Test case generation method and apparatus

Abstract

Embodiments of the present specification provide a test case generation method and device, the method comprising determining a test case generation task, wherein the test case generation task carries a target document to be generated, a task type and a document type; using an information extraction agent to parse the target document according to the document type, and generating a parsing result of the target document; using a test point generation agent to generate an initial test point according to the task type and the parsing result; and using a test case generation agent to generate a first test case corresponding to the test case generation task according to the initial test point and the parsing result. By simulating the analysis logic of a senior test expert through multiple agents, the efficiency and completeness of test case output are greatly improved.

Description

TECHNICAL FIELD

[0001] One or more embodiments of the present specification relate to the technical field of computer technology, and in particular, to a test case generation method. One or more embodiments of the present specification also relate to a test case generation apparatus, a computing device, a computer-readable storage medium, and a computer program product. BACKGROUND

[0002] Current software testing workflows usually cover multiple links such as test case writing, test case execution, and report generation. Among them, the highly dependent manual completion of test case writing has become a bottleneck restricting the improvement of testing efficiency. Although the existing technology has made progress in the field of test execution automation, there are still significant defects in the test case writing link: on the one hand, it is highly dependent on the personal experience and business familiarity of the test personnel, and the writing process needs to manually cross-reference PRD (Product Requirements Document), design documents, historical cases and other multi-dimensional information, which is inefficient and easy to miss; on the other hand, under the multi-element labor mode, the difference in personnel level easily leads to uneven test quality, and there is a lack of unified and automated quality bottoming mechanism, which still needs to invest a large number of senior engineers for manual review, and cannot guarantee the delivery progress and product quality on a large scale. SUMMARY

[0003] Therefore, one or more embodiments of the present specification provide a test case generation method, a test case generation apparatus, a computing device, a computer-readable storage medium, and a computer program product to solve the technical problems of low efficiency and poor quality in the test case writing link in the prior art.

[0004] According to a first aspect of one or more embodiments of the present specification, a test case generation method is provided, comprising:

[0005] determining a test case generation task, wherein the test case generation task carries a target document to be generated, a task type, and a document type;

[0006] using an information extraction agent to parse the target document according to the document type, and generating a parsing result of the target document;

[0007] using a test point generation agent to generate an initial test point according to the task type and the parsing result;

[0008] using a test case generation agent to generate a first test case corresponding to the test case generation task according to the initial test point and the parsing result.

[0009] According to a second aspect of one or more embodiments of the present specification, a test case generation apparatus is provided, comprising:

[0010] a task determination module configured to determine a test case generation task, wherein the test case generation task carries a target document to be generated, a task type, and a document type;

[0011] a document analysis module configured to analyze the target document according to the document type by using an information extraction agent to generate an analysis result of the target document;

[0012] a test point generation module configured to generate an initial test point according to the task type and the analysis result by using a test point generation agent;

[0013] a test case generation module configured to generate a first test case corresponding to the test case generation task according to the initial test point and the analysis result by using a test case generation agent.

[0014] According to a third aspect of one or more embodiments of the present specification, a computing device is provided, comprising:

[0015] a memory and a processor;

[0016] the memory is configured to store computer programs / instructions, and the processor is configured to execute the computer programs / instructions, which realize the steps of the above test case generation method when executed by the processor.

[0017] According to a fourth aspect of one or more embodiments of the present specification, a computer readable storage medium is provided, which stores computer programs / instructions, which realize the steps of the above test case generation method when executed by the processor.

[0018] According to a fifth aspect of one or more embodiments of the present specification, a computer program product is provided, comprising computer programs / instructions, which realize the steps of the above test case generation method when executed by the processor.

[0019] The test case generation method provided by one or more embodiments of the present specification comprises: determining a test case generation task, wherein the test case generation task carries a target document to be generated, a task type, and a document type; using an information extraction agent to parse the target document according to the document type, and generating a parsing result of the target document; using a test point generation agent to generate an initial test point according to the task type and the parsing result; and using a test case generation agent to generate a first test case corresponding to the test case generation task according to the initial test point and the parsing result.

[0020] Specifically, the test case generation method decomposes the test case generation task into multiple levels of task execution such as "parsing result generation", "test point generation", and "test case generation". First, the target document is parsed by calling the information extraction agent to generate a parsing result. Then, the test point generation agent generates multiple initial test points according to the task type and the parsing result. Finally, the test case generation agent generates a first test case for each initial test point. First, the analysis logic of experienced test experts is simulated by multiple agents to automatically read and integrate multi-dimensional information such as PRD and design documents in the target document, directly converting human experience into stable automated output, overcoming the problem of test point omission caused by insufficient or negligence of personnel business familiarity, greatly improving the efficiency and completeness of test case output. Secondly, through the joint processing of multiple agents, it is ensured that the generated test cases follow the built-in best practices and standard processes, eliminating the quality fluctuations of test points caused by personnel level differences, providing stable, reliable, and high-quality automated support, and reducing the excessive dependence on manual review by senior engineers. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present specification or the prior art, the drawings needed in the embodiment or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments described in the present specification, and those skilled in the art can obtain other drawings according to these drawings without creating any inventive labor.

[0022] Figure 1 is an application diagram of a test case generation method provided by an embodiment of the present specification;

[0023] Figure 2 is a flowchart of a test case generation method provided by an embodiment of the present specification;

[0024] Figure 3 is a data mapping diagram of a test case generation method provided by an embodiment of the present specification;

[0025] Figure 4 FIG. 6 is a schematic diagram of a test account preparation and test execution process according to an embodiment of the present specification;

[0026] Figure 5 FIG. 7 is a schematic diagram of a test case generation device according to an embodiment of the present specification;

[0027] Figure 6 FIG. 8 is a block diagram of a computing device according to an embodiment of the present specification. DETAILED DESCRIPTION

[0028] In order to enable persons skilled in the art to better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be clearly and completely described below in conjunction with the drawings in the embodiments of the present specification. Obviously, the described embodiments are only some of the embodiments of the present specification, but not all the embodiments. Based on the embodiments in the present specification, all other embodiments obtained by persons skilled in the art without creative labor should fall within the scope of protection of the present specification.

[0029] The present specification uses specific words to describe the embodiments of the present specification. As "one embodiment", "an embodiment", and / or "some embodiments" means a certain feature, structure or characteristic related to at least one embodiment of the present specification. Therefore, it should be emphasized and noted that the "an embodiment" or "one embodiment" or "one alternative embodiment" mentioned in the present specification twice or more in different positions does not necessarily mean the same embodiment. In addition, the skilled in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples without contradiction.

[0030] The terms used in one or more embodiments of the present specification are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of the present specification. The singular forms "a", "an" and "the" used in one or more embodiments of the present specification and the appended claims are also intended to include plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of the present specification includes any or all possible combinations of one or more associated listed items.

[0031] The terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0032] Although the terms first, second, etc. can be used herein to describe various information, these terms are not intended to denote a temporal sequence, but to distinguish between different pieces of information. For example, a first entity discussed below could be termed a second entity without departing from the scope of one or more embodiments described herein. The first, second, etc. designations can be used herein to distinguish between similar entities without necessarily denoting a temporal sequence. For example, a first server and a second server can refer to two servers in general. To distinguish between these two servers, they can be referred to as a first server and a second server. Of course, these two servers can also be the same server at times.

[0033] Depending on the context, the word "if" can be interpreted to mean "when" or "upon" or "in response to determining" in this document.

[0034] In this document, unless otherwise stated, a "receiving", "sending" of data is not necessarily direct receiving and sending, but can be indirect receiving and sending. For example, A receives data sent by B can be understood as A directly receiving data sent by B, or A indirectly receiving data sent by B through C or other subjects. Similarly, B sends data to A can be understood as B directly sending data to A, or B indirectly sending data to A through C or other subjects. Here, C can be one subject, or two or more subjects.

[0035] In this document, unless otherwise stated, the association relationship between structures can be direct or indirect. For example, when describing "A is connected with B", unless it is explicitly stated that A is directly connected with B, it should be understood that A can be directly connected with B, or A can be indirectly connected with B; for example, when describing "A is on B", unless it is explicitly stated that A is directly on B (AB is adjacent and A is on B), it should be understood that A can be directly on B, or A can be indirectly on B (there are other elements between AB and A is on B). Similarly.

[0036] The user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in one or more embodiments of the present specification are information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards in the relevant region, and provide corresponding operation portal for user to choose authorization or refusal.

[0037] The technical solution provided by one or more embodiments of the present specification can adopt a deep learning model with relatively large model parameter scale, wherein the large model is only one model example, and the one or more embodiments of the present specification do not limit the number of model parameters supported by the deep learning model adopted to meet the actual demand. The deep learning model involved in one or more embodiments of the present specification can be an artificial intelligence-based language model (LM) or a multimodal model (MM), a large language model (LLM), a multi-modal pre-training model, etc.

[0038] In actual application, the large model only needs a small amount of samples to fine-tune the pre-training model and can be applied to different tasks. The large model can be widely applied to natural language processing (NLP) and computer vision fields, and can be applied to computer vision field tasks such as visual question answering (VQA), image caption (IC), image generation, and natural language processing field tasks such as text-based sentiment classification, text summary generation, and machine translation. The main application scenarios of the large model include digital assistants, intelligent robots, search, online education, office software, e-commerce, intelligent design, etc.

[0039] The following explains the terms involved in one or more embodiments of the present specification.

[0040] Credit business (Credit Business): refers to the financial service activities of financial institutions or quasi-financial institutions providing financial support to qualified borrowers by assessing the credit risk, repayment ability and collateral value of borrowers; typical scenarios include personal consumption loans, small and micro enterprise loans, credit card installment payments, etc.

[0041] Test Case: In software engineering, a test case is the smallest executable unit used to verify whether a system's functionality meets the expected requirements. Each test case contains input data, expected results, operation steps, and judgment criteria. In the context of credit business, test cases should cover loan application, risk assessment, approval logic, credit calculation, contract generation, and other core business processes.

[0042] Test Analysis Document: A systematic summary document of the test case design process, execution results, and defect analysis. It includes test objectives, test scope, test strategy, test case coverage, defect statistics, risk assessment, and improvement suggestions. In the context of credit business, this document should reflect the integrity of business rule verification, compliance checks, and abnormal scenario simulation.

[0043] Intelligent Generation: Based on artificial intelligence technologies such as natural language processing, machine learning, and knowledge graph, the process of automatically generating test cases and test analysis documents. Its core goal is to replace repetitive human work through data-driven and rule-based reasoning, improving test efficiency and quality.

[0044] Natural Language Processing (NLP): An important branch of computer science and artificial intelligence, aiming to enable computers to understand, analyze, and generate human language. In this specification, NLP technology is used to extract key information from credit business contracts, risk control rules, and user demand documents as input for test case design.

[0045] Multimodal Data: A collection of heterogeneous data forms such as text, tables, flowcharts, and code. In this specification, multimodal data such as loan contract terms, approval flowcharts, and risk control rule tables are analyzed to generate comprehensive test cases.

[0046] Dynamic Update Mechanism: A mechanism that automatically adjusts the test case library and test analysis document based on changes in credit business rules (such as policy adjustments, model upgrades) or test feedback results (such as defect fixes). This mechanism is implemented through a rule engine and version control to ensure that the test assets remain synchronized with the actual business.

[0047] Rule Engine: An automated decision-making system based on predefined business rules (such as "annual income is more than 2 times the liabilities"). In this specification, the rule engine is used to verify whether the test cases meet the compliance requirements of the credit business and generate a rule conflict analysis report.

[0048] Coverage measures the extent to which test cases cover business processes, functional modules, or code paths. In credit business, coverage should reflect both business rule coverage (e.g., "all loan approval conditions are verified") and abnormal scenario coverage (e.g., "system fault tolerance when illegal parameters are input").

