36 results about "Path coverage" patented technology

The invention particularly relates to a unit test case rapid generation method based on an AI large model. According to the unit test case rapid generation method based on the AI large model, structure information is extracted through a static AST, an execution path is tracked through dynamic instrumentation, and code analysis is achieved; an analysis result is converted into cue words through AI, a multi-scene use case is generated through reasoning, and the multi-scene use case is matched with a language framework; counting row / branch / path coverage, identifying gaps and optimizing loop logic; hot spots and paths are displayed through a visual report, IDE one-key repair and manual correction of a feedback model are supported, intelligent retrieval and reuse of historical cases are realized by relying on FAISS, and the generation accuracy is continuously improved. According to the unit test case rapid generation method based on the AI large model, high-coverage-rate test case generation is achieved through the AI large model, the test case generation efficiency is remarkably improved, the test period is shortened, the software quality is improved, the workload of manually compiling and maintaining the test case is reduced, and the development cost is reduced.

The invention relates to a test case generation method and system based on an industrial agent. The method comprises the steps that an industrial data set is acquired, and a system dependency model is constructed; generating a path coverage test sequence through the first agent; generating a boundary value test data set through the second agent; generating abnormal test scene information through a third agent; and generating a test case set based on the path coverage test sequence, the boundary value test data set and the abnormal test scene information. In conclusion, by constructing the system dependency model and utilizing the multi-agent collaborative analysis process logic path, boundary condition and dependency relationship, the multi-surface coverage of the industrial control system test scene is realized, the automation of the test case generation is realized, the effects of improving the generation efficiency and enhancing the comprehensiveness of the test coverage are achieved, and the method is suitable for industrial control system test. Especially for the coverage of boundary conditions, abnormal scenes and concurrent paths, the manual dependence is reduced, and the reliability and consistency of the test are improved.

The invention relates to a UUV heterogeneous platform global task planning method and system based on full path coverage, and belongs to the technical field of underwater target exploration. According to the method, a three-dimensional environment is reduced to a two-dimensional grid plane, regions are divided according to the number of UUVs, real-time obstacle avoidance and bow-shaped path planning are realized in combination with an improved artificial potential field algorithm, and collaborative operation of A-type (detection disposal type) and B-type (detection-only type) heterogeneous UUVs is supported. The system has a dynamic scheduling mechanism, when the B-type UUV finds a target, the latest idle A-type UUV is scheduled for processing, and task duration and load balancing are optimized while full-area coverage is guaranteed. Simulation examples show that the method is high in planning efficiency, smooth in path, stable in total task time and suitable for detection tasks in complex underwater environments.

The invention belongs to the technical field of software testing, is suitable for the medical field and the financial field, and discloses a service flow chart-based test path generation method, device and equipment and a medium, and the method comprises the steps: obtaining a service flow chart of a service scene, and extracting nodes of the service flow chart and path topology information between the nodes; on the basis of the service importance weight and the historical execution probability of each node, key test paths in the path topology information are identified, and priorities of the key test paths are sorted; constructing a path coverage rate prediction model based on the dynamically adjusted priority of the key path and the historical execution probability of the key path, and predicting the coverage rate of each key path in the business flow chart based on the constructed path coverage rate prediction model; and configuring a test case set based on the predicted coverage rate of each key path in the business flow chart, and dynamically adjusting the configured test case set based on a multi-objective optimization algorithm. According to the invention, the test efficiency of the business scene is improved.

The invention provides a fuzzy test method for driving seed scheduling based on secondary hybrid clustering analysis. Comprising the following steps: step 1, loading and sorting seed files, generating a path coverage bitmap by using an afl-showmap tool, performing de-duplication, quantifying a seed execution path, and constructing a bitmap matrix; 2, performing initial seed classification according to the path similarity by using OPTICS clustering; if necessary, the classification of the noise seeds is further refined through K-Means clustering; and step 3, calculating a rare score according to a result so as to select the most representative seed, allocating appropriate energy to each seed, and dynamically updating a new seed so as to maximize path diversity.

