WO2023138270A1 - Form specification recommendation method, apparatus and system - Google Patents

Abstract

A form specification recommendation method, apparatus and system. The method comprises: firstly, acquiring a code input by a user and/or an annotation of the code (S302); then, on the basis of the code input by the user and/or the annotation of the code, obtaining a plurality of candidate form specifications (S304); displaying the plurality of candidate form specifications to the user (S306); the user being capable of selecting, from the plurality of candidate form specifications, a candidate form specification satisfying the requirements as a target form specification; and using the selected target form specification to verify the code to be verified. According to the method, the plurality of candidate form specifications are automatically generated and recommended to the user, so that the compiling work of the form specification can be simplified, and the acquisition efficiency of the form specification and the development efficiency of the computer system are improved.

Description

Formal specification recommended method, device and system

This application claims the priority of the Chinese patent application with the application number 202210074154.X and the title of the invention "Proforma Specification Recommendation Method, Device and System" submitted to the China Patent Office on January 21, 2022, the entire contents of which are incorporated in this application by reference.

technical field

The present application relates to the field of computer technology, and in particular to a form specification recommendation method, device and system.

Background technique

Formal methods are techniques for formally specifying, developing, and verifying computer systems (software/hardware systems) based on strict mathematical foundations. Designing and developing a reliable computer system usually requires a lot of formal specification support. Among them, the formal specification is the computer system model strictly described by the formal specification language, the properties that the computer system needs to satisfy, or the loopholes to be detected in the computer system, etc. The formal specification is the basis of the formal method and occupies an important position in the formal method.

However, at present, when designing and developing computer systems based on formal methods, developers usually need to manually write a large number of formal specifications. This method of obtaining formal specifications consumes a lot of manpower and time, which will lead to the problems of low efficiency of formal specification acquisition and computer system development.

Contents of the invention

The present application provides a formal specification recommendation method, device and system, which are used to simplify the writing work of the formal specification, improve the efficiency of obtaining the formal specification and the development efficiency of the computer system.

In a first aspect, a method for recommending formal specifications is provided, the method comprising: firstly, acquiring user-input codes and/or code annotations, and then obtaining multiple candidate formal specifications based on the user-input codes and/or code annotations, and displaying the multiple candidate formal specifications to the user.

By implementing the method described in the first aspect above, when the user writes the code and/or formal specification to be verified by the formal specification, the user-input code and/or code comments can be obtained in real time, and then based on the user-input code and/or code comments, multiple candidate formal specifications can be automatically obtained, and the multiple candidate formal specifications can provide users with good references. Save energy and time of users, and improve the acquisition efficiency of formal specifications.

In a possible implementation, the code and/or code-based comments can be specifically implemented to generate candidate formal specifications in the following manner: the code and/or code comments are input into the formal specification generation model for identification, and the formal specification output by the formal specification generation model is used as the candidate formal specification.

In a specific implementation, the formal specification generation model can be a model with high recognition accuracy obtained by training a sample set including a large number of known codes and/or annotations of a large number of known codes. The sample set includes a large number of known codes and/or annotations of a large number of known codes. It can be manually written by users or obtained from the Internet, and is not specifically limited here.

In a specific implementation, after the code and/or code comments are input to the formal specification generation model, the formal specification generation model can obtain multiple formal specifications and the confidence levels of the multiple formal specifications, and the formal specification generation model can directly output the formal specifications whose confidence is greater than the confidence threshold among the multiple formal specifications as candidate formal specifications, or sort them according to the degree of confidence, and output part of the formal specifications with higher confidence as candidate formal specifications, which is not specifically limited here. Both the confidence threshold and the number of output candidate form reductions can be customized by the user according to the actual situation, and are not specifically limited here.

Implement the above implementation, input the code and/or code comments into the formal specification generation model for identification, and use the formal specification output by the formal specification generation model as a candidate formal specification. Since the formal specification generation model is a model with high recognition accuracy trained in advance, the accuracy of the generated candidate formal specification can be guaranteed.

