CN117420985B

CN117420985B - Method for packaging android service function by using JavaScript

Info

Publication number: CN117420985B
Application number: CN202311445484.6A
Authority: CN
Inventors: 阳纯正; 唐宏才; 杨瞻远
Original assignee: Hunan Shuanghong Technology Co ltd
Current assignee: Hunan Shuanghong Technology Co ltd
Priority date: 2023-11-02
Filing date: 2023-11-02
Publication date: 2024-03-29
Anticipated expiration: 2043-11-02
Also published as: CN117420985A

Abstract

The invention relates to the technical field of computers, in particular to a method for packaging android service functions by using JavaScript. The method comprises the following steps: acquiring android system interface data and interface requirement data, and performing format conversion on the android system interface data according to the interface requirement data so as to obtain interface format conversion data; constructing a sub-function relation diagram of the interface format conversion data, thereby obtaining sub-function relation diagram data; and obtaining JavaScript feature data, and carrying out depth parameter mapping association on the sub-function relation diagram data according to the JavaScript feature data, thereby obtaining JavaScript sub-function feature associated data. The invention can realize the compatibility problem between different versions so as to avoid the problem of low practicability caused by a single version.

Description

Method for packaging android service function by using JavaScript

Technical Field

The invention relates to the technical field of computers, in particular to a method for packaging android service functions by using JavaScript.

Background

The method for encapsulating the Android service function by using the JavaScript generally embeds a WebView (a component in Android) in a Web page, and interacts with the WebView through the JavaScript, so that the purpose of calling the Android native function is achieved. The implementation of WebView in different Android versions may vary, and compatibility issues between the different versions need to be handled.

Disclosure of Invention

The invention provides a method for encapsulating android service functions by using JavaScript to solve at least one technical problem.

The application provides a method for packaging android service functions by using JavaScript, which comprises the following steps:

step S1: obtaining android system interface data and interface requirement data, and performing format conversion on the android system interface data according to the interface requirement data to obtain interface format conversion data, wherein the interface requirement data are data format data, data structure data and supplementary information data required by the communication requirement of an android system end, and the supplementary information data comprise protocol version number data, identity authentication data, authorization data, configuration parameter data and security policy data;

step S2: constructing a sub-function relation diagram of the interface format conversion data, thereby obtaining sub-function relation diagram data;

step S3: acquiring JavaScript feature data, and carrying out depth parameter mapping association on sub-function relation diagram data according to the JavaScript feature data so as to obtain JavaScript sub-function feature associated data;

step S4: carrying out android system service encapsulation on the JavaScript sub-function feature associated data according to preset JavaScript encapsulation data so as to obtain android service function encapsulation data;

Step S5: program verification is carried out according to the android service function package data, so that program verification data are obtained, optimization iteration is carried out on the android service function package data according to the program verification data, and then JavaScript package android service function data are obtained.

The invention can expand or modify specific functions according to the needs by decomposing the functions into modularized steps without modifying the whole system in a large scale, and the design can be more flexible in future development and maintenance. The JavaScript is used as the packaging language, so that the service function can be packaged on the android system, the platform-crossing capability is strong, the same function packaging flow can be multiplexed on different platforms, and the development efficiency is improved. Through depth parameter mapping association, the dynamic property and flexibility of JavaScript can be fully utilized, and specific JavaScript features are combined with android service functions, so that the effect of performance optimization is achieved, and the execution of the functions can be more efficient. Program verification is performed in step S5, and problems in function implementation can be found and repaired in time through a systematic verification process, so that stability and reliability of functions are guaranteed, and meanwhile, performance of the functions can be further improved through optimization iteration of the functions according to verification data. The functions are packaged into reusable modules, the workload of repeated development can be greatly reduced, and in similar projects, the packaging of different functions can be quickly realized by modifying and customizing specific steps. The invention can realize the compatibility problem between different versions, so as to avoid the problem of low practicability caused by a single version.

Preferably, step S1 is specifically:

step S11: communicating with an android system interface database through an HTTP request, so as to obtain android system interface data;

step S12: performing JSON analysis on the android system interface data to obtain analysis object data;

step S13: acquiring terminal equipment data, and generating interface requirement data according to the terminal equipment data so as to obtain the interface requirement data;

step S14: and carrying out data format conversion on the analysis object data according to the interface requirement data, thereby obtaining interface format conversion data.

According to the method, the HTTP request is directly communicated with the android system interface database (step S11), so that complicated manual operation and data transmission steps are avoided, and the data obtaining efficiency is greatly improved. In step S12 and step S14, through JSON parsing and data format conversion, it is ensured that data obtained from the android system interface and the terminal device can be accurately and consistently processed and understood, which helps to maintain a unified format of data, and facilitates subsequent processing and analysis. By performing JSON analysis on the android system interface data (step S12), it is ensured that the data obtained from the interface can be accurately analyzed into object data, possible analysis errors are avoided, and accuracy of data processing is ensured. Through the clear steps and the data processing flow, the data processing requirements of different interfaces and devices can be flexibly met, and meanwhile, a foundation is provided for subsequent functional expansion. By adopting an automatic data obtaining and processing mode, the interference of manual operation is reduced, the error risk caused by human errors is reduced, and the reliability of data processing is improved. By directly communicating with the android system interface database (step S11), the latest data can be obtained in real time, and the instantaneity of the system and the timeliness of the data are ensured. By performing accurate processing and format conversion on the obtained data (step S14), high quality of the data is ensured, which is helpful for improving overall performance of the system.

Preferably, step S13 is specifically:

step S131: acquiring terminal equipment data;

step S132: analyzing the terminal equipment data by using a preset intelligent data analysis engine so as to obtain intelligent analysis data of the terminal equipment;

step S133: generating preliminary interface requirement data according to the intelligent analysis data of the terminal equipment, so as to obtain the preliminary interface requirement data;

step S134: adjusting the preset automatic verification parameter data according to the intelligent analysis data of the terminal equipment so as to obtain the automatic verification parameter data;

step S135: and integrating the automatic check parameter data and the preliminary interface requirement data to obtain the interface requirement data.

