Financial service-oriented test behavior analysis method and device

Abstract

The invention discloses a financial service-oriented test behavior analysis method and device, and aims at realizing accurate evaluation of the execution condition of testers and the test coverage rate of a system by automatically collecting and analyzing test data. Through an innovative automatic acquisition and test behavior analysis method, the execution condition of a tester is effectively evaluated, meanwhile, the test coverage rate of the system is accurately measured, and the evaluation accuracy is improved; big data analysis and machine learning technologies are utilized to quickly identify problems and potential defects in the test process, test cases and test strategies are optimized, and the test efficiency is improved; the test coverage rate is comprehensively and systematically evaluated in combination with a system architecture diagram and a test case, and analysis is performed from multiple dimensions such as the code coverage rate, the system coverage rate and the transaction coverage rate; critical paths and potential defects are ensured to be fully tested, and the stability and the safety of the system are improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of software testing, and particularly relates to a test behavior analysis method and device for financial services. BACKGROUND

[0002] In current software testing practice, the execution of testers and the coverage evaluation of system testing are two major challenges. On the one hand, the task allocation, execution progress, error rate and other key information of testers often rely on manual recording and offline analysis, which leads to a cumbersome and subjective evaluation process, and it is difficult to form an objective and accurate evaluation result. On the other hand, the coverage evaluation of system testing usually relies on the experience and intuition of testers, and lacks scientific and systematic analysis methods, which leads to the possibility of missing critical paths and potential defects in testing, affecting the stability and security of the system.

[0003] At present, there is a lack of an automatic scheme for analyzing the test behavior of a financial service system and evaluating the coverage of system testing in the prior art. SUMMARY

[0004] To solve the above problems, the application discloses a test behavior analysis method and device for financial services, which automatically collects and analyzes test data to accurately evaluate the execution of testers and the coverage of system testing.

[0005] To achieve the above purpose, the technical scheme of the application is as follows:

[0006] A test behavior analysis method for financial services, comprising the following steps:

[0007] Step 1, collect the system log to be tested, process and store the log data, draw a transaction system relationship link graph and a system association link graph according to the global serial number, each node in the transaction system relationship link graph is each system with the same transaction code, and the missing link nodes are completed according to the global serial number to form a transaction link model library and a system link model library;

[0008] Step 2: analyze the system and each version of the source code to analyze the code related information in the system, and form a code analysis baseline library based on version management;

[0009]

[0009] Step 3, establish the association relationship between requirements and test cases, transaction codes, write test cases, write function codes, and release the codes to the test environment;

[0010] Step 4, according to the test requirement list, get the total number of affected transaction links and the total number of affected systems, based on the code change range obtained from this version submission, filter out the relevant test cases according to the test requirements and execute the relevant test cases, collect the code coverage data during the test execution process and the covered link transaction and system transaction data during the test execution process;

[0011] Step 5, according to the transaction link model library, system link model library and code analysis baseline library, get the test behavior analysis index data.

[0012] Further, the transaction link graph and the system influence relationship link graph in step 1 are drawn based on the global serial number.

[0013] Further, the process of completing the missing link nodes in step 1 includes: log enhancement supplement for the broken link data, and then circular analysis of the supplemented information to find the incomplete link, and then enhancement supplement again until the link is complete.

[0014] Further, step 2 specifically includes the following sub-steps:

[0015] Step 2.1, automatically calculate the coverage rate of the test by using the coverage rate calculation tool;

[0016] Step 2.2, use static code analysis tools and dynamic test analysis tools to identify the key path and potential defects in the business system;

[0017] Step 2.3, in combination with the data obtained in step 2.2, analyze different system and different version source code to analyze the code related information in the system, and form a code analysis baseline library based on version management;

[0018] Step 2.4, automatically generate a test coverage report according to the coverage calculation result and the key path identification result.

