Low-code platform development method and system based on log system and medium

Abstract

The invention discloses a low-code platform development method and system based on a log system and a medium, belongs to the technical field of software development, and is used for solving the technical problems that existing log information is low in efficiency, consistency and normalization of manually written codes are difficult to guarantee, and potential information in log data is difficult to deeply mine. The method comprises the steps of performing log generation processing on related information of a current development scene based on a pre-trained log intelligent recommendation model to obtain log information; performing level judgment processing on the log information to obtain a level judgment result; according to the level judgment result, performing corresponding deep analysis processing on the log information under different levels to obtain a log analysis strategy; if the log analysis strategy is an error log analysis strategy, performing intelligent code correction on log information in the error log analysis strategy to obtain corrected code information; and based on the correction code information, feeding back and optimizing the log intelligent recommendation model.

Description

Technical Field

[0001] This application relates to the technical field of software development, and in particular, to a low-code platform development method, system, and medium based on a logging system. Background Art

[0002] With the rapid development of the software industry, low-code platforms have received extensive attention because they can quickly build application programs. In the process of software development, the logging system plays a crucial role. It can help developers and operators monitor the running status of application programs, locate problems, and analyze performance bottlenecks. However, there are many deficiencies in the existing low-code platforms in terms of log processing.

[0003] On the one hand, the generation of log information often depends on manually writing code, which is not only inefficient but also error-prone. Different developers may use different logging methods and formats, making it difficult to ensure the consistency and standardization of logs. On the other hand, for a large amount of log data, there is a lack of effective analysis and processing means. Most of the existing analysis tools can only perform simple queries and filtering, and cannot deeply mine the potential information in the log data, such as automatically identifying the severity of problems according to the log level, and analyzing the running trend of the system according to the log files. In addition, when a problem is found in the code, the code cannot be automatically corrected and feedback for optimization, resulting in a long problem-solving cycle and low software development efficiency.

[0004] Therefore, there is an urgent need for an innovative low-code platform development model that supports an intelligent logging system to solve the deficiencies of the existing technology and improve the efficiency of software development and operation. Summary of the Invention

[0005] Embodiments of this application provide a low-code platform development method, system, and medium based on a logging system to solve the following technical problems: the existing log information is inefficient, it is difficult to ensure the consistency and standardization of manually written code, and it is difficult to deeply mine the potential information in the log data.

[0006] Embodiments of this application adopt the following technical solutions:

[0007] On the one hand, an embodiment of the present application provides a low-code platform development method based on a log system, including: performing log generation processing on relevant information of the current development scenario based on a pre-trained log intelligent recommendation model to obtain log information; performing level judgment processing on the log information to obtain a level judgment result; according to the level judgment result, performing corresponding in-depth analysis processing on the log information at different levels to obtain a log analysis strategy; if the log analysis strategy is an error log analysis strategy, performing intelligent code correction on the log information in the error log analysis strategy to obtain corrected code information; based on the corrected code information, feeding back and optimizing the log intelligent recommendation model.

[0008] The embodiment of the present application automatically generates log information through a self-trained log intelligent recommendation model, reducing the workload of developers writing log code and improving development efficiency. The intelligent log analysis processing module can comprehensively analyze the logs, helping developers and operation and maintenance personnel quickly locate problems and improving the maintainability and reliability of the system. The intelligent code correction and model feedback optimization module can automatically correct the code and feedback to optimize the model, forming a closed-loop optimization mechanism and continuously improving the intelligence level of the platform.

[0009] In a feasible implementation manner, before performing log generation processing on relevant information of the current development scenario based on a pre-trained log intelligent recommendation model to obtain log information, the method further includes: collecting a historical log data set; where the historical log data set at least includes: different types of application program data sets and log records in different scenarios; performing cleaning processing on abnormal data in the historical log data set, and extracting key log information of each group of log data in the cleaned historical log data set; where the key log information at least includes: timestamp, operation type, and operation object; through a deep learning network under the Transformer architecture, and based on the learning rate and batch size configured corresponding to the attribute items in the historical log data set, inputting the key log information of each group of log data into the initial log intelligent recommendation model, and obtaining the trained log intelligent recommendation model based on the model convergence degree after multiple rounds of iterative training.

