File quality inspection method and device, computer device and computer readable storage medium

Abstract

The application discloses a file quality inspection method and device, computer equipment and a computer readable storage medium, and relates to the technical field of data processing. The method comprises the following steps: obtaining a target file to be inspected and a corresponding label, if the label contains a preset keyword, performing pre-inspection scanning on the target file to be inspected according to a file type to be inspected of the target file to be inspected, obtaining a pre-inspection scanning result, performing compilation verification on the target file to be inspected based on the pre-inspection scanning result, obtaining a compilation verification result, performing deep analysis on the target file to be inspected based on the compilation verification result, obtaining a deep analysis result, performing test verification on the target file to be inspected based on the deep analysis result, obtaining a test verification result, and further performing multi-source data aggregation to obtain a structured quality inspection report of the target file to be inspected. In this way, early problem discovery is realized for different types of files, and a structured quality inspection report is further generated as a decision basis.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a document quality inspection method, apparatus, computer equipment, and computer-readable storage medium. Background Technology

[0002] In traditional development processes, various assets (source code, shaders, images, audio, etc.) lack unified automated inspection standards, leading to problems being discovered only late in development and resulting in high repair costs. While existing tools can inspect single file types, they lack a complete integrated pipeline, making it impossible to perform comprehensive and rapid quality checks upon code submission. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to overcome the shortcomings of the prior art and provide a document quality inspection method, apparatus, computer equipment and computer-readable storage medium for realizing a complete automated process from tag identification and triggering of the document quality inspection process to pre-inspection scanning, compilation verification, in-depth analysis, test verification and report generation of the target document to be inspected.

[0004] This invention provides the following technical solution: In a first aspect, the present invention provides a document quality inspection method, comprising: Obtain the target file to be inspected and its corresponding tags. If the tags contain preset keywords, perform a pre-inspection scan on the target file to be inspected according to the type of the target file to be inspected, and obtain the pre-inspection scan result. Based on the pre-inspection scan results, the target file to be inspected is compiled and verified to obtain the compilation and verification results. Based on the compilation and verification results, a deep analysis is performed on the target file to be inspected, and the deep analysis results are obtained. Based on the in-depth analysis results, the target file to be inspected is tested and verified to obtain the test and verification results. Based on the pre-inspection scan results, the compilation verification results, the deep analysis results, and the test verification results, multi-source data aggregation is performed to obtain a structured quality inspection report for the target file to be inspected.

[0005] In an optional implementation, the step of performing a pre-inspection scan on the target file to be inspected according to the type of the target file to be inspected, and obtaining the pre-inspection scan result, includes: A preliminary scan of the target document to be inspected is performed based on the compliance verification flow to obtain the first preliminary inspection result; If the file to be inspected is a shader file, then a pre-inspection scan is performed on the shader file based on the shader semantic verification flow to obtain a second pre-inspection result; If the file to be inspected is a resource file, then a pre-inspection scan is performed on the resource file based on the resource integrity verification flow to obtain a third pre-inspection result; The first pre-detection result, the second pre-detection result, and the third pre-detection result are aggregated to obtain the pre-detection scan result.

[0006] In an optional implementation, the step of compiling and verifying the target file to be inspected based on the pre-inspection scan results to obtain the compilation and verification results includes: Obtain the submission type of the target file to be inspected; If the submission type is a functional submission, the target file to be inspected is quickly compiled to obtain the syntax check result; If the submission type is a version-level submission, then the target file to be inspected is fully compiled to obtain the full compilation result; If the submission type is a patch-level submission, then the target file to be inspected is incrementally compiled to obtain the incremental compilation result; The compilation verification result is obtained by summarizing the syntax check result, the full compilation result, and the incremental compilation result.

[0007] In an optional implementation, the step of performing deep analysis on the target file to be inspected based on the compilation verification results to obtain the deep analysis results includes: If the compilation verification result is that the compilation verification is successful, then static code analysis is performed on the target file to be inspected to obtain the quality analysis result; The security vulnerability scan of the target file to be inspected is performed to obtain the security analysis results; Perform an architecture compliance check on the target document to be inspected to obtain the architecture analysis results; Perform developer-specific analysis on the target file to be inspected to obtain developer analysis results; The security analysis results, the architecture analysis results, and the developer analysis results are aggregated and analyzed to obtain the in-depth analysis results.

[0008] In an optional implementation, the step of testing and verifying the target file to be inspected based on the deep analysis results to obtain the test and verification results includes: If the deep analysis result is that the deep analysis passes, then unit testing is performed on the target file to be inspected to obtain the unit test results; The target file to be inspected is subjected to integration testing to obtain integration test results; The target file to be inspected is subjected to developer testing to obtain developer test results; Perform end-to-end testing on the target document to be inspected and obtain end-to-end test results; The unit test results, integration test results, developer test results, and end-to-end test results are aggregated and analyzed to obtain the in-depth analysis results.

[0009] In an optional implementation, after performing multi-source data aggregation based on the pre-inspection scan results, the compilation verification results, the deep analysis results, and the test verification results to obtain a structured quality inspection report for the target file to be inspected, the process includes: Based on the quality inspection report, a quality assessment of the access control system is conducted to obtain a quality score. Access control decisions are determined based on the quality score. The target document to be inspected is processed according to the access control decision.

