit 110 starts processing a prompt template loop (step S121) and determines whether or not there is a data hole in the prompt template acquired from the prompt template input unit 122 (step S122). If there is no data hole in the acquired prompt template (No in step S122), the process returns to step S121 and moves on to processing another prompt template.

[0030] If the acquired prompt template has data holes in it in step S122 (Yes in step S122), the prompt generator 110 extracts the data holes from the results of analyzing and understanding the source code (step S123). Next, prompt generator 110 embeds the data extracted in step S123 into a prompt template, outputs it as a formal prompt (step S124), and ends the prompt template loop (step S125). The prompt generator 110 executes the prompt template loop process from step S121 to step S125 for all the retrieved prompt templates, and then ends the process of embedding the analysis results.

[0031] [Example of consistency assurance information] FIG. 7 shows an example of the consistency assurance information. The consistency assurance information described in this embodiment is information on items that must be common between different source codes, such as between Web page screens or classes. If this consistency is not ensured, the web application will not work. For example, even if a user inputs or selects a number in an input field on a specific web page, the next process that reflects that input or selection will not be executed.

[0032] FIG. 7A shows a case where a prompt without consistency assurance information is generated and output is obtained by processing the generated prompt using the generation AI. In the case of Figure 7A, the prompt is "Implement a function called viewAuthor that passes parameters using the navigate method." The output of the generated AI processing result in response to this prompt is shown in the lower part of Figure 7A. Note that the processing result shown in the lower part of Figure 7A is not appropriate as source code for a web page because it does not contain consistency assurance information.

[0033] FIG. 7B shows a case where a prompt with consistency assurance information is generated and an output is obtained by processing the prompt using the generation AI. In the case of Figure 7B, the prompt says, "Implement a function called viewAuthor using the navigate method to pass the parameter SSCI040Dto as state." This prompt contains the specific parameter "SSCI040Dto," which is the consistency information, and the code to be passed, "state."

[0034] The output of the generated AI processing result from this prompt is shown in the lower part of Figure 7B. The processing result shown in the lower part of Figure 7B has consistency assurance information, so it can be correctly handed over as source code for the web page. In other words, the processing results shown at the bottom of Figure 7B include prompts such as parameters between screens and DTOs commonly used between classes, and are data commonly used between different source codes. The processing results shown at the bottom of Figure 7B also include prompts for things like how to pass parameters, which are processes that require coordination between different source code. This means that the resulting modified web application will function correctly just like the one before the modification.

[0035] [Effects of the first embodiment] As described above, according to the software generation system of this embodiment, when a legacy Web application is modified into a modern Web application by the generation AI, consistency assurance information is included in the prompt, which is the instruction text of the generation AI, so that the Web application can be modified to function correctly. Therefore, various application modification tasks, such as modifying a legacy web application that runs on an MPA as described in Figure 1 into a modern web application that runs on an SPA, can be performed automatically using generative AI.

[0036] In the case of the software generation system of this embodiment, the source code of the Web application created by the processing of the generation AI is displayed on the display unit 104. The source code displayed on the display unit 104 can be checked by, for example, an operator who performs the work of modifying the Web application, and if there are no problems, the source code can be confirmed as a modern Web application as is.

[0037] <Second embodiment> Next, a software generation system according to a second embodiment of the present invention will be described with reference to Figures 8 to 10. In Figures 8 to 10, parts corresponding to those in Figures 1 to 7 described in the first embodiment are given the same reference numerals, and duplicated explanations will be omitted.

[0038] [Overall process flow of the software generation system] FIG. 8 shows the overall flow of processing performed by the software generation system of this embodiment. In Figure 8, the process flow from obtaining the legacy Web application source code 1a, 1b, 1c, ... to creating prompt 11, through the generation AI processing (steps S11, S12, S13) to obtaining generated source code 18, is the same as the process flow shown in Figure 2.