[0010] In a feasible implementation, based on a pre-trained log intelligent recommendation model, log generation processing is performed on the relevant information of the current development scenario to obtain log information, specifically including: when the low-code platform detects application development data, the current code-related information of the current development scenario corresponding to the application development data is obtained in real time; wherein, the current code-related information at least includes: business logic, operation process, and input and output parameters; the current code-related information is input into the log intelligent recommendation model; through the log intelligent recommendation model, all attribute item data in the current code-related information are classified one by one to obtain attribute item classification data; and the attribute item classification data is subjected to detailed information log automatic pulling processing to generate the log information.

[0011] In a feasible implementation, level judgment processing is performed on the log information to obtain a level judgment result, specifically including: through regular expressions, the original log stream in the log information is subjected to format and content parsing processing related to the log file, and text cleaning processing is performed on the free text part of the log file to obtain multimodal log information; wherein, the original log stream includes: structured log fields and unstructured text; the text cleaning includes: domain word segmentation, stop word filtering, stemming, and word embedding conversion; identification and judgment of level identification information are performed on the multimodal log information; wherein, the level identification information is the specific character at the beginning in the log information; if the multimodal log information has the level identification information, based on the level identification information, a first level judgment result under the level identification information is determined; if the multimodal log information does not have the level identification information, keyword mapping processing of the multimodal log information under the rule engine is performed through a preset keyword-level mapping table to obtain a second level judgment result under the content characteristics in the log information; wherein, the rule engine includes: a natural language processing technology engine; wherein, the level judgment result of the log information at least includes: debug level, information level, warning level, and error level.

[0012] In a feasible implementation manner, according to the level judgment result, corresponding in-depth analysis and processing are performed on the log information at different levels to obtain a log analysis strategy, which specifically includes: if the level judgment result is the debug level, the development and debugging operation information of the application code in the log information is obtained, and the development and debugging operation information is sent to the developer port to generate a debug log analysis strategy; if the level judgment result is the error level, through an exception analysis and location algorithm, the abnormal information and call stack information in the log information are used to locate the abnormal code position, and an error log analysis strategy is obtained; wherein, the error log analysis strategy is used for the abnormal functions and abnormal code lines in the application code; if the level judgment result is the warning level, through an association rule mining algorithm, the combination recognition of warning faults is performed on the log record combinations in the log information to obtain a warning log analysis strategy; wherein, the log analysis strategy includes: the debug log analysis strategy, the error log analysis strategy, and the warning log analysis strategy.

[0013] In a feasible implementation manner, if the log analysis strategy is an error log analysis strategy, intelligent code correction is performed on the log information in the error log analysis strategy to obtain corrected code information, which specifically includes: performing code review processing on the log information in the error log analysis strategy to determine abnormal code information; wherein, the abnormal code information at least includes: syntax errors, logical errors, and resource leaks; through the established code quality detection rule library, template matching of abnormal types is performed on both the static code and log information in the abnormal code information to obtain abnormal type information; performing a retrieval process on the correction strategy for the abnormal type information to determine a correction suggestion strategy; according to the correction suggestion strategy, performing intelligent code correction on the log information in the error log analysis strategy, and based on the manual verification result, obtaining the corrected code information.

[0014] In a feasible implementation manner, the code quality detection rule library contains specification templates, best practice templates, and common problem pattern templates for multiple programming languages.

[0015] In a feasible implementation manner, based on the corrected code information, the log intelligent recommendation model is fed back and optimized, which specifically includes: re-inputting the corrected code information and the corresponding relevant running logs into the log intelligent recommendation model; based on the first update data of the corrected code information and the second update data of the relevant running logs, performing iterative update processing on the parameters and algorithms of the log intelligent recommendation model itself, and optimizing and obtaining the log intelligent recommendation model.

[0016] Second aspect, the embodiment of the present application further provides a low-code platform development system based on a log system. The low-code platform development system based on the log system includes: an intelligent log information generation module, configured to perform log generation processing on relevant information of the current development scenario based on a pre-trained log intelligent recommendation model to obtain log information; an intelligent log analysis and processing module, configured to perform level judgment processing on the log information to obtain a level judgment result; according to the level judgment result, perform corresponding in-depth analysis processing on the log information at different levels to obtain a log analysis strategy; an intelligent code correction and model feedback optimization module, configured to, if the log analysis strategy is an error log analysis strategy, perform intelligent code correction on the log information in the error log analysis strategy to obtain corrected code information; and based on the corrected code information, feedback and optimize the log intelligent recommendation model.

