CN117348846A

CN117348846A - Method and device for generating software development requirement document and electronic equipment

Info

Publication number: CN117348846A
Application number: CN202311330576.XA
Authority: CN
Inventors: 倪琳; 吴思遥; 何光宇; 金铸; 陈禹铭
Original assignee: Neusoft Corp
Current assignee: Neusoft Corp
Priority date: 2023-10-13
Filing date: 2023-10-13
Publication date: 2024-01-05

Abstract

The application provides a method and device for generating a software development requirement document and electronic equipment. The method includes the steps of splitting a process from demand to design of software development into a plurality of stages, respectively executing input operation on a first stage and a second stage which are sequentially executed by the plurality of stages, generating demand content corresponding to the first stage and the second stage, and adding the demand content to a corresponding position of a demand document, so that the demand document pointing from demand to design is included in the software development process. In addition, the required content corresponding to the first stage is a first required content, and the required content corresponding to the second stage is a second required content. And the second required content is determined jointly through the second input operation and the first required content output in the first stage, so that the effective series connection of the required content is formed, and the conveying efficiency of the design concept is improved. Thus, when the user wants to know the relation between the requirement and the design, the user only needs to directly read the software development requirement of the existing design from the corresponding position of the requirement document.

Description

Method and device for generating software development requirement document and electronic equipment

Technical Field

The present invention relates to the field of software technologies, and in particular, to a method and an apparatus for generating a software development requirement document, and an electronic device.

Background

The agile software development takes the evolution of the user's demands as the core, and adopts an iterative and progressive method to carry out the software development. The agile software development flow mainly comprises links of demand analysis, product design, development test and the like, namely, the agile software development flow comprises three elements of demand-design-development.

In order to support the management needs from demand tracking to delivery archiving, the unified management of deliverables of each link is often carried out in a warehouse storage mode at the current stage. However, the above mode often uses a demand document from demand to design as a delivery object independent of software development, and the demand document is not incorporated into the software development process, so that an effective series connection between delivery objects is not formed, and it is difficult to efficiently convey the design concept. When the relation between the requirements and the design is required to be known, and whether the existing design can meet the current requirements is analyzed, the mode lacks effective tool support, and related knowledge construction can be performed only through manual exploration, so that great difficulties exist in knowledge transfer and iterative development under the scenes of requirement change analysis, core developer variation and the like.

Disclosure of Invention

The embodiment of the application provides a method and a device for generating a software development requirement document and electronic equipment, wherein the requirement document from requirement to design is incorporated in a development tool, so that effective series connection among deliverables is formed, and a design idea is efficiently conveyed. In addition, when the relation between the demand and the design is known, and whether the existing design can meet the current demand is analyzed, difficulties in knowledge transfer and iterative development in the scenes of demand change analysis, core developer change and the like are reduced.

In a first aspect, an embodiment of the present application provides a method for generating a software development requirement document. In the software development, a process from demand to design is split into a plurality of phases including at least a first phase and a second phase that are sequentially executed, the method comprising:

generating first demand content corresponding to the first stage in response to a first input operation of the software development demand of the first stage, wherein the first demand content is used for representing the software development demand content corresponding to the first input operation in the first stage;

responding to a second input operation of the software development requirement of the second stage, and generating second requirement content corresponding to the second stage according to the first requirement content, wherein the second requirement content is used for representing the software development requirement content corresponding to the second input operation in the second stage;

And adding the first demand content to a first position of the first demand document, and adding the second demand content to a second position of the first demand document to obtain a second demand document.

Optionally, when the first stage generates a user story map stage, the generating, in response to a first input operation of a software development requirement of the first stage, first requirement content corresponding to the first stage includes:

responding to the operation of inputting a target business scene requirement on a first prompt word template, and matching a target flow rule from a vector knowledge base of a pre-stored business scene and a business flow rule, wherein the target flow rule is a business flow rule corresponding to the target business scene requirement; the first prompting word template is a display template for inputting the target business scene requirement in the stage of generating the user story map;

and generating a user story map according to the target flow rule, and taking the user story map as first required content corresponding to the first stage, wherein the user story map comprises a plurality of user stories to be developed corresponding to the software development, and the user stories are used for describing the software development requirements of a user.

Optionally, the method further comprises:

splitting the software development requirement of the user story map generation stage into a plurality of first sub development requirements; the plurality of first sub-development requirements includes the target business scenario requirement;

reconstructing the first prompt word template by using the plurality of first sub-development requirements, so that the reconstructed first prompt word template comprises the plurality of sub-development requirements;

the matching the target process rule from the vector knowledge base of the pre-stored service scene and the service process rule in response to the operation of inputting the target service scene requirement on the first prompt word template comprises the following steps:

and responding to the operation of inputting the target business scene on the reconstructed first prompt word template, and matching the target flow rule from the vector knowledge base.

Optionally, when the second stage generates a user story stage, the generating, in response to a second input operation of the software development requirement of the second stage, second requirement content corresponding to the second stage according to the first requirement content includes:

responding to the operation of inputting the target application theme requirement on a second prompt word template, generating a user story corresponding to the target application theme requirement according to the user story map, and taking the user story corresponding to the target application theme requirement as second requirement content corresponding to the second stage, wherein the second prompt word template is a display template for inputting the target application theme requirement in the stage of generating the user story.

Optionally, the method further comprises:

splitting the software development requirement of the user story generation stage into a plurality of second sub development requirements; the plurality of second sub-development requirements includes the target application theme requirement;

reconstructing the second prompt word template by using the plurality of second sub-development requirements, so that the reconstructed second prompt word template comprises the plurality of sub-development requirements;

the operation of responding to the input of the target application theme requirement on the second prompt word template generates the user story corresponding to the target application theme requirement according to the user story map, and comprises the following steps:

and responding to the operation of inputting the target application theme requirement on the reconstructed second prompt word template, and generating a user story corresponding to the target application theme requirement according to the user story map.

Optionally, after the generating the first demand content corresponding to the first stage, the method further includes:

responding to a first modification operation of the first demand content, and acquiring modified first demand content;

the generating, according to the first demand content, second demand content corresponding to the second stage includes:

Generating the second demand content corresponding to the second stage according to the modified first demand content;

the adding the first demand content to a first location of a first demand document includes:

and adding the modified first demand content to a first position of the first demand document.

Optionally, the method further comprises:

responding to a second modification operation of the second demand content, and acquiring modified second demand content;

the adding the second demand content to the second location of the first demand document includes:

adding the modified second demand content to the second location of the first demand document.

Optionally, the plurality of stages further includes a third stage and a fourth stage, and the method further includes:

responding to a third input operation of the software development requirement of the third stage, and generating third requirement content according to the second requirement content, wherein the third requirement content is used for representing the requirement content of the software development corresponding to the third input operation in the third stage;

responding to a fourth input operation of the software development requirement of the fourth stage, and generating fourth requirement content according to the third requirement content, wherein the fourth requirement content is used for representing the requirement content of the software development corresponding to the fourth input operation in the fourth stage;

Adding the first requirement content to a first position of a first requirement document, and adding the second requirement content to a second position of the first requirement document to obtain a second requirement document, wherein the method comprises the following steps:

and adding the first demand content to a first position of the first demand document, adding the second demand content to a second position of the first demand document, adding the third demand content to a third position of the first demand document, and adding the fourth demand content to a fourth position of the first demand document to obtain a second demand document.