In the invention, the preset intelligent data analysis engine (step S132) is utilized to identify and analyze the terminal equipment data so as to obtain intelligent analysis data, which means that the intelligent analysis data is obtained simply, but advanced data processing is carried out through the intelligent engine, and the intellectualization and the accuracy of the data processing are improved. By performing dynamic adjustment policy generation according to the terminal device intelligent analysis data (step S133), corresponding interface requirement data can be dynamically generated according to real-time terminal device states and data conditions, so that the system can flexibly adjust according to different conditions, and the adaptability and flexibility of the system are improved. By adjusting the preset automatic verification parameter data according to the intelligent analysis data of the terminal equipment (step S134), the verification parameters can be automatically optimized and adjusted according to actual conditions, and the accuracy and the effectiveness of verification are ensured. The data meeting the interface requirements is obtained by integrating the data of the automatic check parameter and the data of the preliminary interface requirements (step S135) and integrating the data which is subjected to intelligent analysis and processing with the check parameters, so that the complexity of subsequent processing is reduced. The steps relate to the processing and dynamic adjustment of the real-time data, so that the system can respond in time under different conditions, and the real-time performance and the adaptability of the system are improved.

Preferably, in step S133, the preliminary interface request data is generated by dynamic adjustment, the dynamic adjustment is processed by a dynamic adjustment policy, and the dynamic adjustment policy is processed by a dynamic adjustment evaluation index generated by evaluation calculation performed by a dynamic adjustment policy evaluation calculation formula, where the dynamic adjustment policy evaluation calculation formula specifically is:

；

assessment index for dynamic adjustment->For the device information index data, < >>For device status data->Load data for device performance->For the device memory index data, < >>For the hardware specification data of the device->For the data of the electric quantity of the equipment, the control unit is used for controlling the control unit to control the power of the equipment>For device sensor data, +.>The method comprises the steps of providing equipment function support degree data, wherein equipment information index data are equipment model number/serial number data, equipment state data are equipment running state data, equipment performance load data are current CPU utilization rate data, equipment memory index data are available memory data, equipment hardware specification data are storage capacity data, equipment electric quantity data are equipment residual available electric quantity data, equipment sensor data are equipment sensor function support degree data, and equipment function support degree data are degree data of functions or characteristics supported by equipment.

The invention constructs a dynamic adjustment strategy evaluation calculation formula which obtains a dynamic adjustment evaluation indexUsed to instruct the system how to adjust policies in different states to optimize the utilization of resources. By dynamically adjusting the policy, it is possible to rely on the real-time device status (/ ->) Performance load (+)>) Memory condition (++>) And the data are transmitted, so that dynamic allocation of system resources is realized, and the response speed and performance of the system are improved. The various parameters in the formula represent various state and performance data of the device, and by comprehensively evaluating the data, the device is prevented from running under high load or unstable state, so that the risk of system breakdown is reduced. By dynamically adjusting the strategy, it is possible toAccording to various states and performance data of the equipment, the working mode of the system is timely adjusted, the system is kept in a stable running state, and the stability of the system is improved. Each parameter in the formula can be adjusted according to the characteristics of different devices, so that the dynamic adjustment strategy can play an optimal effect on the different devices, and the universality of the system is improved. Through dynamic adjustment strategy, resources can be allocated reasonably according to various states and performance data of the equipment, so that energy consumption is reduced, and the battery life of the equipment is prolonged.

Preferably, step S2 is specifically:

step S21: extracting the sub-function node data from the interface format conversion data, thereby obtaining the sub-function node data;

step S22: constructing node dependency relationship of the sub-functional node data so as to obtain node directed graph data;

step S23: defining node attribute of the node directed graph data according to the sub-function node data, so as to obtain node attribute graph data;

step S24: performing sub-function logic modeling on the node attribute map data so as to obtain sub-function logic model data;

step S25: and generating a sub-function relation diagram of the node attribute diagram data according to the sub-function logic model data, thereby obtaining the sub-function relation diagram data.

According to the invention, the function is decomposed into the sub-function nodes (step S21), and the dependency relationship among the nodes is constructed (step S22), so that the function module is clearer, and is convenient for independent development, testing and maintenance. Through definition of the node attribute graph (step S23) and sub-function logic modeling (step S24), association and logic among all the functional nodes are ensured, functional coordination consistency of the whole system is ensured, and functional conflict and data confusion are avoided. Through the generation of the sub-function relation diagram (step S25), the relation among the function nodes is visually presented, so that the whole function structure is clear at a glance, and the executable process is visually presented, so that developers can better understand the relation among functions, the development efficiency is improved, or the development process is optimized, and the process with low load and high fault tolerance is provided. By extracting and modeling the sub-functional node data, the design of each functional module can be finer and more efficient, so that the overall performance of the system is improved, and the problems of potential logic uncertainty or simply realization of simple operation logic or overhigh system overhead caused by overall mapping are avoided.

Preferably, step S24 is specifically:

step S241: extracting input data specification from the node attribute map data, thereby obtaining input data specification data;

step S242: generating a model structure according to the input data specification data and the node attribute graph data, so as to obtain model structure data;

step S243: modeling the node attribute map data according to the model structure data so as to obtain preliminary sub-function logic model data;

step S244: acquiring historical modeling data according to the node attribute map data and a local historical modeling database, so as to obtain the historical modeling data;

step S245: and performing parameter tuning on the preliminary sub-function logic model data according to the historical modeling data, so as to obtain the sub-function logic model data.

According to the invention, through extracting the input data specification of the node attribute map data (step S241), the input data used in the modeling process is ensured to accord with the specification, so that the accuracy and precision of the modeling data are improved. By generating the model structure according to the input data specification data and the node attribute map data (step S242), automatic generation of the model structure is realized, workload of manually designing the model structure is reduced, and modeling efficiency is improved. By using the preset historical modeling database to collect historical modeling data (step S244), the past modeling experience and data can be referenced, so that the current sub-functional logic model is optimized, and the performance and accuracy of the model are improved. By performing parameter tuning on the preliminary subfunction logic model data according to the historical modeling data (step S245), automatic adjustment of model parameters is realized, so that the model is more in line with actual conditions, and the performance and prediction accuracy of the model are improved.