[0010] In an optional implementation, the pre-inspection scan of the target file to be inspected based on the source code compliance verification flow to obtain a first pre-inspection result includes: The target file to be inspected is sequentially subjected to syntax checks, format checks, and basic specification checks to obtain the first pre-inspection result; The pre-detection scan of the shader file based on the shader semantic verification stream yields a second pre-detection result, including: The shader files are categorized to obtain the shader file types; If the shader file type is a shader definition language file, then shader language analysis is performed on the shader definition language file to obtain the language analysis results; If the shader file type is a shader graph file, then the shader graph file is subjected to compliance verification and node connection checks to obtain graph analysis results; Based on the language analysis results and the chart analysis results, a standardization check is performed to obtain the second pre-inspection result; The pre-inspection scan of the resource file based on the resource integrity verification stream yields a third pre-inspection result, including: The resource files are categorized to obtain resource file types; The resource file is pre-checked based on the verification stream corresponding to the resource file type to obtain the third pre-check result.

[0011] Secondly, the present invention provides a document quality inspection system, comprising: The pre-inspection module is used to obtain the target file to be inspected and its corresponding tags. If the tags contain preset keywords, the target file to be inspected is pre-inspected according to the type of the target file to be inspected, and the pre-inspection scan result is obtained. The compilation module is used to compile and verify the target file to be inspected based on the pre-inspection scan results, and obtain the compilation and verification results. The analysis module is used to perform in-depth analysis on the target file to be inspected based on the compilation verification results, and obtain in-depth analysis results; The testing module is used to test and verify the target file to be inspected based on the deep analysis results, and obtain the test and verification results. The aggregation module is used to aggregate multi-source data based on the pre-inspection scan results, the compilation verification results, the deep analysis results, and the test verification results to obtain a structured quality inspection report for the target file to be inspected.

[0012] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the document quality inspection method as described in any of the foregoing embodiments.

[0013] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the document quality inspection method as described in any of the foregoing embodiments.

[0014] This invention discloses a document quality inspection method, apparatus, computer equipment, and computer-readable storage medium. The method involves acquiring a target document to be inspected and its corresponding tags. If the tags contain preset keywords, a pre-inspection scan is performed on the target document according to its type, yielding a pre-inspection scan result. Based on the pre-inspection scan result, the target document is compiled and verified, yielding a compilation verification result. Based on the compilation verification result, the target document is subjected to in-depth analysis, yielding an in-depth analysis result. Based on the in-depth analysis result, the target document is tested and verified, yielding a test verification result. Multi-source data aggregation is performed based on the pre-inspection scan result, the compilation verification result, the in-depth analysis result, and the test verification result to obtain a structured quality inspection report for the target document. In this way, the document quality inspection process is triggered by tags, and pre-inspection scanning, compilation verification, in-depth analysis, and test verification are used to inspect the target document. This enables early problem detection for different types of target documents, eliminates blind spots in detection, and further generates a structured quality inspection report as a basis for document decision-making. Attached Figure Description

[0015] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope of protection of the present invention. In the various drawings, similar components are numbered similarly.

[0016] Figure 1 A flowchart of the document quality inspection method proposed in this embodiment is shown; Figure 2 This illustration shows another flowchart of the document quality inspection method proposed in this embodiment; Figure 3 A schematic diagram of the document quality inspection system proposed in this embodiment is shown.

[0017] Explanation of reference numerals in the attached diagram: 300 - Document Quality Inspection System; 301 - Pre-inspection Module; 302 - Compilation Module; 303 - Analysis Module; 304 - Testing Module; 305 - Aggregation Module. Detailed Implementation

[0018] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0019] The components of the embodiments of the invention described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0020] In the following, the terms “comprising,” “having,” and their cognates, which may be used in various embodiments of the invention, are intended only to indicate a particular feature, number, step, operation, element, component, or combination thereof, and should not be construed as excluding, firstly, the presence of one or more other features, numbers, steps, operations, elements, components, or combinations thereof, or adding the possibility of one or more features, numbers, steps, operations, elements, components, or combinations thereof.

[0021] Furthermore, the terms "first," "second," and "third" are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.

[0022] Unless otherwise specified, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which the various embodiments of the invention pertain. Terms (such as those defined in commonly used dictionaries) shall be interpreted as having the same meaning as in their contextual meaning in the relevant technical field and shall not be interpreted as having an idealized or overly formal meaning, unless clearly defined in the various embodiments of the invention.

[0023] This disclosure provides a document quality inspection method to achieve a complete automated process from tag identification and triggering the document quality inspection process to pre-inspection scanning, compilation verification, in-depth analysis, test verification, and report generation of the target document to be inspected.

[0024] Please see Figure 1 The quality inspection method for this document includes steps S101 to S105, and each step is described in detail below.

[0025] Step S101: Obtain the target file to be inspected and its corresponding tags. If the tags contain preset keywords, perform a pre-inspection scan on the target file to be inspected according to the type of the target file to be inspected, and obtain the pre-inspection scan result.

[0026] In this embodiment, the target file to be inspected can be a code file. In other embodiments, the target file to be inspected can also be a configuration file, log file, etc.

[0027] For the target files to be inspected, developers add corresponding tags themselves. The system obtains the target files to be inspected and their corresponding tags, and determines whether the tags contain preset keywords. If they do, the system performs a differentiated pre-inspection scan on the target files according to their type, triggering the target files to enter the complete quality verification chain. If they do not contain the preset keywords, the system skips the process between the target files and directly merges them.

[0028] Understandably, the mechanism of triggering the entire process through tags containing preset keywords enables flexible and strategic management. For code files awaiting quality inspection, inspection resources can be precisely allocated to critical code, rather than checking all files indiscriminately. While ensuring the quality of critical code, the inspection overhead for non-critical code is significantly reduced, improving the efficiency and resource utilization of the overall process.