[0019] Further, the process of establishing the association between the requirements and the test cases and the transaction codes in step 3 includes the following process:

[0020] Add the code corresponding to the requirement in the test case, comb the related transactions involved in the requirement and the affected systems, and establish the association between the requirement and the transaction code and the test task respectively through natural language processing method according to the requirement description and the test case, transaction and test task related description information; according to the test task, design and write the relevant test case, establish the mapping relationship between the test case and the test task, and the test personnel and the test task; finally, establish the association between the requirement and the test case.

[0021] Further, step 4 includes the following sub-steps:

[0022] Step 4.1, obtaining a requirement list involved in the current version release content;

[0023] Step 4.2, combining the requirement list obtained in step 4.1 and the transaction link model library and the system link model library in the step 1 log analysis module, obtaining the transaction link range and total number and the system link range and total number affected by the requirement list;

[0024] Step 4.3, combining the requirement list and the association between requirements and test cases, obtaining the total number of test cases to be executed associated with the requirements involved in the current version release content;

[0025] Step 4.4, combining the requirement list and the code analysis baseline library in step 2, analyzing the code change of the current version release content;

[0026] Step 4.5, entering the test process and executing the test cases;

[0027] Step 4.6, in the execution process of step 4.5, real-time statistical analysis of the code related data covered in the test execution process; obtaining the number of transactions covered, the number of system links covered, the number of systems covered and the number of test cases covered in the test process.

[0028] Further, in step 5, the test behavior analysis index data includes: test case execution coverage, code coverage, code branch coverage, code line coverage, transaction coverage, system coverage.

[0029] The application also provides a test behavior analysis device for financial services, which is used to realize a test behavior analysis method for financial services, and includes a log analysis module, a code analysis module and a test behavior analysis module. The log analysis module is used to collect data in the test process in real time through an automatic tool, analyze the data after processing, draw and improve the transaction link diagram and the system influence relationship diagram, form the transaction link and system link model library, and analyze the transaction link range and total number and the system link range and total number affected by the requirement list involved in each version release content. The code analysis module is used to automatically calculate the test coverage based on the collected test data and the source code of the financial service system, identify the key path and potential defects in the financial service system, and also automatically generate a test coverage report according to the coverage calculation result and the key path identification result, and analyze the code change range of each version. The test behavior analysis module is used to deeply analyze the collected data and identify the test key indicators.

[0030] The application has the following beneficial effects:

[0031] 1. The present application effectively evaluates the performance of test personnel through innovative automated collection and testing behavior analysis methods, accurately measures system test coverage, improves evaluation accuracy, reduces the influence of manual intervention and subjective judgment, more targetedly focuses on the test range, improves the traceability and transparency of the test process, thereby efficiently and high-qualityly completes the test task under the condition of limited test resources, and solves the main problems faced by current financial business testing work.

[0032] 2. The present application uses big data analysis and machine learning technology to quickly identify problems and potential defects in the testing process, optimize test cases and test strategies, and improve test efficiency.

[0033] 3. The present application combines system architecture diagram and test case to comprehensively and systematically evaluate test coverage, analyzes from multiple dimensions such as code coverage, system coverage, transaction coverage, etc.; ensures that the key path and potential defects are fully tested, and improves the stability and security of the system.

[0034] 4. The present application can provide targeted test optimization suggestions based on data analysis results, such as adjusting test cases, optimizing test strategies, increasing test resources, etc.; intelligently identify bottlenecks and deficiencies in the testing process, and propose practical optimization solutions, which helps to reasonably allocate test resources and improve the overall efficiency of the test work. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 A financial business-oriented test behavior analysis method step schematic diagram is provided for the present application.

[0036] Figure 2 A financial business-oriented test behavior analysis method flowchart is provided for the present application.

[0037] Figure 3 A transaction impact system link schematic diagram is provided. DETAILED DESCRIPTION

[0038] The technical solutions provided by the present application will be described in detail below in combination with specific embodiments, and it should be understood that the following specific embodiments are only used to illustrate the present application and not to limit the scope of the present application.

[0039] The present application provides a financial business-oriented test behavior analysis device, which comprises a log analysis module, a code analysis module and a test behavior analysis module.