[0049] Compliance Check: The process of verifying whether the lending business complies with regulatory requirements or internal risk control standards. This manual uses compliance rules to automatically generate test cases to verify whether the system meets compliance constraints such as "prohibiting the issuance of consumer loans to students."

[0050] To address the technical challenges mentioned above, one approach is to employ a testing platform based on MBT (Model-Based Testing) and a large-model-driven test case generation technology. This platform supports drag-and-drop modeling of business process diagrams, automatically generates both manual and automated test cases, and uses reverse engineering tools to transform manual test cases into automated test cases. In other words, it allows testing experts to create test cases like drawing a process diagram. Figure 1 In this way, the software's operational logic (clicking this button, entering that data, and then a pop-up window) can be drawn on the computer. After viewing this "flowchart," the computer can automatically generate test cases (i.e., step-by-step operation instructions). Furthermore, it can transform test cases that previously required manual testing into scripts that the machine can run automatically. However, this technology suffers from high modeling complexity: it requires professionals to manually construct accurate system models (such as finite state machines or UML (Unified Modeling Language) diagrams), demands a deep understanding of business logic, and significantly increases the cost of modeling complex business rules (such as multiple conditions in credit risk control). It also lacks flexibility: the generated test cases rely on the static definition of the model, making it difficult to dynamically adapt to frequent changes in business rules; coverage of abnormal scenarios (such as inputting illegal parameters) requires additional configuration; automation is limited; maintenance costs are high: the model and test cases need to be updated synchronously; when business logic is adjusted, the model needs to be remodeled and the validity of the generated test cases verified; and reverse engineering capabilities are weak: converting manual test cases to automated test cases requires additional development tools, and the conversion efficiency is low, making it difficult to achieve a seamless "manual-automatic" workflow.

[0051] Another is to use natural language processing-based test case generation technology, which parses business requirement documents through natural language and generates structured test cases, supports business logic mind map generation and multi-scenario coverage analysis. However, this technology has limitations in semantic understanding: it relies on the semantic understanding ability of pre-trained models, and the accuracy of parsing complex business rules (such as risk assessment formulas in credit) is insufficient, the fusion processing ability of multi-modal data (such as contract terms and process diagrams) is weak, and the coverage is insufficient: the generated test cases may miss boundary conditions or abnormal scenarios, and need to be supplemented manually, and the response speed to compliance checks (such as regulatory policy changes) is slow.

[0052] Whether based on an MBT model or an NLP model, there is a common pain point: when processing a large amount of complex information at one time, important points may be missed, understanding may be biased, or irrelevant content may be generated. Figure 1 Directly generating use cases from requirements step by step results in poor effects.

[0053] To solve the above technical defects, one or more embodiments of the present specification provide a data processing method, which divides the test case generation task into multiple explicit steps through a Multi-Agent (Multi-Agent) architecture, and different professional agents cooperate to complete the test case generation. One or more embodiments of the present specification also relate to a data processing apparatus, a computing device, a computer readable storage medium and a computer program product, which are described in detail one by one in the following embodiments.

[0054] The technical solutions provided by the embodiments of the present specification are described in detail below with reference to the accompanying drawings.

[0055] Referring to Figure 1 , Figure 1 An application diagram of a test case generation method provided by one or more embodiments of the present specification is shown.

[0056] Figure 1 The technical solution of the AI (Artificial Intelligence, Artificial Intelligence) assisted test analysis and generation technology shown in the middle is: taking the historical knowledge base as the core, man-machine cooperation.

[0057] Specifically, the system receives a test case generation task, acquires the project document (PRD, design document, etc.) carried in the test case generation task as the original input; first, the information extraction agent is called to analyze the project document according to the document type of the project document, understand the project target, system context and other information, and generate the analysis result of the project document; and after obtaining the analysis result of the project document, the system can also call the test strategy generation agent to generate a preliminary test strategy suggestion based on the analysis result of the project document, for example: "it is detected that this change is to add an interface, it is suggested to focus on function test, interface performance test and security test". The test engineer will adjust the test strategy in a visual interface based on the preliminary test strategy suggestion, such as confirming, modifying or redefining the test strategy.

[0058] Secondly, the system calls the test point generation agent to generate test points according to the test strategy and the analysis result of the project document, and generates one or more test points; and the test point generation agent will refer to the historical test documents in the knowledge base when generating test points, deduce the historical test points, and refer to the design ideas of the historical test points to generate test points. Then the test case generation agent is called to generate test cases according to the test points and the analysis result of the project document, and one or more test cases are generated; similarly, the test case generation agent will also refer to the historical test documents in the knowledge base when generating test cases, refer to the design ideas and step arrangement of the historical test cases to generate test cases. Finally, the test analysis document generation agent is called to generate test documents according to the analysis result of the project document, the test points and the test cases, and generate test documents, i.e. test case analysis reports.

[0059] In order to ensure the correctness and integrity of the test documents, after the test documents are generated, they can also be manually adjusted and confirmed by test experts, for example, the test experts can add, delete or modify the test documents.

[0060] The test case generation method provided by one or more embodiments of the present specification first splits the test case generation task into multiple levels of tasks such as "test point generation" and "test case generation", and processes them step by step by a dedicated intelligent agent. The system can read all the change points and project documents to ensure the completeness of the test range. At the same time, by introducing a historical test knowledge base, the intelligent agent can refer to and imitate the complex scenarios and boundary case designs in the historical test scheme during the generation process, completely overcoming the defects of AI "stiffly talking nonsense" and generating too direct and superficial content, making the generated test points and test cases more close to the actual testing habits, with expert-level testing depth, realizing the precision and depth of test analysis. Secondly, the system can intelligently identify and introduce regression test cases related to the current change point from the historical test knowledge base, effectively solving the problem that traditional automated scripts cannot automatically generate regression test cases due to lack of input, forming an automated bottom protection for core functions, greatly reducing the risk of introducing unknown faults due to modification, and providing high-quality automated regression test protection. Finally, this scheme deposits personal experience into reusable organizational assets, generates output by driving intelligent agents through the knowledge base, ensuring the high quality and consistency of test outputs by different personnel and different projects, solving the quality fluctuation problem caused by level differences in the multi-employment mode, and providing stable and reliable automated quality protection for the test process.

[0061] Reference Figure 2 , Figure 2 A flowchart of a test case generation method provided by an embodiment of the present specification.

[0062] From the perspective of the program, the execution subject of the flow can be a program loaded on an application server or an application terminal. It can be understood that the method can be executed by any device, equipment, platform, device cluster with computing and processing capabilities.

[0063] As shown in Figure 2 , the flow can include the following steps:

[0064] Step 202: Determine the test case generation task.

[0065] Among them, the test case generation task carries the target document to be generated test case, task type and document type.

[0066] Specifically, the application scenarios of the test case generation method provided by one or more embodiments of the present specification include, but are not limited to, iterative development test scenarios: for example, the development team performs an iteration every certain period of time, adds new functions or optimizes existing functions, large-scale refactoring and system migration scenarios: for example, splitting a monolithic application into microservices, data migration, replacing the underlying framework, compliance and security audit scenarios: for example, software in the financial or medical industries needs to meet strict compliance requirements, and compliance verification is required for each release, acceptance test scenarios: for example, before delivery to the customer or business side for acceptance testing, test scenarios covering core business processes need to be prepared. In practical applications, the application scenarios of the test case generation method provided by one or more embodiments of the present specification are not limited to the above-mentioned application scenarios, and can be adapted to any application scenario that requires systematic and high-quality test plan design.

[0067] Among them, the test case generation task can be understood as a core link in the software testing process, which is used to verify whether the software system or its specific part (such as a functional module, a code path) works as expected.

[0068] The target document can be understood as one or more formal documents describing the relevant information of the software system, which is the basis and source of test case design. It is a key reference for testers to understand “what the software should do” and “how to work correctly”.

[0069] The task type can be understood as the nature and scope of this test analysis, which determines the processing logic and scope of the entire Agent workflow. Based on the task type, the system can determine whether to focus on finding change points (iterative testing) or ignore change points for full-scan (full regression), etc. The task type includes, but is not limited to, new function testing (testing newly developed functional modules), iterative / optimization testing (testing modifications to existing functions), repair verification testing (testing the repair of a certain bug), etc.

[0070] The document type can be understood as the form of the target document. The document type determines the reading method and parsing focus of the subsequent information extraction agent and change point recognition agent. The document type includes, but is not limited to, product requirement documents (such as containing detailed business logic, rules and prototype diagrams), design documents (such as containing interface definitions, database changes, technical solutions), UI (User Interface, user interface) / UX (User Experience, user experience) design drawings (the subsequent agent will try to analyze the elements and interaction logic in it, and generate use cases related to front-end interaction and interface verification), other document types, etc.

[0071] A specific implementation of determining a test case generation task is illustrated as follows: a user clicks "create test case generation task" on the visual interface of the system. Upload the document: drag or upload one or more requirements / design documents related to the current iteration into the system. Configure parameters: the system may automatically parse the document and prefill some parameters, and the user needs to confirm or select: task type, associated historical knowledge base, regression strategy, or document type, etc. Start the task: after the user clicks "start generation" or "confirm" control, the test case generation task is created, and the system can create test cases according to the test case generation task.

[0072] Step 204: using the information extraction agent, parsing the target document according to the document type to generate the parsing result of the target document.

[0073] The information extraction agent, also known as information extraction Agent, is used to convert unstructured natural language documents into structured machine-readable data. The specific implementation techniques for parsing the target document according to the document type using the information extraction Agent include but are not limited to rule-based and template-based parsing techniques, natural language processing and named entity recognition techniques, multi-modal information extraction techniques, large language model information extraction and structured output techniques, etc.

[0074] The parsing result includes but is not limited to system and project identification, key business terms and entities, document structure information, etc. The system and project identification can be understood as: which large system (such as "consumer credit core system", "enterprise risk control platform") does the current project or function belong to; project / iteration background: an overview of the target, business problems to be solved, etc. Key business terms and entities can be understood as: key business objects (such as "loan application", "credit", "interest rate", "user"), states (such as "audited", "lent") and rules mentioned in the target document. Document structure information can be understood as the chapter, title, list, etc. in the target document, and the structured relationship of understanding the content.

[0075] A specific implementation of "using the information extraction agent to parse the target document according to the document type to generate the parsing result of the target document" is illustrated as follows: after the user clicks the "start" button, the system receives and executes the test case generation task. Specifically, the system first loads all the target documents uploaded by the user and calls the information extraction Agent. At this time, the information extraction Agent is activated, and the information extraction Agent is used to quickly scan and parse the uploaded target document, and the parsing result of the target document is generated according to the key information extracted by the information extraction Agent.

[0076] Assumption: A user uploads a PRD with the title of “AA Product Discounted Rate Function Iteration”.

[0077] After a quick scan by the information extraction agent, the following structured information (i.e., the parsing result) is output:

[0078] System name: Consumer Credit Core System.

[0079] Project background: Add a discounted rate function for high-quality users for the AA product to improve conversion rates.

[0080] Key entities: [discounted rate, high-quality user, credit score, interest rate]

[0081] Associated modules: user credit assessment module, interest rate calculation module, order generation module.

[0082] Then, the subsequent change point identification agent knows to look for modification points related to “discounted rate” in the document. When the knowledge base needs to be referenced, the system will also intelligently recommend the historical test knowledge base of the “Consumer Credit Core System” based on the name. The test point generation agent also knows that the use cases it generates are about “interest rate” calculation and will use business terms such as “credit score”.

[0083] Step 206: Use the test point generation agent to generate initial test points based on the task type and the parsing result.

[0084] After using the information extraction agent to generate the parsing result of the target document, the test point generation agent can generate initial test points based on the task type and the parsing result.