The invention discloses an intelligent question answering system and method based on generative AI, and belongs to the technical field of natural language processing. The system comprises a problem input interface, an extensible classification matching module (including a dynamic threshold decision tree and semantic extension verification), a strategic output engine, a strategy cooperation controller (a main-auxiliary strategy hard binding mechanism) and an output control layer. According to the method, through semantic feature analysis, adaptive classification mapping, strategy cooperation control and conflict solving processes, the problems of fuzzy classification, strategy cooperation deficiency and classification system stiffness in the prior art are solved. The innovation points comprise a dynamic threshold decision tree (reference value = alpha * (1-beta * ln (classification number)), alpha belongs to [0.48, 0.52], beta belongs to [0.08, 0.12]), a strategy collaboration mechanism bound with RBAC authority, and a JSON template classification expansion interface. Through 100,000-level real data tests, the key action missing rate is reduced by 29.8%, the executable path coverage rate is improved by 37.1%, and the method is suitable for multi-strategy collaborative scenes such as law compliance, supply chain management and medical diagnosis.

The invention discloses a logic replication decision-making method and device, equipment and a storage medium. Comprising the following steps: extracting key parameters of each copy unit, and calculating the revenue density of each copy unit according to the key parameters; generating a copy unit list based on the revenue density, and selecting candidate combinations from the copy unit list; and calculating a cost value of each candidate combination to determine an optimal copy unit combination. The key parameters of the copy unit are extracted through the script tool, the key parameters comprise the multiplexing frequency, the time sequence margin lifting amount and the resource overhead, and errors and low efficiency caused by manual extraction are avoided. The replication value is quantified through a revenue density function, a list is generated, then the area cost and the time sequence loss are fused, the comprehensive optimization cost of the candidate combinations is quantified, the dimension difference of different dimension indexes is eliminated, and the advantages and disadvantages of the candidate combinations have the comparable basis. The dynamic balance of the area and the time sequence is realized, the problem that the copy area is out of control or the critical path coverage is incomplete is avoided, and the logic copy optimization effect is guaranteed.

The invention discloses a deep learning auxiliary test case generation method for parallel program path coverage, and aims to improve the test efficiency of parallel program path coverage. The method comprises the following specific steps: (1) sampling a test input set conforming to uniform distribution characteristics, obtaining a parallel program execution path of each test input, and forming a representative training sample set; (2) training a deep learning model by using the representative training sample set, estimating a candidate test case of the target path by means of the model, and judging whether the candidate case can cover the target path or not; and (3) updating the deep learning model based on the candidate test cases not covering the target paths and the execution paths of the candidate test cases, and repeatedly iterating application and updating of the deep learning model until a test case set covering all the target paths is generated.

The invention relates to the technical field of information security, in particular to a sequential cipher security analysis method for a composite structure, which comprises the following steps of: acquiring round function parameters of a target cipher algorithm, nonlinear component definition and a key scheduling rule, and generating a standardized configuration data set; dividing the complete algorithm into a plurality of independent low wheel structures along the boundary of the nonlinear component according to the wheel function parameters, and constructing a directed graph data model containing state transition constraints to compress a state space; calling a password component vulnerability feature library to quantify a sub-structure active S box distribution abnormal value and a linear diffusion layer vulnerability path, and generating a risk weight priority ranking list; and inputting the state sequence data into the target application environment simulation platform, and generating a reproducible vulnerability report when the output deviation exceeds a security threshold. Through hierarchical compression, risk guidance and multi-environment verification, the technical defect of insufficient path coverage of a high-round-number lightweight algorithm in a resource-limited scene is solved.

The application discloses a path coverage test method based on agent-assisted variable multi-objective optimization, aiming at improving the effectiveness and efficiency of test case generation. The specific steps are as follows: (1) a variable multi-objective optimization model of the test case generation problem of the parallel program path coverage is established and updated, which is used for guiding the generation of the test case with high effectiveness; (2) after a certain number of iterations of solving the variable multi-objective optimization model based on an intelligent optimization algorithm, each difficult-to-cover target sub-path is determined, and a sample set is formed, laying a foundation for improving the test efficiency of the difficult-to-cover target sub-path; (3) based on the formed sample set, the agent model corresponding to each difficult-to-cover target sub-path is managed, and the superior evolution individual is selected, which is used for actually executing the parallel program, and the execution times and costs of the program are greatly reduced.