Optionally, the third stage generates a unified modeling language activity diagram for the stage, the fourth stage generates an application program interface stage,

the generating third requirement content according to the second requirement content in response to the third input operation on the third stage software development requirement comprises the following steps:

responding to the operation of inputting the target business scene requirement on a third prompt word template, generating a unified modeling language activity diagram according to the second requirement content generated in the second stage, and taking the unified modeling language activity diagram as the third requirement content in the third stage, wherein the unified modeling language activity diagram is used for describing the activity flow of the second requirement content; the third prompting word template is a display template for inputting the target business scene requirement in the generation of the unified modeling language activity diagram stage;

The generating fourth requirement content according to the third requirement content in response to the fourth input operation on the fourth stage software development requirement comprises the following steps:

responding to the operation of inputting a target activity requirement on a fourth prompt word template, generating an application program interface corresponding to the target activity requirement according to the same modeling language activity diagram, using the application program interface to communicate a server side with a client side, and using the application program interface as fourth requirement content of a fourth stage; the fourth prompting word template is a display template for inputting the target activity requirement in the generation application program interface stage.

In a second aspect, an embodiment of the present application provides a generating device of a software development requirement document, in the software development, a process from requirement to design is split into a plurality of phases, where the plurality of phases includes at least a first phase and a second phase that are sequentially executed, and the device includes:

a first generation unit, configured to generate first required content corresponding to the first stage in response to a first input operation of a software development requirement for the first stage, where the first required content is used to represent required content of software development corresponding to the first input operation in the first stage;

A second generating unit, configured to generate, according to the first demand content, second demand content corresponding to the second stage in response to a second input operation for a software development demand of the second stage, where the second demand content is used to represent the demand content of the software development corresponding to the second input operation in the second stage;

and the adding unit is used for adding the first requirement content to a first position of the first requirement document, and adding the second requirement content to a second position of the first requirement document to obtain a second requirement document.

In a third aspect, embodiments of the present application provide an electronic device including a memory and a processor;

the memory is coupled with the processor;

the memory stores program instructions that, when executed by the processor, cause the electronic device to perform the method of any of the first aspects.

The beneficial effects are that:

the embodiment of the application provides a method and device for generating a software development requirement document and electronic equipment. When the method is executed, the process from the requirement to the design of the software development is split into a plurality of stages, input operation is respectively executed on a first stage and a second stage which are sequentially executed by the plurality of stages, the corresponding requirement content is generated, and then the requirement content is added to the corresponding position of the requirement document, so that the requirement document pointing from the requirement to the design is included in the process of the software development. In addition, the required content corresponding to the first stage is a first required content, and the required content corresponding to the second stage is a second required content. The second demand content is determined by the second input operation and the first demand content output in the first stage, so that the first demand content and the second demand content are effectively connected in series, and the conveying efficiency of the design concept is improved. Therefore, when the user wants to know the relation between the requirement and the design, and analyze whether the existing design can meet the current requirement, the user only needs to directly read the software development requirement of the existing design from the corresponding position of the requirement document. In summary, compared with the method of constructing related knowledge through manual exploration, the method and the device can reduce difficulty in knowledge transfer and iterative development in the scenes of demand change analysis, core developer variation and the like.

Drawings

FIG. 1 is a flowchart of a method for generating a software development requirement document according to an embodiment of the present application;

FIG. 1A is a schematic diagram of obtaining multiple stages after splitting according to an embodiment of the present application;

FIG. 2 is a schematic diagram of a first display of a plurality of alert word templates according to an embodiment of the present application;

FIG. 3 is a schematic diagram of a fourth alert word template according to an embodiment of the present disclosure;

FIG. 4 is a schematic diagram of two kinds of user confirmation first requirement content according to an embodiment of the present application;

FIG. 5 is a flowchart of a second method for generating a software development requirement document according to an embodiment of the present application;

FIG. 6 is a second schematic illustration of a plurality of alert templates provided in an embodiment of the present application;

fig. 7 is a schematic diagram of an implementation principle of LLM based on a plug-in vector knowledge base according to an embodiment of the present application;

FIG. 8 is a schematic diagram of displaying a demand document according to an embodiment of the present disclosure;

FIG. 9 is a flowchart of a third method for generating a software development requirement document according to an embodiment of the present application;

FIG. 10 is a schematic illustration of a disassembly provided in an embodiment of the present application;

fig. 11 is a schematic structural diagram of a generating device 110 of a software development requirement document according to an embodiment of the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.

The term "and/or" is herein merely an association relationship describing an associated object, meaning that there may be three relationships, e.g., a and/or B, may represent: a exists alone, A and B exist together, and B exists alone.

The terms first and second and the like in the description and in the claims of embodiments of the present application are used for distinguishing between different objects and not necessarily for describing a particular sequential order of objects. For example, the first and second demand contents are only used to indicate demand contents of different phases, not to describe a specific order of demand contents.

In the embodiments of the present application, words such as "exemplary" or "such as" are used to mean serving as examples, illustrations, or descriptions. Any embodiment or design described herein as "exemplary" or "for example" should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "such as" is intended to present related concepts in a concrete fashion.

In the description of the embodiments of the present application, unless otherwise indicated, the meaning of "a plurality" means two or more. For example, the plurality of processing units refers to two or more processing units; the plurality of systems means two or more systems.

The agile software development is software development which takes the evolution of the user's demands as the core and adopts an iterative and progressive method. The agile software development does not pursue the early perfect design, but develops the core function of the product in a very short period, and is continuously and iteratively upgraded according to new requirements in the subsequent generation period to perfect the product. Therefore, compared with other software development, such as waterfall type software development, agile software development has the characteristic of being short in development period and adaptable to various requirement changes, and is now the first choice of many users and enterprises.

The agile software development flow comprises links of demand analysis, product design, development test and the like, namely three elements of demand-design-development. Because agile software development procedures emphasize that "working software is above an exhaustive document," users put most effort in the iterative development process, but have inadequate visibility into the process from demand tracking to delivery management. At present, a warehouse storage model is often adopted to uniformly manage deliverables from demand to development. That is, deliverables from demand to design, such as product demand models, user Interface (UI) design scripts, development plans, application program interfaces, and review campaigns of links, are stored in a designated area for unified management. However, the above modes do not incorporate the requirement documents from requirement to design into the software development process, and the deliverables are independent and not effectively connected in series, so that it is difficult to efficiently convey the design concept. When the relation between the requirements and the design is required to be known, and whether the existing design can meet the current requirements is analyzed, the mode lacks effective tool support, and related knowledge construction can be performed only through manual exploration, so that great difficulties exist in knowledge transfer and iterative development under the scenes of requirement change analysis, core developer variation and the like.

In view of the above, the embodiments of the present application provide a method for generating a software development requirement document, by splitting a process from requirement to design into a plurality of stages, performing an input operation on a first stage and a second stage sequentially performed on the plurality of stages, respectively, generating requirement content corresponding thereto, and then adding the requirement content to a corresponding position of the requirement document, thereby incorporating the requirement document from requirement to design in the software development process. In addition, the required content corresponding to the first stage is a first required content, and the required content corresponding to the second stage is a second required content. The second demand content is determined jointly by the second input operation and the first demand content generated in the first stage. Therefore, the first demand content and the second demand content are effectively connected in series, and the conveying efficiency of the design concept is improved. Therefore, when a user wants to know the relation between the requirement and the design, and whether the existing design can meet the current requirement is analyzed, only the software development requirement of the existing design is needed to be directly read from the corresponding position of the requirement document. Therefore, compared with a mode of constructing related knowledge through manual exploration, the embodiment of the application can reduce the difficulty of knowledge transfer and iterative development in the scenes of demand change analysis, core developer change and the like.