[0085] However, a PRD or design document contains a lot of content, including a large amount of background information, summaries, unchanged parts of the design, references, etc. These are “noise” for test analysis. If the test point generation agent directly generates test points based on the task type and the parsing result, the test point generation agent will attempt to generate test points for all the functions described in the document, including those that already exist and are not touched at all this time. Therefore, the initial test points generated will contain a large number of redundant test points. To avoid this situation, before using the test point generation agent to generate initial test points based on the task type and the parsing result, first use the change point identification agent to generate the change point, and then use the test point generation agent to generate test points based on the change point. The specific implementation is as follows:

[0086] The utilization information extraction agent further includes:

[0087] The utilization change point identification agent analyzes the change of the parsed result according to the task type, obtains a change analysis result, and generates a change point according to the change analysis result.

[0088] The utilization test point generation agent generates an initial test point according to the task type and the parsed result, and includes:

[0089] The utilization test point generation agent generates an initial test point according to the change point and the parsed result.

[0090] The change point identification agent can also be referred to as a change point identification agent, which is used to find all places where changes have occurred in the parsed result of the target document.

[0091] The task type can tell the change point identification agent which mode or filter should be used to analyze the parsing result, and determines the granularity and strictness of the analysis of the change point identification agent. That is, according to the task type, the system can determine how to call the change point identification agent, that is, guide the change point identification agent how to work. For example, if the task type is new function test, the system will call the change point identification agent and give the change point identification agent an instruction: "regard the document content as a new function, and extract all function points". At this time, the change point identification agent will use the powerful reasoning ability of the large language model (LLM) to carefully read the parsing result; it will look for keywords in the parsing result that imply "change" or "add", such as: "added XX function", "optimized YY process", "supports ZZ on the basis of the original", "abandoned old version of AA method"; by understanding these semantics, the LLM can infer which parts are likely to be the change points of this time. If the task type is iteration / optimization test, the system will call the change point identification agent and give the change point identification agent an instruction: "strictly compare and analyze to identify the added and modified points". The specific implementation is: the information extraction agent parses the system name, document ID, version number and other key information from the current document; the system uses these information to automatically find the previous version of the document in the version library; after finding it, it is automatically pulled down as the old version benchmark; at this time, the change point identification agent receives the new and old versions of the document at the same time, and performs accurate difference comparison. If the task type is repair verification test, the system will call the change point identification agent and give the change point identification agent an instruction: "focus on the change points related to Bug repair". At this time, the change point identification agent first parses the Bug ticket, extracts key information (such as defect phenomenon, reproduction steps) and locks the associated code commit (Commit). Then, pull and analyze the code difference (Diff), and accurately locate the modified files, functions and specific logic. Finally, comprehensive generation of high-precision change points, such as "XX function adds YY condition judgment to fix ZZ problem", directly points out the code level modification that needs to be verified, ensuring that the verification activity is focused and efficient.

[0092] Specifically, the change point identification agent analyzes the parsed results according to the task type, and can achieve specific technical implementations such as difference comparison (if there is an old version of the document, text or structured difference comparison is performed), semantic understanding (the meaning of the content is understood by using NLP technology. For example, it can be judged that "a certain level field is added" is a functional change, but "a wrong word is modified" is not), rule matching (screening is performed according to predefined rules (such as: all sentences starting with "add", "modify", and "delete" are important), association speculation (if the interface parameter is modified, it can be speculated that the related calling party may be affected), and the like.

[0093] The change analysis result can be understood as the output of the change point identification agent analyzing the parsed results according to the task type. For example, the change analysis result can be a record: "In the 'user management' chapter, it is mentioned that a 'user status field' is added.

[0094] The change point can be understood as the result output by the change point identification agent after a series of processing such as cleaning, merging, and classifying of the change analysis result. The change point identification agent usually generates a change point list, and each item in the change point list is a clear change point description. For example: [change point 1: the'member level' field is added to the user table, change point 2: the 'level' parameter is added to the query interface].

[0095] Then, after using the change point identification agent to analyze the parsed results according to the task type, obtaining the change analysis result, and generating the change point according to the change analysis result, a test point generation agent can be used to generate an initial test point according to the change point and the parsed results.

[0096] The test point generation agent can also be called a test point generation agent, which is used for test point design.

[0097] The specific implementation is that the test point generation agent receives the change point (for example, "user table adds'member level' field") in the change point list and the analysis result. Specifically, the change point list is a condensed and targeted list. Only "what has changed?" is recorded. For example: [change point 1: user table adds'member level' field, change point 2: query interface adds 'level' parameter]. Through the change point, it is only known that the "member level" field is to be tested, but what is the business rule of this field? What are the enumeration values? This will not be reflected in the change point. The analysis result is complete structured information extracted from the original requirement target document, which provides rich business background and details; including: system background, business terminology, specific function rules, interface definition, data dictionary, etc. For example: in its data dictionary part, it may explicitly record: member level: enumeration type, optional value: [1-level 1, 2-level 2, 3-level 3, 4-level 4]. Then, the test point generation agent can know "what has changed" according to the change point and "what it is and why it exists" according to the analysis result, and can generate test points on the basis of fully understanding the change point.

[0098] When generating test points, the test point generation agent will traverse each change point in the "change point list" and perform the following operations: for a single change point (for example, "add'member level' field"), the test point generation agent reads the analysis result to obtain the detailed business rules of the current change point, and specificizes the general test mode to the current business scenario. From the analysis result, it is confirmed that the member level is an enumeration field, and its valid values are: 1, 2, 3, and 4; then a comprehensive and accurate test idea is generated, and a structured test point is generated. For example, the test point is: verify that the member level field can successfully write valid values (1, 2, 3, 4) and persist to the database.

[0099] The test case generation method provided by one or more embodiments of the present specification first filters out the newly added, modified, and deleted content (signals) in this iteration from the analysis result by using the change point identification agent, and uses these contents as the only input for subsequent analysis. The test point generation agent will no longer need to process the structured analysis result of a whole document full of noise, resulting in test points that contain a large number of invalid tests on existing functions and unchanged parts, solving the signal and noise problem. Secondly, the regression test identification agent can also analyze which existing functions will be affected by the change points generated by the change point identification agent, so as to accurately recall the relevant regression test cases.

[0100] In order to ensure the accuracy and integrity of the subsequent generation of test points according to the change points, after the generation of the change points, the analysis report corresponding to the change points can be generated first, so as to adjust the change points according to the analysis report. The specific implementation mode is as follows:

[0101] After the change point identification agent generates the change points according to the task type, the change analysis result is obtained, and the change point corresponding analysis report is generated according to the change analysis result.

[0102] The change analysis agent is used for text analysis of the change points, and the analysis report corresponding to the change points is generated.

[0103] The change analysis agent, also known as change analysis Agent, is used for text analysis of the change points, and the analysis report corresponding to the change points is generated.

[0104] The change analysis agent is used for text analysis of the change points, and the analysis report corresponding to the change points is generated.

[0105] Firstly, the change analysis agent first analyzes the change point text, such as deep text analysis, influence range refinement, associated knowledge linking, etc. Among them, the deep text analysis is not only to extract keywords, but also to understand the intention and nature of the change. For example, the original change point is "modify the parameter list of bb function", and the deep text analysis is "this change is to support the new compound interest calculation rule. The core change is in the interest calculation module, and the bb parameter is added, which will directly affect the final calculated user interest amount. Influence range refinement can be understood as a more specific and clear description of the influence of the change point. For example, the original change point is "user table adds cc field", and the analysis is "this change will affect the following aspects: database: user table structure change, add a column, backend: user service creation, update, query interface needs to be adapted. The associated knowledge linking can be understood as associating the change point with the existing system knowledge to provide context. For example, in the analysis report, the change point "optimize the performance of the payment interface" is directly associated with the "payment timeout" fault ticket and the "payment core process" architecture diagram in history, to explain the background and value of this optimization.

[0106] Then, the change analysis agent generates the analysis report corresponding to the change point according to the above analysis result. And the text description analysis report can be converted into intuitive visual chart.

[0107] Of course, in actual application, the change point identification Agent can also be directly used to generate the change point after the generation of the change point, and the text analysis of the change point is directly generated to generate the analysis report corresponding to the change point. The specific implementation mode is the same as that of the change analysis Agent, which will not be repeated here.

[0108] The test case generation method provided by one or more embodiments of the present specification greatly reduces the threshold for testers to understand code changes, and provides solid data support for their efficient and targeted (i.e., "secondary processing and customization") test design.

[0109] In practical applications, in order to make the generated test points more in line with user cognition, the historical test knowledge base can be referred to when generating test points. The specific implementation is as follows:

[0110] The test case generation task also carries an associated historical knowledge base;

[0111] The test point generation agent generates initial test points according to the change points and the analysis results, including:

[0112] The test point generation agent retrieves the associated historical knowledge base according to the change points to obtain a first search result, and generates initial test points according to the first search result and the analysis results.

[0113] The associated historical knowledge base can be understood as one or more knowledge bases selected by the user from the existing multiple historical knowledge bases of the system and most relevant to the current task. The associated historical knowledge base stores historical test assets and experience, including historical test plans, test cases, discovered bugs, business rule documents, etc., and is usually processed by vectorization and other technologies for easy intelligent search. Different projects or business lines (such as "consumer credit", "enterprise risk control", "payment settlement") have completely different testing habits, business rules and special terms, and the subsequent test point generation agent will "imitate" the historical test cases in the specified associated historical knowledge base to ensure that the generated test points meet the business language and testing specifications of the current project.

[0114] The associated historical knowledge base is selected by the user when creating a test case generation task. According to the associated historical knowledge base, the system can determine that "in this task, all Agents that need to refer to historical experience can search for answers in this specified associated historical knowledge base." For example, when the process reaches the test point generation stage, the test point generation agent starts working and knows which associated historical knowledge base it is authorized to use. It will directly take the change points in the "change point list" as search keywords and search and match in the specified associated historical knowledge base.

[0115] With the above example, the agent is generated using the test point, the associated historical knowledge base is retrieved according to the change point, the first retrieval result is obtained, and the initial test point is generated according to the first retrieval result and the analysis result.

[0116] The specific operation steps are: the test point generation agent traverses each change point in the "change point list" and performs the following operations: for a single change point (taking "adding a'member level' field" as an example).

[0117] First, retrieve the associated knowledge base according to the change point: take the core concept of the current change point (such as "add field", "member level", "enumerated field") as the keyword, and initiate vector retrieval or keyword retrieval to the associated historical knowledge base to retrieve all relevant test cases and test solutions for testing "add field" and "enumerated type field" in history.

[0118] Obtain the first retrieval result: retrieve N most similar historical test cases (for example: test cases for testing "user name" field and "user status" field in history).

[0119] Generate initial test points according to the first retrieval result and the analysis result: the test point generation reads the analysis result to obtain the detailed business rules of the current change point, and specificizes the general test mode to the current business scenario. From the analysis result, it is confirmed that the member level is an enumerated field, and its valid values are: 1, 2, 3, 4. Then, the test point generation agent reasons: for example, imitate history: "when testing the 'user name' field in history, what aspects did they test?" -> for example: valid value, invalid value, boundary value, database storage, front-end display. Apply rules: "what are the standard enumeration field test strategies?" -> for example: equivalence class partitioning (valid / invalid), boundary value (maximum value, minimum value). Integrate business: "how should I design it in combination with the specific business rules of'member level'?" -> for example: values 1, 2, 3, 4 represent different levels, and the business logic needs to be verified. Finally, generate comprehensive and accurate test ideas, and convert the test ideas into structured natural language descriptions to form a structured test point. For example, test point 1: verify that the member level field can successfully write valid values (1, 2, 3, 4) and persist to the database. Test point 2: verify that when inputting illegal values (such as 0, 5, 'abc'), the system gives an explicit error prompt. Test point 3: verify that the interface returns and the front-end interface correctly display the business meaning corresponding to the member level (such as 1 displayed as 'ordinary'). Test point 4: verify the system processing logic when the member level field is empty.