[0040] The log analysis module is used for collecting data in the test process in real time through an automatic tool, including test task allocation, execution progress, error logs, etc., analyzing the data after processing, drawing and improving the transaction link diagram and the system influence relationship diagram, forming a transaction link, system link model library; in addition, the log analysis module will analyze the impact of the transaction link range (total number of transaction links) and the impact of the system (total number of systems) according to the test requirement list, and obtain the impact of the transaction link range and the total number, the system link and the business system range and the total number of the requirement list affected by each version release content. The log analysis module specifically implements part of the step 1 and step 4 contents in the test behavior analysis method for financial business.

[0041] The code analysis module is used for automatically calculating the coverage rate of the test based on the collected test data and the source code of the financial business system, including statement coverage rate, branch coverage rate, etc.; and using an algorithm to identify the key path and potential defects in the financial business system, to ensure that these areas are fully tested; and generating a test coverage report according to the coverage calculation result and the key path identification result, for the test team and management to refer. In addition, the code analysis module will analyze the code change range of each version, and obtain the total number of changed code lines, the number of changed classes, the number of methods, the number of branches, etc. The code analysis module specifically implements part of the step 2 and step 4 contents in the test behavior analysis method for financial business.

[0042] The test behavior analysis module is used for establishing the association relationship between requirements and test cases and transaction codes, and the association relationship between test tasks and test cases, test personnel, and deeply analyzing the collected data to identify key indicators such as the behavior mode of the test personnel, execution efficiency, error distribution, etc. In addition, the test behavior analysis module will provide real-time feedback to the test personnel according to the analysis result, to help them adjust the test strategy and improve the test efficiency and quality. The test behavior analysis module specifically implements the step 3 and step 5 contents in the test behavior analysis method for financial business.

[0043] The present application provides a test behavior analysis method for financial business, as shown in Figure 1 、 Figure 2 , including the following steps:

[0044] Step 1: log analysis of the system to be tested, ETL (Extract-Transform-Load) processing of the log of the system to be tested, combining artificial supervised learning, rectification processing of different application scenarios and data, constantly improving the system transaction link library, the completeness of the system link library, and finally forming a transaction link, system link model library, laying the foundation for subsequent requirement impact range analysis. Specifically, the following sub-steps are included:

[0045] Step 1.1: Collect the system logs on the physical environment of the system under test and the system on the cloud environment through ETL technology. The data extraction log collection mainly collects the system execution logs in real time through the FileBeat technology, and the data filtering and cleaning loads the filtered data into the Kafka message queue through Flink / Logstash.

[0046] Step 1.2: Store the filtered and cleaned data in step 1.1 in the distributed real-time data storage ElasticSearch for subsequent data analysis.

[0047] Step 1.3: Analyze the data stored in step 1.2, mainly draw the system link diagram according to the transaction code in the global serial number in the data. The global serial number definition rules include the previous node code, the current node code, the subsequent node code, the transaction code, and the system code, such as 10010021002TY001SYS001XXXXXXXX, etc. rules. Through the transaction code in the global serial number, it is known which systems are affected by the transaction, and thus the global transaction system relationship link diagram is drawn, as shown in Figure 3 Figure 3 In the diagram, TYXXX is the transaction code, and SYSXXX is the system code. In the link diagram, according to the order of transaction data generation, systems with the same transaction code are connected in series as nodes in the link, and there can be multiple branches in the link. The transaction system relationship link diagram nodes and connection relationship are stored in the database. In addition, based on the previous and subsequent nodes of the global serial number in the log data, the system association link diagram can be drawn. The system association link diagram is not limited by the transaction code, and is more complex and comprehensive than the transaction system relationship link diagram. The system association link diagram can be used as a reference and supplement for the transaction system relationship link diagram. The system association link diagram is also stored in the database.