[0017] Third aspect, the embodiment of the present application further provides a non-volatile computer storage medium, characterized in that the storage medium is a non-volatile computer-readable storage medium, and the non-volatile computer-readable storage medium stores at least one program, and each program includes instructions, and when the instructions are executed by a terminal, the terminal is caused to execute a low-code platform development method based on a log system according to any one of the above embodiments.

[0018] The present application provides a low-code platform development method, system and medium based on a log system. Compared with the prior art, the embodiment of the present application has the following beneficial technical effects:

[0019] 1. Automatic log generation: Through a pre-trained log intelligent recommendation model, relevant log information of the development scenario can be automatically generated, reducing the burden of manually writing logs.

[0020] 2. Quick problem positioning: Performing level judgment processing on the log information can help developers quickly identify the urgency of the logs and give priority to handling key problems.

[0021] 3. In-depth log analysis: Performing in-depth analysis on log information at different levels helps developers deeply understand code behavior and potential problems.

[0022] 4. Intelligent error repair: For error logs, the system can automatically correct the code, improving code quality and stability and reducing the debugging time of developers.

[0023] 5. Code quality improvement: Through intelligent code correction, some common programming errors can be automatically repaired, thereby improving the overall code quality.

[0024] 6. Development efficiency improvement: The automatic log analysis and code correction process can significantly improve development efficiency and reduce manual intervention.

[0025] 7. Continuous learning and optimization: Based on the feedback of the corrected code information, the log intelligent recommendation model can be continuously optimized to more accurately adapt to different development scenarios.

[0026] 8. User experience improvement: By providing more intelligent development tools, the user experience of developers can be enhanced, enabling them to work efficiently even in a low-code environment.

[0027] 9. Cost savings: Reducing the time for manual code review and error fixing can save labor costs and improve the efficiency of the development team.

[0028] 10. Error prevention: By analyzing error logs and automatically correcting them, it helps prevent similar errors that may occur in the future and enhances the robustness of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings described below are only some embodiments recorded in the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. In the drawings:

[0030] Figure 1 is a flowchart of a method for developing a low-code platform based on a log system provided by an embodiment of the present application;

[0031] Figure 2 is a schematic diagram of the process of an intelligent log information generation module provided by an embodiment of the present application;

[0032] Figure 3 is a schematic diagram of the structure of a device for developing a low-code platform based on a log system provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] In order for those skilled in the art to better understand the technical solutions in the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of them. Based on the embodiments of this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present application.

[0034] An embodiment of the present application provides a method for developing a low-code platform based on a log system, as Figure 1 shown, the method for developing a low-code platform based on a log system specifically includes steps S101 - S105:

[0035] S101. Process the relevant information of the current development scenario based on a pre-trained log intelligent recommendation model to obtain log information.

[0036] Specifically, it is necessary to first collect a historical log dataset. Among them, the historical log dataset at least includes: application program datasets of different types and log records under different scenarios.

[0037] Furthermore, perform cleaning processing on abnormal data in the historical log dataset, and extract the key log information of each group of log data in the cleaned historical log dataset. Among them, the key log information at least includes: timestamp, operation type, and operation object.

[0038] Furthermore, through a deep learning network under the Transformer architecture, and based on the learning rate and batch size configured corresponding to the attribute items in the historical log dataset, input the key log information of each group of log data into the initial log intelligent recommendation model, and obtain the trained log intelligent recommendation model based on the convergence degree of the model after multiple rounds of iterative training.

[0039] Furthermore, if the low-code platform detects application development data, then obtain the current code-related information of the current development scenario corresponding to the application development data in real time. Among them, the current code-related information at least includes: business logic, operation process, and input and output parameters. Input the current code-related information into the log intelligent recommendation model.

[0040] Furthermore, through the log intelligent recommendation model, perform classification processing on all attribute item data in the current code-related information one by one to obtain attribute item classification data. And perform detailed information log automatic pulling processing on the attribute item classification data to generate log information.