[0120] The test case generation method provided by one or more embodiments of the present specification can effectively avoid the problem of generating too direct and simple content by allowing the test point generation agent to retrieve and refer to historical test schemes in real time when generating test points, automatically covering boundary values, abnormal scenarios, compatibility, security and other test dimensions that are easily overlooked by beginners, greatly improving the depth and comprehensiveness of test point generation; and converting personal experience into reusable organizational assets to ensure that test points generated by different personnel at different times can maintain a high and stable quality level, solving the problem of quality fluctuations caused by personnel level differences and ensuring the output quality and consistency of test points; finally, the test design is enhanced by reasoning based on historical experience, and the generated test points are more in line with actual testing habits, greatly reducing the workload of manual supplementation and correction, and realizing intelligent generation of test points.

[0121] Step 208: Using the test case generation agent, a first test case corresponding to the test case generation task is generated according to the initial test point and the analysis result.

[0122] Then, after generating the test point, the test case generation agent can be used to generate a first test case corresponding to the test case generation task according to the initial test point and the analysis result.

[0123] Similarly, to ensure that the generated test case is more in line with user cognition, the historical test knowledge base can also be referred to when generating the test case. The specific implementation is as follows:

[0124] The test case generation task also carries an associated historical knowledge base;

[0125] The test case generation agent generates a first test case corresponding to the test case generation task according to the initial test point and the analysis result, including:

[0126] The test case generation agent retrieves the associated historical knowledge base according to the initial test point to obtain a second retrieval result, and generates a first test case corresponding to the test case generation task according to the second retrieval result and the analysis result.

[0127] The test case generation agent, also known as test case generation Agent, can be understood as converting abstract test point test ideas into specific and executable operation steps.

[0128] The second retrieval result can be understood as a set of historical test cases related to the initial test point retrieved from the associated historical knowledge base.

[0129] Specifically, the initial test points, the analysis result, and the associated knowledge base are input into a test case generation agent, and each initial test point is traversed in sequence by the test case generation agent, and the following operations are performed for the current initial test point being traversed:

[0130] First, the test case generation agent analyzes the intention of the current initial test point, understands the verification target of the current initial test point, for example: when the test point verifies the input of an illegal value (such as 0, 5, 'abc'), the system gives an explicit error prompt. The test case generation agent understands its intention as "exception handling" and "invalid input verification". And based on the intention of the current initial test point, similar historical test cases are retrieved from the associated historical knowledge base for subsequent learning of their design patterns, step structures, and expression styles; for example, a historical test case of "testing illegal name value" is retrieved: steps: enter '1' in the name field, click save, expected result: prompt "illegal name input".

[0131] Second, the test case generation agent uses the analysis result to convert the abstract intention into specific test data that conforms to the current business. For example, from the analysis result, it knows that the valid values of "member level" are 1, 2, 3, and 4. Therefore, it can intelligently deduce that the illegal values / test data are 0, 5, 99, -1, "ABC", "" (empty string), and NULL; the boundary values / test data are 1 (lower limit) and 4 (upper limit); and the valid values / test data are 2 and 3 (for other test cases).

[0132] Finally, the test case generation agent fills the information intelligently deduced from the analysis result into a standard test case template using the historical test case as a reference template, and generates the final executable first test case. For example: test case ID: TC_LEVEL_002; title: verification of system processing when illegal integer value 5 is input in the member level field; precondition: the user has logged in and entered the information editing page; test steps: enter the number 5 in the member level input box and click the "save" button. Expected result: the system prompts an error message: "member level value range is 1-4". Data is not successfully saved.

[0133] Finally, the test case generation agent automatically completes this work that requires a lot of manual work and is highly repetitive by imitating the design patterns of historical test cases, injecting specific business rules and data, thereby improving test efficiency.

[0134] The test case generation method provided by one or more embodiments of the present specification, when generating a test case, the test case generation Agent introduces a correlation historical knowledge base as a reference, realizes the "standardization, high-quality automatic generation" of the test case, and solves the quality and efficiency bottleneck in the case writing stage. First, the test case generation Agent directly references the steps, data and assertion method of the historical case, and the "first test case" generated is detailed, executable, avoids common problems such as ambiguous step description and lack of checkpoints, and improves the accuracy and operability of the test case steps. Secondly, the knowledge base deposits the team's inherent test design habits and expression style, so that the AI-generated cases are highly consistent with historical cases in step organization, language description, and data naming, greatly improving the readability and maintainability of the test set, and unifying the team style and specifications. Finally, through the test case generation Agent, the test engineer only needs to review and optimize the AI-generated test cases with high completion degree, without starting from scratch, so as to focus on more complex test scenario design and strategy analysis, greatly improving the test case generation efficiency.

[0135] Then, after the first test case is generated, a test analysis document generation agent can be called to generate a test analysis report corresponding to the test case generation task. The specific implementation manner is as follows:

[0136] After the test case generation agent generates the first test case corresponding to the test case generation task according to the initial test point and the analysis result, the method further includes the following steps:

[0137] The test analysis document generation agent generates a test case analysis report corresponding to the test case generation task according to the analysis result, the change point, the initial test point and the first test case.

[0138] The test analysis document generation agent, also known as the test analysis document generation Agent, is an intelligent agent responsible for compiling, formatting and presenting, which is used to integrate the generation results of the upstream information extraction agent, the change point identification agent, the test point generation agent and the test case generation agent into a final document with clear structure, professionalism and deliverability. That is, the test analysis document generation Agent is used for information integration, structure organization and format rendering.

[0139] Specifically, after obtaining the analysis result, the change point, the initial test point and the first test case, the analysis result, the change point, the initial test point and the first test case are input into the test analysis document generation Agent, and the test analysis document generation Agent integrates the four parts into a complete test analysis document (such as a word or pdf format document).

[0140] Specifically, the test analysis document generation agent first establishes a standard document framework template, which defines the chapter structure of the test analysis document. Secondly, the test analysis document generation agent fills in the corresponding chapters of the template like filling in the blanks, such as Chapter 1: Revision History (automatically generated version, date, author information by the test analysis document generation agent), Chapter 2: Project Background and Test Scope (fill in: system name, project background, test objectives generated by the information extraction agent), Chapter 3: Change Point Analysis (fill in: change point list generated by the change point identification agent), Chapter 4: Functional Test Design (fill in: test point generation agent output and detailed test case set generated by the test case generation agent, etc.). Then, the test analysis document generation agent applies predefined document styles (such as applying corresponding fonts and font sizes for different levels of headings, tabulating test cases, ensuring uniform formatting, etc.), ensuring that the output is visually appealing and uniform, and may also perform consistency checks and optimizations, etc. Finally, the test analysis document generation agent renders the final arranged content into a test case analysis report in the target format (such as.docx,.pdf). At the same time, the test case analysis report can be saved to a specified location and the user can be notified that the test case generation task is complete.

[0141] The test case generation method provided by one or more embodiments of the present specification integrates the intelligent output of all previous agents into a final, directly usable result using the test analysis document generation agent, forming an automated closed loop; avoids the problem of non-uniform format and time-consuming manual integration, and the output always maintains professional standards, improving the professionalism and readability of the test case analysis report. And through the template method, it can ensure that each generated test analysis document is complete in structure and covers all necessary aspects without major omissions, ensuring the integrity of the test case analysis report.

[0142] In order to improve the understanding efficiency, communication efficiency and review quality of the test analysis result, after generating the test case analysis report, the test case analysis report can also be visually processed. The specific implementation method is as follows:

[0143] After the test analysis document generation agent generates the test case analysis report corresponding to the test case generation task according to the analysis result, the change point, the initial test point and the first test case, the method further comprises:

[0144] The visual image generation agent performs graphical processing on the test case analysis report to obtain a visual image analysis report corresponding to the test case analysis report.

[0145] The visualization image generation agent, which can also be referred to as a visualization image generation Agent, is a professional agent responsible for data visualization and graphical expression. The structured text and data information in the test case analysis report are converted into graphs, charts and diagrams that are easier for users to understand and remember quickly.

[0146] Graphical processing can be understood as the processing of analyzing, refining and visually converting the test case analysis report by the visualization image generation Agent, including but not limited to information extraction: identifying key entities (such as systems, modules, interfaces) and relationships (such as calling, dependency, containing) from the report, visual mapping: determining which chart is most suitable (such as architecture diagram, flowchart, mind map, timing diagram), layout and rendering: automatically generating the final image file (visualization image analysis report) and other graphical processing.

[0147] The visualization image analysis report can be understood as the graphical version of the test case analysis report, including but not limited to mind map, system architecture diagram, flowchart, use case distribution pie chart and other forms.

[0148] Specifically, the visualization image generation agent is used to perform graphical processing on the test case analysis report to obtain a visualization image analysis report corresponding to the test case analysis report. The visualization image generation Agent can be understood as automatically converting the test case analysis report into a visual mind map (convenient for macro review, quick memory and display, and complementing the detailed linear document to improve readability and maintainability), and outputting an additional visualization image analysis report.

[0149] The test case generation method provided by one or more embodiments of the present specification uses the visualization image generation agent to generate a visualization image analysis report. A clear visualization image analysis report, such as a mind map, can enable project members (product, development, testing, manager) to grasp the test scope, focus and overall logic at a glance within a few minutes, greatly improving information absorption rate. In addition, the visualization image analysis report provides a common and unambiguous focus for discussion, making communication more accurate and efficient, and making it easier to find logical flaws or coverage deficiencies in the design. At the same time, the graphical information is more in line with human memory patterns, and the key test strategies and inter-system dependency relationships are visually retained as visual assets for subsequent audits and traceability.

[0150] In order to intelligently analyze the impact range of each change point, the historical function cases that need to be regression tested can be automatically found from the associated historical knowledge base to prevent the case of "modifying A function and breaking B function". The specific implementation mode is as follows:

[0151] The test case generation task also carries an associated historical knowledge base and a regression strategy;

[0152] The change point identification intelligent agent further comprises:

[0153] The regression test identification intelligent agent performs regression analysis on the change point according to the regression strategy, obtains a regression analysis result, and determines a system associated component according to the regression analysis result; and

[0154] The third retrieval result is obtained by searching the associated historical knowledge base according to the system associated component, and a second test case is determined according to the third retrieval result, wherein the system associated component is a component having a direct or indirect dependency relationship with the change point, and the second test case is a historical test case having a matching relationship with the system associated component.

[0155] The regression strategy can be understood as a regression test range selected by a user when creating a test case generation task and needed to be covered in this test. Through the regression strategy, how many historical test cases should be found by the regression test identification intelligent agent to test together can be guided. The regression strategy can include a conservative type (only regression of use cases of core modules directly associated with the change point and having a strong dependency), a standard type (regression of all use cases of the function module where the change point is located and associated modules upstream and downstream thereof), and an aggressive type (regression of use cases of all core business processes of the system). The regression strategy needs to balance between test sufficiency and execution efficiency. The more conservative the regression strategy is, the fewer regression test cases are generated, and the faster the execution is. The more aggressive the regression strategy is, the higher the coverage rate is, but the longer the time consumption is. The regression strategy can be selected according to actual needs.

[0156] The regression test identification intelligent agent can also be referred to as a regression test identification Agent, which is used to predict which existing and seemingly unrelated functions may be damaged by a code change, thereby preventing software from being rolled back.

[0157] The regression analysis result can be understood as an output of the regression test identification Agent performing regression analysis on the change point according to the regression strategy. The regression analysis result can be understood as process data, and the system associated component can be understood as a series of software components (microservices, modules, database tables, interfaces, etc.) that will be directly or indirectly affected by the current change point.

[0158] The third retrieval result can be understood as all historical regression test cases for testing the system associated components retrieved from the associated historical knowledge base by taking the system associated components as keywords.

[0159] The second test case, which can be understood as a regression test case that needs to be executed in the current round of testing, can be understood as a subset of the third search result.