[0029] In one specific embodiment, step S101 includes: performing a pre-inspection scan on the target file to be inspected based on a compliance verification flow to obtain a first pre-inspection result; if the type of the file to be inspected is a shader file, then performing a pre-inspection scan on the shader file based on a shader semantic verification flow to obtain a second pre-inspection result; if the type of the file to be inspected is a resource file, then performing a pre-inspection scan on the resource file based on a resource integrity verification flow to obtain a third pre-inspection result; and aggregating the first pre-inspection result, the second pre-inspection result, and the third pre-inspection result to obtain the pre-inspection scan result.

[0030] In this embodiment, a pre-inspection scan is performed on the target file to be inspected based on the compliance verification flow to obtain a first pre-inspection result. Optionally, for the code file to be inspected, if the file type is a source code file, such as a C# file, a pre-inspection scan is performed on the source code file based on the source code compliance verification flow to achieve automatic checks on code style, standards, security risks, etc., and obtain a first pre-inspection result, thereby improving the interception rate of source code defects at the submission stage.

[0031] An exemplary source code compliance verification flow includes: performing syntax checks, format checks, and basic specification checks on the source code files in sequence to obtain the first pre-check result.

[0032] The core of the syntax checking phase is Roslyn compilation checking, which ensures that the code is free of compilation errors. It also sequentially performs reference verification, dependency analysis, and generates repair suggestions. The format checking phase mainly uses dotnet-format checks and EditorConfig verification to ensure consistent code style. The basic style check phase comprehensively utilizes multiple tools such as StyleCop analysis, Roslynator checks, and dependency analysis to review code style, quality, and reference relationships, and classifies warning levels. This phase also performs reference verification and code cleanup checks, and generates repair suggestions.

[0033] Each inspection stage produces a clear result status, including failure, serious error, warning, or pass. Specifically, it produces syntax check results (including syntax errors), format check results (including format issues), and specification check results (including specification violations or specification warnings).

[0034] After all results are collected, they are aggregated and intelligently evaluated. Errors and warnings are categorized based on the aggregated results, and a final assessment is made. Specifically: if an error is detected, the process check fails, a failure report is generated, and the process terminates immediately, while remedial guidance is provided; if no errors are detected but there are formatting warnings, the conclusion is that manual review is required, a review report is generated, and the process enters the waiting-for-manual-approval stage. After a manual approval decision, a first pre-inspection result is generated; if rejected, the process terminates immediately; if no errors are detected and only formatting warnings are issued, the check passes, the conclusion is "warning passed," and a warning report is generated as the first pre-inspection result; if no errors are detected and no warnings are issued, the check passes directly, and a pass report is generated as the first pre-inspection result.

[0035] If the file to be inspected is a shader file, a pre-inspection scan is performed on the shader file based on the shader semantic verification stream to automatically check the shader syntax, semantics, etc., and obtain a second pre-inspection result, thereby reducing shader-related real device crashes.

[0036] As an example, the shader semantic verification flow includes: classifying shader files to obtain shader file types; if the shader file type is a shader definition language file (ShaderLab file), then performing shader language analysis on the shader definition language file to obtain language analysis results; if the shader file type is a shader graph file (ShaderGraph file), then performing compliance verification and node connection checks on the shader graph file to obtain graph analysis results; and performing specification checks based on the language analysis results and graph analysis results to obtain a second pre-check result.

[0037] If the shader file type is pure HLSL code or header file, it will directly enter the direct HLSL analysis; if it is other irrelevant file types, it will be skipped and will not enter the pre-check process.

[0038] The shader language analysis includes ShaderLab syntax checking. It then examines its included HLSL files and extracts HLSL code blocks for direct HLSL analysis. Compliance verification includes ShaderGraph loading and validation using Unity.ShaderGraphValidator.

[0039] All validated language analysis and graph analysis results will be aggregated and sequentially subjected to four core compliance checks: formatting compliance check, naming compliance check, structural compliance check, and performance warning check. Each check will output specific results, such as syntax errors, pass / fail status, or corresponding warning messages.

[0040] Subsequently, all inspection results are aggregated and a final evaluation is performed. If the evaluation is all passed, the Shader check passes; if the evaluation has warnings but no errors, the conclusion is warning but pass; if the evaluation has errors, the Shader check fails. For cases where the check passes or has warnings but passes, the process will generate a quality report as a second pre-inspection result and allow progress to the next stage; if the check fails, the process will terminate immediately.

[0041] If the file to be inspected is a resource file, a pre-inspection scan is performed on the resource file based on the resource integrity verification flow to achieve automatic inspection of the integrity of different resource files and obtain a third pre-inspection result, thus preventing runtime crashes or functional failures caused by resource missing or abnormalities.

[0042] An exemplary resource integrity verification flow includes: classifying resource files to obtain resource file types; performing a pre-check scan on the resource files according to the verification flow corresponding to the resource file type to obtain a third pre-check result.

[0043] For example, if the resource file type is an image detection resource, a pre-detection scan is performed on the image resource file based on the image detection verification stream to obtain the image inspection result; if the resource file type is an audio resource file, a pre-detection scan is performed on the audio resource file based on the audio detection verification stream to obtain the audio inspection result; if the resource file type is a prefab resource file, a prefab resource file is performed on the prefab resource file based on the prefab detection verification stream to obtain the prefab inspection result; if the resource file type is a text resource file, a pre-detection scan is performed on the text resource file based on the text detection verification stream to obtain the text inspection result; if the resource file type is a material resource file, a pre-detection scan is performed on the prefab resource file based on the material detection verification stream to obtain the material inspection result; if the resource file type is another type of resource file, a pre-detection scan is performed on the other resource file based on the basic detection verification stream to obtain the basic inspection result.