In a possible implementation manner, multiple candidate formal specifications can be obtained by implementing code and/or code-based comments in the following manner: first, calculate the matching degree between the code and/or code comments and multiple rules included in the rule base, and then, according to the calculation result, use the formal specification corresponding to the rule whose matching degree is greater than the matching degree threshold among the multiple rules as a candidate formal specification.

Wherein, multiple rules included in the rule base correspond to multiple formal specifications, specifically, the corresponding relationship may be one-to-one. The matching degree threshold can be customized by the user according to the actual situation, for example, 80% or 75%, etc., which is not specifically limited here.

Optionally, after calculating the matching degree between the code and the multiple rules included in the rule base, the multiple rules may also be sorted according to the calculation result, and then the form specification corresponding to the part of the multiple rules with a higher matching degree is used as a candidate form specification. In a specific implementation, the number of selected candidate form specifications can be customized according to the actual situation, for example, 5 or 10 can be customized, which is not specifically limited here.

In a specific implementation, multiple rules in the rule base and the formal specifications corresponding to the multiple rules can be obtained in advance according to various application scenarios for specialized rule designers.

Implement the above implementation method, use the formal specification corresponding to the code and/or code comment matching degree greater than the matching degree threshold in the rule base as the candidate formal specification, because the multiple rules in the rule base and the formal specification corresponding to the multiple rules can be found in advance by a special rule designer according to various application scenarios, therefore, the accuracy of the generated candidate formal specification can be guaranteed.

In a possible implementation manner, multiple candidate form specifications can be obtained by implementing code-based and/or code-based comments in the following manner: multiple candidate form specifications are obtained based on the intermediate representation corresponding to the code and/or structured comments of the code, wherein the above-mentioned structured comments of the code are obtained by structurally processing the comments of the code.

Since the intermediate representation corresponding to the code is easier to analyze than the code itself, and the obtained analysis results are usually more accurate, the structured comment of the code is easier to process than the comment of the code (referring to the comment before the structured processing), and the obtained processing result is usually more accurate. Therefore, implementing the above implementation method, based on the intermediate representation corresponding to the code and/or the structured comment of the code, multiple candidate form specifications can be obtained, which can improve the accuracy of the obtained candidate form specifications, and can make the multiple candidate form specifications finally recommended to users more accurate.

In a possible implementation manner, the acquisition of the code input by the user and/or the comment of the code may be specifically achieved in the following manner: the acquisition of the code input by the user and/or the comment of the code is triggered by a keyword.

In a possible implementation manner, the acquisition of the code and/or comment of the code input by the user can be triggered by keywords in the following manner: monitor whether the user has input a keyword, and obtain the code and/or comment of the code input by the user when it is detected that the user has input the keyword.

Implementing the above-mentioned implementation mode can only monitor the code and/or code annotations entered by the user after the keyword is entered, and then recommend candidate form specifications to the user based on the user-input code and/or code annotations. In this way, user experience can be optimized and computing resources can be saved.

In a possible implementation manner, obtaining the code and/or code annotations input by the user may also be achieved in the following manner: first, receiving the time range input by the user, and then obtaining the code and/or code annotations entered by the user within the time range.

Implementing the above-mentioned implementation method can only obtain the user-input code and/or code annotations within the user's expected time range, and then recommend candidate formal specifications to the user based on the user-input code and/or code annotations. In this way, user experience can be optimized and computing resources can be saved.

In a possible implementation manner, displaying multiple candidate form specifications to the user may specifically be implemented in the following manner: displaying the sorted multiple candidate form specifications to the user, wherein among the multiple candidate form specifications, the candidate form specification that best matches the code and/or code comments is ranked first.

Implement the above implementation method, and arrange the candidate form specification that best matches the code and/or code comments in the first place among the multiple candidate form specifications, so that the user can conveniently and quickly view the above-mentioned optimal candidate form specification from the multiple candidate form specifications, and can optimize user experience.