[0160] Specifically, the regression test identification agent is utilized to perform regression analysis on the change point according to the regression strategy, to obtain a regression analysis result, and to determine system associated components according to the regression analysis result; and the associated historical knowledge base is searched according to the system associated components, to obtain a third search result, and a second test case is determined according to the third search result. Specific implementation examples of the second test case are as follows:

[0161] First, the change point, the regression strategy, and the associated historical knowledge base are input into the regression test identification agent. Assuming that the change point is "modify the 'calculate points' method in the 'order service'", the regression strategy is "this time is a small version iteration, and the 'core path + high-risk module' regression strategy is adopted"; the regression test identification agent will identify the directly affected modules (such as the order service and the points service) and the possibly affected modules (such as the user level calculation) in combination with code dependencies, call chains, database associations, and the like. At this time, the regression strategy starts to play a role: it filters or weights the impact range. For example, if the strategy is "core path", only the modules related to the main process such as "order" and "payment" are retained; if the strategy is "high risk", the "points calculation" module with a high historical defect rate is preferentially retained.

[0162] Second, the regression test identification agent uses the system associated components (such as "order", "payment", and "points calculation") obtained in the previous step as keywords to search for all associated test cases in the associated historical knowledge base, that is, to find those historical test cases whose verification targets or test module fields match the affected system associated components; for example, test case A: verify that the points calculation is correct after ordering (associated with the order service).

[0163] Finally, the regression test identification agent directly uses these historical test cases as the second test case and introduces them into the current test plan.

[0164] The test case generation method provided by one or more embodiments of the present specification utilizes a regression test identification agent to perform dependency analysis, and the regression test identification agent can comprehensively identify all potential affected system associated components, avoiding omissions. Secondly, the regression test identification agent can also perform automatic retrieval based on the associated historical knowledge base, and instantly recall all historical test cases related to these system associated components from the massive historical test cases, forming a high-confidence regression test case (second test case). This not only completely liberates the test personnel from tedious and error-prone manual impact analysis work, but also provides a reliable automated safety net for software quality through technical means, greatly reducing the risk of regression defects after release, and providing a solid guarantee for rapid iteration.

[0165] In addition, in order to ensure the comprehensiveness of the test analysis, not only the functions are covered, but also the quality attributes are covered. After generating the change point, the system will also call a non-functional test agent to generate a third test case. The specific implementation manner is as follows:

[0166] The test case generation task also carries the associated historical knowledge base;

[0167] The change point identification agent, according to the task type, performs change analysis on the analysis result to obtain a change analysis result, and according to the change analysis result, generates a change point, and further comprises:

[0168] The non-functional test agent performs feature analysis on the change point, extracts the key features of the change point, matches the key features with a preset non-functional test rule library, and in the case of successful matching, generates a third test case according to the associated historical knowledge base and the analysis result.

[0169] The non-functional test agent, also known as the non-functional test Agent, can be understood as an intelligent agent responsible for quality attribute test analysis, and is used to analyze whether the change point introduces risks in performance, security, reliability, compatibility, etc. Specifically, the non-functional test focuses on "how" the system works, rather than "what" it does. Unlike functional testing (verifying whether the software correctly performs a specific function according to the requirement specification), non-functional testing evaluates the quality attributes, performance characteristics, and user experience of the software, ensuring the quality, efficiency, reliability, and user experience of the software in the real use environment. Non-functional testing includes but is not limited to performance testing, security testing, usability testing, compatibility testing, reliability testing, maintainability testing, etc.

[0170] Feature analysis can be understood as that the non-functional test agent scans the change point and identifies its technical characteristics; the key features of the change point can be understood as that after the non-functional test agent scans the change point and identifies its technical characteristics, the technical keywords and technical features that may cause non-functional risks are extracted. For example, if the change point is to add an aa interface, the extracted key features can include: keywords "add interface, payment, external call, etc.", and features "high-frequency operation, capital loss sensitive, network I / O (Input / Output)".

[0171] The preset non-functional test rule library can be understood as the "IF-THEN" rule library preset in the non-functional test agent, which is used to store non-functional test examples. For example: IF (the change point contains "add interface" AND "payment") THEN trigger "performance test" and "security test" requirements.

[0172] Specifically, the key features of the change point are extracted by using the non-functional test agent to analyze the features of the change point, and the key features are matched with the preset non-functional test rule library. In the case of successful matching, the third test case is generated according to the associated historical knowledge base and the analysis result. It can be understood that the key features of the change point are extracted by using the non-functional test agent to analyze the features of the change point, and the key features are matched with the non-functional test examples in the preset non-functional test rule library. In the case of successful matching, the non-functional test requirements are determined according to the matched non-functional test examples, and the third test case is generated according to the non-functional test requirements, the associated historical knowledge base and the analysis result.

[0173] Following the above example, the specific example implementation of "using the non-functional test agent to analyze the features of the change point, extracting the key features of the change point, matching the key features with the preset non-functional test rule library, and generating the third test case according to the associated historical knowledge base and the analysis result in the case of successful matching" is as follows:

[0174] The change point, the associated historical knowledge base, and the analysis result are input into the non-functional test agent.

[0175] First, the non-functional test agent analyzes each change point, extracts technical keywords and technical features from it, and obtains the key features of each change point: keywords "add interface, payment, external call, etc.", and features "high-frequency operation, capital loss sensitive, network I / O".

[0176] The non-functional test agent matches the extracted key features with the preset non-functional test rule library. Once a match is successful, a matching non-functional test example is obtained. Then, the non-functional test requirements in the non-functional test example are determined, and a third test case is generated according to the non-functional test requirements, in combination with the associated historical knowledge base and the analysis result.

[0177] Taking the above example, the matching non-functional test example is: IF change point contains new interface AND payment -> THEN trigger performance test requirements, security test requirements, and asset loss test requirements; IF change point contains high-frequency operation -> THEN trigger stress test requirements and load test requirements.

[0178] At this time, the non-functional test agent can determine that the performance test requirements, security test requirements, asset loss test requirements, stress test requirements, and load test requirements need to be triggered according to the matching non-functional test examples.

[0179] Taking the performance test requirements as an example, the non-functional test agent queries the associated historical knowledge base to see what the historical performance test standards for the similar "payment interface" are. If it is found that there is a historical performance test requirement that "the performance target of aa interface is TP99 < 200ms", the non-functional test agent generates an initial test requirement according to the historical performance test requirement found by the search: "performance test: it is suggested to perform performance test on aa interface, and the performance target is referred to the historical standard: TP99 ≤ 200ms". Then, the initial test requirement is adjusted according to the analysis result to generate the final test requirement, i.e., the third test case.

[0180] Specifically, the analysis result is used to provide a more accurate test context. For example, the non-functional test agent learns from the technical design part of the analysis result that the request body of the interface contains the fields of userId, amount, and channel. Then, the non-functional test agent adjusts the initial test requirement according to the analysis result to generate a more specific test requirement, such as: "performance test: use different userId and amount to construct a request to test aa interface, and verify that the TP99 response time ≤ 200ms." Based on this, it can be understood that the third test case is a non-functional test requirement, which is only a targeted test instruction or verification standard.

[0181] The test case generation method provided by one or more embodiments of the present specification generates non-functional test cases by using a non-functional test agent. By matching the feature analysis of the change point with the rule library, the non-functional test agent can actively identify the non-functional risks (such as new interfaces that may not withstand pressure, and new fields that may not be checked for security) implied in technical changes like an expert, and automatically propose accurate test requirements. Not only does it overcome the defect that manual review is easy to miss non-functional requirements, but it also makes special tests such as performance and security no longer optional projects considered in the later stage of the project, but rather actions that are synchronized with functional development, greatly reducing the high online failure cost caused by non-functional defects. That is, through the early, automatic, and systematic intervention of non-functional testing, quality assurance is improved from the "functionally correct" level to the "operationally reliable" level, effectively preventing major incidents such as online performance and security.

[0182] To address the risks that may arise when the system is accessed by multiple users at the same time, a fourth test case for addressing such problems can be generated using a concurrency analysis agent. The specific implementation is as follows:

[0183] After generating the initial test points using the test point generation agent according to the change point and the analysis result, the method further includes:

[0184] Using the concurrency analysis agent, the initial test points are subjected to concurrency analysis to obtain a concurrency analysis result, and in the case where it is determined according to the concurrency analysis result that the initial test points involve concurrency scenarios, a concurrency test point is generated;

[0185] Using the test case generation agent, a fourth test case corresponding to the test case generation task is generated according to the concurrency test point and the analysis result.

[0186] The concurrency analysis agent, also known as a concurrency analysis Agent, is an agent specifically responsible for analyzing potential risks in scenarios where multiple threads and multiple users access at the same time. It is used to scan initial test points to find scenarios that may cause resource contention and data competition.

[0187] The concurrency analysis result can be understood as a judgment conclusion indicating which initial test points involve concurrency risks, and can also include corresponding risk types (such as data competition and deadlock). For example, the concurrency analysis result is that the initial test point "user balance deduction" involves a "data competition" risk.

[0188] The concurrency scenario can be understood as an operation scenario in which shared resources (such as database rows, cache values, and files) may be accessed and modified by multiple threads / processes / users at the same time.

[0189] Concurrent test point, which can be understood as a test point added after the test project based on the concurrent scenario determined by the concurrent analysis agent when the initial test point exists in the concurrent scenario, is used to describe the direction and goal of the concurrent test. For example, the consistency of inventory data is verified when multiple users simultaneously purchase the same product.

[0190] In a specific implementation, the test point generation agent generates an initial test point list, and the concurrent analysis agent scans the initial test points in the list. When it identifies a scenario such as "user balance deduction", it modifies the test point list and adds new test points. For example, the concurrent test point is to verify that the data accuracy is not damaged when multiple users simultaneously deduct the balance. When the test case generation agent starts working for this concurrent test point, it generates the corresponding test case for the concurrent test point, i.e., the fourth test case. Following the above example, for the concurrent test point, the concurrent test case is generated: a script or instruction is simulated using a concurrent test tool to simulate 100 users simultaneously initiating a request.

[0191] Specifically, the test case generation agent generates the specific implementation of the fourth test case corresponding to the test case generation task according to the concurrent test point and the analysis result. For details, refer to the specific generation method of the initial test case in the above embodiment, which is not repeated here.

[0192] The test case generation method provided by one or more embodiments of the present specification generates concurrent test points and the fourth test case corresponding to the concurrent test point using the concurrent analysis agent after generating the initial test point, which realizes active and automatic identification and test coverage of high-concurrency scenarios, effectively prevents online major accidents (such as capital loss and data disorder) caused by concurrency problems, and improves the quality assurance capability to the production level.

[0193] In addition, in order to analyze whether the interface or operation has idempotency (i.e., the result is the same whether the request is made once or multiple times), to prevent business errors caused by repeated requests, the fifth test case can be generated using the idempotency analysis agent after identifying the change point and identifying the interface change point. The specific implementation is as follows:

[0194] The test case generation task also carries an associated historical knowledge base;

[0195] After the change point identification agent analyzes the changes in the analysis result according to the task type and obtains a change analysis result, and generates the change point according to the change analysis result, the method further includes:

[0196] The idempotency analysis agent is used to determine a target change point from the change points according to the change points, the analysis result, and / or the associated historical knowledge base, and generate a test case generation suggestion corresponding to the target change point.

[0197] The test case generation agent is used to generate a fifth test case corresponding to the test case generation task according to the test case generation suggestion and the analysis result.

[0198] The idempotency analysis agent, also known as idempotency analysis Agent, is an intelligent agent responsible for analyzing whether an interface or operation has idempotency (i.e., multiple repeated execution of the same operation results in consistent results as a single execution). It is used to identify operations that require idempotency guarantee and design verification strategies.

[0199] The target change point can be understood as a subset of all change points. It refers to those change points that modify or add "write operation" interfaces, and the execution results of these interfaces will change the system state (such as creating an order, paying a deduction, updating a state). For example, adding an aa interface can be understood as a target change point.