[0044] The image detection and verification process includes: sequentially performing image format verification, size check, color space check, compression setting check, and file size check, and generating image inspection results.

[0045] The audio detection and verification stream includes: audio format verification, encoding setting check, duration check, and file size limit check, and finally generates audio inspection results.

[0046] The prefab inspection and verification flow includes: checking component integrity, performing reference verification and complexity analysis (such as nesting depth checking), and executing script injection checks, and finally generating prefab inspection results.

[0047] The text detection and verification process includes: text encoding check, content verification (such as sensitive information check), and file naming convention check, ultimately generating the text inspection result.

[0048] The material inspection and verification process includes: material property checking, shader reference verification, and checking rendering pipeline compatibility, ultimately generating material inspection results.

[0049] The basic detection and verification process includes: checking file naming conventions and directory structure, and generating basic check results.

[0050] After all the detection and verification flows have completed their checks, their outputs will be aggregated. Subsequently, the system will perform resource dependency analysis on all resources and estimate their impact on the project's package size.

[0051] Based on the aggregation and analysis results, the process proceeds to the final evaluation stage. If the evaluation is complete, the resource check passes; if the evaluation shows warnings but no errors, the conclusion is "warning but pass"; if the evaluation shows errors, the resource check fails. For checks that pass or show warnings but pass, the process generates a quality report as a third pre-inspection result and allows progress to the next stage. If the check fails, the process terminates immediately.

[0052] Further, the first, second, and third pre-inspection results are aggregated according to preset aggregation criteria to obtain the aggregated pre-inspection result. Then, the process enters the final evaluation stage. If the evaluation shows all results are passed, the pre-inspection scan passes; if the evaluation shows warnings but no errors, the conclusion is a warning but the pre-inspection scan passes; if the evaluation shows errors, the pre-inspection scan fails. For cases where the pre-inspection scan passes or shows a warning but passes, the process generates a pass report or a warning report as the pre-inspection scan result and allows entry into the next stage. If the check fails, a failure report is generated as the pre-inspection scan result, and the process terminates immediately.

[0053] Step S102: Based on the pre-inspection scan results, compile and verify the target file to be inspected to obtain the compilation and verification results.

[0054] In this embodiment, if the pre-inspection scan result is that the pre-inspection scan passes, the target file to be inspected is then subjected to layered compilation verification to obtain the compilation verification result. This can verify the functional integrity of the code and the integrability between modules, and capture semantic errors.

[0055] In one specific embodiment, step S102 includes: obtaining the submission type of the target file to be inspected; if the submission type is a function-level submission, then performing a fast compilation on the target file to be inspected to obtain a syntax check result; if the submission type is a version-level submission, then performing a full compilation on the target file to be inspected to obtain a full compilation result; if the submission type is a patch-level submission, then performing an incremental compilation on the target file to be inspected to obtain an incremental compilation result; summarizing the syntax check result, the full compilation result, and the incremental compilation result to obtain the compilation verification result.

[0056] In this embodiment, the submission type of the target file to be inspected is obtained. Optionally, the submission type can be obtained after the pre-inspection scan of the target file has passed.

[0057] If the submission type is a functional submission, the fast verification mode is triggered. The .NET SDK + Roslyn compiler is used to quickly compile and analyze the code in the target file to be inspected, and obtain the syntax check results. The feedback is extremely fast and there is no need to start Unity.

[0058] If the commit type is a version-level commit, then the target files to be inspected will be fully compiled through Unity Batch Mode, such as a full Unity compilation, including script compilation checks, resource dependency verification, and serialization checks, to obtain the Unity compilation results.

[0059] If the submission type is a patch-level submission, the cached files stored in the previous full verification mode are used to perform change file analysis, dependency graph construction, incremental compilation execution, and compilation cache update for the scripts changed in the target file to be inspected, so as to compile an independent assembly and merge it into the main assembly, thereby realizing Unity incremental compilation and obtaining incremental compilation results.

[0060] Furthermore, the syntax check results, Unity compilation results, and incremental compilation results are summarized. Based on the compilation verification decision, the summarized results are evaluated. If all evaluations are passed, the compilation verification passes; if there are warnings but no errors, the conclusion is warnings but compilation verification passes; if there are errors, the compilation verification fails. For cases where compilation verification passes or passes with warnings, the process generates a pass report or a warning report as the compilation verification result and allows progress to the next stage. If compilation verification fails, a failure report is generated as the compilation verification result, and the process terminates immediately.

[0061] Step S103: Perform in-depth analysis on the target file to be inspected based on the compilation verification results to obtain in-depth analysis results.

[0062] In this embodiment, if the compilation verification result is that the compilation verification is successful, multiple parallel analysis processes are used to perform in-depth analysis on the target file to be inspected, and in-depth analysis results are obtained to maximize efficiency and coverage, thereby exposing potential risks in a multi-dimensional and in-depth manner.

[0063] In one specific embodiment, step S103 includes: if the compilation verification result is that the compilation verification is passed, then performing static code analysis on the target file to be inspected to obtain a quality analysis result; performing security vulnerability scanning on the target file to be inspected to obtain a security analysis result; performing architecture compliance checks on the target file to be inspected to obtain an architecture analysis result; performing developer-specific analysis on the target file to be inspected to obtain a developer analysis result; and performing aggregate analysis on the security analysis result, the architecture analysis result, the architecture analysis result, and the Unity analysis result to obtain the deep analysis result.

[0064] In this embodiment, if the compilation verification result is successful, static code analysis is performed on the target file to be inspected to check code complexity, duplication rate and maintainability. Tools such as SonarQube are used to identify problems such as excessively long methods and high cyclomatic complexity, and the quality analysis results are obtained.