In a possible implementation manner, the method provided in the first aspect further includes the following steps: acquiring a target form specification selected by the user from multiple candidate form specifications, and then using the target form specification to verify the code to be verified.

Implementing the above implementation method and directly using the selected target form specification to verify the code to be verified can improve the verification efficiency of the code, thereby improving the development efficiency of the computer system.

In a second aspect, an apparatus for recommending formal specifications is provided, which includes: an acquisition module, configured to acquire codes input by users and/or annotations of the codes; a protocol recommendation module, configured to obtain multiple candidate formal specifications based on the code and/or the annotations of the code, and display the multiple candidate formal specifications to the user.

In a possible implementation, the specification recommendation module can specifically implement code and/or code-based comments to obtain multiple candidate formal specifications in the following manner: input the code and/or code comments into the formal specification generation model for identification, and use the formal specification output by the formal specification generation model as a candidate formal specification.

In a possible implementation, the specification recommendation module can specifically implement code and/or code-based annotations to obtain multiple candidate formal specifications in the following manner: first, calculate the matching degree between the code and/or code annotations and multiple rules included in the rule base, and then, according to the calculation result, use the formal specification corresponding to the rule whose matching degree is greater than the matching degree threshold among the multiple rules as a candidate formal specification.

In a possible implementation, the protocol recommendation module can specifically implement code-based and/or code-based annotations to obtain multiple candidate formal specifications: based on the corresponding intermediate representation of the code and/or code structured annotations, the candidate formal specifications are generated, wherein the above-mentioned structured annotations of the code are obtained by structurally processing the annotations of the code.

In a possible implementation manner, the acquisition module may specifically acquire the code and/or code annotation input by the user in the following manner: trigger the acquisition of the code and/or code annotation entered by the user through keywords.

In a possible implementation manner, the acquisition module can specifically realize the acquisition of the code and/or comment of the code input by the user through keywords in the following manner: monitor whether the user has entered a keyword, and obtain the code and/or comment of the code input by the user when it is detected that the user has input the keyword.

In a possible implementation manner, the obtaining module may also obtain the code and/or code annotations input by the user in the following manner: first, receive the time range input by the user, and then obtain the code and/or code annotations entered by the user within the time range.

In a possible implementation manner, the specification recommendation module may specifically display multiple candidate form specifications to the user in the following manner: display the sorted multiple candidate form specifications to the user, wherein among the multiple candidate form specifications, the candidate form specification that best matches the code and/or code comments is ranked first.

In a possible implementation manner, the device provided by the second aspect further includes a verification module. When the obtaining module is also used to obtain a target form specification selected by the user from multiple candidate form specifications, the verification module is used to use the target form specification to verify the code to be verified.

In a third aspect, a computing device is provided, and the computing device includes a processor and a memory; the processor is configured to execute instructions stored in the memory, so that the computing device implements the method provided in the first aspect or any possible implementation manner of the first aspect.

In a fourth aspect, a form specification recommendation system is provided, the system includes a terminal device and a server; the terminal device is used to obtain the code and/or code annotations input by the user; the terminal device is used to send the code and/or code annotations to the server; the server is used to obtain multiple candidate form specifications according to the code and/or code annotations, and send the multiple candidate form specifications to the terminal device; the terminal device is used to display multiple candidate form specifications to the user.

In a fifth aspect, there is provided a computer-readable storage medium, where instructions are stored in the computer-readable storage medium, and the instructions are used to implement the method provided in the above-mentioned first aspect or any possible implementation manner of the first aspect.

A sixth aspect provides a computer program product, including a computer program. When the computer program is read and executed by a computing device, the computing device executes the method provided in the first aspect or any possible implementation manner of the first aspect.