[0041] As a feasible implementation manner, use an intelligent log information generation module and combine a self-trained log intelligent recommendation model to generate log information. The intelligent recommendation model can be a model trained by the user himself or an existing log model such as hunyuan-lite, etc. Through learning and analysis of a large amount of historical log data, the intelligent recommendation model can understand the log record requirements under different business scenarios, and automatically generate standardized and normalized log information according to the specific development scenario, business logic, and operation process. For example, in the order processing module of an e-commerce system, the intelligent recommendation model can automatically generate log records containing detailed information, such as order number, operation time, operator, order status, etc., according to different states of the order, such as creation, review, and delivery.

[0042] In one embodiment, Figure 2A schematic flow diagram of an intelligent log information generation module provided by an embodiment of this application is as follows. Figure 2 As shown in the figure, first, a set of log data of an e-commerce platform in different time periods and different business scenarios is collected, including log records of operations such as user login, product search, and order processing. The collected log data is cleaned to remove invalid data, duplicate data, etc., and then feature extraction is performed, such as extracting key information (timestamp, operation type, operation object, etc.) in the log. A deep learning algorithm (such as the Transformer architecture) is selected to train a log intelligent recommendation model, and appropriate hyperparameters (such as learning rate, batch size, etc.) are set to make the model converge through multiple rounds of training. When developing a new e-commerce application on a low-code platform, when a developer creates a user login function module, the low-code platform obtains relevant information (such as login interface, user verification logic, etc.) and inputs it into the trained log intelligent recommendation model. The model generates log information such as "[User Login] Time: [Specific Time] Username: [Username] Login Success / Failure (Failure Reason: [Specific Reason])" for developers to reference and use.

[0043] S102. Perform level judgment processing on the log information to obtain a level judgment result.

[0044] Specifically, using regular expressions, the original log stream in the log information is parsed for the format and content of the log file, and the free text part of the log file is cleaned to obtain multimodal log information. The original log stream includes: structured log fields and unstructured text. Text cleaning includes: domain word segmentation, stop word filtering, stemming, and word embedding conversion.

[0045] Furthermore, perform identification and judgment on the level identification information of the multimodal log information. The level identification information is the specific character at the beginning of the log information.

[0046] If the multimodal log information has level identification information, based on the level identification information, a first level judgment result under the level identification information is determined. If the multimodal log information does not have level identification information, through a preset keyword-level mapping table, keyword mapping processing is performed on the multimodal log information under the rule engine to obtain a second level judgment result under the content features in the log information. The rule engine includes: a natural language processing technology engine.

[0047] Among them, the level judgment result of the log information at least includes: debug level, information level, warning level, and error level.

[0048] In one embodiment, an identification rule for log levels can be defined. For example, specific characters or keywords are used at the beginning of log information to represent log levels (such as "DEBUG:", "INFO:", "WARN:",

[0049] "ERROR:", etc.). For new log data, the log level is determined by parsing the identification information in the log header. If the level cannot be determined through the identification information, natural language processing techniques are used to analyze the log content to determine whether it contains keywords related to errors, thereby determining the log level.

[0050] In one embodiment, by analyzing, the level identification information contained in the log is identified. If there is no clear identification information, the log level is judged according to the content characteristics of the log (such as containing keywords like "error", "exception", etc.).

[0051] S103. According to the level judgment result, corresponding in-depth analysis and processing are performed on the log information at different levels to obtain a log analysis strategy.

[0052] Specifically, if the level judgment result is the debug level, the development and debugging operation information of the application code in the log information is obtained, and the development and debugging operation information is sent to the developer port to generate a debug log analysis strategy.

[0053] Furthermore, if the level judgment result is the error level, through an exception analysis and location algorithm, the abnormal information and call stack information in the log information are used to locate the abnormal position of the relevant code, and an error log analysis strategy is obtained. Among them, the error log analysis strategy is used for the abnormal functions and abnormal code lines in the application code.

[0054] Furthermore, if the level judgment result is the warning level, through an association rule mining algorithm, the combination identification of warning faults is performed on the log record combinations in the log information to obtain a warning log analysis strategy.

[0055] Among them, the log analysis strategy includes: a debug log analysis strategy, an error log analysis strategy, and a warning log analysis strategy.