[0200] The test case generation suggestion can be understood as a structured test instruction or scheme that describes "what to test" and "why", rather than "how to test". For example, the test case generation suggestion is: suggest idempotency testing for aa interface: use the same request parameters (especially the same order number) to send N times of requests in succession, and verify that only one order is successfully created and the same result is returned.

[0201] The fifth test case can be understood as an executable and step-by-step detailed test case generated by the test case generation agent according to the test case generation suggestion. It includes but is not limited to specific request parameters, the number of repeated sending, and detailed assertions for verifying database state and interface return result.

[0202] The specific implementation of "using the idempotency analysis agent to determine a target change point from the change points according to the change points, the analysis result, and / or the associated historical knowledge base, and generate a test case generation suggestion corresponding to the target change point; using the test case generation agent to generate a fifth test case corresponding to the test case generation task according to the test case generation suggestion and the analysis result" is illustrated.

[0203] Firstly, the idempotency analysis agent analyzes each change point according to the change point, the parsing result and / or the associated historical knowledge base to determine a target change point from the change point. For example, each change point can be analyzed by semantic analysis, rule base matching, historical knowledge query, etc. to select a target change point from the change point, wherein the semantic analysis is to judge the business intention by analyzing the name of the interface, the HTTP method and the interface description extracted from the parsing result in the change point. The create, pay and submit operations usually require idempotency. The rule base matching is that the idempotency analysis agent has a rule base containing a plurality of rule examples, such as the rule example: IF (interface method = POST) AND (interface name contains ['create','submit', 'pay']) THEN idempotency risk = high, and whether the change point is the target change point is determined by the rule base matching. The historical knowledge base query is to query the historical test knowledge base to check whether the change point has a similar interface historical use case marked as requiring idempotency test.

[0204] Secondly, after the idempotency analysis agent determines the target change point, a structured test suggestion corresponding to each target change point, i.e. a test case generation suggestion, can be generated, and an attribute or label can be added to the corresponding test point.

[0205] Finally, the test case generation agent and the parsing result generate the fifth test case. In actual implementation, when the target change point has the attribute or label added to the corresponding test point, the test case generation agent determines that the test point has the attribute or label when generating detailed test cases for the test point, and generates the corresponding idempotency test case, i.e. the fifth test case, when generating the first test case for the test point.

[0206] The test case generation method provided by one or more embodiments of the present specification automatically identifies all "write operation" change points (target change points) by using the idempotency analysis agent, intelligently generates accurate test suggestions, ensures complete test coverage, realizes automatic and standardized verification of the key quality attribute of idempotency, fundamentally eliminates serious online faults such as data duplication and multiple fund deductions caused by repeated requests, and greatly improves the reliability and robustness of the system.

[0207] In addition, if the test case generation method provided by one or more embodiments of the present specification is applied to the financial field, in order to ensure compliance, the resource identification intelligent agent can be used to analyze the resources of the change point, and the change point involving resources can be marked to attract the attention of the tester in the subsequent process. The specific implementation manner is as follows:

[0208] The change point recognition intelligent agent performs change analysis on the parsing result according to the task type, obtains a change analysis result, and generates the change point, and further comprises:

[0209] The resource recognition intelligent agent performs resource analysis on the change point, obtains a resource analysis result, and acquires a resource change point from the change point according to the resource analysis result, and labels a security level for the resource change point, wherein the resource change point is any one or more of the change points;

[0210] The test point generation intelligent agent generates an initial test point according to the change point and the parsing result, and further comprises:

[0211] The resource test point corresponding to the resource change point is acquired from the initial test point, and a resource label is set for the resource test point, wherein the resource test point is any one or more of the initial test points.

[0212] The resource recognition intelligent agent can also be referred to as a resource recognition agent, which is an intelligent agent responsible for analyzing change points from the perspective of security and operation. It is used to identify changes related to sensitive or critical system resources and assess their security risks.

[0213] Resource analysis can be understood as the process of resource recognition agent scanning change points and identifying resource objects operated or affected. The resource analysis result can be understood as an output of the analysis, which is a list containing the identified resources and their context. Resources include but are not limited to data (such as user mobile number, amount), infrastructure (server configuration, network policy), etc.

[0214] The security level can be understood as the risk rating label set by the resource recognition agent for each resource change point.

[0215] The resource test point can be understood as a test point in the initial test point for testing the "resource change point". The resource label is a label set for it, which is usually associated with the corresponding "security level".

[0216] Specifically, the resource recognition agent first performs resource analysis on the change point through preliminary filtering based on keywords and patterns, deep analysis based on semantics and context, and correlation analysis based on dependency relationship, obtains a resource analysis result, acquires a resource change point from the change point according to the resource analysis result, and sets a risk level label for the resource change point, i.e. sets a security level for the resource change point. At the same time, the resource test point corresponding to the resource change point is acquired from the initial test point, and the corresponding resource label is set for the resource test point, so as to attract the attention of testers in the subsequent.

[0217] The test case generation method provided by one or more embodiments of the present specification can automatically identify risk changes related to sensitive resources (such as funds, privacy data, and permissions) through analysis of resource identification agents, and accurately mark the security level. This makes test design no longer purely function-oriented, but incorporates a security perspective. Finally, all related test points are labeled with "resource labels" to form a risk test set. Subsequent testers can prioritize and focus on high-security level testing to ensure asset loss. By significantly advancing the intervention time of security testing from the later penetration testing to the test design stage of the development stage, "security left shift" is achieved, which can detect potential vulnerabilities earlier and at a lower cost, greatly reducing the risk of online security incidents and asset loss due to improper resource operation.

[0218] In specific implementation, if a change point involves the operation of multiple microservices or systems, in order to ensure the eventual consistency of data, a sixth test case can be generated by using a distributed transaction agent. The specific implementation is as follows:

[0219] The change point identification agent, according to the task type, performs change analysis on the analysis result to obtain a change analysis result, and according to the change analysis result, generates a change point after the change point.

[0220] A distributed transaction agent is used to analyze transactions of the change point to obtain a transaction analysis result, and according to the transaction analysis result, a transaction change point is obtained from the change point, wherein the transaction change point is any one or more of the change points.

[0221] A test point generation agent is used to generate a transaction test point according to the transaction change point and the analysis result.

[0222] A test case generation agent is used to generate a sixth test case corresponding to the test case generation task according to the transaction test point and the analysis result.

[0223] The distributed transaction agent can also be referred to as a distributed transaction agent, which is an intelligent agent responsible for analyzing cross-service and cross-database operation processes. It is used to identify change points that may disrupt the eventual consistency of data in a distributed environment.

[0224] Transaction analysis can be understood as the process of distributed transaction agent scanning change points to determine whether it participates in a business operation across multiple system boundaries (such as determining whether an operation involves multiple microservices or databases by analyzing code call chains, message queue dependencies, or database access patterns). The transaction analysis result can be understood as the output of the distributed transaction agent analysis, including the identified distributed transaction flow and its boundary information.

[0225] Transaction change point can be understood as a subset of change points, referring to those change points that are part of a distributed transaction process, such as the change point that modifies the 'order service' to call the 'inventory service' to deduct inventory after creating an order. This is a transaction change point because it is a cross-service transaction: "create order - deduct inventory".

[0226] Transaction test point can be understood as a special test point generated by the distributed transaction agent for transaction change points. It focuses on data consistency and fault recovery. For example, the transaction test point is to verify whether the order service can correctly roll back or initiate compensation operations when the inventory service fails to deduct inventory, ensuring final data consistency.

[0227] The sixth test case can be understood as the final, executable distributed transaction test scheme, including specific steps and assertions to simulate various intermediate state failures (such as network timeout, service downtime, database exception).

[0228] In specific implementation, first, the distributed transaction agent performs transaction analysis on change points to identify the specific function points or service interfaces affected by the current requirement change, and obtains the transaction analysis result. When the transaction analysis result identifies that the change point crosses multiple microservices (such as order service and inventory service), and there is data dependency or state change between these services (such as order deduction), the change point is considered as a transaction change point. Second, the test point generation agent will generate transaction test points based on the transaction change point and the analysis result, such as verifying whether the order service can correctly roll back order creation when the inventory service call times out or returns failure. That is, the distributed transaction agent determines whether there are multiple change points and they constitute a cross-service, state-changing business process. Only when multiple service change points cooperatively complete a business goal (such as "order and deduct inventory"), is it considered that there is a distributed transaction risk, and then triggering consistency analysis, using the test point generation agent to generate corresponding transaction test points. Finally, using the test case generation agent, according to the transaction test point and the analysis result, the sixth test case corresponding to the test case generation task is generated.

[0229] The test case generation method provided by one or more embodiments of the present specification generates the sixth test case by using a distributed transaction agent, realizes the automated test design for the complex challenge of data consistency under a distributed system, effectively prevents the whole system data disorder caused by partial failure, and provides a key guarantee for the reliability of the microservice architecture. That is, the distributed transaction agent automatically identifies the transaction flow (transaction change point) across services, and intelligently generates test cases (sixth test cases) for various abnormal branches (such as compensation logic in Saga, TCC mode), realizing full coverage testing of the distributed transaction life cycle.

[0230] In actual application, when generating the test case analysis report corresponding to the test case generation task, the test analysis document generation agent will not only consider the parsing result, the change point, the initial test point, the first test case, but also cover the second test case, the third test case, the fourth test case, the fifth test case, the sixth test case and the like output by the above-mentioned other agents, that is, the output results of any of the above-mentioned agents, which will be considered and included in the test case analysis report of the test analysis document generation agent when generating the test case analysis report corresponding to the test case generation task, so as to ensure the completeness and accuracy of the test case analysis report.

[0231] The test case generation method provided by one or more embodiments of the present specification decomposes the test case generation task into multiple levels of task execution such as "parsing result generation", "test point generation", "test case generation" and the like. First, the target document is parsed by calling the information extraction agent to generate a parsing result, then the test point generation agent is called to generate multiple initial test points according to the task type and the parsing result, and finally the test case generation agent is called to generate a first test case for each initial test point. First, the analysis logic of a senior test expert is simulated by multiple agents to automatically read and integrate multi-dimensional information such as PRD, design documents and the like in the target document, directly converting human experience into stable automated output, overcoming the problem of test point omission caused by insufficient or negligence of personnel business familiarity, greatly improving the efficiency and completeness of test case output. Secondly, through the joint processing of multiple agents, it is ensured that the generated test cases all follow the built-in best practices and standard processes, eliminating the quality fluctuations of test points caused by personnel level differences, providing stable, reliable and high-quality automated guarantee, and reducing the excessive dependence on manual review of senior engineers.

[0232] In specific implementation, the test case generation method provided by one or more embodiments of the present specification can be understood as a technical basis supporting three major usage modes from basic, advanced to senior in the business domain.

[0233] In the basic stage, the test case generation method can be understood as a test document generation engine. Through accurate change point identification, test point derivation and case generation, the test document is automatically written in a short time (such as 20 minutes) to achieve 70% of the test case coverage and 80% of the test case recall rate, which provides high-quality drafts for efficient secondary processing for testers, and lays the foundation for intelligent "online".

[0234] In the advanced stage, the test case generation method is deeply integrated with data construction capability and used to drive intelligent data construction. When generating test cases (such as asset disposal test cases of the gold consumption system), the system can automatically analyze the "data preparation" step and call the external data factory tool interface to generate corresponding complex business data (such as asset packages in a specific state). This not only transfers the test service to the R&D joint debugging stage, but also solves the efficiency bottleneck of test data preparation in complex scenarios; for example, when the test case generation Agent generates a test case, it analyzes the preconditions of the test case, such as the requirement of "logging in as an 11-level member user". The Agent will automatically call the API of the "intelligent data construction platform" and send the instruction: "Please construct a user account with a level of '11 level' and return its login information". Subsequently, the account information returned by this API will be directly written into the 'precondition' or 'test data' field of the test case.