The following describes the method for generating the software development requirement document in full detail according to the embodiment of the present application with reference to the accompanying drawings.

Example 1

Referring to fig. 1, a flowchart of a method for generating a first software development requirement document according to an embodiment of the present application is provided. The execution subject of the method can be a generating device of a software development requirement document, and the generating device can be a software development platform. The software development platform comprises a management system (hereinafter referred to as management system) for software development requirements and a display interface (hereinafter referred to as display interface) connected with the management system, wherein the method specifically comprises the following steps:

s11: the first demand content is generated in response to a first input operation of the first stage software development demand.

In the embodiment of the application, firstly, a user or a generating device of a software development requirement document splits the process from requirement to design into a plurality of stages. Exemplary description: referring to fig. 1A, a schematic diagram of a plurality of stages obtained after splitting is provided in an embodiment of the present application. As shown in fig. 1A, the process from demand to design may be split into four phases that execute sequentially, respectively, to generate a user story map phase, to generate a user story phase, to generate a unified modeling language (Unified Modeling Language, UML) activity diagram phase, and to generate an application program interface phase.

In an embodiment of the present application, the plurality of stages includes at least a first stage and a second stage that are sequentially executed. For example, when the first stage is a user story map generation stage, the second stage is a user story generation stage, when the first stage is a user story generation stage, the second stage is a UML activity diagram generation stage, etc., or when the first stage is a UML activity diagram generation stage, the second stage is an application program interface generation stage. The specific stage forms of the first stage and the second stage are not limited in this application.

The software development requirement refers to requirements of functions, indexes and the like which need to be met by a software product to be developed. In the embodiment of the application, the software development requirement may be a requirement including a target service scene, an application theme and the like. The user can conduct agile software development according to the software development requirement.

In the embodiment of the application, the user may first perform a first input operation on the display interface, and the management system generates the first required content corresponding to the first stage in response to the first input operation. The first input operation refers to an operation of inputting a software development requirement of the first stage on a display interface corresponding to the first stage.

In one possible implementation, the display interface corresponding to the first stage has an input box, and the user may perform the first input operation in the input box.

Illustratively, when the first stage is a user story map stage, the input box of the display interface corresponding to the first stage is a first alert word template. The first prompt word template is a display template for inputting the requirements of the target business scene in the stage of generating the user story map. Referring to fig. 2, a first display schematic diagram of a plurality of alert word templates according to an embodiment of the present application is provided. And (a) in fig. 2, a display interface schematic diagram of a first prompt word template corresponding to a user story map stage is generated. A first input operation for at least one of a target business scenario requirement and an input target business process requirement may be performed on the first prompt word template.

Illustratively, when the first stage is a user story stage, the input box of the display interface corresponding to the first stage is a second alert word template. The second prompting word template is a display template for inputting the theme requirement of the target application in the stage of generating the user story. Fig. 2 (b) is a schematic diagram of a display interface for generating a second alert word template corresponding to a user story stage. A first input operation for the target application theme requirement and/or the input target business scenario requirement may be performed on the second hint word template.

Illustratively, when the first stage is a stage of generating the UML activity diagram, the input box of the display interface corresponding to the first stage is a third alert word template. The third prompting word template is a display template for inputting the requirements of the target business scene in the stage of generating the UML active diagram. As shown in fig. 2 (c), a schematic diagram of a display interface of a third alert word template corresponding to the UML activity diagram stage is generated. A first input operation for at least one of a target business scenario requirement and an input target business process requirement may be performed on the third prompt word template.

It should be noted that the first stage is only schematically shown, and the first stage described in the embodiments of the present application may be other stages, and may be adjusted by those skilled in the art as required.

In one possible implementation, the management system generates the corresponding first required content according to a first input operation on the display interface. The first required content refers to required content of software development corresponding to the first input operation in the first stage.

For example, when the first input operation is to input at least one of the target business scenario requirement or the target business process requirement in the first prompt word template, the first requirement content is a user story map corresponding to the target business scenario requirement. For example, if the target business scene requirement is a medical scene requirement, the first requirement content is a user story map corresponding to the medical scene requirement.

The user story map refers to a map formed by arranging all user stories to be developed corresponding to the target business scene requirements according to importance, execution priority and the like. The user story map can intuitively display all user stories to be developed and display the logical relations among all the user stories to be developed. Thus, facilitating software development by the user.

A user story refers to a particular task or goal that a user desires to accomplish, describing the user's software development needs. For example, as an online shopper, I want to view detailed information of the merchandise and user ratings in order to make a purchase decision.

For example, when the first input operation is to input a target application theme requirement in the second prompt word template and/or input a target service scene requirement, the first requirement content refers to a user story corresponding to the target application theme requirement and corresponding to the target service scene requirement. For example, if the target application theme requirement is an oral cavity and the target service scene requirement is a medical scene, the first requirement content is a user story corresponding to oral cavity medical treatment.

Illustratively, when the first input operation inputs the target business scenario requirement at the third hint word template, and/or inputs the language or the like used for writing, the first requirement content refers to the UML activity diagram corresponding to the target business scenario requirement. The UML activity diagram is used to describe the activity flow of a user story corresponding to a target application theme requirement.

In one possible implementation, the generating means may generate the first required content based on a large language model (Large Language Model, LLM) from the input content of the first input operation. In the embodiment of the application, LLM refers to LLM trained through a large amount of text data related to a software project to be developed. Therefore, the generated first demand content is more fit with the demands of the user, and the user experience is improved.

S12: in response to a second input operation to the second stage software development requirement, second requirement content is generated.

In this embodiment of the present application, the user may perform a second input operation on the display interface, and the management system generates the second required content corresponding to the second stage based on the second input operation and the first required content generated in the first stage. Thus, the first demand content and the second demand content are effectively associated, and the conveying efficiency of the design concept is improved.

The second input operation is an input operation executed on the display interface corresponding to the second stage. In the embodiment of the present application, the second input operation refers to an operation of inputting a software development requirement of the second stage.

In one possible implementation, the display interface corresponding to the second stage has an input box, and the user may perform the second input operation on the input box. The input box may be a prompt word template.

Illustratively, when the second stage is a user story stage, the input box of the display interface corresponding to the second stage is a second prompt word template. The second input operation may be an operation of inputting the target application theme requirement in the second prompt word template and/or inputting the target service scene requirement.

Optionally, when the second stage is a stage of generating the UML activity diagram, the input box of the display interface corresponding to the second stage is a third prompting word template. The second input operation may be an operation of inputting a target service scene requirement in the third prompting word template and/or inputting a language adopted by writing, etc.

Optionally, when the second stage is the generation application program interface stage, the input box of the display interface corresponding to the second stage is the fourth prompting word template. The fourth hint word is aimed at generating a display template for the application program interface to input the fourth stage software development requirement. Wherein the second input operation may be to input the target activity requirement in the fourth prompt word template and/or the application program interface format requirement adopted by the target activity. Referring to fig. 3, a fourth prompting word template schematic diagram provided in this embodiment of the present application is provided, where a user may directly input a target service scene requirement at a position (1) in the display interface shown in fig. 3, and input an application program interface format requirement adopted by a target activity at a position (2) in the user interface shown in fig. 3.

It should be noted that the second stage is only schematically shown, and the second stage described in the embodiments of the present application may be other stages, and may be adjusted by those skilled in the art as required.

In this embodiment of the present application, the second required content corresponds to a second input operation, and the second required content refers to a required content of software development corresponding to a second output operation in a second stage.