Description of drawings

Fig. 1 is a schematic diagram of an application scenario involved in the present application;

Fig. 2 is a schematic structural diagram of a formal specification recommendation system provided by the present application;

Fig. 3 is an interactive schematic diagram of a formal specification recommendation method provided by the present application;

Fig. 4A is a graphical user interface exemplarily shown in this application;

Fig. 4B is a graphical user interface exemplarily shown in this application;

Fig. 5 is an interactive schematic diagram of another formal specification recommendation method provided by the present application;

Fig. 6 is an exemplary graphical user interface provided by the present application;

FIG. 7 is a schematic structural diagram of a formal specification recommendation device provided by the present application;

FIG. 8 is a schematic structural diagram of a computing device provided by the present application.

Detailed ways

The technical solution provided by this application will be described below with reference to the accompanying drawings. In order to make the technical solution provided by this application clearer, before describing the technical solution provided by this application in detail, explanations of relevant terms are firstly made.

Formal specification: also known as formal specification or formal description, is a computer system model strictly described by a formal specification language or the properties that a computer system needs to satisfy. The former is a model specification, and the latter is a property specification. Like other modeling methods of computer systems, formal specification can use different/same formal specification language to describe computer system from different angles, which is the so-called multi-dimensional perspective specification method.

Formal specification language: refers to a language defined by strict recursive grammar rules, including model specification language, property specification language, etc. Among them, the model specification language uses mathematical structure to describe the state change or event trajectory of the system, which directly defines the structure, functional behavior and even non-functional behavior of the described system model; the property specification language is based on the program logic system, and describes a set of properties through logical formulas to define the desired system behavior. The property specification does not directly define the specific behavior of the system.

Intermediate representation (IR): In the field of compilation, the compiler is usually divided into a front-end and a back-end. Among them, the front-end will perform lexical analysis, syntax analysis, and semantic analysis on the input program, and then generate an intermediate representation, which is the intermediate representation. The back-end will optimize the intermediate representation and then generate the object code. The intermediate representation and the source code are semantically equivalent. Because the source code usually includes various comments, it is very inconvenient to analyze. When performing code analysis, it is usually not directly analyzed based on the source code, but will be converted into an intermediate representation for analysis, so that the analysis results obtained will be more accurate. At present, the commonly used intermediate representations include three address code (three address code, abbreviated as TAC or 3AC), control flow graph (control flow graph, CFG) and static single assignment (static single assignment, SSA), among them, SSA can be divided into minimum SSA, pruned SSA, maximum SSA and so on.

Structured data: Refers to data that can be described by numbers or unified data models, with strict length and format. Compared with unstructured data, structured data has the advantages of easy storage, management and processing, and the processing results obtained are usually more accurate.

Neural networks (neural networks): It can be composed of neural units (also called neurons). A neural unit can refer to an operation unit that takes the variable x _s and the intercept b as input, and the output of the operation unit can be:

Among them, s=1, 2, ... n, n is a natural number greater than 1, W _s is the weight of x _s , and b is the bias of the neuron unit. f is the activation function of the neural unit, which is used to introduce nonlinear characteristics into the neural network to convert the input signal in the neural unit into an output signal. The output signal of this activation function can be used as the input of the next convolutional layer. The activation function may be a sigmoid function or other functions, which are not limited here. A neural network is a network formed by connecting many of the above-mentioned single neural units, that is, the output of one neural unit can be the input of another neural unit. The input of each neural unit can be connected with the local receptive field of the previous layer to extract the features of the local receptive field. The local receptive field can be an area composed of several neural units. There are many kinds of neural networks. Common neural networks include random forest, support vector machine (SVM), graph neural networks (GNN), convolutional neural networks (CNN), etc.

Model training: refers to the process of training a neural network to obtain a trained model that can be used to complete a specific task. Training neural network refers to the use of existing data to make the neural network fit the law of the existing data through a certain method, so as to determine the parameters in the neural network. Training a neural network requires the use of a data set. According to whether the data in the data set is labeled (that is, whether the data has a specific type or name), the training of the neural network can be divided into supervised training and unsupervised training.

The following briefly describes the application scenarios involved in the embodiments of the present application.