Preferably, in step S245, the parameter tuning is performed by a parameter tuning weight calculation formula, where the parameter tuning weight calculation formula specifically includes:

;

optimizing weight data for parameters, +.>For the first attribute data in the node attribute map data, < >>For the second attribute data in the node attribute map data, < > for the first attribute data>First performance influencing item for preliminary subfunction logic model data,/for the first performance influencing item>Second performance influencing item for preliminary subfunction logic model data,/for the first performance influencing item>For the parameter item to be tuned, +.>Is a constant term of circumference ratio, +.>Is a natural index term, < >>Is a model behavior control item, wherein the first attribute number in the node attribute graph dataThe method comprises the steps that node type data, node capacity data and node load data are included, second attribute data in node attribute diagram data comprise node transmission rate data and node power consumption data, a first performance influence item of preliminary sub-function logic model data comprises model input rate data and model response time data, a second performance influence item of the preliminary sub-function logic model data comprises model output quality data and model efficiency data, a parameter item to be regulated comprises model parameter importance degree data, and a model behavior control item is parameter data affecting model behaviors in a scene.

The invention constructs a parameter tuning weight calculation formula, and the calculation formula enables the tuning process to be more comprehensive and accurate through factors such as node attributes, model performances, parameters to be tuned and the like, and can be better adapted to actual scenes. Parameters in the formulaAnd->Representing the first and second attribute data in the node attribute map data, respectively, means that this formula considers the specific information of the node attribute in the tuning process, thereby reflecting the system characteristics more accurately. Parameters in the formula>And->The first performance influence item and the second performance influence item which represent the preliminary subfunction logic model data represent that in the tuning process, not only node attributes are considered, but also influence of model performance on results is considered, so that tuning is more comprehensive and accurate. +.>Representing the parameter items to be tuned, the specific parameter values influence the tuning result in the tuning process, so that the tuning is more flexible and controllable.

Preferably, step S3 is specifically:

step S31: acquiring JavaScript feature data;

step S32: carrying out structuring treatment on the JavaScript feature data so as to obtain JavaScript feature structured data;

step S33: generating parameter association mapping table data according to the JavaScript feature structured data;

Step S34: carrying out label processing on the parameter association mapping table data according to JavaScript feature label data corresponding to the JavaScript feature structured data, thereby obtaining JavaScript feature mapping table data;

step S35: and carrying out depth parameter mapping association on the sub-function relation diagram data according to the JavaScript feature mapping table data, thereby obtaining JavaScript sub-function feature association data.

By acquiring the JavaScript feature data (step S31), the invention fully utilizes the JavaScript as a powerful script language, and provides abundant information and support for subsequent function encapsulation. By carrying out structuring processing on the JavaScript feature data (step S32), the JavaScript feature data is converted into data with good organization structure, so that subsequent processing is more efficient and targeted. Through generating parameter association mapping table data according to the JavaScript feature structured data (step S33), association between JavaScript features and functional parameters is realized, and an important basis is provided for subsequent functional packaging. By performing label processing on the JavaScript feature label data corresponding to the JavaScript feature structured data (step S34), the accuracy and precision of the mapping table are improved, so that the association is finer. The depth parameter mapping association is carried out on the sub-function relation diagram data through the JavaScript feature mapping table data (step S35), so that the efficient association between JavaScript features and the function diagram is realized, and important support is provided for the realization of functions. The invention enables the realization of the function to be more robust, and the realization of the function to be more flexible and intelligent through the structured processing and mapping association of the JavaScript feature.

Preferably, step S4 is specifically:

step S41: acquiring packaging requirement data;

step S42: making a service packaging strategy according to the packaging requirement data, thereby obtaining service packaging strategy data;

step S43: and carrying out service calling logic coding on the JavaScript sub-function feature associated data according to the service encapsulation policy data and preset JavaScript encapsulation data, thereby obtaining android service function encapsulation data.

According to the invention, the package requirement data is acquired (step S41), so that the accuracy and the comprehensiveness of the basic data in the package process are ensured, and an important basis is provided for the subsequent service package strategy formulation. By making a service encapsulation policy according to the encapsulation requirement data (step S42), a flexible encapsulation policy is made according to specific requirements, so that the encapsulation process better meets the actual requirements. By performing service calling logic coding on the JavaScript sub-function feature associated data according to the service encapsulation policy data and preset JavaScript encapsulation data (step S43), automatic coding on the service calling logic is realized, and the efficiency of the encapsulation process is improved. The invention makes the function packaging process more intelligent, and makes the packaging process more efficient and flexible through the establishment of the service packaging strategy and the automatic coding of the service calling logic. By flexibly formulating the service packaging strategy, the function is customized according to different packaging requirements, and the customizable performance of the function is improved, so that the service packaging strategy is better suitable for different scenes and requirements.

Preferably, the program verification data includes functional program verification data and abnormal program verification data, and step S5 is specifically:

step S51: performing function program simulation verification on the android service function package data to obtain function program verification data;

step S52: performing abnormal program simulation verification on the android service function package data to obtain abnormal program verification data;

step S53: performing program verification decision tree construction according to the functional program verification data and the abnormal program verification data, so as to obtain a program verification decision tree model;

step S54: and optimizing and iterating the android service function packaging data by using the program verification decision tree model, so as to obtain the JavaScript packaged android service function data.

According to the method, through the step S51 and the step S52, the function program simulation and the abnormal program simulation are carried out on the android service function package data, so that the comprehensive verification of functions including the processing of normal conditions and abnormal conditions is realized. Through step S53, a program verification decision tree model is constructed according to the functional program verification data and the abnormal program verification data, and the model can make an effective decision on the execution path of the program. And (3) optimizing and iterating the android service function package data by using the program verification decision tree model (step S54), so that the accuracy of program verification can be improved, and the execution of the program is more reliable. The invention realizes the automatic process of program verification, builds a decision tree model by simulating functions and abnormal conditions, and optimizes the program, thereby improving the efficiency of program verification. Through comprehensive program verification, the reliability of the functional package is ensured, the normal condition and the abnormal condition are processed, and the practicability and the stability of the functional package are improved.

Preferably, in step S53, the program verification decision tree construction is processed by a program verification decision tree error calculation formula, where the program verification decision tree error calculation formula specifically is:

；

validating decision tree error data for a program, +.>Verifying the quantity data of the data for the program, +.>Verifying data order items for a program,/->Is a constant term of circumference ratio, +.>Is->Program verification data->Is->Program verification decision tree model predictive data, +.>Verifying error tolerance data for a functional program, +.>For error adjustment item, ++>The error tolerance data is validated for the abnormal program,for the whole error adjustment term, +.>Verifying a micro disturbance term of a decision tree model for a program, < ->Is an error control term.