[0235] Referring to Figure 3 , Figure 3 FIG. 1 shows a data mapping diagram in a test case generation method according to an embodiment of the present specification.

[0236] Figure 3 The process of integrating multiple sources of chaotic user raw data into unified and reliable identifiers through a strategy model is described, which is an indispensable infrastructure for building user data intelligence in the advanced stage. Only after completing this "advanced" step of data governance, can high-quality, correlatable data be provided for intelligent analysis, precision marketing and AI applications at the upper level.

[0237] Among them, Figure 3 "Production / development data" is raw data, which can be understood as raw, redundant and inconsistent data about users scattered in various business systems, such as multiple different IDs, protocol numbers, etc. for the same user. "Strategy model" is the core rule, which can be understood as a data integration standard that defines that no matter how many different identifiers from different sources, they must be mapped to a unique and unified user subject. "New account" can be understood as user master data or standard ID generated after rule cleaning and integration.

[0238] The accounts mentioned in the strategy model (attributes include credit account contract number, user number, and certificate number), agreements (attributes include agreement number, user number, and expiration date), bills (attributes include credit account contract number, bill number, and bill), and assets (attributes include credit account contract number, asset, and institution) are all specific data entities generated by the user in different business scenarios. They are all components of "N" and are objects that need to be integrated into "1" (the user). For example, "account" is the financial attribute of the user, "agreement" is the contractual relationship of the user, "bill" is the consumption result of the user, and "asset" is the holding value of the user. The role of the strategy model is to associate and normalize these scattered entities to the unique user identity.

[0239] In the advanced stage, this test case generation method is the key to realizing the "intelligent construction and execution" closed loop. By structurally decomposing the generated test cases into three elements: "data preparation, business operation, and result verification", and converting them into machine-recognizable instructions, AI can automatically execute the test cases (such as automatically calling interfaces and verifying database results), ultimately liberating the role of the test personnel from a repetitive executor to a strategy maker and result reviewer, and achieving full-intelligent iteration of the test workflow. That is, the system can not only generate documents and data, but also directly drive the execution engine to run the test and return the results.

[0240] Referring to Figure 4 , Figure 4 FIG. 1 shows a test case generation method according to an embodiment of the present specification.

[0241] Figure 4 The test case generation method includes case acquisition, account selection & data preparation, intelligent execution, and result verification conversion.

[0242] Among them, the goal of case acquisition is to automatically convert human demand documents (system specification documents) into a set of executable and understandable test plans and specific operation instructions. By inputting the artificial system specification document, AI automatically generates test cases (AI test case generation), and from all the test cases generated by AI, according to the specific goal of this test (such as regression testing and smoke testing), a subset of test cases that need to be executed is filtered and extracted (extracted test cases). The three constitute an intelligent pipeline from demand input to executable test plan, in which AI greatly improves the conversion efficiency and coverage from demand to test case.

[0243] Then the extracted test cases are disassembled into a series of specific and executable atomic operation steps.

[0244] Account selection & data preparation, to intelligently select or manufacture the most suitable test account and data according to the upstream disassembled operation steps, to prepare for test execution. For example, matching existing platform accounts (the system directly searches for a ready-made account in the platform's test account pool that fully meets the test case requirements), matching existing platform to generate new accounts (when most of the conditions of the existing account meet the requirements, but the individual key states do not meet the requirements, the system selects it as a "template" or "basis" to transform it by performing a series of operations to generate a completely new account that meets the requirements), and generating new accounts based on knowledge input (when encountering completely new business or extreme test scenarios, the previous two strategies fail (there is no similar existing account in the platform), the system will completely fabricate a new account that meets the requirements based on its understanding of the business and data model (knowledge) from scratch).

[0245] Intelligent execution, to automatically and reliably execute the test cases and accounts output by the previous two parts in the real environment. For example, intelligent test case matching before gold consumption (in the loan application (pre-loan) approval process in the consumer finance (gold consumption) field, the test platform can intelligently identify the purpose of a newly submitted test case and automatically match, assemble and execute a complete and compliant test pipeline), and platform general link tool matching (when a test task (such as a disassembled test case) arrives, the test platform can automatically find, select and assemble existing and general test tools and services on the platform to form a complete and executable test pipeline).

[0246] Result verification and conversion, to solve the problem of "whether the result is correct". The system state after execution is verified in all directions and depth, and the final report is generated. For example, DB (Database, database) data verification (checking whether the data changes caused in the database after the interface call are correct), TR interface return value verification (checking whether the direct response of the tested interface (API) meets the expected requirements), and link call relationship verification (verifying whether a single interface call triggers the correct downstream service call chain).

[0247] The test case generation method provided by one or more embodiments of the present specification realizes the analysis of multi-modal data by the combination of natural language processing and a rule engine, and effectively improves the semantic understanding ability of complex business rules by means of a structured rule library and a dynamic knowledge graph supplement, so that the test case coverage rate is improved from 70% to more than 90%; the change analysis function based on LLM tracks business rule changes in real time and automatically updates the test case library, breaking through the limitations of the MBT platform relying on static model maintenance, reducing the test case maintenance cost by 50%, and adapting to high-frequency business iteration scenarios; the agent framework is combined with the input analysis and code generation capability of LLM, and through chained prompting and vector database optimization, the whole process automation from requirement document analysis to test script generation is realized, and the processing efficiency of large-scale requirement documents is improved by 4 times; by integrating concurrent analysis, asset loss analysis, distributed transaction analysis and other modules, the test dimensions not involved in the traditional scheme are covered, and the comprehensiveness of the test is improved by means of rule reasoning and abnormal scenario simulation, reducing the workload of manual supplementary test cases.

[0248] Corresponding to the method embodiments described above, the present specification also provides test case generation device embodiments, Figure 5 A structural schematic diagram of a test case generation device provided by one embodiment of the present specification is shown. As shown in the figure, Figure 5 The device comprises:

[0249] The task determination module 502 is configured to determine a test case generation task, wherein the test case generation task carries a target document to be generated, a task type and a document type.

[0250] The document analysis module 504 is configured to use an information extraction agent to analyze the target document according to the document type, and generate an analysis result of the target document.

[0251] The test point generation module 506 is configured to use a test point generation agent to generate an initial test point according to the task type and the analysis result.

[0252] The test case generation module 508 is configured to use a test case generation agent to generate a first test case corresponding to the test case generation task according to the initial test point and the analysis result.

[0253] Optionally, the device further comprises:

[0254] The change point generation module is configured to utilize a change point identification agent, perform change analysis on the parsing result according to the task type, obtain a change analysis result, and generate a change point according to the change analysis result.

[0255] The test point generation module 506 is further configured to:

[0256] The test point generation module 506 is further configured to:

[0257] The test point generation module 506 is further configured to:

[0258] The test point generation module 506 is further configured to:

[0259] The test point generation module 506 is further configured to:

[0260] The test point generation module 506 is further configured to:

[0261] The test point generation module 506 is further configured to:

[0262] The test point generation module 506 is further configured to:

[0263] The test point generation module 506 is further configured to:

[0264] The apparatus further includes:

[0265] The regression test module is configured to utilize a regression test identification agent, perform regression analysis on the change point according to the regression strategy, obtain a regression analysis result, and determine a system associated component according to the regression analysis result; and

[0266] The regression test module is configured to utilize a regression test identification agent, perform regression analysis on the change point according to the regression strategy, obtain a regression analysis result, and determine a system associated component according to the regression analysis result; and

[0267] The test point generation module 506 is further configured to:

[0268] The apparatus further comprises:

[0269] The non-functional test module is configured to perform feature analysis on the change point by using a non-functional test agent, extract key features of the change point, match the key features with a preset non-functional test rule library, and in the case of successful matching, generate a third test case according to the associated historical knowledge base and the analysis result.

[0270] Optionally, the apparatus further comprises:

[0271] The concurrency analysis module is configured to perform concurrency analysis on the initial test point by using a concurrency analysis agent, obtain a concurrency analysis result, and in the case of determining that the initial test point has a concurrency scenario according to the concurrency analysis result, generate a concurrency test point.

[0272] The test case generation agent is used to generate a fourth test case corresponding to the test case generation task according to the concurrency test point and the analysis result.

[0273] Optionally, the test case generation task further carries an associated historical knowledge base.

[0274] The apparatus further comprises:

[0275] The idempotency analysis module is configured to determine a target change point from the change point according to the change point, the analysis result, and / or the associated historical knowledge base by using an idempotency analysis agent, and generate a test case generation suggestion corresponding to the target change point.

[0276] The test case generation agent is used to generate a fifth test case corresponding to the test case generation task according to the test case generation suggestion and the analysis result.

[0277] Optionally, the apparatus further comprises:

[0278] The resource identification module is configured to perform resource analysis on the change point by using a resource identification agent, obtain a resource analysis result, and obtain a resource change point from the change point according to the resource analysis result, and mark a security level for the resource change point, wherein the resource change point is any one or more of the change points.

[0279] The apparatus further comprises:

[0280] The resource label setting module is configured to obtain a resource test point corresponding to the resource change point from the initial test point, and set a resource label for the resource test point, wherein the resource test point is any one or more of the initial test points.

[0281] Optionally, the apparatus further comprises:

[0282] a distributed transaction module configured to utilize a distributed transaction agent to perform transaction analysis on the change points, to obtain a transaction analysis result, and to obtain a transaction change point from the change points according to the transaction analysis result, wherein the transaction change point is any one or more of the change points;

[0283] a test point generation agent configured to generate a transaction test point according to the transaction change point and the analysis result;

[0284] a test case generation agent configured to generate a sixth test case corresponding to the test case generation task according to the transaction test point and the analysis result.

[0285] Optionally, the apparatus further comprises:

[0286] a change analysis module configured to utilize a change analysis agent to perform text analysis on the change points, to generate an analysis report corresponding to the change points.

[0287] Optionally, the apparatus further comprises:

[0288] a test analysis document generation module configured to utilize a test analysis document generation agent to generate a test case analysis report corresponding to the test case generation task according to the analysis result, the change points, the initial test points, and the first test case.

[0289] Optionally, the apparatus further comprises:

[0290] a visual image generation module configured to utilize a visual image generation agent to perform graphic processing on the test case analysis report, to obtain a visual image analysis report corresponding to the test case analysis report.

[0291] It can be understood that each of the above modules refers to a computer program or program segment for executing one or more specific functions. In addition, the distinction of each of the above modules does not mean that the actual program code must also be separate.

[0292] For ease of description, the above apparatus is described in various modules or units in terms of functions. Of course, the functions of each module or unit can be implemented in one or more software and / or hardware when implementing one or more of the present specification, and modules implementing the same function can be implemented by a combination of multiple sub-modules or sub-units. The above-described apparatus embodiments are only illustrative, for example, the division of the units is only a logical function division, and actual implementation can have another division mode, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0293] The above is a schematic scheme of a test case generation apparatus of the present embodiment. It should be noted that the technical scheme of the test case generation apparatus belongs to the same concept as the technical scheme of the test case generation method described above, and the technical scheme of the test case generation apparatus which is not described in detail can be referred to the description of the technical scheme of the test case generation method.

[0294] Referring to Figure 6 , Figure 6 A structural block diagram of a computing device 600 is shown according to one or more embodiments of the present specification. The components of the computing device 600 include, but are not limited to, a memory 610 and a processor 620. The processor 620 is connected to the memory 610 through a bus 630, and a database 650 is used to save data.

[0295] The computing device 600 also includes an access device 640 that enables the computing device 600 to communicate via one or more networks 660. Examples of such networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or combinations of such networks, such as the Internet. The access device 640 can include one or more of any type of network interface (for example, a network interface card (NIC)) such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, a Near Field Communication (NFC) interface, or the like.