[0048] ​Step 1.4: Due to the incomplete data collected by step 1.2, which may be missing due to partial system or module not outputting as required, data loss during transmission, etc., the link drawn in step 1.3 may be incomplete. If the link graph is disconnected, in order to ensure the completeness and consistency of the output link business, based on the link drawn in step 1.3, combined with supervised learning algorithm model (mainly relying on the link pre-node and subsequent node in the global flow number to infer, continuously optimize), the transaction impact system relationship link graph is analyzed to obtain the incomplete link graph, and then combined with manual decision-making, the missing link node (disconnected link) is completed (the main reason for missing is that the source code has not been logged for this node, or data is lost during data collection, etc.). For link completion, first of all, the source code is enhanced by non-invasive code logging technology (Sandbox), and the disconnected link data is enhanced and supplemented. The supplemented information is analyzed again (multi-round training reasoning), and the system link integrity is continuously trained, so as to achieve the completeness of the link, and improve the coverage rate of the system link library. Finally, the transaction link model library and the system link model library are formed, which lay the foundation for subsequent demand impact range analysis. The transaction link model library stores the improved transaction impact system relationship link graph, and the system link model library stores the system correlation link graph.

[0049] Step 2: The code syntax tree analysis model is used to analyze different system and different version source code, and the total number of code lines involved in the current system, the involved classes, branches and context call management relationship and other information are analyzed, and finally the code analysis baseline library based on version management is formed, which prepares for subsequent code impact range analysis and code analysis during test execution. Specifically, the following sub-steps are included:

[0050] Step 2.1: Based on the collected test data and business system source code, use coverage calculation tools (such as JaCoCo, Emma, etc.) to automatically calculate the test coverage.

[0051] Step 2.2: Use static code analysis tools (such as SonarQube, FindBugs, etc.) and dynamic test analysis tools (such as TestNG, JUnit, etc.) to identify the key path and potential defects in the business system.

[0052] Step 2.3: Based on the data processed in step 2.2, different system and different version source code are analyzed for file directory parsing, configuration file parsing, file type parsing, file filtering processing, file dependency processing, and effective code line number analysis. The total number of code lines involved in the current system, the involved classes, branches, and context call management relationships are analyzed, and finally a code analysis baseline library based on version management is formed. The analysis results are stored in relational, non-relational, or file databases, preparing for subsequent code impact analysis and code analysis during test execution.

[0053] Step 2.4: Based on the coverage calculation results and key path identification results, a test coverage report is automatically generated. This report mainly shows the total file data, total line number, coverage, covered code, and uncovered code, as well as the execution link of the covered code to the developers, testers, and core management personnel. This helps developers clearly understand the code execution logic and quickly locate problems and solve defects raised by testers, while effectively improving the understanding of test case coverage by testers, constantly improving test cases, and enhancing the correctness and integrity of project business logic.

[0054] Step 3: Establish the association between requirements and test cases and transaction codes, as well as the association between test tasks and test cases and testers. When adding test cases, this process is mainly implemented through two ways. Way one: by adding a requirement coding attribute (consistent with the business requirement coding) in the test case to form a related mapping relationship. Way two: through natural language (NLP) processing technology, the requirement description and case description related information are analyzed, and with the continuous training of the natural language generation model (private domain model) as an aid, the automatic matching of the related dependency mapping association is realized through technical means. This process requires the combination of way one and way two, and more manual intervention is needed in the early stage. With the continuous accumulation and improvement of the knowledge base, the fully automated way is finally realized. Similarly, the requirement and transaction code relationship mapping and the requirement and test task association mapping implementation logic are consistent with the above. The specific steps include:

[0055] Step 3.1: This step combines the requirement analysis stage, and the requirement analysis personnel analyzes the detailed requirements, sorts out the related transactions involved in the requirements, and the affected systems, and then uploads them to the test behavior analysis system. Through natural language (NLP) processing technology, the requirement description and case description related information are analyzed, and with the continuous training of the natural language generation model (private deployment based on the general model) as an aid, the automatic matching of the business requirement and transaction code dependency mapping association is realized through technical means.

[0056] Step 3.2, the implementation principle and use of related technical means for the demand and test task association mapping are similar to step 3.1.