[0065] Security vulnerability scanning is performed on the target documents to be inspected, detecting security risks such as hard-coded sensitive information and SQL injection, and obtaining security analysis results.

[0066] The architecture compliance check is performed on the target documents for quality inspection, which verifies dependencies, design pattern compliance, and package structure standardization, and obtains the architecture analysis results.

[0067] Perform developer-specific analysis on the target files for quality inspection, such as Unity-specific analysis, review MonoBehaviour specifications, performance hotspots, and resource management best practices to obtain Unity analysis results.

[0068] Furthermore, the security analysis results, architecture analysis results, and Unity analysis results are aggregated and analyzed. The aggregated analysis results are then evaluated. If all evaluations are passed, the deep analysis passes; if the evaluation shows warnings but no errors, the conclusion is a warning but compilation verification passes; if the evaluation shows errors, the deep analysis fails. For cases where the deep analysis passes or shows warnings but passes, the process generates a pass report or a warning report as the deep analysis result and allows progress to the next stage. If the deep analysis fails, a failure report is generated as the deep analysis result, and the process is immediately terminated.

[0069] Understandably, the intelligent incremental detection mechanism and hierarchical verification strategy significantly optimize detection efficiency. A traditional full Unity compilation takes 2-5 minutes, while the fast detection mode in this embodiment only requires 2-5 seconds to provide feedback on basic issues. Intelligent incremental compilation utilizes multi-level caching, reducing conventional detection time by 70-80%. The time developers receive immediate feedback is reduced from an average of 15 minutes to less than 3 minutes, effectively reducing context switching costs and improving development smoothness.

[0070] Step S104: Based on the in-depth analysis results, the target file to be inspected is tested and verified to obtain the test and verification results.

[0071] In this embodiment, if the deep analysis result is that the deep analysis passes, the target file to be inspected is tested and verified based on the layered testing strategy to obtain the test verification results, which cover the comprehensive verification from micro logic to the complete user flow, ensuring the correctness and quality of the code functionality.

[0072] In one specific embodiment, step S104 includes: if the deep analysis result is that the deep analysis passes, then performing unit testing on the target file to be inspected to obtain unit test results; performing integration testing on the target file to be inspected to obtain integration test results; performing developer testing on the target file to be inspected to obtain developer test results; performing end-to-end testing on the target file to be inspected to obtain end-to-end test results; and performing aggregate analysis on the unit test results, the integration test results, the developer test results, and the end-to-end test results to obtain the deep analysis result.

[0073] In this embodiment, if the deep analysis result is "passed," unit testing is performed on the target file to be inspected. The NUnit framework is used to test each independent function and method, verifying the correctness of input and output, boundary condition handling, and exception scenarios, thus obtaining unit test results and achieving independent verification of the core business logic. This stage requires code coverage of no less than 80%, focusing on computational logic, data transformation, and business rules. Test cases cover normal paths, abnormal paths, and boundary conditions to ensure the correctness of each independent unit.

[0074] Furthermore, integration testing is performed on the target documents to be inspected, testing the data flow, interface calls, and event transmission between components to obtain integration test results, thereby verifying inter-module collaboration and ensuring correct integration of each module. This stage simulates real data exchange to verify inter-service communication, database operations, and third-party API integration. The focus is on checking interface consistency, data format conversion, and error handling mechanisms.

[0075] Furthermore, considering the characteristics of Unity projects, developer testing is performed on the target files to be inspected. This includes Unity testing to verify the MonoBehaviour lifecycle, physics system interaction, UI responsiveness, and resource management, as well as performance benchmarking and rendering tests to obtain Unity test results. Performance benchmarking monitors frame rate stability and memory usage to ensure game performance meets standards. Rendering tests verify shader correctness and visual effects, while physics tests ensure accurate collision detection and motion simulation.

[0076] Furthermore, end-to-end testing is conducted on the target documents to be inspected, simulating real user operation processes to obtain end-to-end test results. Verification is performed on the entire path from game launch to core function completion, including scene switching, user interaction, and platform compatibility. Multi-platform compatibility testing verifies adaptability to different resolutions and devices, while user experience testing ensures user-friendliness and smooth workflow.

[0077] Furthermore, the unit test results, integration test results, Unity test results, and end-to-end test results are aggregated and analyzed. The aggregated test results are evaluated. If all results are deemed passed, the test verification is considered successful. If there are warnings but no errors, the conclusion is that the test verification passes despite warnings. If there are errors, the test verification fails. For cases where the test verification passes or passes despite warnings, the process generates a pass report or a warning report as the test verification result and allows the process to proceed to the next stage. If the test verification fails, a failure report is generated as the test verification result, and the process is terminated immediately.

[0078] Understandably, through in-depth security scanning and architecture compliance checks, this implementation identifies potential risks at an early stage. Compared to traditional solutions that only discover security issues during the testing phase, this solution advances the discovery time of security vulnerabilities by an average of 3-4 weeks. Architecture violation detection prevents the implicit accumulation of technical debt, and project maintenance costs are expected to decrease by 40%. Multi-platform compatibility verification ensures a quality benchmark of "code once, work on multiple platforms," ​​reducing rework in later adaptation work by 60%.

[0079] Meanwhile, comprehensive and multi-dimensional quality control was achieved through pre-screening, compilation verification, in-depth analysis, and testing verification. By integrating over 20 professional tools, including C# syntax checking, shader verification, resource file inspection, security scanning, and architecture analysis, into a unified pipeline, the problem of scattered tools and inconsistent standards in traditional solutions was solved. Actual application data shows that after adopting this implementation, the code defect rate decreased by an average of 68%, the accuracy of technical debt identification increased by 85%, and the team's coding standard compliance rate increased from less than 60% to over 95%.