[0039] This embodiment includes a source code analysis unit 21. The source code analysis unit 21 receives the source codes 1a, 1b, 1c, . . . and the consistency assurance information 14. The source code analysis unit 21 analyzes the source code 1a, 1b, 1c, ... and analyzes source code related to consistency assurance from the source code 1a, 1b, 1c, .... When analyzing the source code related to consistency assurance, the source code analysis unit 21 obtains and executes consistency assurance information 14. The source code analysis unit 21 may acquire a response 12 as a result of executing the process of the generated AI according to the prompt 11 (step S11).

[0040] The consistency assurance information 14 here includes a target prompt, information, and an information source, as shown in FIG. The target prompt is the code associated with the prompt template. The information is about specific processing that requires consistency, such as the "button that generates the screen event," "the screen to transition to," "parameters to pass to the destination," "how the parameters are passed," and "method name of the Service class used in the screen event." The information source can be "legacy" for legacy web apps or "modern" for modern web apps.

[0041] Then, the source code analysis unit 21 obtains a prompt 15 from a portion of the analysis result 22 obtained by the source code analysis unit 21 (such as the target prompts 1-1, 1-2, 1-3, ... shown in Figure 8) and the response 12 resulting from the processing of the generated AI in step S11.

[0042] In addition, the source code analysis unit 21 obtains a prompt 17 that modifies the Component class based on the remaining part of the analysis result 22 obtained by the source code analysis unit 21 (such as the target prompts 2-1, 2-2, ... shown in Figure 8) and the response 16 obtained by processing the generation AI in step S12. Other processes performed by the software generation system of this embodiment shown in FIG. 8 are the same as the process flow shown in FIG.

[0043] [Configuration of software generation system] FIG. 9 shows an example of the configuration of a software generation system 100 according to this embodiment. The software generation system 100 includes a prompt generation unit 110, a consistency processing unit 120′, a source code generation unit 130, and a display unit 104. In this embodiment, the configuration of the consistency processing unit 120′ differs from that of the consistency processing unit 120 in the first embodiment shown in FIG.

[0044] The configuration of the consistency processing unit 120' will be described. The consistency processing unit 120' is the same as the consistency processing unit 120 shown in FIG. 3 in the first embodiment in that it includes a consistency assurance information input unit 121 and a prompt template input unit 122. The consistency processing unit 120 ′ includes a screen generation target input unit 124 , a source code input unit 126 , and a source code analysis unit 21 .

[0045] A list of screens 123 of legacy Web applications is input to the screen-to-be-generated input unit 124. The screen-to-be-generated input unit 124 obtains, from the screen list 123, a screen to be generated 125 related to consistency assurance. A screen 125 to be generated is input to a source code input unit 126. The source code input unit 126 extracts source code related to consistency assurance that constitutes the screen 125 to be generated from a source code database 140, and obtains source code 127 to be analyzed.

[0046] The analysis target source code 127 is sent to the source code analysis unit 21, which analyzes the contents of the source code to obtain an analysis result 22. The analysis result 22 is sent to the prompt generation unit 110. The prompt generator 110 generates the prompt 11 using the analysis result 22, the consistency assurance information 14', and the prompt template 13'. Other processes of the software generation system 100 shown in FIG. 9 are the same as those of the software generation system 100 shown in FIG.

[0047] [Source code analysis example] FIG. 10 shows an example of analysis by the source code analysis unit 21. The source code analysis unit 21 analyzes what kind of screen transition occurs in each function of the source code. For example, as shown in Fig. 10, when the source code to be analyzed is input, the source code analysis unit 21 outputs the "function name," "destination screen," "parameters," and "implementation form" as the analysis results.

[0048] [Effects of the second embodiment] According to the software generation system of this embodiment, as in the first embodiment, when a legacy web app is modified into a modern web app using the generation AI, consistency assurance information is included in the prompt, which is the instruction text for the generation AI, so the web app can be modified to function correctly. Here, the prompt generation unit 110 obtains the source code analysis results from the source code analysis unit 21 and generates a prompt that understands the source code, thereby enabling the generation of a prompt to ensure appropriate consistency based on the analysis of an actual legacy Web application. This makes it possible to more reliably ensure consistency between legacy and modern web apps.

[0049] <Third embodiment> Next, a software generation system according to a third embodiment of the present invention will be described with reference to Figures 11 and 12. In Figures 11 and 12, parts corresponding to those in Figures 1 to 10 described in the first and second embodiments are given the same reference numerals, and duplicated explanations will be omitted.