The development of computer science mainly involves the development of hardware and software, and one of the core issues in the development of software/hardware is how to ensure that they are reliable and safe. Nowadays, the hardware performance is getting higher and higher, the computing speed is getting faster and faster, the system structure is becoming more and more complex, and the function of software is becoming more and more powerful and complex. How to develop a reliable computer (software/hardware) system has become a huge challenge for the development of computer science, especially now that computer systems have been widely used in many safety systems related to the national economy and people's livelihood, such as high-speed train control systems, aerospace control systems, nuclear reactor control systems, medical equipment systems, etc. Any error in these systems may lead to catastrophic consequences.

In practical applications, a reliable and secure computer system is usually designed and developed based on formal methods. When designing and developing computer systems based on formal methods, developers usually need to manually write a large number of formal specifications on terminal devices, as shown in Figure 1. In a specific implementation, the terminal device used by the developer may be a personal computer, a tablet computer, a mobile notebook, etc., which have the function of writing a formal specification. FIG. 1 illustrates that the terminal device 100 is a personal computer as an example, but this application does not specifically limit it.

The above-mentioned method of manually writing the formal specification by the developer consumes a lot of manpower and time, which will lead to the problems of low efficiency of obtaining the formal specification and low efficiency of computer system development.

In order to solve the above problems, the present application provides a formal specification recommendation method, device and system, which can automatically obtain multiple candidate formal specifications based on the code and/or code comments that the user has input during the process of manually writing the code and/or formal specification to be verified by the user (that is, the above-mentioned developer), and display the obtained multiple candidate formal specifications to the user. The multiple candidate formal specifications can provide good reference for the user. The specification is used as the target form specification, and the selected target form specification is used to verify the code to be verified. The user does not need to write the entire form specification, which can save the user's energy and time, and can improve the acquisition efficiency of the form specification and the development efficiency of the computer system.

In order to facilitate a clearer understanding of the formal specification recommendation method, device, and system provided in this application, detailed descriptions will be made below in conjunction with the corresponding drawings.

First, please refer to Fig. 2, Fig. 2 is a schematic structural diagram of a formal specification recommendation system provided by the present application, the system may include a terminal device 100 and a server 300, and the terminal device 100 and the server 300 are connected through a network 200, and the network 200 may be a wide area network, a local area network, a point-to-point connection, etc., or any combination thereof.

The server 300 can provide users with shared resource services, and the shared resource services can include format specification recommendation services, object storage services (object storage service, OBS), cloud backup services (cloud backup and recovery, CBR), data management services (data admin service, DAS), etc. This application does not limit the types of shared resource services that the server 300 can provide.

Exemplarily, the server 300 may include a plurality of protocol recommendation nodes, and each protocol recommendation node specifically includes hardware, a virtualization service, and a protocol recommendation server application.

Hardware includes computing resources, storage resources, and network resources. Computing resources can adopt heterogeneous computing architectures, for example, central processing unit (central processing unit, CPU) + graphics processing unit (graphics processing unit, GPU) architecture, CPU+AI chip, CPU+GPU+AI chip architecture, etc., are not specifically limited here. Storage resources may include memory, among others. Here, computing resources may be divided into multiple computing unit resources, storage resources may be divided into multiple storage unit resources, and network resources may be divided into multiple network unit resources. Therefore, the image processing platform can be freely combined on the basis of unit resources according to the user's resource requirements, so as to provide resources according to the user's needs. For example, computing resources can be divided into computing unit resources of 5u, and storage resources can be divided into storage unit resources of 10G, then the combination of computing resources and storage resources can be, 5u+10G, 5u+20G, 5u+30u,..., 10u+10G, 10u+20G, 10u+30u,....