For example, when the second input operation is to input the target application theme requirement in the second prompt word template and/or input the target service scene requirement, the second requirement content refers to a user story corresponding to the target application theme requirement and corresponding to the target service scene requirement. For example, if the target application theme requirement is an oral cavity and the target service scene requirement is a medical scene, the second requirement content is a user story corresponding to oral cavity medical treatment.

Illustratively, when the second input operation is an operation of inputting a target business scenario requirement in the third prompt word template, and/or inputting a language adopted by writing, etc., the second requirement content refers to a UML activity diagram corresponding to the target business scenario requirement.

Illustratively, when the second input operation is to input the target service scene requirement on the fourth prompting word template and/or the application program interface format requirement adopted by the target activity, the second requirement content refers to an application program interface corresponding to the target service scene and the application program interface format requirement adopted by the target activity. The application program interface is used for realizing communication between the server and the client, namely, communication between a product for realizing software development and the client.

In one possible implementation, the management system may generate the second demand content based on a large language model (Large Language Model, LLM) from the input content of the second input operation. In the embodiment of the application, LLM refers to LLM trained through a large amount of text data related to a software project to be developed. Therefore, the generated second demand content is more fit with the demands of the user, and the user experience is improved.

S13: and adding the first demand content to a first position of the first demand document, and adding the second demand content to a second position of the first demand document to obtain a second demand document.

The requirements document is used to store deliverables for multiple phases of software development requirements splitting. The requirement document can be located in the management system or outside the management system, but is directly related to the management system, and the document position of the requirement document is not particularly limited in the application.

In an embodiment of the present application, the first requirement document and the second requirement document represent a first location in the same document, and/or whether the second location has the requirement content added. The first position of the first demand document is not added with the first demand content, and the second position of the first demand document is not added with the second demand content, wherein the second demand document refers to the first demand document with the first demand content added at the first position and the second demand content added at the second position.

For example, after the first demand content is generated, the first demand content may be directly added to the first location of the first demand document according to a preset mapping relationship between the first demand content and the first location in the first demand document. Therefore, the mapping relation between the position and the required content is used for avoiding unordered addition of the required content in the required document, and the readability and the storage accuracy of the required document are facilitated.

After the first demand content is generated, the first demand content is directly sent to the display interface for display, and the user confirms whether the first demand content displayed on the display interface is correct. For example, if the displayed first demand content meets the user demand, the first demand content displayed on the display interface is confirmed to be correct. If the displayed first demand content does not meet the user demand, confirming that the first demand content displayed on the display interface is incorrect.

Referring to fig. 4, a schematic diagram of confirming first requirement content for two users according to an embodiment of the present application is provided. Fig. 4 (a) is a schematic diagram of a user confirmation first requirement content provided in an embodiment of the present application. When the user confirms that the first required content is correct, the user can click a confirm button to perform subsequent operations. When the user considers that the first required content is incorrect, the user can directly perform the first modification operation on the first required content, such as modifying the first required content, or adding the first required content, or subtracting the first required content, etc. After the modification is completed, the "confirm" button may be clicked. The management system may add the modified first demand content to the first location of the first demand document according to a preset mapping relationship between the first demand content and the first location in the first demand document. Therefore, the first demand content result stored in the first position of the first demand document is more accurate and meets the demands of the user by means of reconfirming and modifying again. And the operation is convenient, and the user experience is good.

Illustratively, the first place adds a first demand document of the first demand content, which may be presented to the user via the display interface. Fig. 4 (b) is a schematic diagram of confirmation of the first requirement content for another user according to the embodiment of the present application. The user may directly perform a first modification operation on the first required content on the first location document in which the first required content is added at the first location, for example, modifying the first required content, or adding the first required content, or subtracting the first required content, etc. After editing is completed, a save button, or other buttons, may be clicked, corresponding to modifying the generated first desired content. Therefore, the method can intuitively enable the user to determine whether the required document added with the first required content is accurate or not, and improve the user experience.

In this embodiment of the present application, after the second demand content is generated, the second demand content may be directly added to the second location of the first demand document according to a preset mapping relationship between the second demand content and the second location in the first demand document. Therefore, the mapping relation between the position and the required content is used for avoiding unordered addition of the required content in the required document, and the readability and the storage accuracy of the required document are facilitated.

After the second demand content is generated, the second demand content is directly sent to the display interface for display, and the user confirms whether the second demand content displayed on the display interface is correct. For example, when the displayed second requirement content meets the requirement of the user, the second requirement content is confirmed to be correct, and the 'confirm' button is clicked to execute the subsequent operation. When the displayed second demand content does not meet the user demand, determining that the second demand content is incorrect, and the user can directly perform a second modification operation on the second demand content of the display interface, for example, modifying the second demand content, or adding the second demand content, or reducing the second demand content, etc. After the modification is completed, the "confirm" button may be clicked. The management system may add the modified second demand content to the second location of the first demand document according to a preset mapping relationship between the second demand content and the second location in the first demand document. Therefore, the second requirement content result stored in the second position of the first requirement document is more accurate and meets the requirement of the user by means of reconfirming and modifying again. And the operation is convenient, and the user experience is good.

For example, the first requirement document with the second requirement content added at the second position can be presented to the user through the display interface, and the user can directly perform the second modification operation on the second requirement content on the first requirement document, such as modifying the second requirement content, or adding the second requirement content, or reducing the second requirement content, etc. After the second modification is completed, a save button, or other buttons, may be clicked, corresponding to the second desired content generated by the modification. Therefore, through the mode of displaying the demand document, a user can intuitively determine whether the second demand content stored at the second position of the first demand document is accurate or not, the operation is convenient, and the user experience is improved.

The embodiment of the application can firstly add the first requirement content to the first position of the first requirement document, and the requirement document after adding the second requirement content to the second position of the first requirement document is used as the second requirement document. The first demand content may be added to the first location of the first demand document at the same time, and the demand document after the second demand content is added to the second location of the first demand document may be used as the second demand document. The second demand content may also be added to the second location of the first demand document first, and the demand document after the first demand content is added to the first location of the first demand document may be used as the second demand document. The present application is not particularly limited.

It should be noted that, the first stage may be a first stage of a development cycle, and may be other stages, and the application is not particularly limited, and the second stage may be a last stage of the development cycle, and may also be other stages, and the application is not particularly limited.

The embodiment of the application provides a method for generating a software development demand document, which comprises the steps of splitting a process from demand to design in software development into a plurality of stages, respectively executing input operations on a first stage and a second stage which are sequentially executed by the plurality of stages, generating demand content corresponding to the first stage and the second stage, and adding the demand content to a corresponding position of the demand document, so that the demand document pointing from demand to design is incorporated in the software development process. In addition, the required content corresponding to the first stage is a first required content, and the required content corresponding to the second stage is a second required content. The second demand content is determined jointly by the second input operation and the first demand content generated in the first stage. Therefore, the first demand content and the second demand content are effectively connected in series, and the conveying efficiency of the design concept is improved. Therefore, when a user wants to know the relation between the requirement and the design, and whether the existing design can meet the current requirement is analyzed, only the software development requirement of the existing design is needed to be directly read from the corresponding position of the requirement document. Therefore, compared with a mode of constructing related knowledge through manual exploration, the embodiment of the application can reduce the difficulty of knowledge transfer and iterative development in the scenes of demand change analysis, core developer change and the like.

Example two

As shown in connection with FIG. 1A, the software development can be split from demand to design into a first phase, a second phase, a third phase, and a fourth phase. The method comprises a first stage for generating a user story map stage, a second stage for generating a user story stage, a third stage for generating a UML activity diagram stage and a fourth stage for generating an application program interface stage. The software development method provided in the embodiment of the present application is described in detail below based on the above four stages. Fig. 5 is a flowchart of a second method for generating a software development requirement document according to an embodiment of the present application, and fig. 6 is a second display schematic diagram of a plurality of alert word templates according to an embodiment of the present application.