[0056] As a feasible implementation, the intelligent log analysis and processing module is utilized to support intelligent analysis and processing based on the log level and log files. It can automatically identify the log levels (such as debug level, information level, warning level, error level, etc.) and adopt different processing strategies according to different levels. For logs at the error level, the module can quickly locate the position where the problem occurs and the relevant code, and provide detailed error descriptions and possible solutions. At the same time, the module can also conduct in-depth analysis of log files, such as counting the number of logs in different time periods, analyzing the time trend of logs, and mining associated information in logs, thereby helping developers and operation and maintenance personnel comprehensively understand the running status of the system.

[0057] In one embodiment, different processing strategies are adopted for logs at different levels. For example, for logs at the debug level, they are mainly used in the development and debugging stages to provide detailed running information to facilitate developers to track the execution process of the program; for logs at the error level, problems need to be quickly located and fixed, so it is necessary to analyze the logs to find the cause and location of the problems and provide corresponding solutions.

[0058] In one embodiment, for logs at the error level, an exception analysis algorithm is used to locate the position where the problem occurs and the relevant code. For example, by analyzing the exception information and call stack information thrown in the logs, the function and line number of the code where the problem lies are determined. Data mining algorithms can also be used for in-depth analysis of a large number of log files. For example, the time series analysis method is used to analyze the change trend of the number of logs in different time periods to judge whether there are performance problems during high-load periods of the system; the association rule mining algorithm is used to find combinations of log records that often appear simultaneously to discover possible system failure modes.

[0059] S104. If the log analysis strategy is the error log analysis strategy, then perform intelligent code correction on the log information in the error log analysis strategy to obtain corrected code information.

[0060] Specifically, it is also necessary to conduct code review processing on the log information in the error log analysis strategy to determine abnormal code information. Among them, the abnormal code information at least includes: syntax errors, logical errors, and resource leaks.

[0061] Furthermore, through the established code quality detection rule library, template matching of abnormal types is performed on both the static code in the abnormal code information and the log information to obtain abnormal type information. The code quality detection rule library contains specification templates, best practice templates, and common problem pattern templates for various programming languages.

[0062] Furthermore, perform retrieval processing on the correction strategy for the abnormal type information to determine the correction suggestion strategy.

[0063] Further, according to the correction suggestion strategy, intelligent code correction is performed on the log information in the error log analysis strategy, and based on the manual verification result, the corrected code information is obtained.

[0064] In one embodiment, problems existing in the code, such as syntax errors, logical errors, resource leaks, etc., can be detected by analyzing log data and code review first. Then, a code quality detection rule library is established, which includes specifications, best practices, and common problem patterns of various programming languages. By statically analyzing the log data and the code, it is detected whether the code violates these rules, so as to determine the problems existing in the code. For different types of problems, the corresponding correction strategies are retrieved from the rule library to generate specific correction suggestions. That is, according to the detected problems, combined with relevant programming specifications and best practices, specific correction suggestions are put forward. For example, for syntax errors, correct code examples can be provided; for logical errors, the reasons for the errors can be analyzed and the modified code logic can be given. For example, for syntax errors in the code, correct code syntax examples are provided; for resource leak problems, it is recommended to add the corresponding resource release code. Developers can manually modify the code according to the correction suggestions, or use an automated code correction tool to automatically modify the code according to the correction suggestions.

[0065] S105. Based on the corrected code information, feedback and optimize the log intelligent recommendation model.

[0066] Specifically, the corrected code information and the corresponding relevant running logs are first re-input into the log intelligent recommendation model.

[0067] Further, based on the first update data of the corrected code information and the second update data of the relevant running logs, iterative update processing of the parameters and algorithms of the log intelligent recommendation model is performed to optimize and obtain the log intelligent recommendation model.

[0068] As a feasible implementation manner, an intelligent code correction and model feedback optimization module is used. This module can intelligently correct the code and feedback the correction result to the model for optimization. When the intelligent log analysis processing module detects problems in the code, such as exceptions or non-compliant situations in the log records, it will automatically analyze the reasons for the problems and put forward corresponding correction suggestions according to the types and severities of the problems. Developers can correct the code according to the suggestions, or directly use the automatic correction function provided by the module. The corrected code and the relevant running logs will be fed back to the intelligent log information generation module and the intelligent log analysis processing module again to further optimize the parameters and algorithms of the model, improving the accuracy and intelligence of the model.