Virtualization service is a service that builds the resources of multiple physical hosts into a unified resource pool through virtualization technology, and flexibly isolates mutually independent resources according to user needs to run user application codes. The virtualization service may include a virtual machine (virtual machine, VM) service, a bare metal server (bare metal server, BMS) service, and a container (container) service. Wherein, the VM service may be a service in which virtual machine (virtual machine, VM) resource pools are virtualized on multiple physical hosts through a virtualization technology to provide VMs for users on demand. BMS service is a service that virtualizes BMS resource pools on multiple physical hosts to provide users with BMS on demand. Container Service is a service that virtualizes container resource pools on multiple physical hosts to provide users with containers on demand. VM is a simulated virtual computer, that is, a logical computer. BMS is an elastically scalable high-performance computing service. Its computing performance is the same as that of traditional physical machines, and it has the characteristics of safe physical isolation. Container is a kernel virtualization technology that can provide lightweight virtualization to achieve the purpose of isolating user space, processes and resources. It should be understood that the VM service, BMS service, and container service in the above virtualization services are only specific examples. In practical applications, the virtualization service may also be other lightweight or heavyweight virtualization services, which are not specifically limited here.

The protocol recommendation server application can be used to receive the user-input code and/or code comments sent by the terminal device 100 when the user manually writes the code and/or the formal specification to be verified on the terminal device 100, and then automatically obtain a plurality of candidate formal specifications based on the user-input code and/or code annotations, and return the generated multiple candidate formal specifications to the terminal device 100, and be displayed to the user by the terminal device 100. When the user views multiple candidate formal specifications, he can select the target formal specification from the multiple candidate formal specifications according to his own needs.

In a specific implementation, the server 300 can provide an interface to the user, and the protocol recommendation server application can receive the code and/or code comments sent by the terminal device 100 through the interface, and the protocol recommendation server can send the multiple candidate format protocols to the terminal device 100 through the interface after obtaining multiple candidate format protocols. Specifically, the interface may be a console (console) or an application program interface (application program interface, API) of the server 300, and the console may specifically be an application program or a web page for the user to interact with the server 300. It should be understood that the above examples are for illustration, and the present application does not make specific limitations.

Next, with reference to the interactive schematic diagram of a form specification recommendation method provided by the present application shown in FIG. 3 , the process of implementing form specification recommendation by the form specification recommendation system shown in FIG. 2 is described in more detail. As shown in Figure 3, the formal specification recommendation method provided by this application may include the following steps:

S301: The terminal device 100 monitors the code writing operation performed by the user in real time.

Among them, the code written by the user in the code writing operation can be the code and/or the formal specification to be verified by the formal specification. In the specific implementation, the user can use any programming language (such as java, C, C++, etc.) For example, codes for displaying time or calculating running kilometers, etc.), the formal specification written by the user may be a data model specification, a data function specification, or an interactive communication protocol specification, etc., which are not specifically limited here.

It can be understood that no matter whether the user is writing the code to be verified by the formal specification or writing the formal specification, the user needs to input characters one by one during the writing process, and the user can also modify the input characters during the process of inputting characters, so as to obtain the final code or formal specification to be verified by the formal specification.

S302: The terminal device 100 acquires the code and/or the comment of the code input by the user.

Wherein, the code refers to the characters that the user has input during the process of writing the code to be verified by the formal specification and/or the formal specification.

The exemplary graphical user interface shown in Figure 4A and Figure 4B, the code shown in Figure 4A is the complete and correct form specification that the user wants to obtain, and the code "amount" shown in Figure 4B is the code that the user has entered when writing the form specification shown in Figure 4A.

S303: The terminal device 100 sends the code and/or the comment of the code input by the user to the server 300 .

S304: The server 300 obtains a plurality of candidate form specifications based on the code and/or the comment of the code sent by the terminal device 100.

Next, the process of obtaining multiple candidate formal specifications based on codes and/or code annotations is described in detail by taking the example of obtaining multiple candidate formal specifications based on codes.

In a specific embodiment of the present application, the server 300 can refer to one or more combinations of the following methods to obtain multiple candidate form specifications based on the code:

Mode 1, the server 300 may pre-store a rule base, and multiple rules in the rule base correspond to multiple formal specifications. When the server 300 receives the code sent by the terminal device 100, the server 300 may acquire the degree of matching between the code and the multiple rules included in the rule base, and then use the formal specification corresponding to the rule whose matching degree is greater than the first matching degree threshold among the multiple rules as a candidate formal specification.