Referring to fig. 5, a flowchart of a second method for generating a software development requirement document according to an embodiment of the present application is provided. The method specifically comprises the following steps:

s51: and responding to a first input operation required by the target business scene on the first prompt word template, generating a user story map and displaying.

In the embodiment of the present application, the first input operation refers to an operation including, but not limited to, inputting a target business scenario requirement, for example, the first input operation may also be an operation of inputting a target business process requirement.

An exemplary illustration, as shown in fig. 6 (a), is a schematic diagram of a first alert word template. The first hint word template may be: please draw a user story map. The user may enter the target traffic scenario requirements at the location of "".

The management system can generate a user story map according to the target business scene requirement and send the user story map to the display interface for display. For example, the management system may match, according to the target service scenario requirement, a target service flow rule from a vector knowledge base that stores service scenarios and service flows in advance, where the target service flow rule refers to a rule of a target service flow corresponding to the target service scenario. After the matching is successful, the management system generates a user story map corresponding to the flow rule based on the LLM according to the flow rule. That is, the embodiment of the application can use LLM of the plug-in vector knowledge base to generate the user story map. In this way, the user story map generated by vector knowledge base association is facilitated to be controllable in the vertical field of vector knowledge support. In addition, by using the LLM mode, users can pay attention to the development stage and meanwhile, clear demand expression and detailed design documents are ensured.

Referring to fig. 7, a schematic diagram of an implementation principle of LLM based on a plug-in vector knowledge base according to an embodiment of the present application is provided. The method comprises the following steps: firstly, a business scene from the requirement to the design process is developed by software, business process rules are stored in a local knowledge file, then the file is loaded and read, the read file is subjected to text segmentation, the text after segmentation is subjected to text vectorization, and the text after vectorization is stored in a vector knowledge base, so that a vector knowledge base for storing the business scene and the business process rules is obtained. And then inputting the problem through a display interface, vectorizing the problem, and carrying out vector matching in a vector knowledge base to find out k most similar texts, namely the most similar Top-k texts. Top-k text is added as context to the alert word template along with the question. Namely, the Top-k text searched in the knowledge base forms known information context, and the prompting word template is adjusted to { known information }: context, problem: the LLM generates the required content in the constraint range according to the adjusted prompt word template. The constraint scope refers to the content of the request aiming at the query in the known information scope. This helps to generate controllable demand content.

S52: and confirming whether the story map of the user is correct. If the user story map is confirmed to be correct, S54 is executed, and if the user story map is confirmed to be incorrect, S53 is executed.

In the embodiment of the application, the user can confirm whether the user story map is correct according to the user story map displayed on the display interface. The user story map accurately means that the user stories in the user story map are obtained to correspond to the target business scenes, and the obtained user story map is attached to the user demands.

If the user story map is confirmed to fit the user requirement, and the user story in the user story map corresponds to the target service scene, the generated user story map is indicated to be correct. Otherwise, the generated user story map is indicated to be incorrect.

S53: the user story in the user story map is modified.

When the user story map is confirmed to be incorrect, the user can directly modify the user stories in the user story map on the display interface. The modification mode can be to replace the original user story in the user story map, delete the original user story, add a new user story and the like.

By modifying or confirming the generated user story map, the method and the system are beneficial to improving the accuracy of the generated user story map and enabling the generated user story map to be more fit with user requirements.

S54: the user story map or the modified user story map is added to the first location of the first demand document.

When the user story map is confirmed to be correct, the user story map is added to the first location of the first demand document. Exemplary, referring to fig. 8, a schematic display diagram of a requirement document is provided in an embodiment of the present application. Fig. 8 (a) is a schematic diagram of the first requirement document. The user story map or the modified user story map is added at the first location 801, resulting in the result shown in fig. 8 (b).

S55: and responding to a second input operation on the target application theme requirement on a second prompt word template, generating a user story and displaying.

In the embodiment of the application, the second input operation is an operation including, but not limited to, inputting a target application theme requirement.

The management system generates a user story corresponding to the target application theme requirement based on the user story map or the modified user story map described in step S54 according to the input target application theme requirement. Specifically, if the content of the first position of the first requirement document is a user story map, the management system generates a user story corresponding to the target application theme requirement based on the user story map. If the content of the first position of the first requirement document is a modified user story map, the management system generates a user story corresponding to the target application theme requirement based on the modified user story map.

Optionally, the second input operation may be an operation of inputting a target application theme requirement and inputting a target service scenario requirement.

For example, see (b) in fig. 6 for a second cue word template schematic. The second prompting word template is as follows: and writing the user stories according to the scenes and the application topics only according to the generated user story map. Scene: * Theme: * ****. A user may input a target traffic scenario requirement at a "location" and input a target application theme requirement at a "location".

In the embodiment of the application, the management system generates the user story corresponding to the target application theme requirement based on the user story map or the modified user story map according to the target service scene requirement and the target application theme requirement input on the second prompt word template. And the management system sends the generated user story to a display interface for display.

Alternatively, the management system may generate the user story based on LLM from the user story map above, as well as the target business scenario requirements and target application theme requirements. Generating user stories through LLM helps users to keep their efforts in the development phase and ensures clear demand expressions and detailed documentation. Furthermore, based on the user story map above, the content that facilitates LLM generation is controllable.

In addition, other manners may be adopted in the embodiments of the present application to generate the user story, which is not specifically limited in the present application.

S56: confirm whether the user's story is correct. If the user story is confirmed to be correct, S58 is executed, and if the user story map is confirmed to be incorrect, S57 is executed.

In the embodiment of the application, the user can confirm whether the user story is correct according to the user story displayed on the display interface.

The user story is exactly that the obtained user story corresponds to the target application theme, and the obtained user story fits the user requirement. If the user story is confirmed to fit the user requirement, and the user story in the user story corresponds to the target service scene, the generated user story is indicated to be correct. Otherwise, the generated user story is indicated as incorrect.

Therefore, through interaction judgment, the accuracy of the generated user stories is improved, and the generated user stories are attached to the development requirements of users.

S57: the user story is modified.

When the user story is confirmed to be incorrect, the user can perform modification operation on the user story of the display interface directly. The specific modification operation may be to replace the original user story, or to delete the original user story, etc.

According to the embodiment of the application, the accuracy of the generated user story can be improved by modifying or confirming the generated user story, so that the generated user story is more fit with the user requirement.

S58: the user story or the modified user story is added to a second location referring to the first demand document.

When the user story is confirmed to be correct, the user story is added to the second location of the first demand document. When the user story is confirmed to be incorrect, the modified user story is added to the second location of the first demand document. Illustratively, referring to fig. 8 (c), a user story or a post-modified user-derived schematic diagram of the demand document is added to the second location 802 of the first demand document.

S59: in response to entering a third input operation on a third hint word template, a UML activity diagram is generated and presented.

In the embodiment of the present application, the third input operation refers to an operation of inputting the target business scenario requirement in the third stage.

For example, referring to fig. 6 (c), a third hint word template is shown, where the third hint word module is: from the generated user stories, a UML activity map expressed in the build unified modeling language (Plant Unified Modeling Language, plant UML) language is composed. The user may enter the target traffic scenario requirements at the "location".

The third prompting word template shown in (c) in fig. 6 is only schematically represented, and the third prompting word template in the embodiment of the present application may also be in other template forms, for example, a user may customize a language written by inputting. The template form of the third hint word template is not particularly limited in this application.