[0296] In one embodiment of the present application, the above-mentioned components of the computing device 600, as well as other components not shown in FIG. 6, can be connected to each other by a bus. It should be understood that Figure 6 the components of the computing device 600 can be connected to each other by a bus. It should be understood that Figure 6 The computing device structure diagram shown is merely for the purpose of example, and is not a limitation on the scope of the present application. Those skilled in the art can add or replace other components as needed.

[0297] The computing device 600 can be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (for example, a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (for example, a smartphone), a wearable computing device (for example, a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 600 can also be a mobile or stationary server.

[0298] The processor 620 is configured to execute computer program / instructions that implement the steps of the above-mentioned data processing method when executed by the processor.

[0299] The above is a schematic scheme of the computing device of the embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the data processing method described above belong to the same concept, and the details of the technical scheme of the computing device that are not described in detail can be referred to the description of the technical scheme of the test case generation method.

[0300] An embodiment of the present specification also provides a computer readable storage medium storing computer programs / instructions, which, when executed by a processor, implement the steps of the test case generation method described above.

[0301] Each of the embodiments in the present specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment mainly describes the difference from other embodiments. In particular, for the computer readable storage medium embodiment, since it is basically similar to the test case generation method embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the test case generation method embodiment.

[0302] An embodiment of the present specification also provides a computer program product comprising computer programs / instructions, which, when executed by a processor, implement the steps of the test case generation method described above.

[0303] The above is a schematic scheme of the computer program product of the embodiment. It should be noted that the technical scheme of the computer program product and the technical scheme of the test case generation method described above belong to the same concept, and the details of the technical scheme of the computer program product that are not described in detail can be referred to the description of the technical scheme of the test case generation method.

[0304] Each of the embodiments in the present specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment mainly describes the difference from other embodiments. In particular, for the computer readable storage medium embodiment, since it is basically similar to the test case generation method embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the test case generation method embodiment.

[0305] The above described embodiments of the present description have been described. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order and still achieve desirable results. Additionally, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve the desired results. In certain implementations, multitasking and parallel processing can be advantageous.

[0306] In the 1990s, it was possible to distinguish whether an improvement in a technology was a hardware improvement (e.g., an improvement in the circuit structure of a diode, transistor, switch, etc.) or a software improvement (an improvement in a method flow). However, as technology has advanced, many improvements in method flows today can be considered as direct improvements in hardware circuit structures. Designers almost always obtain a corresponding hardware circuit structure by programming an improved method flow into a hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented using a hardware entity module. For example, a programmable logic device (PLD) (e.g., a field programmable gate array (FPGA)) is an integrated circuit whose logic function is determined by user programming of the device. A designer programs a digital system "integrated" on a PLD by himself / herself, without having to ask a chip manufacturer to design and manufacture a special integrated circuit chip. Furthermore, instead of manually manufacturing an integrated circuit chip, this programming is now mostly implemented using "logic compiler" software, which is similar to a software compiler used when developing a program, and the original code before compilation is also written in a specific programming language, which is called a hardware description language (HDL), and there are many types of HDL, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc., and the most commonly used are VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also be aware that it is only necessary to logically program a method flow using the above-mentioned hardware description languages and program it into an integrated circuit to easily obtain a hardware circuit that implements the logical method flow.

[0307] The controller can be implemented in any suitable way, for example, the controller can take the form of a microprocessor or processor and a computer readable medium storing computer readable program code, such as software or firmware, executable by the (micro)processor, logic gates, switches, an application specific integrated circuit (ASIC), a programmable logic controller and an embedded microcontroller, examples of which include but are not limited to the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320, the memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also know that, in addition to implementing the controller in pure computer readable program code, it is also possible to implement the controller in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers and embedded microcontrollers, etc. to perform the same functions by logically programming the method steps. Such a controller can therefore be considered as a hardware component, and the means included therein for performing various functions can also be considered as structures within the hardware component. Alternatively, the means for performing various functions can even be considered as both a software module implementing the method and a structure within the hardware component.

[0308] The systems, apparatuses, modules or units illustrated by the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer can be a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0309] For the sake of description, the above apparatuses are described in various units with functions respectively. Of course, the functions of the units can be implemented in one or more software and / or hardware in the implementation of the present application.

[0310] Those skilled in the art will understand that one or more embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0311] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof.

[0312] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof.

[0313] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof. Figure 1 Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof.

[0314] In one typical configuration, the computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0315] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) and / or cache memory, non-volatile memory, such as read-only memory (ROM), EPROM, and / or flash memory. The memory is an example of computer-readable media.

[0316] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0317] The present application can be described in the general context of computer- executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform particular tasks or implement particular abstract data types. The present application can also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communication network. In a distributed computing environment, program modules can be located in both local and remote computer storage media including memory storage devices.

[0318] The above only describes the embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various changes and variations. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the scope of claims of the present application.

Claims

1. A test case generation method applied to a multi-agent architecture, comprising: determining a test case generation task, wherein the test case generation task carries a target document to be generated, a task type and a document type, the target document comprising multi-modal data, and the task type is used to guide a working mode of a change point identification agent; using an information extraction agent to parse the target document according to the document type, and generating a parsing result of the target document; using a change point identification agent to perform change analysis on the parsing result according to the task type, obtaining a change analysis result corresponding to the task type, and generating a change point according to the change analysis result, wherein the task type comprises new function test, iteration / optimization test and repair verification test; using a test point generation agent to generate an initial test point according to the change point and the parsing result; using a test case generation agent to generate a first test case corresponding to the test case generation task according to the initial test point and the parsing result. 2.The test case generation method of claim 1, wherein the test case generation task further carries an associated historical knowledge base. The test point generation agent generates an initial test point according to the change point and the parsing result, comprising: using the test point generation agent to retrieve the associated historical knowledge base according to the change point, obtaining a first retrieval result, and generating an initial test point according to the first retrieval result and the parsing result. 3.The test case generation method of claim 1, wherein the test case generation task further carries an associated historical knowledge base. The test case generation agent generates a first test case corresponding to the test case generation task according to the initial test point and the parsing result, comprising: using the test case generation agent to retrieve the associated historical knowledge base according to the initial test point, obtaining a second retrieval result, and generating a first test case corresponding to the test case generation task according to the second retrieval result and the parsing result. 4.The test case generation method of any one of claims 1-3, wherein the test case generation task further carries an associated historical knowledge base and a regression strategy. After the change point identification agent performs change analysis on the parsing result according to the task type, obtains a change analysis result, and generates a change point according to the change analysis result, the method further comprises: using a regression test identification agent to perform regression analysis on the change point according to the regression strategy, obtaining a regression analysis result, and determining a system associated component according to the regression analysis result; and retrieving the associated historical knowledge base according to the system associated component, obtaining a third retrieval result, and determining a second test case according to the third retrieval result, wherein the system associated component is a component having a direct or indirect dependency relationship with the change point, and the second test case is a historical test case having a matching relationship with the system associated component.

5. The test case generation method according to any one of claims 1-3, wherein the test case generation task further carries an associated historical knowledge base; the change point identification agent further comprises, after performing change analysis on the parsed result according to the task type to obtain a change analysis result and generating a change point according to the change analysis result, the following steps of: performing feature analysis on the change point by a non-functional test agent, extracting a key feature of the change point, matching the key feature with a preset non-functional test rule base, and in the case of successful matching, generating a third test case according to the associated historical knowledge base and the parsed result.

6. The test case generation method according to any one of claims 1-3, wherein the test point generation agent, after generating an initial test point according to the change point and the parsed result, further comprises the following steps of: performing concurrency analysis on the initial test point by a concurrency analysis agent to obtain a concurrency analysis result, and in the case of determining that the initial test point has a concurrency scenario according to the concurrency analysis result, generating a concurrency test point; generating a fourth test case corresponding to the test case generation task according to the concurrency test point and the parsed result by the test case generation agent.

7. The test case generation method according to any one of claims 1-3, wherein the test case generation task further carries an associated historical knowledge base; the change point identification agent further comprises, after performing change analysis on the parsed result according to the task type to obtain a change analysis result and generating a change point according to the change analysis result, the following steps of: performing idempotency analysis on the change point, the parsed result and / or the associated historical knowledge base by an idempotency analysis agent to determine a target change point from the change point and generate a test case generation suggestion corresponding to the target change point; generating a fifth test case corresponding to the test case generation task according to the test case generation suggestion and the parsed result by the test case generation agent.

8. The test case generation method according to any one of claims 1-3, wherein the change point identification agent, after performing change analysis on the parsed result according to the task type to obtain a change analysis result and generating a change point according to the change analysis result, further comprises the following steps of: performing resource analysis on the change point by a resource identification agent to obtain a resource analysis result, and according to the resource analysis result, obtaining a resource change point from the change point and marking a security level for the resource change point, wherein the resource change point is any one or more of the change points; the test point generation agent, after generating an initial test point according to the change point and the parsed result, further comprises the following steps of: obtaining a resource test point corresponding to the resource change point from the initial test point, and setting a resource label for the resource test point, wherein the resource test point is any one or more of the initial test points.

9. The test case generation method according to any one of claims 1-3, after the change point identification agent analyzes the change of the parsed result according to the task type to obtain a change analysis result and generates the change point according to the change analysis result, further comprising: a transaction analysis agent that analyzes the change point according to a transaction analysis result to obtain a transaction analysis result, and acquires a transaction change point from the change point according to the transaction analysis result, wherein the transaction change point is any one or more of the change points; a test point generation agent that generates a transaction test point according to the transaction change point and the parsed result; a test case generation agent that generates a sixth test case corresponding to the test case generation task according to the transaction test point and the parsed result.

10. The test case generation method according to any one of claims 1-3, after the change point identification agent analyzes the change of the parsed result according to the task type to obtain a change analysis result and generates the change point according to the change analysis result, further comprising: a change analysis agent that analyzes the change point according to a text analysis result to generate an analysis report corresponding to the change point.

11. The test case generation method according to claim 1, after the test case generation agent generates a first test case corresponding to the test case generation task according to the initial test point and the parsed result, further comprising: a test analysis document generation agent that generates a test case analysis report corresponding to the test case generation task according to the parsed result, the change point, the initial test point, and the first test case.

12. The test case generation method according to claim 11, after the test analysis document generation agent generates the test case analysis report corresponding to the test case generation task according to the parsed result, the change point, the initial test point, and the first test case, further comprising: a visual image generation agent that performs graphic processing on the test case analysis report to obtain a visual image analysis report corresponding to the test case analysis report.

13. A test case generation device applied to a multi-agent architecture, comprising: a task determination module configured to determine a test case generation task, wherein the test case generation task carries a target document to be generated, a task type, and a document type, the target document includes multi-modal data, and the task type is used to guide the working mode of a change point identification agent; a document parsing module configured to use an information extraction agent to parse the target document according to the document type to generate a parsed result of the target document; The change point generation module is configured to utilize a change point identification agent, perform change analysis on the parsing result according to the task type, obtain a change analysis result corresponding to the task type, and generate a change point according to the change analysis result, wherein the task type includes new function testing, iteration / optimization testing, and repair verification testing; The test point generation module is configured to utilize a test point generation agent, generate an initial test point according to the change point and the parsing result; The test case generation module is configured to utilize a test case generation agent, generate a first test case corresponding to the test case generation task according to the initial test point and the parsing result.

14. A computing device, comprising: a memory and a processor; the memory is configured to store computer programs / instructions, and the processor is configured to execute the computer programs / instructions, which, when executed by the processor, implement the steps of the test case generation method according to any one of claims 1 to 12.

15. A computer readable storage medium storing computer programs / instructions, which, when executed by a processor, implement the steps of the test case generation method according to any one of claims 1 to 12.

16. A computer program product comprising computer programs / instructions, which, when executed by a processor, implement the steps of the test case generation method according to any one of claims 1 to 12.

Images