The invention discloses a near-end strategy enhanced ant colony optimization path coverage method, and belongs to the technical field of intelligent path planning. Core mechanisms such as a dynamic hybrid decision mechanism, a feedback pheromone updating strategy, a path redundancy suppression method and pheromone concentration double-threshold constraint are introduced, so that the overall coverage rate and convergence rate of the path are remarkably improved, repeated access behaviors are effectively suppressed, the robustness and the adaptivity of the algorithm are enhanced, and the method is suitable for large-scale popularization and application. Therefore, a high-quality planning path which is comprehensive in coverage, simple in path and rapid in response is provided for applications such as an autonomous mobile robot, an unmanned inspection system and intelligent cleaning equipment, and finally efficient and reliable full-path coverage autonomous operation is achieved.

The invention discloses a local weighted linear regression assisted parallel program path coverage test method, and aims to improve the efficiency and effectiveness of test case generation. The method comprises the following specific steps: (1) generating a certain number of program inputs for executing parallel programs to obtain a target function value set, and defining sample input and output pairs to form a multi-output sample set; (2) calculating a Gaussian distance between a population evolution individual and sample input for estimating each-dimensional target function value component to obtain a target function estimation value; and (3) selecting a representative evolution individual for actually executing a parallel program, and judging whether an evolution individual of which the target function value component is 1 exists or not, namely, whether a test case covering a target path is generated or not.

This invention provides a fuzzy testing mutation scheduling optimization method based on the mayfly algorithm, belonging to the field of fuzzy testing. It includes: Step 1, constructing a lightweight mayfly population for each seed in the fuzzy testing process, where individuals in the population represent a probability distribution of a set of mutation operators; Step 2, selecting mutation operators for each seed in the queue according to a uniform probability distribution, performing mutation, and simultaneously collecting path coverage and crash detection as feedback information, and calculating fitness scores through normalization and weighting; Step 3, simulating diverse behavioral patterns of the mayfly population based on individual fitness scores, dynamically updating the probability distribution of mutation operators, and optimizing the probability distribution; Step 4, finally applying the optimized probability distribution to the next round of seed mutation to generate more effective test cases, and repeatedly executing the above steps to achieve continuous optimization of the mutation scheduling strategy, improving fuzzy testing coverage and vulnerability discovery capabilities.

The invention discloses a JavaScript context awareness test generation method and a JavaScript context awareness test generation system based on a large model. A JavaScript engineering project to be analyzed is used as input, a function recognition graph, a function context dependency graph and a third-party interface semantic graph are constructed based on a static analysis technology, and information such as a definition position, a module structure, a calling relation, a parameter type and external dependency of a modeling function is systematically established. On the basis of the structured modeling result, prompt words in a unified format are further constructed and used for guiding a large language model to generate unit test cases with semantic consistency and performability. After the test codes are generated, syntax verification, assertion recognition and dependency completion are carried out on the result, and a test file conforming to an engineering structure is automatically generated. According to the method, the generation accuracy, the path coverage rate and the engineering adaptation capability of the test case can be remarkably improved, and the method is suitable for various JavaScript projects and mainstream Node.js application scenes and has good expansibility and universality.

The application discloses a kind of access control vulnerability detection method and system based on state deviation analysis, which comprises: the source code of web application is as input, the expected behavior logic contained in the site map and code is extracted by static analysis, then the generated site map is as the guide of dynamic analysis, and the login credentials of multiple roles and multiple users are input, to obtain HTTP request and response under different login states;Web application is modeled using finite state machine FSM, and the discovery of access control vulnerabilities is formalized as the difference between the expected FSM behavior model and the actual FSM behavior model, identifying access control vulnerabilities and generating a vulnerability report.The present application can detect complex logic that is difficult to find by static analysis, and the targeted functional testing also greatly improves the detection efficiency, enabling full path coverage.