The invention constructs a program verification decision tree error calculation formula, which calculates error data @ by the calculation formula) The accuracy of program verification can be quantitatively evaluated by comparing errors between actual program verification data and decision tree model prediction data. />、/>、/>、/>The tolerance and adjustment items are introduced into the isoparameter, so that the sensitivity and tolerance of error calculation can be adjusted, < >>And->The equal parameters control the influence of the micro disturbance of the model, ensure the stability of error calculation and can be used for describing the influence of the change on the overall error.

The invention has the beneficial effects that: through the accurate processing of the interface requirement data, the format conversion of the android system interface data is realized, so that the readability and operability of the interface data are improved, and the subsequent processing is more efficient. The sub-function relation diagram in the invention establishes a foundation for the subsequent steps, and specific sub-function relation diagram data is generated by analyzing and extracting interface format conversion data, so that the dependency relationship of the system on each function node is clearer, and the problem of low program operation efficiency caused by integral logic mapping or simple grammar tree conversion is avoided. The depth parameter mapping association in the invention provides powerful support for the JavaScript feature data of the system, and the JavaScript feature data is associated with the sub-function relation diagram data, so that the JavaScript feature is accurately called, and the encapsulation of the android service function is more intelligent. According to the method and the device for encapsulating the JavaScript sub-function feature associated data, the JavaScript sub-function feature associated data is encapsulated through the preset JavaScript encapsulated data, so that the service function of the android system can meet specific requirements during encapsulation, and the applicability and the customization of the system are improved. The program verification and optimization iteration are key steps of the method, and the system can comprehensively verify and optimize the android service function through the generation of program verification data and the construction of the decision tree model, so that the packaged function is more stable and efficient, and a solid foundation is laid for improving the system performance.

Drawings

Other features, objects and advantages of the present application will become more apparent upon reading of the detailed description of non-limiting implementations made with reference to the following drawings in which:

FIG. 1 illustrates a flow chart of steps of a method for encapsulating android service functionality using JavaScript, in accordance with one embodiment;

FIG. 2 shows a step flow diagram of step S1 of an embodiment;

FIG. 3 shows a step flow diagram of step S13 of an embodiment;

FIG. 4 shows a step flow diagram of step S2 of an embodiment;

fig. 5 shows a step flow diagram of step S24 of an embodiment.

Detailed Description

The following is a clear and complete description of the technical method of the present patent in conjunction with the accompanying drawings, and it is evident that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, are intended to fall within the scope of the present invention.

Furthermore, the drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. The functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor methods and/or microcontroller methods.

It will be understood that, although the terms "first," "second," etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.

Referring to fig. 1 to 5, the present application provides a method for encapsulating android service functions by using JavaScript, which includes the following steps:

Specifically, the android system interface data is obtained by sending an HTTP request, for example, using an HTTP communication library such as OkHttp. Then, the interface data can be parsed using a JSON parsing library, such as Gson.

Specifically, for example, the interface requirement data includes the following: data format: JSON format, data structure: fields including user ID, name, mailbox, etc., and supplementary information: protocol version number, identity authentication information, authorization information, configuration parameters, security policies, etc. And according to the format defined by the interface requirement data, the system carries out format conversion on the original data acquired from the android system. For example, the original data is converted into data conforming to JSON format.

in particular, the sub-functional relationship graph is constructed, for example, using a graph theory-related library or algorithm, such as a topological ordering algorithm. Graph theory algorithm (topological ordering, etc.): for establishing dependencies between sub-functions, ensuring execution in the correct order.

Specifically, for example, data processing and mapping algorithms, such as a deep learning model or a complex mapping function, are used to correlate JavaScript feature data with sub-functional relationship diagram data.

specifically, javaScript feature association data is encapsulated as Android system specific services, for example, with specific services and APIs provided by Android, such as various kinds and methods in Android sdks.

Specifically, program verification is performed using, for example, an automated test tool, such as a JUnit or an Appium. And according to the verification result, the android service function data can be optimally packaged through a feedback control algorithm.

Specifically, for example, the network connection state of the android system is obtained, an OkHttp library is used to send an HTTP request to a corresponding system interface, and returned data is obtained. The interface requirement data may include information such as URL of the request, method of the request, etc. The interface format conversion data includes functions to obtain network connection status and functions to obtain device information, and a graph theory algorithm, such as topology sequencing, is used to construct a sub-function relationship graph to ensure that the functions are executed in the correct order. The JavaScript feature data comprises information such as network connection speed, equipment model number, version number, javaScript grammar tree and the like, and the features are associated with the subfunction relation diagram data by using a data processing and mapping algorithm such as a deep learning model, wherein the JavaScript sentence block-subfunction relation diagram corresponds to the JavaScript grammar tree corresponding to the equipment model number-version number. The function that gets the network connection state is packaged as an android system specific service, and the relevant class and method provided in the android sdk, such as ConnectivityManager, are used to implement the packaging of this function. The encapsulated android service functionality is tested using an automated test tool such as JUnit or app. And according to the test result, feeding back to the optimization iteration process, and adjusting and improving the package data through an algorithm.

Preferably, step S1 is specifically:

specifically, for example, using modern HTTP request libraries, such as OkHttp, communication with the android system interface is performed by sending HTTP requests. OkHttp library: the method is used for executing HTTP requests, supporting various types of requests such as GET, POST and the like, and providing an efficient network communication function.

specifically, data obtained from the android system interface is parsed into actionable objects using, for example, JSON parsing library, such as Gson. Gson library: for converting JSON formatted data into Java objects, a convenient API is provided to handle JSON data.

specifically, for example, device APIs (such as the sensor API of Android) are used to obtain terminal device data, and then corresponding data is generated according to interface requirement data. AndroidSensorAPI: an interface is provided for accessing sensor data of the device, and data of various sensors such as acceleration, gyroscopes and the like can be obtained.