[0069] In addition, the embodiment of the present application also provides a low-code platform development system based on a log system, such asFigure 3 As shown in , the low-code platform development system 300 based on the log system specifically includes:

[0070] An intelligent log information generation module, which is used to generate log information by processing the relevant information of the current development scenario based on a pre-trained log intelligent recommendation model;

[0071] An intelligent log analysis and processing module, which is used to perform a level judgment process on the log information to obtain a level judgment result; according to the level judgment result, perform corresponding in-depth analysis processing on the log information at different levels to obtain a log analysis strategy;

[0072] An intelligent code correction and model feedback optimization module, which is used to, if the log analysis strategy is an error log analysis strategy, perform intelligent code correction on the log information in the error log analysis strategy to obtain corrected code information; based on the corrected code information, feedback and optimize the log intelligent recommendation model.

[0073] In the embodiments of the present application, the log intelligent recommendation model trained by itself is used to automatically generate log information, reducing the workload of developers writing log code and improving the development efficiency. The intelligent log analysis and processing module can comprehensively analyze the logs, helping developers and operation and maintenance personnel quickly locate problems and improving the maintainability and reliability of the system. The intelligent code correction and model feedback optimization module can automatically correct the code and feedback to optimize the model, forming a closed-loop optimization mechanism and continuously improving the intelligence level of the platform.

[0074] The embodiments in the present application are all described in a progressive manner. The same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the embodiments of the device and the non-volatile computer storage medium, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0075] The device and medium provided in the embodiments of the present application correspond one by one to the method. Therefore, the device and medium also have beneficial technical effects similar to those of the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the device and medium will not be elaborated here.

[0076] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0077] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0078] The memory may include non-permanent memory in the computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.

[0079] Computer-readable media include permanent and non-permanent, removable and non-removable media and can store information by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0080] The above description has been made of specific embodiments of the present application. In some cases, the acts or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require the particular order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0081] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various changes and modifications can be made to the embodiments of the present application.

Claims

1. A development method for a low-code platform based on a logging system, characterized in that, The method includes: Based on a pre-trained log intelligent recommendation model, perform log generation processing on the relevant information of the current development scenario to obtain log information; Perform level judgment processing on the log information to obtain a level judgment result; According to the level judgment result, perform corresponding in-depth analysis processing on the log information at different levels to obtain a log analysis strategy; If the log analysis strategy is an error log analysis strategy, perform intelligent code correction on the log information in the error log analysis strategy to obtain corrected code information; Based on the corrected code information, feedback and optimize the log intelligent recommendation model.

2. A low-code platform development method based on a log system according to claim 1, characterized in that, Before performing log generation processing on the relevant information of the current development scenario based on a pre-trained log intelligent recommendation model to obtain log information, the method further includes: Collect a historical log dataset; wherein, the historical log dataset at least includes: application program datasets of different types and log records in different scenarios; Perform cleaning processing on abnormal data in the historical log dataset, and extract the key log information of each group of log data in the cleaned historical log dataset; wherein, the key log information at least includes: timestamp, operation type, and operation object; Through a deep learning network under the Transformer architecture, and based on the learning rate and batch size configured corresponding to the attribute items in the historical log dataset, input the key log information of each group of log data into the initial log intelligent recommendation model, and obtain the trained log intelligent recommendation model based on the model convergence degree after multiple rounds of iterative training.

3. A low-code platform development method based on a log system according to claim 2, characterized in that Performing log generation processing on the relevant information of the current development scenario based on a pre-trained log intelligent recommendation model to obtain log information specifically includes: If the low-code platform detects application development data, then obtain the current code-related information of the current development scenario corresponding to the application development data in real time; wherein, the current code-related information at least includes: business logic, operation process, and input-output parameters; Input the current code-related information into the log intelligent recommendation model; Through the log intelligent recommendation model, perform one-by-one classification processing on all attribute item data in the current code-related information to obtain attribute item classification data; and perform detailed information log automatic pulling processing on the attribute item classification data to generate the log information.