The embodiment of the application provides a vulnerability detection method and a computing device. Test cases are generated according to a target optimization PASS of a compiler, which can ensure that the test cases are highly related to the characteristics of the target optimization PASS. Basic blocks of vulnerabilities are determined based on intermediate representation data obtained after the test cases are compiled, so that the position of potential vulnerabilities is located more accurately. The test cases are sliced according to the control flow characteristics of the basic blocks of vulnerabilities, and initial seeds generated based on this can have more targeted execution paths to reach the basic blocks of vulnerabilities, thereby significantly improving the directionality and efficiency of fuzz testing, increasing the vulnerability triggering probability and detection accuracy. By fusing the target distance and the optimization path coverage, high-value seeds are selected for mutation, and efficient and accurate detection of the Rust compiler optimization vulnerabilities is realized.

This invention relates to a symbolic execution vulnerability detection method based on the real runtime context of WebAssembly, belonging to the field of software technology. The method includes: 1) Real runtime context clustering and classification: based on multi-dimensional clustering dimensions and recursive similarity matching, massive amounts of real runtime states are classified and redundantly removed, retaining representative context samples; 2) Symbolic context generation: the representative clustered contexts are modeled and standardized using multi-type parameters to construct an initial symbolic context pool, providing input for symbolic execution; 3) Dynamic pruning of symbolic execution: based on path coverage-oriented context sorting and dynamic pruning strategies, symbolic execution and vulnerability detection are implemented on the WebAssembly binary code. This invention achieves accurate and efficient detection of WebAssembly program vulnerabilities by integrating the real runtime context of WebAssembly in a browser environment with a symbolic execution framework, solving the dual problems of distorted results caused by analyzing only WebAssembly code and low efficiency caused by analyzing the entire real context in traditional methods.

The application discloses a test case generation method based on reinforcement learning, which is used for solving the problems of strong manual dependence and low coverage rate of the existing state machine graph test case generation method. The method comprises the following steps: obtaining a state machine graph of a system to be tested, extracting states, transition conditions and assignment operations, and parsing a test path into an operation node queue; assigning a wide value interval to an initial input variable; constructing a reinforcement learning environment model based on dynamic promotion, encoding node progress, variable interval and constraint into a state space, and narrowing the interval boundary into an action space; training an intelligent agent by using deep reinforcement learning, dynamically narrowing the variable interval in node promotion, and obtaining an optimal narrowing strategy; and solving a new test path by using the trained intelligent agent, and outputting a variable feasible region and a representative test case satisfying the constraints of the whole path. The application can be widely applied to software system testing of state machine modeling, and improves path coverage rate and test case generation efficiency.

The invention belongs to the field of taint path searching, and particularly relates to a taint path searching method, system and device and a storage medium, and the method comprises the steps: marking all Source basic blocks and Sink basic blocks according to a marking rule, and generating a Source point list; creating a search instance, marking a Source point variable as a preset state, adding a hash value of the Source point variable into a path hash chain, analyzing an instruction in a basic block to extract a variable identifier for pollution state analysis, and performing path splitting according to an updated dynamic hash table and a global variable pollution abstract table; according to the global variable pollution abstract table and the path hash chain, performing search optimization on the search instance and the split instance to obtain an optimized instance; and verifying the optimization instance to obtain a taint path search result. The method has the effects of effectively reducing the space-time complexity of the path space and providing a feasible optimized path for static taint analysis on the premise of ensuring core vulnerability path coverage.

The invention discloses a test case generation method and device, equipment and a computer storage medium. The method comprises the steps of performing static analysis on a to-be-tested program to obtain program structure information; executing an initial test based on the program structure information, and collecting initial coverage information; selecting a target path from the to-be-covered path set based on a dynamic evaluation mechanism and the initial coverage information; performing symbolization processing on the target path, generating a path constraint and performing constraint solution; when constraint solving succeeds, generating a test case covering the target path; and when the constraint solution fails, generating a guide signal based on the unsuccessfully solved part, and executing a guide type search based on the guide signal to generate a test case covering the target path. According to the method, the problems that complex branches are difficult to cover due to low random test efficiency, symbolic execution path explosion and constraint solution are difficult, a search technology is easy to fall into local optimum, and initial exploration efficiency is low can be solved, and test case generation efficiency and path coverage capability are improved.