Specifically, the parsing object data is converted into a format conforming to the interface requirements, for example, using a data format conversion function provided by the Java programming language. Java data format conversion: the functions of Java data type conversion, formatting and the like can be utilized to perform corresponding format conversion on the analyzed data according to interface requirements.

Preferably, step S13 is specifically:

step S131: acquiring terminal equipment data;

specifically, sensor data of the terminal device is obtained using, for example, a sensor api provided by the Android platform, such as an acceleration sensor, a gyro sensor, or the like. AndroidSensorAPI: an interface is provided for accessing sensor data of the device, and data of various sensors such as acceleration, gyroscopes and the like can be obtained.

Specifically, for example, device number data, device memory data, device CPU data, and device power data in the terminal device data are acquired.

specifically, for example, the data analysis engine, such as TensorFlow, pyTorch, is used to perform recognition analysis of the deep learning model on the terminal device data.

specifically, for example, algorithms, such as genetic algorithm, simulated annealing algorithm, etc., are used to generate a dynamic adjustment strategy according to the terminal device intelligent analysis data and preset parameters.

Specifically, for example, according to the intelligent analysis data and preset parameters of the terminal equipment, a dynamic adjustment algorithm is applied to generate an adaptive strategy so as to meet the interface requirements.

specifically, for example, according to intelligent analysis data of the terminal equipment, the adjustment processing of the parameter data is performed in combination with preset automatic verification parameters, such as linear regression, polynomial fitting and the like.

Specifically, for example, the system presets a set of preset verification parameters, where the parameters may include a threshold value, a specification value, and the like, to determine whether the data of the terminal device meets the expected conditions. Parameter adjustment is carried out according to the analyzed data: by comparing the parsed data with preset verification parameters, the system determines whether these parameters need to be adjusted. For example, if the data provided by the terminal device exceeds a preset threshold, the system needs to adjust the verification parameters accordingly to ensure correct verification. If the data specification provided by the device is illegal, the data specification processing is performed according to specific violation conditions, such as missing/excessive/normal form non-specification.

Specifically, for example, according to the automatic verification parameter data and the preliminary interface requirement data, data integration is performed to generate final interface requirement data, such as weighted average, data fusion and the like.

Specifically, final data meeting the interface requirements may be generated, for example, by combining the auto-verification parameter data and the preliminary interface requirement data.

Specifically, for example, terminal equipment data such as load capacity, computing capacity, sensor parameters and the like are analyzed, for example, preset JSON or XML format mapping is performed on the terminal equipment data, so as to obtain terminal equipment intelligent analysis data, identification is performed according to the terminal equipment intelligent analysis data, so as to obtain the terminal equipment intelligent analysis data, for example, a use scene, sensor representation meaning, equipment performance and corresponding primary function interface requirement data are obtained, for example, an identification mode is established through a preset identification model, and the identification model is constructed through a neural network algorithm or a decision tree algorithm. According to the terminal equipment data obtained through analysis, preset automatic verification parameters can be dynamically adjusted. For example, if the higher computing power of the terminal device is detected, the relevant verification parameters are adjusted accordingly, so as to improve the detection sensitivity.

；

assessment index for dynamic adjustment->For the device information index data, < >>For device status data->Load data for device performance->For the device memory index data, < >>For the hardware specification data of the device->For the data of the electric quantity of the equipment, the control unit is used for controlling the control unit to control the power of the equipment>For device sensor data, +.>The system is characterized in that the system is equipment function support degree data, wherein equipment information index data is equipment model/serial number data, equipment state data is equipment running state data, equipment performance load data is current CPU utilization rate data, equipment memory index data is available memory data, and equipment is hardThe piece specification data is storage capacity data, the equipment electric quantity data is equipment residual available electric quantity data, the equipment sensor data is equipment sensor function support degree data, and the equipment function support degree data is equipment support function or characteristic degree data.

Specifically, for example, the device information index data (B) is device model/serial number data such as: the device model is 'SamsungGalaxyS 21', and the serial number is '123456789'; device status data (C) device operational status data, such as: the equipment is in a normal running state at present; device performance load data (A) current CPU utilization data, such as: CPU utilization is 60%; device memory index data (E) available memory data, such as: the available memory is 2GB; device hardware specification data (F) storage capacity data, such as: the storage capacity is 128GB; device power data (G) device remaining available power data, such as: the residual electric quantity is 50%; device sensor data (H) device sensor function support degree data such as: the device supports acceleration sensors, gyroscopes, etc.; device function support degree data (I) degree data of functions or characteristics supported by the device, for example: the device supports high-definition video playing, 5G network and the like. The device sensor data (H) and the device function support degree data (I) define a numerical map as follows: device sensor data (H) map: support the acceleration sensor: 1, support gyroscopes: 2, do not support any sensors: 0; device function support degree data (I) mapping: high definition video play support degree: from 1 to 5 (1 means unsupported, 5 means very high supported), 5G network supported: from 1 to 5 (1 means not supported, 5 means very high degree of support); the system converts descriptive text information into indexes with numerical meanings in a matching mode. For example: if one device supports an acceleration sensor and a gyroscope, and the support degree for high-definition video playing is 4, and the support degree for a 5G network is 5, the system maps the acceleration sensor and the gyroscope as follows: and the device sensor data (H) is 3 (supporting an acceleration sensor and a gyroscope), and the device function supporting degree data (I) is [4,5] (the supporting degree of high-definition video playing is 4,5G and the supporting degree of a network is 5).

The invention constructs aEvaluating a calculation formula for dynamic adjustment strategy, the calculation formula obtaining a dynamic adjustment evaluation indexUsed to instruct the system how to adjust policies in different states to optimize the utilization of resources. By dynamically adjusting the policy, it is possible to rely on the real-time device status (/ ->) Performance load (+)>) Memory condition (++>) And the data are transmitted, so that dynamic allocation of system resources is realized, and the response speed and performance of the system are improved. The various parameters in the formula represent various state and performance data of the device, and by comprehensively evaluating the data, the device is prevented from running under high load or unstable state, so that the risk of system breakdown is reduced. By means of the dynamic adjustment strategy, the working mode of the system can be adjusted in time according to various states and performance data of the equipment, the system is kept in a stable running state, and the stability of the system is improved. Each parameter in the formula can be adjusted according to the characteristics of different devices, so that the dynamic adjustment strategy can play an optimal effect on the different devices, and the universality of the system is improved. Through dynamic adjustment strategy, resources can be allocated reasonably according to various states and performance data of the equipment, so that energy consumption is reduced, and the battery life of the equipment is prolonged.