[0080] Step S105: Based on the pre-inspection scan results, the compilation verification results, the deep analysis results, and the test verification results, multi-source data aggregation is performed to obtain a structured quality inspection report for the target file to be inspected.

[0081] In this embodiment, the pre-inspection scan results, compilation verification results, in-depth analysis results, and test verification results are uniformly converted into a standardized format to further generate a structured quality inspection report for the target file to be inspected. The structured quality inspection report includes: (1) specific problem locations, severity levels, and remediation suggestions for developers to refer to; (2) a visual quality dashboard, which displays the current status and historical trends of various quality indicators in the form of charts, making it easier for team managers to grasp the overall quality situation; and (3) a team notification summary, which needs to extract key issues and risk items and push them in real time through channels such as Slack and WeChat Work.

[0082] It should be noted that the system possesses intelligent analysis capabilities, automatically comparing the differences between each submitted result and historical benchmark results to identify trends of quality improvement or degradation; simultaneously, it correlates with the scope of code changes to accurately assess the quality performance of newly added code, all of which are updated in a structured quality inspection report. All reports support interactive viewing on the web and can be filtered and sorted by multiple dimensions such as issue type, severity, and module.

[0083] Understandably, visual quality dashboards and intelligent report generation have transformed traditional quality management models. All quality metrics are displayed in real-time, with historical trends readily apparent. The system automatically identifies common team problems and recommends targeted training content, reducing onboarding time for new members by 50%. A technical debt tracking mechanism ensures continuous improvement, leading to a steady increase in the team's overall coding proficiency and fostering a quality culture that can be accumulated and passed on.

[0084] Please see Figure 2 In one specific embodiment, step S105 is followed by steps S201 to S203, and each step is described in detail below.

[0085] Step S201: Conduct a quality access control assessment based on the quality inspection report to obtain a quality score.

[0086] In this embodiment, the quality inspection report is evaluated based on multi-dimensional indicators to obtain a corresponding quality score.

[0087] Step S202: Determine the access control decision based on the quality score.

[0088] In this embodiment, the access control decision corresponding to the quality score is determined based on preset access control rules.

[0089] As an example, the preset access control rule mapping relationship adopts a four-level hierarchical standard: Access control decision level A (pass decision: automatically pass document quality inspection) requires a quality score greater than or equal to the first threshold, such as 90 points, a test pass rate greater than or equal to the second threshold, such as 95%, and no serious security issues; Access control decision level B (warning pass decision: generate a warning but document quality inspection passes) allows for non-critical issues, and the quality score must be within the first preset range, such as 75-89 points; Access control decision level C (human review decision: document quality inspection pass requires human review) is for quality scores within the second preset range, such as 60-74 points, or complex situations requiring human judgment; Level D (reject decision: document quality inspection fails, code merging is directly rejected) is applicable to submissions with a quality score below the third threshold, such as 60 points, or with serious security vulnerabilities.

[0090] Understandably, the multi-level access control decision system represents an intelligent upgrade to quality control. Traditional solutions rely on manual judgment, resulting in inconsistent standards and low efficiency. This embodiment uses a quantitative scoring model to transform code quality into a measurable quality index of 0-100 points and establishes an AD four-level automatic decision-making mechanism. The system's false alarm rate is controlled within 5%, with an accuracy rate of 92%, while also supporting dynamic threshold adjustments based on historical data, making quality control both rigorous and reasonable.

[0091] Step S203: Process the target document to be inspected according to the access control decision.

[0092] In this embodiment, the target files to be inspected are processed according to the access control decision. For example, the decision triggers automatic code merging and enters the deployment process, while recording the pass details; the warning pass decision allows merging while recording the problem in the technical debt list and requiring the development of a remediation plan; the manual review decision suspends the automated process, creates a code review task and assigns a reviewer, and can only continue after the review is passed; the rejection decision immediately terminates the process, returns a detailed error report and remediation guidance, and prevents code merging.

[0093] Understandably, through a layered execution strategy and intelligent resource scheduling, this embodiment significantly reduces resource consumption while ensuring detection quality. Traditional solutions require independent environment configuration for each inspection, resulting in resource utilization of less than 30%. This embodiment, through containerized deployment and toolchain optimization, reduces the average CPU utilization of the CI / CD server by 40% and memory usage by 35%. Enterprise-level deployment cases show that annual hardware and cloud service costs are reduced by an average of 250,000-300,000 yuan.

[0094] The document quality inspection method proposed in this embodiment obtains the target document to be inspected and its corresponding tags. If the tags contain preset keywords, a pre-inspection scan is performed on the target document according to its type, yielding a pre-inspection scan result. Based on the pre-inspection scan result, the target document is compiled and verified, yielding a compilation verification result. Based on the compilation verification result, the target document is subjected to in-depth analysis, yielding an in-depth analysis result. Based on the in-depth analysis result, the target document is tested and verified, yielding a test verification result. Multi-source data aggregation is performed based on the pre-inspection scan result, the compilation verification result, the in-depth analysis result, and the test verification result to obtain a structured quality inspection report for the target document. In this way, the document quality inspection process is triggered by tags, and pre-inspection scanning, compilation verification, in-depth analysis, and test verification are used to inspect the target document, enabling early problem detection for different types of target documents, eliminating detection blind spots, and further generating a structured quality inspection report as a basis for document decision-making.