The invention discloses a spanning tree symbolic execution reasoning vulnerability micro-patch method, system and device and a medium, and relates to the technical field of vulnerability detection and repairing, and the method comprises the steps: taking a vulnerability basic block in a target program as a root node, reversely traversing a control flow diagram, constructing a tree structure, executing symbolic solution on each path in a tree, and obtaining the vulnerability micro-patch. Extracting path constraint conditions and variable symbol states of vulnerability points to form a path constraint-state pair set, clustering the path constraint-state pair set according to the structure isomorphism of the variable symbol states to generate semantic equivalent vulnerability variant clusters, and extracting a common defect mode for each vulnerability variant cluster to obtain a common defect pattern; and generating a micropatch structure containing the protection logic and outputting the micropatch structure. According to the method, the variant cluster is generated based on variable state isomorphism clustering, the public defect mode is extracted to generate the precise micropatch, a full-automatic closed loop from vulnerability detection to semantic driven repair is formed, and the problems of incomplete path coverage, weak variant recognition capability and poor repair adaptability are solved.

The application belongs to the field of taint path search, and particularly relates to a taint path search method, system, device and storage medium, the method comprising: according to a marking rule, marking all Source basic blocks and Sink basic blocks to generate a Source point list; creating a search instance, marking a Source point variable as a preset state, adding a hash value of the Source point variable to a path hash chain, analyzing a variable identifier in an instruction in a basic block to perform pollution state analysis, and performing path splitting according to an updated dynamic hash table and a global variable pollution abstract table; according to the global variable pollution abstract table and the path hash chain, performing search optimization on the search instance and the split instance to obtain an optimized instance; and verifying the optimized instance to obtain a taint path search result. The application can effectively reduce the space-time complexity of a path space under the premise of guaranteeing core vulnerability path coverage, and can provide a feasible optimized path effect for static taint analysis.

The application discloses a path coverage test data generation method and system based on a reinforcement learning selection strategy, and relates to the technical field of software testing. The method solves the problems of long calculation time, low generation efficiency, unstable exploration process and redundant test data in the existing test data generation of large-scale path coverage or complex programs. The method comprises the following steps: acquiring all executable paths and numbers of a program to be tested; defining the executable paths as agent states and data selection as agent actions, and all states forming a whole state space; generating multiple groups of data according to an initial state obtained by an agent from initial data, and generating new data when the whole state space is not covered; selecting the data by a reinforcement learning selection strategy combining Q learning and a greedy method, defining an action reward function for the path coverage, and retaining the data with the maximum reward value as the new test data; and continuously updating until the test data covers the whole state space. The application is applied to the field of software product testing.

This invention discloses a JavaScript context-aware test generation method and system based on a large model. Taking the JavaScript project to be analyzed as input, and relying on static analysis technology, it constructs a function identification graph, a function context dependency graph, and a third-party interface semantic graph, systematically modeling information such as function definition locations, module structures, call relationships, parameter types, and external dependencies. Based on this structured modeling result, it further constructs a unified format of prompt words to guide the large language model in generating unit test cases with semantic consistency and executability. After generating test code, it supports syntax validation, assertion recognition, and dependency completion, automatically generating test files that conform to the project structure. This invention can significantly improve the accuracy of test case generation, path coverage, and project adaptability, and is applicable to various JavaScript projects and mainstream Node.js application scenarios, possessing good scalability and versatility.

The application relates to the technical field of Internet software development test, and particularly provides a test device for Internet software development, which comprises a network construction subsystem, a path selection subsystem and a report generation subsystem. The network construction subsystem identifies key logical nodes by calculating the difference between adjacent logical chains and combining a dynamic threshold value, and forms a multilevel logical chain network. The path selection subsystem defines a transition probability of random sampling based on the multilevel logical chain network, and generates a plurality of test paths through Monte Carlo simulation. The weight of each path is determined by the number of logical nodes covered by the path, and a plurality of paths with the highest weight are selected as preliminary test paths. The report generation subsystem constructs positive and negative example pairs in the multilevel logical chain network. The test paths are optimized so that the paths cover key nodes in the logical chain network. The test paths in the multilevel logical chain network are weighted and aggregated to generate a final test report. The application realizes automatic test path generation based on actual software running characteristics.