[0057] Step 3.3, at the same time of demand establishment, the test team designs and writes relevant test cases according to the assigned test tasks, the test task and test case satisfy one-to-many mapping relationship, and the test personnel and test task satisfy one-to-many mapping relationship. Based on the association between test task and test case, the association between demand and test case is finally realized.

[0058] Based on the above steps, the mapping relationship between demand and test case is finally obtained.

[0059] In addition to the above method, in the case where the test case has been initially designed, the coding and description statements have been determined, we can also directly establish the mapping relationship (one-to-many) between test requirements and test cases through step 3.1 method, such as test personnel Zhang San (A0001), Li Si (A0002), and Wang Wu (A0003), Zhang San (A0001) is responsible for the demand (demand coding rule: X0001, X0002, …), and the demand X0001 contains the use case (use case coding rule: X0001_T0001, X0001_T0002, X0001_T0003 …).

[0060] Step 3.4, at the same time of demand establishment, the R&D team writes the corresponding function code according to the demand document content proposed by the demand team.

[0061] Step 3.5, after the development of the demand, the R&D manager / operation engineer releases the demand related code to the corresponding test environment / production environment.

[0062] Step 4: Test case execution, R&D team releases code based on requirements to test environment (must mark release version and related requirement [unique code] list), at this time, the log analysis module will analyze the impact of the transaction range (total number of transactions) and the impact of the system (total number of systems) based on the test requirement list (including business requirement code), combined with the test requirement and the mapping relationship of the transaction code, the transaction impact system relationship link diagram, and form the statistical benchmark data of this submission (mainly including which transactions are affected by this requirement submission, and which system links are affected). At this time, the code analysis module will also analyze the code change range based on this version submission (mainly based on the difference analysis of the pre-submission version code and the content of this submission code) to obtain the total number of changed code lines, the number of changed classes, the number of methods, the number of branches, etc. Testers select relevant test cases based on the related requirements provided by R&D personnel and execute relevant test cases; at the same time, the code analysis module collects, monitors and analyzes the code coverage data and test behavior analysis module during the test execution process. The specific steps are as follows:

[0063] Step 4.1: Combine the code test information of the R&D team in step 3.4, and analyze the requirements (unique code) list involved in the current version release content based on the description in the code test information. Note: Code test information input should meet the standard requirements and clearly describe the test business requirements and business requirement codes.

[0064] Step 4.2: Combine the requirement list obtained in step 4.1 and the transaction link model library in step 1 log analysis module, and analyze the transaction range and total number affected by the requirements list based on the current version release content through the log analysis model in the log analysis module based on the dependency relationship between requirements and transactions. At the same time, analyze the system link range and total number affected by the test requirement list content based on the current version, and use it as the denominator for calculating the transaction coverage rate in the subsequent execution process. Specifically, based on the requirement code obtained in step 4.1, the transaction code involved can be obtained according to the mapping relationship between requirements and transaction codes obtained in step 3.1, and the specific systems involved by the transaction code can be obtained according to the transaction impact system link diagram.

[0065] Step 4.3: Combine the requirement list obtained in step 4.1 and step 3 to analyze the test cases and their total number based on the requirements involved in the current version release content.

[0066] Step 4.4: Based on the code analysis module, the code change impact analysis based on the code syntax tree is performed to analyze the code changes in the current version release content, including the total number of code changes, the number of classes affected, the number of methods, the number of branches, the context association relationship, and other information, which is used as the denominator for calculating the transaction coverage rate in the subsequent execution process.

[0067] Step 4.5: Based on the requirements list obtained in step 4.1, the association between requirements and test tasks, test tasks and test cases, and test tasks and test personnel, the test tasks to be executed and the test personnel executing the tasks can be obtained. The test team enters the test process and executes the test cases under their respective tasks. Combined with the analysis results of step 4.3, the test team members perform the test cases (this process includes automated test cases, manual test cases, UI test cases, and non-functional test cases). By comparing the actual execution cases with the total number of planned execution cases, it can be objectively analyzed whether the test personnel have completed the tasks according to the plan, avoiding the problem of subjective judgment in traditional methods.