In a specific implementation, the correspondence between multiple rules and multiple formal specifications may be a one-to-one correspondence. The first matching degree threshold can be customized by the user according to the actual situation, for example, 80% or 75%, etc., which is not specifically limited here.

For example, assuming that the code input by the user is code 1, the rule base includes rules A to rule E, where rule A corresponds to form a, rule B corresponds to form b, rule C corresponds to form c, rule D corresponds to form d, rule E corresponds to form e, and the first matching degree threshold is 50%. , the formal specification d corresponding to rule D and the formal specification e corresponding to rule E are determined as candidate formal specifications.

Optionally, after calculating the matching degree between the code and the multiple rules included in the rule base, the multiple rules may also be sorted according to the calculation result, and then the form specification corresponding to the part of the multiple rules with a higher matching degree is used as a candidate form specification. In a specific implementation, the number of selected candidate form specifications can be customized according to the actual situation, for example, 5 or 10 can be customized, which is not specifically limited here.

Continuing to use the code input by the user as code 1, the rule base includes rules A to rule E, where rule A corresponds to form a, rule B corresponds to form b, rule C corresponds to form c, rule D corresponds to form d, rule E corresponds to form e, and the matching degree threshold is 50%. By sorting rules A to E, the order of multiple rules can be obtained: rule E, rule D, rule B, rule A, and rule C, so two candidate formal specifications can be obtained: formal specification d corresponding to rule D and formal specification e corresponding to rule E.

In specific implementation, multiple rules and the formal specifications corresponding to the multiple rules can be obtained in advance by specialized rule designers according to various application scenarios. In this way, the formal specifications corresponding to the rules in the rule base whose code matching degree is greater than the first matching degree threshold are determined as candidate formal specifications, or the formal specifications corresponding to some rules with a relatively high matching degree are determined as candidate formal specifications, which can ensure the accuracy of the determined candidate formal specifications.

Method 2: When the server 300 receives the code sent by the terminal device 100, it inputs the code into a pre-trained formal specification generation model with high recognition accuracy for recognition, and then uses the formal specification output by the formal specification generation model as a candidate formal specification. Among them, the formal specification generation model can be trained using a sample set including a large number of known codes.

In a specific implementation, after the server 300 inputs the code into the formal specification generation model, the formal specification generation model can obtain multiple formal specifications and the confidences of the multiple formal specifications. The formal specification generation model can directly output the formal specifications with a confidence greater than the confidence threshold among the multiple formal specifications as candidate formal specifications, or sort them according to the degree of confidence, and output some formal specifications with higher confidence as candidate formal specifications, which are not specifically limited here. Both the confidence threshold and the number of output candidate form reductions can be customized by the user according to the actual situation, and are not specifically limited here.

In the specific embodiment of this application, the formal specification generation model can be expressed as:

y=f(x)

Among them, y is the formal specification, x is the code, and f() is the mapping relationship between the code and the formal specification.

The training process of the formal reduction generative model is described in detail below.

In a specific implementation, a machine learning model can be trained using a sample set including a large number of known codes to obtain a formal specification generation model. Among them, a large number of known codes included in the sample set can be manually written by the user or obtained from the Internet, which is not specifically limited here; machine learning models can include but are not limited to decision trees, support vector machines, deep learning, etc., deep learning models can use common models that generate codes from natural language processing, such as bidirectional encoder representations from transformers based on transformers (BERT for short), Seq2Seq, transformers, etc., which are not specifically limited in this application; The manner in which the sample set is used to train the machine learning model to obtain the form reduction model can be supervised training or unsupervised training, which is not specifically limited in this application.

Take the example of using a sample set to perform supervised training on a machine learning model to obtain a formal specification generation model. When the training method is supervised training, a large number of known codes included in the sample set can be used as input data samples. In addition, the sample set can also include known formal specifications corresponding to each code in a large number of known codes. A large number of known formal specifications can be used as output data samples. A large number of known formal specifications can be manually written by users or obtained from the Internet. No specific restrictions are made here.