[0050] [Overall process flow of the software generation system] FIG. 11 shows the overall flow of processing performed by the software generation system of this embodiment. In Figure 8, the process flow from obtaining the source code 1a, 1b, 1c, ... of the legacy Web application and creating the prompt 11 to obtaining the generated source code 18 through the generation AI processing (steps S11, S12, S13) is the same as the process shown in Figure 8.

[0051] In this embodiment, the generated source code 18 obtained by the process by the generation AI in step S13 is added as the source code to be analyzed by the source code analysis unit 21. That is, as shown in FIG. 11, the generated source code 18 (source codes 3a, 3b, 3c, ...) obtained by the processing by the generation AI in step S13 is returned to the source code analysis unit 21, where it is analyzed. The source code analysis unit 21 analyzes the generated source codes 3a, 3b, 3c, . . . and the source codes 1a, 1b, 1c, . . . of the legacy Web application, and obtains an analysis result 22 from both analyses. Other processes performed by the software generation system of this embodiment shown in FIG. 11 are the same as the process flow shown in FIG.

[0052] [Configuration of software generation system] FIG. 12 shows an example of the configuration of a software generation system 100 according to this embodiment. The software generation system 100 includes a prompt generation unit 110, a consistency processing unit 120″, a source code generation unit 130, and a display unit 104, and the configuration of the consistency processing unit 120″ differs from that of the consistency processing unit 120′ shown in FIG. 9 in the second embodiment.

[0053] The configuration of the consistency processing unit 120″ will be explained. The consistency processing unit 120″ is the same as the consistency processing unit 120′ shown in FIG. 9 in the second embodiment in that it includes a consistency assurance information input unit 121, a prompt template input unit 122, a screen to be generated input unit 124, and a source code input unit 126. The consistency processing unit 120 ″ includes a generation order determination unit 128 . A legacy Web application screen list 123 is input to the generation order determination unit 128. The generation order determination unit 128 determines the order of the screens from the input screen list, and obtains a sorted screen list 129.

[0054] The sorted screen list 129 is input to the screen to be generated input unit 124. The screen to be generated input unit 124 obtains the screen to be generated 125 related to consistency assurance from the sorted screen list 129. The source code input unit 126 retrieves source code related to consistency assurance that constitutes the screen 125 to be generated from the source code database 140, and obtains source code 127 to be analyzed. Here, the source code database 140 stores the source code generation result 131 generated by the source code generation unit 130. Therefore, the source code to be analyzed 127 reflects the source code of the modern Web application.

[0055] 12, the consistency assurance information 14′ obtained by the consistency assurance information input unit 121 is also supplied to the source code analysis unit 21. As a result, the source code analysis unit 21 analyzes the source code related to the information necessary for consistency assurance. The analysis results from source code analysis unit 21 are supplied to prompt generation unit 110 . The other configurations of the software generation system 100 shown in FIG. 12 are the same as those of the software generation system 100 shown in FIG.

[0056] [Dependencies and generation order of consistency information] FIG. 13 is a diagram illustrating how appropriate processing is performed by the consistency processing unit 120''. Since consistency assurance information also exists in the source code generated by the source code generation unit 130, the order in which the source code is generated must be determined based on the dependency relationships between classes and screen transition information. For example, if there is a transition relationship between screens, the source code of the transition source screen is required to generate the transition destination screen. Therefore, it is preferable to generate each screen in order from the end of the screen transition relationship.

[0057] In the example shown in FIG. 13, it is assumed that there is a transition from screen A to screen B and a transition from screen B to screen C in the Web application, as shown in the upper part of FIG.

[0058] At this time, as shown in the lower part of Figure 13, the source code generation unit 130 generates source code for screen A in the first step, generates source code for screen B in the next step, and generates source code for screen C in the next step, thereby obtaining generated source code for each screen.

[0059] The dependency relationships between the screens are determined by the framework structure of the modern Web application, and therefore, it is preferable that the source code analysis unit 21 analyzes the framework structure of the modern Web application.