[0068] Step 4.6: During the execution process in step 4.5, the code analysis module and the test behavior analysis module process and dynamically analyze the data in the execution process. At this time, the code analysis module (code module involves system -> module -> class -> method -> branch -> line level) analyzes the total number of covered code lines, the number of covered classes, the number of covered code lines, the number of covered branches, the context relationship, and the code function during the test execution process. At this time, the log analysis module collects and analyzes the log data during the test execution process, analyzes the data processing process in step 1, and analyzes the number of transactions covered, the number of system links covered, the number of systems covered, and the number of use cases covered.

[0069] Step 5: Test behavior analysis index analysis, based on the model library in step 1, the baseline library in step 2, and the real-time collected data in step 4, real-time analysis and processing are performed to analyze the test execution case progress of the test personnel, the code test coverage, the related transaction coverage, and the related system coverage. Specific sub-steps include:

[0070] Step 5.1: Based on the data in Step 1, Step 2, and the real-time information collected in Step 4, the code analysis module and the behavior analysis module perform line index analysis management from the dimensions of test management personnel and test execution personnel. Management personnel (test supervisor, test group leader, quality manager) dimension 1: based on the overall use case execution coverage, code coverage (code branch coverage, code line coverage, code function coverage), transaction coverage, system coverage, and other indicators of this requirement; Implementation personnel dimension 2: based on the test personnel dimension analysis of use case execution coverage, code coverage (code branch coverage, code line coverage, code function coverage), transaction coverage, system coverage, and other indicators. Index calculation formula: Use case execution coverage (refers to the proportion of the number of test cases that have been executed to the total number of test cases) = (number of test cases executed [Note: corresponds to data in Step 4.6] / total number of test cases planned [Note: data in Step 4.3]) x 100%; Code coverage is a broad concept that describes the proportion of code actually executed during testing. Code coverage can be further divided into different types, such as code line coverage = (number of executed statements [Note: corresponds to data in Step 4.6] / total number of statements [Note: corresponds to data in Step 4.4]) x 100%, branch coverage = (number of executed branches [Note: corresponds to data in Step 4.6] / total number of branches [Note: corresponds to data in Step 4.4]) x 100%, code function coverage = (number of executed functions [Note: corresponds to data in Step 4.6] / total number of functions [Note: corresponds to data in Step 4.4]) x 100%, etc.; Transaction coverage (refers to the proportion of executed transaction scenarios to total transaction scenarios) = (number of executed transaction scenarios [Note: corresponds to data in Step 4.6] / total number of transaction scenarios [Note: corresponds to data in Step 4.2]) x 100%; System coverage = (number of covered systems [Note: corresponds to data in Step 4.6] / total number of affected systems [Note: corresponds to data in Step 4.2]) x 100%;

[0071] Step 5.2: While the test personnel execute test cases, the test management personnel (test supervisor) can observe the execution progress based on the test team personnel dimension and the test code coverage, test transaction coverage, system coverage, and other related indicators based on the requirement dimension. Thus, the behavior of test personnel can be effectively managed and the quality of the system can be effectively controlled through objective indicators.

[0072] It should be noted that the above content only illustrates the technical idea of the present application, and cannot limit the protection scope of the present application. For ordinary skilled in the art, without departing from the principle of the present application, a number of improvements and refinements can be made, which fall within the protection scope of the claims of the present application.

Claims

1. A test behavior analysis method for a financial service, characterized by, The method comprises the following steps: Step 1, collecting logs of a system to be tested, processing and storing log data, and drawing a transaction-influenced system relationship link graph and a system correlation link graph according to a global serial number, each node in the transaction-influenced system relationship link graph being each system with the same transaction code, and completing missing link nodes according to the global serial number to form a transaction link model library and a system link model library; Step 2, analyzing the system and each version source code to analyze code-related information in the system and form a code analysis baseline library based on version management; Step 3, establishing an association between requirements and test cases, transaction codes, writing test cases, writing function codes, and publishing the codes to a test environment; Step 4, obtaining the total number of transaction links and the total number of system influences according to a test requirement list, obtaining a code change range based on the version submission, selecting and executing relevant test cases according to the test requirements, and collecting code coverage data during test execution and covered link transaction and system transaction data during test execution; Step 5, obtaining test behavior analysis index data according to the transaction link model library, the system link model library, and the code analysis baseline library.