[0095] Furthermore, this disclosure provides a document quality inspection system 300, please refer to [link to relevant documentation]. Figure 3 ,include: The pre-inspection module 301 is used to obtain the target file to be inspected and its corresponding tags. If the tags contain preset keywords, the target file to be inspected is pre-inspected according to the type of the target file to be inspected, and the pre-inspection scan result is obtained. Compilation module 302 is used to compile and verify the target file to be inspected based on the pre-inspection scan results, and obtain compilation and verification results; Analysis module 303 is used to perform in-depth analysis on the target file to be inspected based on the compilation verification results, and obtain in-depth analysis results; Test module 304 is used to test and verify the target file to be inspected based on the deep analysis results, and obtain the test and verification results; The aggregation module 305 is used to aggregate multi-source data based on the pre-inspection scan results, the compilation verification results, the deep analysis results, and the test verification results to obtain a structured quality inspection report for the target file to be inspected.

[0096] In one embodiment, the pre-inspection module 301 is further configured to perform a pre-inspection scan on the target file to be inspected based on a compliance verification flow to obtain a first pre-inspection result; if the type of the file to be inspected is a shader file, then perform a pre-inspection scan on the shader file based on a shader semantic verification flow to obtain a second pre-inspection result; if the type of the file to be inspected is a resource file, then perform a pre-inspection scan on the resource file based on a resource integrity verification flow to obtain a third pre-inspection result; and aggregate the first pre-inspection result, the second pre-inspection result, and the third pre-inspection result to obtain the pre-inspection scan result.

[0097] In one embodiment, the compilation module 302 is further configured to obtain the submission type of the target file to be inspected; if the submission type is a function-level submission, the target file to be inspected is quickly compiled to obtain a syntax check result; if the submission type is a version-level submission, the target file to be inspected is fully compiled to obtain a full compilation result; if the submission type is a patch-level submission, the target file to be inspected is incrementally compiled to obtain an incremental compilation result; and the syntax check result, the full compilation result, and the incremental compilation result are summarized to obtain the compilation verification result.

[0098] In one embodiment, the analysis module 303 is further configured to, if the compilation verification result is a successful compilation verification, perform static code analysis on the target file to be inspected to obtain a quality analysis result; perform security vulnerability scanning on the target file to be inspected to obtain a security analysis result; perform architecture compliance checks on the target file to be inspected to obtain an architecture analysis result; perform developer-specific analysis on the target file to be inspected to obtain a developer analysis result; and perform aggregate analysis on the security analysis result, the architecture analysis result, the architecture analysis result, and the Unity analysis result to obtain the deep analysis result.

[0099] In one embodiment, the testing module 304 is further configured to, if the deep analysis result is a deep analysis pass, perform unit testing on the target file to be inspected to obtain unit test results; perform integration testing on the target file to be inspected to obtain integration test results; perform developer testing on the target file to be inspected to obtain developer test results; perform end-to-end testing on the target file to be inspected to obtain end-to-end test results; and perform aggregate analysis on the unit test results, the integration test results, the developer test results, and the end-to-end test results to obtain the deep analysis result.

[0100] In one embodiment, the system further includes a decision module, used to perform a quality access control assessment based on the quality inspection report to obtain a quality score; determine an access control decision based on the quality score; and process the target document to be inspected based on the access control decision.

[0101] In one embodiment, the pre-inspection module 301 is further configured to perform syntax checking, format checking and basic specification checking on the source code file in sequence to obtain a first pre-inspection result; The pre-inspection module 301 is further configured to classify the shader files to obtain the shader file type; if the shader file type is a shader definition language file, then perform shader language analysis on the shader definition language file to obtain language analysis results; if the shader file type is a shader graph file, then perform compliance verification and node connection checks on the shader graph file to obtain graph analysis results; and perform specification checks based on the language analysis results and the graph analysis results to obtain the second pre-inspection result. The pre-inspection module 301 is also used to classify the resource files to obtain resource file types; and to perform a pre-inspection scan on the resource files according to the verification stream corresponding to the resource file types to obtain the third pre-inspection result.

[0102] The system provided in this embodiment can perform the steps of the document quality inspection method described above, and will not be repeated here to avoid duplication.

[0103] The document quality inspection system proposed in this embodiment acquires the target document to be inspected and its corresponding tags. If the tags contain preset keywords, a pre-inspection scan is performed on the target document according to its type, yielding a pre-inspection scan result. Based on the pre-inspection scan result, the target document is compiled and verified, yielding a compilation verification result. Based on the compilation verification result, the target document is deeply analyzed, yielding a deep analysis result. Based on the deep analysis result, the target document is tested and verified, yielding a test verification result. Multi-source data aggregation is performed based on the pre-inspection scan result, the compilation verification result, the deep analysis result, and the test verification result to obtain a structured quality inspection report for the target document. In this way, the document quality inspection process is triggered by tags, and pre-inspection scanning, compilation verification, deep analysis, and test verification are used to inspect the target document, enabling early problem detection for different types of target documents, eliminating detection blind spots, and further generating a structured quality inspection report as a basis for document decision-making.

[0104] Furthermore, this disclosure provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the computer program, when executed by the processor, implements the above-described document quality inspection method.

[0105] The device provided in this embodiment can perform the steps of the document quality inspection method described above, and will not be repeated here to avoid repetition.

[0106] This disclosure provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned document quality inspection method.

[0107] In this embodiment, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0108] The computer-readable storage medium provided in this embodiment can implement the document quality inspection method described above. To avoid repetition, it will not be described again here.

[0109] In all examples shown and described herein, any specific values ​​should be interpreted as merely exemplary and not as limitations; therefore, other examples of exemplary embodiments may have different values.