Preferably, step S2 is specifically:

specifically, the subfunction node data is extracted from the interface data in XML format, for example, using XPath technology. XPath: a query language for navigating and extracting information in XML documents. Using XPath, the required sub-function node data can be located and extracted in the XML document, which makes the data extraction process more efficient and accurate.

specifically, for example, using a graph theory algorithm, dependency relationships between nodes are established from the sub-function node data. Graph theory algorithm: such as Depth First Search (DFS), breadth First Search (BFS), etc.

specifically, for example, a corresponding attribute is defined for each node, such as a type of node, an input parameter, an output parameter, and the like. Data structure definition: data structures such as classes, structures, etc. may be used to represent attributes of nodes. Node attribute definition: node name: calculator, node type: operation node, input parameters: input 1 (Operand 1): representing the first Operand, input 2 (Operand 2): representing the second operand, output parameters: results (Result): node directed graph data representing the sum of two input parameters: assume a simple directed graph comprising two nodes: an "input node" and a "calculator node". Input node: type (2): input nodes, output parameters: two operands (Operand 1 and Operand 2) entered by the user through the UI, calculator node: type (2): operation node, input parameters: operand1 and Operand2 obtained from input nodes, output parameters: and calculating the Result.

specifically, the execution logic of each sub-functional node is modeled using a method such as a UML activity diagram or Petri net. UML activity diagram: for describing the flow and behavior of activities in the system.

Specifically, a sub-functional relationship graph is constructed from sub-functional logic model data, for example, using a graph theory algorithm. Such as using a shortest path algorithm, topological ordering, etc.

Preferably, step S24 is specifically:

specifically, input data conforming to a specific specification is extracted from node attribute map data, for example, using a regular expression. The regular expression: a powerful pattern matching tool for matching strings. By using regular expressions, required input data can be efficiently extracted from node attribute map data, ensuring that it meets specifications.

specifically, for example, according to the input data specification and the node attribute map data, a corresponding model structure is constructed by using a model generation algorithm. Such as decision trees, neural networks, etc.

specifically, the node property map data is modeled with a modeling tool (e.g., tensorFlow, pyTorch, etc.), for example, from the model structure data. Such as model modeling using a framework of TensorFlow, pyTorch or the like.

specifically, historical modeling data relating to node attribute map data is extracted from a local historical modeling database, for example, using a database query language (e.g., SQL).

Specifically, the preliminary sub-functional logic model data is parameter-optimized according to the historical modeling data by using an optimization algorithm such as a genetic algorithm or gradient descent. Optimization algorithms such as genetic algorithm, gradient descent, etc.: for finding the optimal solution in the parameter space.

Specifically, for example, node attribute map data: node name: temperature converter, node type: temperature conversion node, input parameters: degree Celsius (Celsius): the temperature value to be converted is in degrees centigrade, and the output parameters are as follows: fahrenheit (Fahrenheit): the converted temperature value, in degrees fahrenheit, requires the extraction of degrees Celsius (Celsius) as an input data specification, which is specified in degrees Celsius. Model structure generation is performed according to input data specification data and node attribute map data, and the model structure generation may involve a temperature conversion formula, for example: fahrenheit=celsius×9/5+32, modeling the node attribute map data according to the model structure data, and converting the input degrees Celsius by using the formula to obtain the corresponding degrees Fahrenheit. According to the node attribute map data and a local historical modeling database, historical modeling data acquisition is carried out, and if conversion data from a few degrees celsius degree to a few degrees fahrenheit degree exist in the historical modeling database, the data can be used for further optimizing a model, such as adjusting a calculation formula based on a position or a space structure. By comparing historical modeling data to the degrees Fahrenheit values output by the model, parameters of the model may be adjusted to improve conversion accuracy.

;

optimizing weight data for parameters, +. >For the first attribute data in the node attribute map data, < >>For the second attribute data in the node attribute map data, < > for the first attribute data>First performance influencing item for preliminary subfunction logic model data,/for the first performance influencing item>Second performance influencing item for preliminary subfunction logic model data,/for the first performance influencing item>For the parameter item to be tuned, +.>Is a constant term of circumference ratio, +.>Is a natural index term, < >>The method comprises the steps that a first attribute data in node attribute graph data comprises node type data, node capacity data and node load data, a second attribute data in the node attribute graph data comprises node transmission rate data and node power consumption data, a first performance influence item of preliminary sub-function logic model data comprises model input rate data and model response time data, a second performance influence item of the preliminary sub-function logic model data comprises model output quality data and model efficiency data, a parameter item to be regulated comprises model parameter importance degree data, and the model behavior control item is parameter data affecting model behavior in a scene.

Specifically, for example, node attribute map data: first attribute data (a) node type data=2, second attribute data (b) node transmission rate data=100 Mbps, first attribute data (c) node capacity data=200 GB, second attribute data (d) node power consumption data=50 Watts, preliminary sub-function logic model data: a first performance impact term (c) model input rate data=10 requests/s, a second performance impact term (d) model output quality data=0.95 (full fraction 1); model efficiency data=80% (100% full), to-be-tuned optimal parameter term (x): model parameter importance data, for example: importance degree of parameter 1=0.3, importance degree of parameter 2=0.7, model behavior control term (f): parameter data affecting model behavior in scene=0.5, constant term: the parameter items to be tuned can be learning rate or regularization items, and also can be parameter items affecting corresponding primary sub-functional logic model data, such as the allowable efficiency or system load degree of function-code tree/graph in the primary sub-functional logic model.

Preferably, step S3 is specifically:

step S31: acquiring JavaScript feature data;

specifically, javaScript feature data is obtained through a debugging tool, for example, using a developer tool provided by a modern browser, such as ChromeDevTools.