In the embodiment of the present application, the management system may generate the UML activity diagram corresponding to the target service scene requirement according to the target service scene requirement input by the third input operation, based on the user story in step S58, or the modified user story.

Specifically, when the content added at the second position of the first requirement document is a user story, the management system generates a UML activity diagram corresponding to the target business scene requirement based on the user story. When the content added at the second position of the third requirement document is the modified user story, the management system generates a UML activity diagram corresponding to the target business scene requirement based on the modified user story, so that the accuracy of the generated UML activity diagram is improved.

Alternatively, the management system may generate a UML activity map corresponding to the target business scenario requirements based on the user stories using LLM. UML activity diagrams generated by LLM help users to keep their efforts in the development phase and ensure clear demand expression and detailed documentation. Furthermore, based on the user stories above, the content that facilitates LLM generation is controllable.

The embodiment of the present application may also generate the UML activity map in other manners, and the embodiment of the present application is not particularly limited.

S510: and confirming whether the generated UML activity diagram is correct. If the UML activity pattern is confirmed to be correct, S512 is performed, and if the UML activity pattern is confirmed to be incorrect, S511 is performed.

Illustratively, the UML activity diagram includes a start node, an activity node, a judgment node, an end node, and the like. The user can determine whether the UML activity diagram corresponds to the target service scene or not according to the content of each node, whether a preset language expression is adopted or not, and the like.

When the user determines that the UML activity diagram corresponds to the target business scene requirement and adopts a preset language expression, the generated UML activity diagram can be determined to be correct. Otherwise, determining that the generated UML activity diagram is incorrect.

S511: the UML activity diagram is modified.

When the user confirms that the UML activity diagram is incorrect, contents such as a start node, an activity node, a judgment node, an end node and the like in the UML activity diagram can be modified on the display interface. The modification mode can be to replace original node content, delete original node content or new node content, etc. This ensures that the UML activity diagram added to the requirements document is correct.

S512: the UML activity diagram or the modified UML activity diagram is added to a third location of the first demand document.

When the UML activity diagram is confirmed to be correct, the UML activity diagram is added to a third location of the first requirement document. When the UML activity diagram is confirmed to be incorrect, the modified UML activity diagram is added to a third location of the first requirements document. Illustratively, referring to (d) in fig. 8, a UML activity diagram or a schematic diagram of the demand document obtained after the modified UML activity diagram is added to the third location 803 of the first demand document.

S513: and responding to a fourth input operation on the target activity requirement on a fourth prompt word template, generating an application program interface and displaying.

In this embodiment of the present application, the fourth input operation may specifically be an operation of inputting a target service scenario requirement and a target activity requirement.

Referring to fig. 6 (d), a fourth prompting word template is shown, where the fourth prompting word module is: based on the generated UML activity map, a markdown format (application program interface, application Interface, API) design is written. The user may input the target traffic scenario at "/input the target activity.

In the embodiment of the present application, the management system may generate the application program interface corresponding to the target activity requirement based on the UML activity diagram in step S512 or the modified UML activity diagram according to the target business scenario requirement and the target activity requirement input by the fourth input operation.

Wherein when the content added at the third location of the first demand document is a UML activity diagram, the management system generates an application program interface based on the UML activity diagram. When the content added at the third location of the first demand document is the modified UML activity diagram, the management system generates an application program interface based on the modified UML activity diagram.

Alternatively, the management system may generate the application program interface from the UML activity diagram based on LLM. The manner in which the application interface is generated is not particularly limited in this application.

S514: and confirming whether the generated application program interface is correct. If the application program interface is confirmed to be correct, S516 is executed, and if the application program interface is confirmed to be incorrect, S515 is executed.

The application program interface includes application program names, responses, error responses, descriptions, and the like. The user may confirm the application interface content, for example, whether the application interface content corresponds to a target business scenario, whether it corresponds to a target activity, whether it adopts a markdown format, and so on. If the user confirms that the generated application program interface meets the content, the generated application program interface can be confirmed to be correct. Otherwise, the application program interface is confirmed to be incorrect.

S515: the application program interface is modified.

When the application program interface is confirmed to be incorrect, the user can directly modify names, parameters, descriptions, responses and the like in the application program interface on the display interface. The operation is simple, and the accuracy of the generated application program interface is ensured through further verification of the user.

S516: and adding the application program interface or the modified application program interface to a fourth position of the first requirement document.

When the application program interface is confirmed to be correct, the application program interface is added to the fourth position of the first requirement document. And when the application program interface is confirmed to be incorrect, adding the modified application program interface to a fourth position of the first requirement document to obtain a second requirement document. Illustratively, referring to (e) in FIG. 8, an application program interface or modified application program interface is added to the fourth location 804 of the first requirement document, resulting in a second requirement document schematic. According to the embodiment of the application, a user story map is added at a first position, a user story is added at a second position, a UML activity map is added at a third position, and a first requirement document of an API interface is added at a fourth position to serve as a second requirement document.

According to the embodiment of the application, the user story map, the user story, the UML activity map and the API interface are sequentially used, the input of each stage is composed of the output of the previous stage and the problem of the current stage, and the user can interactively modify and confirm the generated result of each stage, so that the whole software development process has clear requirement arrangement and detailed document design, and the accuracy and feasibility of the development process of a piece of engineering are improved.

Example III

Further, for each stage of input operation, there are cases where software development requirements are complex and cannot be directly implemented end-to-end. When the LLM processes complex software development requirements, accurate results are difficult to obtain. Based on the above problems, the embodiment of the application provides a software development method for disassembling the self-planning complex software development requirement.

The self-planning complex software development requirement dismantling is to disassemble the final generation result on the premise that the generation result of each sub development requirement meets the requirement. The following formula is referred to, and is an implementation manner for generating the demand content after splitting the complex software development demand:

P(c|a,X)≡P(c|b,a)·P(b|a,X)

wherein X is given prompt, a is input of complex software development requirement, c is generated result, b is intermediate generated result after planning and disassembling complex software development requirement, P (c|a, X) represents final result generated by input of complex software development requirement under given prompt, P (c|b, a) represents result generated by taking disassembled intermediate generated result as input under input of complex software development requirement, and P (b|a, X) represents intermediate result generated by input of complex software development requirement under given prompt. It follows that the resolution of complex tasks into simple tasks does not affect the value of the solution.

The following will describe the stages shown in the second embodiment in detail.

Referring to fig. 9, a flowchart of a method for generating a third software development requirement document according to an embodiment of the present application is provided. The method specifically comprises the following steps:

s91: and generating a user story map stage.

In the generation of the user story map phase, it is first determined whether the software development requirements of the generation of the user story map phase are complex. If so, reconstructing the first prompt word template. If not, the user directly performs the first input operation on the first prompt word template.

For example, the user may determine the number of software development requirements for the user story map stage, and if the number exceeds a preset number threshold, determine that the software development requirements for the stage are complex. And then for complex software development requirements, splitting based on self-planning complex software development requirement dismantling rules. The disassembly mode can be as follows: the software development requirement of the stage is split into a plurality of first sub development requirements based on a pre-constructed splitting rule. Wherein the plurality of first sub-development requirements includes a target business scenario requirement. And then reconstructing a first prompt word template by utilizing the split multiple first sub development requirements. And the user executes a first input operation on the reconstructed first prompt word template.