4. A development method of a low-code platform based on a log system according to claim 1, characterized in that, Performing level judgment processing on the log information to obtain a level judgment result specifically includes: Through a regular expression, perform format and content parsing processing of the relevant log file on the original log stream in the log information, and perform text cleaning processing on the free text part of the log file to obtain multi-modal log information; wherein, the original log stream includes: structured log fields and unstructured text; the text cleaning includes: domain word segmentation, stop word filtering, stemming, and word embedding conversion; Perform identification and judgment on the level identification information of the multi-modal log information; wherein, the level identification information is the specific character at the beginning in the log information. If the multimodal log information has the level identification information, based on the level identification information, determine a first level judgment result at the level identification information; If the multimodal log information does not have the level identification information, through a preset keyword-level mapping table, perform keyword mapping processing on the multimodal log information under the rule engine to obtain a second level judgment result under the content features in the log information; wherein, the rule engine includes: a natural language processing technology engine; Wherein, the level judgment result of the log information at least includes: debug level, information level, warning level, and error level.

5. A development method of a low-code platform based on a log system according to claim 1, characterized in that, According to the level judgment result, perform corresponding in-depth analysis processing on the log information at different levels to obtain a log analysis strategy, specifically including: If the level judgment result is the debug level, obtain the development and debugging operation information of the application code in the log information, and send the development and debugging operation information to the developer port to generate a debug log analysis strategy; If the level judgment result is the error level, through an exception analysis and location algorithm, perform relevant code exception location on the exception information and call stack information in the log information to obtain an error log analysis strategy; wherein, the error log analysis strategy is used for exception functions and exception code lines in the application code; If the level judgment result is the warning level, through an association rule mining algorithm, perform combined identification of warning faults on the log record combination in the log information to obtain a warning log analysis strategy; Wherein, the log analysis strategy includes: the debug log analysis strategy, the error log analysis strategy, and the warning log analysis strategy.

6. A low-code platform development method based on a log system according to claim 1, characterized in that If the log analysis strategy is the error log analysis strategy, perform intelligent code correction on the log information in the error log analysis strategy to obtain corrected code information, specifically including: Perform code review processing on the log information in the error log analysis strategy to determine exception code information; wherein, the exception code information at least includes: syntax error, logical error, and resource leakage; Through the established code quality detection rule library, perform template matching of exception types on both the static code and log information in the exception code information to obtain exception type information; Perform retrieval processing on the correction strategy for the exception type information to determine a correction suggestion strategy; According to the correction suggestion strategy, perform intelligent code correction on the log information in the error log analysis strategy, and based on the manual verification result, obtain the corrected code information.

7. A method for developing a low-code platform based on a log system according to claim 6, characterized in that, The code quality detection rule library contains specification templates, best practice templates, and common problem pattern templates for multiple programming languages.

8. A low-code platform development method based on a log system according to claim 1, characterized in that Based on the corrected code information, feedback and optimize the log intelligent recommendation model, specifically including: Re-enter the corrected code information and the corresponding relevant running logs into the log intelligent recommendation model; Based on the first updated data of the corrected code information and the second updated data of the relevant operation logs, perform iterative update processing on the parameters and algorithms of the log intelligent recommendation model itself to optimize and obtain the log intelligent recommendation model.

9. A low-code platform development system based on a logging system, characterized in that, The system includes: An intelligent log information generation module, configured to perform log generation processing on the relevant information of the current development scenario based on a pre-trained log intelligent recommendation model to obtain log information; An intelligent log analysis and processing module, configured to perform level judgment processing on the log information to obtain a level judgment result; according to the level judgment result, perform corresponding in-depth analysis processing on the log information at different levels to obtain a log analysis strategy; An intelligent code correction and model feedback optimization module, configured to, if the log analysis strategy is an error log analysis strategy, perform intelligent code correction on the log information in the error log analysis strategy to obtain corrected code information; based on the corrected code information, feedback and optimize the log intelligent recommendation model.

10. A non-volatile computer storage medium, characterized in that, The storage medium is a non-volatile computer-readable storage medium, and the non-volatile computer-readable storage medium stores at least one program, and each program includes instructions, and when the instructions are executed by a terminal, the terminal is caused to execute a low-code platform development method based on a log system according to any one of claims 1-8.

Images