2. The financial services oriented test behavioral analysis method of claim 1, wherein, The transaction link graph and the system influence relationship link graph in step 1 are drawn based on the global serial number.

3. The financial services oriented test behavioral analysis method of claim 1, wherein, The process of completing the missing link nodes in step 1 comprises: log enhancement supplementing of broken link data, re-circulating analysis of the supplemented information to obtain incomplete links, and re-enhancing supplementing until the links are complete.

4. The financial services oriented test behavioral analysis method of claim 1, wherein, Step 2 specifically comprises the following sub-steps: Step 2.1, automatically calculating the coverage rate of the test by using a coverage rate calculation tool; Step 2.2, identifying the key path and potential defects in the business system by using a static code analysis tool and a dynamic test analysis tool; Step 2.3, analyzing different system and version source codes based on the data obtained in step 2.2 to analyze code-related information in the system and form a code analysis baseline library based on version management; Step 2.4, automatically generating a test coverage report by using a report generation tool according to the coverage rate calculation result and the key path identification result.

5. The financial services oriented test behavioral analysis method of claim 1, wherein, The process of establishing the association between requirements and test cases, transaction codes, and writing test cases in step 3 comprises the following steps: The codes corresponding to the requirements are added to the test cases, the related transactions involved in the requirements and the affected systems are sorted out, the association between the requirements and the transaction codes and the test tasks is established by using a natural language processing method according to the requirement description and the test case, transaction, and test task related description information, the mapping relationship between the test cases and the test tasks and the test personnel and the test tasks is established according to the test task, and finally the association between the requirements and the test cases is established.

6. The financial services oriented test behavioral analysis method of claim 1, wherein, Step 4 comprises the following sub-steps: Step 4.1, obtaining a requirement list involved in the current version release content; Step 4.2, combining the requirement list obtained in step 4.1 and the transaction link model library and system link model library in the step 1 log analysis module, obtaining the transaction link range and total number affected by the requirement list, and the system link and total number; Step 4.3, combining the requirement list and the association between requirements and test cases, obtaining the total number of test cases to be executed associated with the requirements involved in the current version release content; Step 4.4, combining the requirement list and the code analysis baseline library in step 2, analyzing the code changes in the current version release content; Step 4.5, entering the test process and executing the test cases; Step 4.6, during the execution of step 4.5, real-time statistical analysis of the code-related data covered in the test execution process; Obtaining the number of transactions covered, the number of system links covered, the number of systems covered, and the number of test cases covered in the test process.

7. The financial services oriented test behavioral analysis method of claim 1, wherein, In step 5, the test behavior analysis index data includes: test case execution coverage, code coverage, code branch coverage, code line coverage, transaction coverage, and system coverage.

8. A test behavior analysis apparatus for a financial service, characterized by comprising: The test behavior analysis method for financial services according to any one of claims 1-7, comprising a log analysis module, a code analysis module, and a test behavior analysis module, the log analysis module is used to collect data in the test process in real time through an automated tool, process the data, analyze, draw and improve the transaction link diagram and the system influence relationship diagram, form the transaction link and system link model library, and analyze the transaction link range and total number affected by the requirement list involved in each version release content, and the system link range and total number; The code analysis module is used to automatically calculate the test coverage based on the collected test data and the source code of the financial business system, identify the key path and potential defects in the financial business system, and also automatically generate a test coverage report according to the coverage calculation result and the key path identification result, and analyze the code change range of each version; the test behavior analysis module is used to perform in-depth analysis on the collected data to identify test key indicators.

Images