[0110] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0111] The above-described embodiments are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A document quality inspection method, characterized in that, include: Obtain the target file to be inspected and its corresponding tags. If the tags contain preset keywords, perform a pre-inspection scan on the target file to be inspected according to the type of the target file to be inspected, and obtain the pre-inspection scan result. Based on the pre-inspection scan results, the target file to be inspected is compiled and verified to obtain the compilation and verification results. Based on the compilation and verification results, a deep analysis is performed on the target file to be inspected, and the deep analysis results are obtained. Based on the in-depth analysis results, the target file to be inspected is tested and verified to obtain the test and verification results. Based on the pre-inspection scan results, the compilation verification results, the deep analysis results, and the test verification results, multi-source data aggregation is performed to obtain a structured quality inspection report for the target file to be inspected.

2. The document quality inspection method according to claim 1, characterized in that, The step of performing a pre-inspection scan on the target file to be inspected according to the type of the target file to be inspected, and obtaining the pre-inspection scan result, includes: A preliminary scan of the target document to be inspected is performed based on the compliance verification flow to obtain the first preliminary inspection result; If the file to be inspected is a shader file, then a pre-inspection scan is performed on the shader file based on the shader semantic verification flow to obtain a second pre-inspection result; If the file to be inspected is a resource file, then a pre-inspection scan is performed on the resource file based on the resource integrity verification flow to obtain a third pre-inspection result; The first pre-detection result, the second pre-detection result, and the third pre-detection result are aggregated to obtain the pre-detection scan result.

3. The document quality inspection method according to claim 1, characterized in that, The compilation and verification of the target file to be inspected based on the pre-inspection scan results, to obtain the compilation and verification results, includes: Obtain the submission type of the target file to be inspected; If the submission type is a functional submission, the target file to be inspected is quickly compiled to obtain the syntax check result; If the submission type is a version-level submission, then the target file to be inspected is fully compiled to obtain the full compilation result; If the submission type is a patch-level submission, then the target file to be inspected is incrementally compiled to obtain the incremental compilation result; The compilation verification result is obtained by summarizing the syntax check result, the full compilation result, and the incremental compilation result.

4. The document quality inspection method according to claim 1, characterized in that, The deep analysis of the target file to be inspected based on the compilation verification results yields the following results: If the compilation verification result is that the compilation verification is successful, then static code analysis is performed on the target file to be inspected to obtain the quality analysis result; The security vulnerability scan of the target file to be inspected is performed to obtain the security analysis results; Perform an architecture compliance check on the target document to be inspected to obtain the architecture analysis results; Perform developer-specific analysis on the target file to be inspected to obtain developer analysis results; The security analysis results, the architecture analysis results, and the developer analysis results are aggregated and analyzed to obtain the in-depth analysis results.

5. The document quality inspection method according to claim 1, characterized in that, The step of testing and verifying the target file to be inspected based on the deep analysis results, and obtaining the test and verification results, includes: If the deep analysis result is that the deep analysis passes, then unit testing is performed on the target file to be inspected to obtain the unit test results; The target file to be inspected is subjected to integration testing to obtain integration test results; The target file to be inspected is subjected to developer testing to obtain developer test results; Perform end-to-end testing on the target document to be inspected and obtain end-to-end test results; The unit test results, integration test results, developer test results, and end-to-end test results are aggregated and analyzed to obtain the in-depth analysis results.

6. The document quality inspection method according to claim 1, characterized in that, After aggregating multi-source data based on the pre-inspection scan results, the compilation verification results, the deep analysis results, and the test verification results to obtain a structured quality inspection report for the target file to be inspected, the process includes: Based on the quality inspection report, a quality assessment of the access control system is conducted to obtain a quality score. Access control decisions are determined based on the quality score. The target document to be inspected is processed according to the access control decision.

7. The document quality inspection method according to claim 2, characterized in that, The pre-inspection scan of the target document to be inspected based on the compliance verification flow yields a first pre-inspection result, including: The target file to be inspected is sequentially subjected to syntax checks, format checks, and basic specification checks to obtain the first pre-inspection result; The pre-detection scan of the shader file based on the shader semantic verification stream yields a second pre-detection result, including: The shader files are categorized to obtain shader file types; If the shader file type is a shader definition language file, then shader language analysis is performed on the shader definition language file to obtain the language analysis results; If the shader file type is a shader graph file, then the shader graph file is subjected to compliance verification and node connection checks to obtain graph analysis results; Based on the language analysis results and the chart analysis results, a standardization check is performed to obtain the second pre-inspection result; The pre-inspection scan of the resource file based on the resource integrity verification stream yields a third pre-inspection result, including: The resource files are categorized to obtain resource file types; The resource file is pre-checked based on the verification stream corresponding to the resource file type to obtain the third pre-check result.

8. A document quality inspection system, characterized in that, include: The pre-inspection module is used to obtain the target file to be inspected and its corresponding tags. If the tags contain preset keywords, the target file to be inspected is pre-inspected according to the type of the target file to be inspected, and the pre-inspection scan result is obtained. The compilation module is used to compile and verify the target file to be inspected based on the pre-inspection scan results, and obtain the compilation and verification results. The analysis module is used to perform in-depth analysis on the target file to be inspected based on the compilation verification results, and obtain in-depth analysis results; The testing module is used to test and verify the target file to be inspected based on the deep analysis results, and obtain the test and verification results. The aggregation module is used to aggregate multi-source data based on the pre-inspection scan results, the compilation verification results, the deep analysis results, and the test verification results to obtain a structured quality inspection report for the target file to be inspected.

9. A computer device, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, implements the document quality inspection method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the document quality inspection method as described in any one of claims 1 to 7.

Images