If the number does not exceed the preset number threshold, determining that the software development requirement of the stage is not complex, and the LLM end-to-end direct implementation can be utilized. For example, assuming that the preset number threshold is 3, for the stage of generating the user story map, the number of software development requirements is only the target business scene requirements, and the number is smaller than the preset number threshold, so that the method can be directly realized by utilizing LLM end-to-end. Therefore, the user story map may be generated directly using steps S51-S54 without disassembly, and the generated user story may be added to the first location of the first demand document.

S92: a user story phase is generated.

In the generate user story phase, it is first determined whether the software development requirements to generate the user story phase are complex. If so, reconstructing a second prompt word template. If not, the user directly performs a second input operation on the second prompt word template.

For example, the user may determine the number of software development requirements in the user story stage, and if the number of software development requirements exceeds a preset number threshold, determine that the software development requirements in the stage are complex. And then split the complex software development requirements. The disassembly mode can be as follows: the stage software development requirement is split into a plurality of second sub development requirements based on pre-built splitting rules. Wherein the plurality of second sub-development requirements includes a target application theme requirement. And then reconstructing a second cue word template by using the plurality of second sub-development requirements. And the user executes the first input operation on the reconstructed second prompting word template.

For example, assuming that the preset number threshold is 3, it is found through analysis that the software development requirements of the user story stage include multiple requirements of service scenes, application topics, roles in a return format, completed activities, realization values, acceptance conditions and the like, and the preset number threshold is exceeded. If LLM is directly utilized to process the target business scene requirement and the target application theme requirement, the situation that an accurate user story cannot be generated due to complex software development requirements exists. At this time, the software development requirements may be disassembled.

Referring to fig. 10, a schematic diagram of disassembly is provided in an embodiment of the present application. The software development requirement of the user story generation stage can be disassembled into three parts of requirements of input, return format and acceptance condition. The input includes "business scenario and application theme", the return format is "as" i want to "in order to" i "and the acceptance condition is" Given under what conditions "what actions are taken while" and what results are obtained ".

And reconstructing the second prompting word template by utilizing the split content to obtain a reconstructed second prompting word template.

Illustratively, the second alert word template before reconstruction is: and writing the user stories according to the scenes and the application topics only according to the generated user story map. Scene: * Theme: * ****. The reconstructed second prompting word template is as follows: and writing the agile user stories according to the scenes and the application topics only according to the generated user story map. Scene: * Theme: * ****. The return format of the user story is: as a < role >, i want < function/activity completed > in order to < goal/implementation value >. Acceptance conditions: given < what conditions > white < what actions are taken > th < what results are obtained >.

The user may perform a second input operation on the reconstructed second alert word template, including inputting the target traffic scenario requirement at ", inputting the target application theme requirement at" ", and inputting the corresponding content at the <" > location. Then, the operations of steps S55 to S58 are performed to generate a user story.

Thus, the complex software development requirement is disassembled into small sub development requirements, and the accuracy of the generated user story is improved.

S93: a UML activity diagram phase is generated.

In the generate UML activity diagram stage, it is first determined whether the software development requirements for generating UML activity diagrams are complex. If so, reconstructing a third prompting word template. If not, the user directly performs a third input operation on the third prompting word template.

For example, the user may determine the number of software development requirements for the UML activity diagram stage, and if the number of software development requirements exceeds a preset number threshold, determine that the software development requirements for the stage are complex. And then split the complex software development requirements. The disassembly mode can be as follows: based on a pre-constructed splitting rule, splitting and generating the software development requirement of the UML activity diagram stage. And then reconstructing a third prompting word template by utilizing the split content. And the user executes a third input operation on the reconstructed third prompting word template.

Illustratively, assuming that the preset number threshold is 3, it is found through analysis that the software development requirements of the UML activity diagram generation stage include a plurality of software development requirements of a start node, an activity node, a judgment node, an end node, and the like. Exceeding a preset number threshold. If the LLM is directly used for processing the content of the third input operation, there are cases where an accurate UML activity diagram cannot be generated due to the complex software development requirements. At this time, the software development requirements may be disassembled. Specifically, the software development requirement can be disassembled according to the format of the plant uml language, including: a start node, an active node, a decision node, an end node, etc.

And reconstructing the third prompting word template by utilizing the split content to obtain a reconstructed third prompting word template.

Illustratively, the third alert word template before reconstruction is: and compiling a UML activity diagram expressed in the plant UML language according to the generated user stories. The reconstructed third prompting word template is as follows: and compiling a UML activity diagram expressed in the plant UML language according to the generated user stories. The return format of the UML activity diagram is: start execution (multiple executions if there are multiple active nodes and decision nodes), and end @ reduce.

The user may perform a third input operation on the reconstructed third alert word template, including inputting the target service scene requirement at "@" a startuml location, where the corresponding content is input. Then, the operations of steps S59 to S512 are performed to generate a UML activity map.

S94: an application program interface stage is generated.

In the generation application program interface stage, it is first determined whether the software development requirements of the generation application program interface stage are complex. If the fourth prompting word template is complex, reconstructing the fourth prompting word template. If not, the user directly performs a fourth input operation on the fourth prompting word template.

For example, the user may determine the number of software development requirements for generating the application program interface stage, and if the number of software development requirements exceeds the preset number threshold, determine that the software development requirements for the stage are complex. And then split the complex software development requirements. The disassembly mode can be as follows: the software development requirements of this stage are split based on pre-built splitting rules. And then reconstructing a fourth prompting word template by utilizing the split content. And the user executes a fourth input operation on the reconstructed fourth prompting word template.

Illustratively, assuming that the preset number threshold is 3, it is found through analysis that the software development requirements of the generation application program interface stage include a plurality of software development requirements such as names, descriptions, parameters, responses, and error responses of the application program interface. Exceeding a preset number threshold. If the LLM is directly used for processing the content of the fourth input operation, the situation that an accurate application program interface cannot be generated due to complex software development requirements exists. At this time, the software development requirements may be disassembled. Specifically, the complex software development requirements can be broken down into names, descriptions, transmission methods, responses, error responses, and the like of the application program interfaces.

And reconstructing a fourth prompting word template by utilizing the split content to obtain the reconstructed fourth prompting word template.

Illustratively, the fourth cue word template before reconstruction is: and writing an API design adopting a markdown format according to the generated UML activity diagram. The reconstructed fourth prompting word template is as follows: and writing an API design adopting a markdown format according to the generated UML activity diagram. The return format of the API design is madkDown table format, and list heads are respectively: API route name, transfer method, description, request, response, error response.

The user may perform a second input operation on the reconstructed second prompting word template, where the second input operation includes inputting a target service scene in "×", inputting a target activity in "×", and inputting corresponding content in a corresponding position of the list head. Then, the operations of steps S513 to S516 are performed to generate an application program interface.

Thus, the complex software development requirement is disassembled into small sub-development requirements, and the accuracy of the generated application program interface is improved.

According to the embodiment of the application, the accuracy of the result of the demand content generated at each stage in the software development method is further improved through the self-planning disassembly of the complex software development demands. In addition, the embodiment of the application also provides a device for generating the software development requirement document.

Referring to fig. 11, a schematic structural diagram of a generating device 100 of a software development requirement document according to an embodiment of the present application is provided. In software development, the process from demand to design is split into multiple phases, including at least a first phase and a second phase, which are executed sequentially. The apparatus 110 specifically includes:

the first generating unit 1101 is configured to generate, in response to a first input operation for a software development requirement of a first stage, first required content corresponding to the first stage, where the first required content is used to represent required content of software development corresponding to the first input operation in the first stage.

The second generating unit 1102 is configured to generate, according to the first demand content, second demand content corresponding to a second stage in response to a second input operation for the software development requirement of the second stage, where the second demand content is used to represent the demand content of the software development corresponding to the second input operation in the second stage.