Specifically, for example, javaScript feature data includes JavaScript syntax tree feature data and JavaScript regular function feature data, using Abstract Syntax Tree (AST): the AST is a tree structure, and represents a syntax structure of JavaScript codes, and the system parses the JavaScript codes into the AST by using a tool through a preset instruction, and extracts feature data therefrom. Babel is a popular JavaScript compiler by which the system converts the collected JavaScript code into backward compatible code, and it also provides the function of parsing the JavaScript code into AST.

specifically, the data is structured, for example, using JavaScript objects or JSON format.

specifically, each feature is associated with a corresponding parameter, for example, when processing JavaScript feature data. For example, feature names are mapped to corresponding parameters.

In particular, the parameters are further classified or labeled, for example, according to the labels or descriptions of the features, for subsequent associative mapping.

Specifically, the JavaScript feature is deep mapped with the data of the sub-functional relationship diagram, for example, using a data processing algorithm such as a machine learning model or natural language processing technique.

Specifically, text processing is realized, for example, by using a pre-trained NLP model, such as BERT or GPT-3, and depth parameter mapping association is completed, so that JavaScript sub-function feature association data is obtained. Processing the feature map data as follows: and acquiring the keywords and the description of each feature from the JavaScript feature mapping table. Mapping using NLP model: for the description in each sub-functional relationship diagram, the NLP model is used to compare it to the features in the feature map. A similarity calculation method (e.g., cosine similarity) may be used to determine the degree of similarity between the description and the feature. A threshold is selected to determine when to match a description with a feature. Establishing association data: the successfully matched sub-function descriptions are associated with the features and recorded.

By acquiring the JavaScript feature data (step S31), the JavaScript feature data fully utilizes the JavaScript feature of a powerful scripting language, and provides abundant information and support for subsequent function encapsulation. By carrying out structuring processing on the JavaScript feature data (step S32), the JavaScript feature data is converted into data with good organization structure, so that subsequent processing is more efficient and targeted. Through generating parameter association mapping table data according to the JavaScript feature structured data (step S33), association between JavaScript features and functional parameters is realized, and an important basis is provided for subsequent functional packaging. By performing label processing on the JavaScript feature label data corresponding to the JavaScript feature structured data (step S34), the accuracy and precision of the mapping table are improved, so that the association is finer. The depth parameter mapping association is carried out on the sub-function relation diagram data through the JavaScript feature mapping table data (step S35), so that the efficient association between JavaScript features and the function diagram is realized, and important support is provided for the realization of functions. The invention enables the realization of the function to be more robust, and the realization of the function to be more flexible and intelligent through the structured processing and mapping association of the JavaScript feature.

Preferably, step S4 is specifically:

step S41: acquiring packaging requirement data;

specifically, the requirements are explicitly encapsulated, e.g., using methods such as requirements documentation, interface specifications, etc., which are converted into an operable data format.

specifically, for example, it is assumed that the package requirement data includes the following information: the requirements 1, 2, the maximum memory usage limit and 3, the data security guarantee, the corresponding service encapsulation policy may include using an efficient algorithm to guarantee the fast response time, setting a threshold of the memory usage, and implementing encryption measures to guarantee the data security.

Specifically, for example, assume that JavaScript sub-function feature association data includes the following information: the method is characterized by comprising the steps of 1, accessing network resources, 2, data processing, and writing service calling logic by using corresponding JavaScript codes according to a service packaging strategy. For example, the Ajax library is used for network request, and the JavaScript algorithm is used for data processing.

specifically, the Android service function package data is run in a simulation environment, for example, using a simulation environment such as an Android simulator or a virtual machine, to obtain function program verification data. Assuming that the functional program verification data requires that the service can maintain stable response speed under the condition of poor network, simulating low network speed condition in a simulation environment, and observing the response condition of the service.

specifically, for example, a simulated environment is created, abnormal conditions such as network interruption, server crash, etc. are introduced in the environment, and the behavior of the service is observed. The abnormal program verification data is assumed to require that the service can correctly handle abnormal situations under the condition of unstable network, simulate network interruption or server faults in a simulation environment, and observe the coping capability of the service.

Specifically, for example, using a decision tree algorithm, a program verification decision tree model is constructed from functional program verification data and abnormal program verification data. And constructing a decision tree model according to the functional program verification data and the abnormal program verification data so as to perform corresponding processing according to the input condition in practical application.

Specifically, optimization iteration is performed on the android service function package data according to the output result of the program verification decision tree model, for example, so as to improve the performance and stability of the service. According to the output result of the program verification decision tree model, service logic can be adjusted in a targeted manner or some optimization strategies can be introduced to meet the program verification requirements. If the hypothesis program verification decision tree model outputs the following results: if the inputs are all significant digits, an addition is performed. If one or both inputs are invalid, an error prompt is returned. Case 1: if the results of the decision tree model output indicate that if the inputs are all significant numbers, then it is optimal to perform the addition directly, in which case the system does not need to be additionally optimized, since the model already gives the best way to handle. Case 2: for optimization of a particular invalid input, assume that the decision tree model outputs the following: if the number 1 is valid but the number 2 is invalid, an error prompt is returned. In this case, the system is optimized for the particular situation. For example, the system may add an additional check to the program that if a number 2 is detected to be invalid, an error prompt may be returned in advance without having to continue the addition operation. Case 3: a caching mechanism is introduced, and if the result output by the decision tree model indicates that some specific input combinations have been processed before, and the result can be cached to improve performance, the system presets a caching mechanism. For example, if the program validates the decision tree model finds that if both digits 1 and 2 are valid and their sum has been calculated before, the system caches the results so that the cached results are returned directly, without having to be recalculated, the next time the same input occurs. Case 4: optimization exception handling if the results output by the decision tree model indicate that the probability of certain specific exceptions occurring is very low, the system may take some optimization measures to improve performance. For example, if the decision tree model finds that a particular invalid input condition hardly occurs, the system places the code that handles that condition in a lower priority branch, thereby avoiding the performance overhead under normal conditions.

Specifically, for example, a simulated user may make a request to a helper via a voice command, the system may parse the request, query the corresponding weather data, and return the results to the user via a speech synthesis engine via the simulated user's voice input. The user is simulated to issue ambiguous voice commands or erroneous commands, such as too fast speech, unclear pronunciation, etc., to test the system's ability to respond in abnormal situations. According to the data of the function program simulation and the abnormal program simulation, a decision tree is constructed, and the decision tree can judge the intention of a user according to the input voice instruction and select corresponding operations (such as inquiring weather, prompting the user to reenter, etc.). And optimizing the functions of voice recognition, weather inquiry, voice synthesis and the like of the system according to the output of the program verification decision tree, and if the functions are transmitted to a cloud platform for voice recognition function searching.