[0060] [Effects of the third embodiment] According to the software generation system of this embodiment, as in the first and second embodiments, when a legacy Web app is modified into a modern Web app by the generation AI, consistency assurance information is included in the prompt, which is the instruction text of the generation AI, so the Web app can be modified to function correctly. Here, a prompt is generated that reflects the generated modern web app, making it possible to generate a more appropriate web app.

[0061] <Modification> It should be noted that the embodiments described so far have been described in detail to clearly explain the present invention, and are not necessarily limited to those having all of the configurations described. For example, in each of the above-mentioned embodiments, the process of modifying a legacy Web application as shown in Figure 1 to make it a modern Web application was described, but the present invention can also be applied to generating a new Web application by changing some of the operations, screens, etc. of a legacy Web application. Also, as shown in FIG. 3, the processing of the generated AI is performed by an external generated AI execution server 191, but this is just one example, and if possible, the processing of the generated AI may be performed by a computer as the software generation system 100.

[0062] In addition, in the configuration diagrams shown in Figures 3, 9, 12, etc., only control lines and information lines that are considered necessary for explanation are shown, and not all control lines and information lines in the product are necessarily shown. In reality, it can be assumed that almost all components are interconnected. Furthermore, the processing flows shown in the flowcharts of FIGS. 4, 5, and 6 are also examples, and as long as the processing results are the same, the order of some of the processing may be changed or multiple processes may be executed simultaneously. Furthermore, in each of the above-described embodiments, the software generation system 100 shown in Figures 3, 9, and 12 is configured as a computer, but in this case, the program to be executed by the computer may be prepared in a storage device (memory or storage) within the computer system, or may be stored on a recording medium such as an external memory, IC card, SD card, optical disk, or the like, and transferred. [Explanation of symbols]

[0063] 1...Legacy Web application program, 1a, 1b, 1c, 3a, 3b, 3c...Source code, 2...P layer, 3...Modern Web application program, 4...P layer, 5...API, 11...Prompt, 12...Response, 13, 13'...Prompt template, 14, 14'...Integrity assurance information, 15...Prompt, 16...Response, 17...Prompt, 18...Generated source code, 21...Source code analysis unit, 22...Analysis results, 100...Software generation system, 101...CPU, 102...Memory, 103...Storage, 104...Display unit, 105...Communication interface, 110...Prompt generation unit, 111...Prompt, 120, 120', 120"...Integrity processing unit, 121...Integrity assurance information input unit, 122...Prompt template input unit, 123...Screen list, 124... Screen to be generated input unit, 125... Screen to be generated, 126... Source code input unit, 127... Source code to be analyzed, 128... Generation order determination unit, 129... Sorted screen list, 130... Source code generation unit, 131... Source code generation result, 140... Source code database, 191... Generation AI execution server

Claims

1. a consistency assurance information input unit for obtaining consistency assurance information, which is information for ensuring consistency between multiple screens or classes of a Web application program; a prompt template input unit for obtaining a prompt template that is an instruction for generating or modifying the Web application program when the Web application program is generated or modified; a prompt generation unit that generates a prompt in which the consistency assurance information is embedded in the prompt template based on the consistency assurance information input to the consistency assurance information input unit; a source code generation unit that generates source code for a portion of the Web application program to be generated or modified based on the prompt generated by the prompt generation unit. Software generation system.

2. the web application program is a legacy web application program that has already been created, a source code analysis unit that analyzes the source code of the legacy Web application program; The prompt generation unit generates a prompt in which the consistency assurance information is embedded, based on the consistency assurance information input to the consistency assurance information input unit, the prompt template input to the prompt template input unit, and the analysis result of the source code analysis unit. The software generation system of claim 1 .

3. The generated source code generated by the source code generation unit is also input to the source code analysis unit and analyzed. The software generation system of claim 2 .

4. The consistency assurance information is data commonly used among different source codes or data that needs to be linked among different source codes. The software generation system of claim 1 .

5. The source code generated by the source code generation unit modifies the presentation layer of the Web application program from one that performs a multi-page application to one that performs a single-page application. The software generation system of claim 1 .

Images