The adding unit 1103 is configured to add the first requirement content to a first location of the first requirement document, and add the second requirement content to a second location of the first requirement document, so as to obtain a second requirement document.

Alternatively, when the first stage generates the user story map stage, the first generating unit 1101 is specifically configured to: responding to the operation of inputting the target business scene requirement on the first prompting word template, and matching a target flow rule from a vector knowledge base of a pre-stored business scene and business flow rules, wherein the target flow rule is a business flow rule corresponding to the target business scene requirement; the first prompt word template is a display template for inputting the requirements of the target business scene in the stage of generating the user story map. And generating a user story map according to the target flow rule, and taking the user story map as first required content corresponding to the first stage, wherein the user story map comprises a plurality of user stories to be developed corresponding to the software development, and the user stories are used for describing the software development requirements of the user.

Optionally, the apparatus 110 further comprises a first disassembly reconstruction unit, in particular for: splitting the software development requirement of the user story map generation stage into a plurality of first sub development requirements; the plurality of first sub-development requirements includes a target business scenario requirement; and reconstructing the first prompt word template by using the plurality of first sub-development requirements, so that the reconstructed first prompt word template comprises the plurality of sub-development requirements. And responding to the operation of inputting the target business scene on the reconstructed first prompting word template, and matching the target flow rule from the vector knowledge base.

Optionally, when the second stage is a user story stage, the second generating unit 1102 is specifically configured to: responding to the operation of inputting the target application theme requirement on a second prompt word template, generating a user story corresponding to the target application theme requirement according to a user story map, taking the user story corresponding to the target application theme requirement as second requirement content corresponding to a second stage, wherein the second prompt word template is a display template for inputting the target application theme requirement in the stage of generating the user story.

Optionally, the apparatus 110 further comprises a second disassembly reconstruction unit, in particular for: splitting the software development requirement for generating the user story stage into a plurality of second sub development requirements; the plurality of second sub-development requirements includes a target application theme requirement; reconstructing a second cue word template by using the plurality of second sub-development requirements, such that the reconstructed second cue word template includes the plurality of sub-development requirements; and responding to the operation of inputting the target application theme requirement on the reconstructed second prompting word template, and generating a user story corresponding to the target application theme requirement according to the user story map.

Optionally, the apparatus 110 further comprises a first modifying unit for: and responding to the first modification operation of the first demand content, and acquiring the modified first demand content. The second generating unit 1102 is further configured to: and generating second demand content corresponding to the second stage according to the modified first demand content. The adding unit 1103 is further configured to add the modified first requirement content to the first location of the first requirement document.

Optionally, the apparatus 110 further comprises a second modifying unit, in particular for: and responding to a second modification operation of the second required content, and acquiring the modified second required content. The adding unit 1103 is further configured to: the modified second demand content is added to the second location of the first demand document.

Optionally, the plurality of stages further includes a third stage and a fourth stage, and the apparatus 110 further includes:

and the third generating unit is used for responding to a third input operation of the software development requirement of the third stage and generating third requirement content according to the second requirement content, wherein the third requirement content is used for representing the requirement content of the software development corresponding to the third input operation in the third stage. And the fourth generating unit is used for responding to a fourth input operation of the software development requirement of the fourth stage and generating fourth requirement content according to the third requirement content, wherein the fourth requirement content is used for representing the requirement content of the software development corresponding to the fourth input operation in the fourth stage. The adding unit 1103 is further configured to: the method comprises the steps of adding first demand content to a first position of a first demand document, adding second demand content to a second position of the first demand document, adding third demand content to a third position of the first demand document, and adding fourth demand content to a fourth position of the first demand document to obtain a second demand document.

Optionally, the third stage is a stage of generating a unified modeling language activity diagram, the fourth stage is a stage of generating an application program interface, and the third generating unit is specifically configured to: responding to the operation of inputting the target business scene requirement on the third prompting word template, generating a unified modeling language activity diagram according to the second requirement content generated in the second stage, and taking the unified modeling language activity diagram as the third requirement content in the third stage, wherein the unified modeling language activity diagram is used for describing the activity flow of the second requirement content; the third prompting word template is a display template for inputting the requirements of the target business scene in the stage of generating the unified modeling language activity diagram.

The fourth generation unit is specifically configured to: responding to the operation of inputting the target activity requirement on the fourth prompting word template, generating an application program interface corresponding to the target activity requirement according to the same modeling language activity diagram, utilizing the application program interface to communicate the server side with the client side, and taking the application program interface as fourth requirement content of a fourth stage; the fourth prompting word template is a display template for inputting the target activity requirement in the generation of the application program interface stage.

The embodiment of the application provides a device for generating a software development requirement document. The device is specifically used for: the method includes the steps of splitting a process from demand to design of software development into a plurality of stages, respectively executing input operation on a first stage and a second stage which are sequentially executed by the plurality of stages, generating demand content corresponding to the first stage and the second stage, and adding the demand content to a corresponding position of a demand document, so that the demand document pointing from demand to design is included in the software development process. In addition, the required content corresponding to the first stage is a first required content, and the required content corresponding to the second stage is a second required content. The second demand content is determined by the second input operation and the first demand content output in the first stage, so that the first demand content and the second demand content are effectively connected in series, and the conveying efficiency of the design concept is improved. Therefore, when the user wants to know the relation between the requirement and the design, and analyze whether the existing design can meet the current requirement, the user only needs to directly read the software development requirement of the existing design from the corresponding position of the requirement document. In summary, compared with the method of constructing related knowledge through manual exploration, the method and the device can reduce difficulty in knowledge transfer and iterative development in the scenes of demand change analysis, core developer variation and the like.

The present application still further provides an electronic device comprising: memory and a processor. A memory having a computer program stored thereon. And a processor for executing the computer program in the memory to implement some or all of the steps in the method for generating a software development requirement document described in the foregoing embodiment.

The present application also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements some or all of the steps in the method for generating a software development requirement document described in the foregoing embodiment.

It should be noted that, in the present specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment is mainly described in a different point from other embodiments. In particular, for the apparatus and device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments for relevant points. The apparatus and device embodiments described above are merely illustrative, wherein elements illustrated as separate elements may or may not be physically separate, and elements presented as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

Claims

1. A method of generating a software development requirement document, wherein in the software development, a process from requirement to design is split into a plurality of phases including at least a first phase and a second phase that are sequentially executed, the method comprising:

2. The method of claim 1, wherein when the first stage generates a user story map stage, the generating first demand content corresponding to the first stage in response to a first input operation of a software development demand for the first stage comprises:

3. The method according to claim 2, wherein the method further comprises:

4. The method of claim 2, wherein when the second stage generates a user story stage, the generating second demand content corresponding to the second stage according to the first demand content in response to a second input operation for the second stage's software development demand, comprises:

5. The method according to claim 4, wherein the method further comprises:

6. The method of claim 1, wherein after the generating the first demand content corresponding to the first stage, the method further comprises:

7. The method according to any one of claims 1-6, further comprising:

8. The method of claim 1, wherein the plurality of stages further comprises a third stage and a fourth stage, the method further comprising:

9. A generation apparatus of a software development requirement document, characterized in that in the software development, a process from requirement to design is split into a plurality of stages including at least a first stage and a second stage which are sequentially executed, the apparatus comprising:

10. An electronic device comprising a memory and a processor;

the memory is coupled with the processor;

the memory stores program instructions that, when executed by the processor, cause the electronic device to perform the method of any of claims 1-9.