；

Specifically, for example, a large amount of program verification data (n samples) including actual observed values is collectedAnd decision tree model predictor +.>. And (3) calculating the overall error, namely summing the errors of all samples to obtain the total error sigma (E_i). And (3) analyzing the error trend, namely further adjusting and optimizing the total error according to different parameters. For example, the +.>,/>,/>,/>,/>,/>And parameters, so as to achieve the purpose of optimizing the model. Specific numerical values are given according to actual conditions. For example: / >=0.1 (assuming that the tolerance of the function program simulation error is 0.1), +.>=0.01 (assuming that the error adjustment term is 0.01),/>=0.05 (assuming that the tolerance of the abnormal program simulation error is 0.05), ++>=2.0 (assuming that the global error adjustment term is 2.0), +.>=0.001 (assuming that the program verification decision tree model perturbation term is 0.001), +.>=0.005 (assuming that the error control term is 0.005).

Through the accurate processing of the interface requirement data, the format conversion of the android system interface data is realized, so that the readability and operability of the interface data are improved, and the subsequent processing is more efficient. The sub-function relation diagram in the invention establishes a foundation for the subsequent steps, and specific sub-function relation diagram data is generated by analyzing and extracting interface format conversion data, so that the dependency relationship of the system on each function node is clearer, and the problem of low program operation efficiency caused by integral logic mapping or simple grammar tree conversion is avoided. The depth parameter mapping association in the invention provides powerful support for the JavaScript feature data of the system, and the JavaScript feature data is associated with the sub-function relation diagram data, so that the JavaScript feature is accurately called, and the encapsulation of the android service function is more intelligent. According to the method and the device for encapsulating the JavaScript sub-function feature associated data, the JavaScript sub-function feature associated data is encapsulated through the preset JavaScript encapsulated data, so that the service function of the android system can meet specific requirements during encapsulation, and the applicability and the customization of the system are improved. The program verification and optimization iteration are key steps of the method, and the system can comprehensively verify and optimize the android service function through the generation of program verification data and the construction of the decision tree model, so that the packaged function is more stable and efficient, and a solid foundation is laid for improving the system performance.

The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The method for encapsulating the android service function by using the JavaScript is characterized by comprising the following steps of:

step S1, including:

step S14: performing data format conversion on the analysis object data according to the interface requirement data to obtain interface format conversion data, wherein the interface requirement data are data format data, data structure data and supplementary information data required by the android system end communication requirement, and the supplementary information data comprise protocol version number data, identity authentication data, authorization data, configuration parameter data and security policy data;

step S2, including:

step S25: generating a sub-function relation diagram of the node attribute diagram data according to the sub-function logic model data, thereby obtaining sub-function relation diagram data;

Step S3, including:

step S31: acquiring JavaScript feature data;

step S35: carrying out depth parameter mapping association on the sub-function relation diagram data according to the JavaScript feature mapping table data so as to obtain JavaScript sub-function feature association data;

step S4, including:

step S41: acquiring packaging requirement data;

step S43: carrying out service calling logic coding on the JavaScript sub-function feature associated data according to the service encapsulation policy data and preset JavaScript encapsulation data, thereby obtaining android service function encapsulation data;

step S5, including:

2. The method according to claim 1, wherein step S13 is specifically:

step S131: acquiring terminal equipment data;

step S135: data integration is carried out on the automatic verification parameter data and the preliminary interface requirement data, so that the interface requirement data is obtained;

In step S133, the preliminary interface request data is generated through dynamic adjustment, the dynamic adjustment is processed through a dynamic adjustment policy, the dynamic adjustment policy is processed through a dynamic adjustment evaluation index generated by evaluation calculation through a dynamic adjustment policy evaluation calculation formula, and the dynamic adjustment policy evaluation calculation formula specifically includes:

the method comprises the steps of dynamically adjusting an evaluation index, wherein D is equipment information index data, C is equipment state data, A is equipment performance load data, E is equipment memory index data, F is equipment hardware specification data, G is equipment electric quantity data, H is equipment sensor data, I is equipment function support degree data, wherein the equipment information index data is equipment model/serial number data, the equipment state data is equipment running state data, the equipment performance load data is current CPU utilization rate data, the equipment memory index data is available memory data, the equipment hardware specification data is storage capacity data, the equipment electric quantity data is equipment residual available electric quantity data, the equipment sensor data is equipment sensor function support degree data, and the equipment function support degree data is the degree data of functions or characteristics supported by equipment.

3. The method according to claim 1, wherein step S24 is specifically:

4. The method according to claim 3, wherein the parameter tuning in step S245 is tuned by a parameter tuning weight calculation formula, wherein the parameter tuning weight calculation formula is specifically:

w is parameter tuning weight data, a is first attribute data in node attribute map data, b is second attribute data in node attribute map data, c is a first performance influence item of preliminary subfunction logic model data, d is a second performance influence item of preliminary subfunction logic model data, x is a parameter item to be tuned, pi is a round rate constant item, e is a natural index item, f is a model behavior control item, wherein the first attribute data in node attribute map data comprises node type data, node capacity data and node load data, the second attribute data in node attribute map data comprises node transmission rate data and node power consumption data, the first performance influence item of preliminary subfunction logic model data comprises model input rate data and model response time data, the second performance influence item of preliminary subfunction logic model data comprises model output quality data and model efficiency data, the parameter item to be tuned comprises model parameter importance degree data, and the model behavior control item is parameter data affecting model behavior in a scene.

5. The method according to claim 1, wherein the program verification decision tree construction in step S53 is processed by a program verification decision tree error calculation formula, wherein the program verification decision tree error calculation formula is specifically:

E _p for program verification decision tree error data, n is the number data of the program verification data, i is the program verification data sequence item, pi is the circumference rate constant item, f _i The data is validated for the i-th program,for the ith predicted data of the program verification decision tree model, ρ is the functional program verification error tolerance data, ε is the error adjustment term, σ is the abnormal program verification error tolerance data, m is the overall error adjustment term, and +_>And verifying a small disturbance term of the decision tree model for a program, wherein tau is an error control term.