A binary file instruction recovery method, system, electronic device and storage medium, the method obtains the address of the suspected function entry through the data space of the binary file; traverses the program path according to the program control flow path, for the part not traversed in the program, the next address adjacent to the control flow transfer instruction is selected to continue the traversal, and the first instruction is analyzed combined with the predetermined strategy; the range between the next address adjacent to the control flow transfer instruction and the address of the first instruction is taken as the function gap, and the address of the suspected function entry falling into the function gap is determined as the legal starting address; starting from the legal starting address, the instructions whose addresses can be aligned with the function gap range are taken as the recovered second instructions. The correctness of the identified instruction semantics is guaranteed, and the efficiency of processing the binary file and the full path coverage of the binary file can be improved.

The invention discloses a path semantic perception-based large language model test case generation method, which comprises the following steps of: analyzing a to-be-tested item by utilizing a static analysis technology, and constructing an item-level dependency call graph to extract code elements and dependency relationships thereof; obtaining a dependency entity of the function to be tested according to the project dependency graph, and analyzing an execution path of the dependency entity; generating semantic intention description of each path by utilizing a large language model based on a thinking chain path analysis method, and generating test cases consistent with path behaviors according to the semantic intention description; importing the generated test case into a test framework for execution, obtaining a path coverage report and identifying an uncovered path; and according to a coverage feedback result, generating a supplementary test case for the uncovered path by using a large language model, and realizing iterative optimization of the test case. According to the method, path-level refined test case generation can be realized, the test coverage rate and the generation accuracy are improved, manual intervention is reduced, and the efficiency and the intelligent level of automatic testing are improved.

This invention discloses a robot control software testing method integrating DQN and PSO. First, based on the path association characteristics in the robot control software, paths are divided into a high-association path set and a low-association path set. Then, for the low-association path set, a suitable sample set is updated, and the parameters of the DQN network are fine-tuned to achieve reuse of the DQN model for high-association paths by low-association paths. Finally, based on the PSO algorithm, starting with an initial population with high contribution, a test case set covering the target path is generated. This method, through the differentiation of path association characteristics and the efficient reuse of the DQN model, significantly improves testing efficiency while ensuring the sufficiency of robot control software testing, such as path coverage integrity and control accuracy compliance, providing technical support for the reliability verification of robot control strategy software.

The invention discloses a method and device for determining the coverage degree of structured query statements in a test case. The method relates to the field of software testing and artificial intelligence, and comprises the following steps: determining a preset database of a target program, and extracting a first feature of a dynamic structured query statement branch in the preset database and a second feature of a logic execution path of a storage process; inputting the first feature into a first model to obtain a first probability, and inputting the second feature into a second model to obtain a second probability; calculating a branch coverage degree based on the first probability, calculating a path coverage degree based on the second probability, and calculating a statement coverage degree based on the number of structured query statements executed by the test case and the number of static structured query statements; and combining the branch coverage, the path coverage and the statement coverage to obtain the test coverage of the structured query statement of the target program. Through the method and the device, the problem of inaccurate measurement of the coverage degree of the structured query statement in the test case in related technologies is solved.

The application discloses a kind of based on artificial intelligence multimodal data fusion processing method and system, it is related to data processing technical field, including: respectively with each entity as starting point, according to target expansion depth is expanded, obtains the candidate subgraph of each entity;The semantic vector of each subgraph is obtained, according to and extraction;The path vector and path coverage of each path in are obtained, according to and extraction;With original question structured as Prompt after input pre-training large language model, obtain question and answer result;The application is favorable in guaranteeing the accuracy of specific field knowledge question and answer while, give consideration to semantic coverage, the balance of computing resource control and reasoning efficiency.