Specifically, the i-th known code among a large number of known codes can be used as an input data sample S _i , and the known form specification corresponding to the i-th known code can be used as an output data sample s _i . It can be understood that through the above combination, a large number of input data samples and a large number of output data samples can be obtained, and there is a one-to-one correspondence between the large number of input data samples and the large number of output data samples.

After obtaining a large number of input data samples and a large number of output data samples, a large number of input data samples can be used as the input of the machine learning model in turn, and the output data samples corresponding to each input data sample can be used as a reference for the output value of the machine learning model, and the loss function is used to calculate the loss value between the output value of the machine learning model and the above data output samples, and then adjust the parameters of the machine learning model according to the loss value. In a specific implementation, the machine learning model can be iteratively trained by using the above-mentioned large number of input data samples and a large number of output data samples, and the parameters of the machine learning model can be continuously adjusted until the machine learning model can accurately output the same output value as the output data sample corresponding to the input data sample according to the input data sample, so as to obtain a trained formal specification generation model.

Mode 3, the server 300 may pre-store a one-to-one correspondence between a large number of codes and a large number of formal specifications, wherein each formal specification is used to verify its corresponding code. When the server 300 receives the code sent by the terminal device 100, the server 300 may obtain the degree of matching between the code and a large number of pre-stored codes, and then use the formal specification corresponding to the code whose matching degree is greater than the second matching degree threshold among the large number of codes as a candidate form specification. Wherein, the second matching degree threshold can be customized by the user according to the actual situation, for example, 80% or 75% can be customized, which is not specifically limited here.

It should be understood that the process of obtaining candidate formal specifications based on code comments and the process of obtaining candidate formal specifications based on codes and code comments are similar to the above-mentioned process of obtaining candidate formal specifications based on codes. For details, please refer to the relevant description above. For the sake of brevity, details will not be repeated here.

It should be understood that the above-listed implementation methods for obtaining candidate formal specifications are only examples. In specific implementations, other methods that can obtain candidate formal specifications based on code and/or code comments also belong to the scope of protection of this application.

S305: The server 300 sends multiple candidate form protocols to the terminal device 100.

In a possible embodiment, what the server 300 sends to the terminal device 100 may be a plurality of sorted candidate form specifications, wherein among the multiple candidate form specifications, the candidate form specification that best matches the code and/or code comments is ranked first, so that the user can conveniently and quickly view the optimal candidate form specification from the multiple candidate form specifications, and optimize user experience.

Optionally, after sorting the multiple candidate form specifications, the server 300 may also send to the terminal device 100 only the candidate form specification that best matches the code and/or the comment of the code among the multiple candidate form specifications, or only send to the terminal device 100 a part of the candidate form specifications that are ranked higher among the multiple candidate form specifications, which is not specifically limited here.

In a specific embodiment of the present application, the server 300 can refer to any one of the following ways to sort multiple candidate form specifications:

Mode 1', in the case that multiple candidate formal specifications are obtained by adopting the above method 1 in step S304, the multiple candidate formal specifications can be sorted based on the degree of matching between the rules corresponding to the multiple candidate formal specifications and the code and/or code comments.

For example, multiple candidate form specifications may be sorted in descending order of matching degree, and the candidate form specification corresponding to the highest matching degree is determined to be the optimal candidate form specification and arranged first.

Mode 2', in the case that multiple candidate formal specifications are obtained by adopting the above-mentioned mode 2 in step S304, the multiple candidate formal specifications may be sorted based on the confidence levels corresponding to the multiple candidate formal specifications.

For example, multiple candidate form specifications can be sorted in descending order of confidence, and the candidate form specification with the highest corresponding confidence is determined to be the best candidate form specification and arranged first.

Mode 3', in the case that multiple candidate form protocols are obtained by adopting mode 3 in the above step S304, the multiple candidate form protocols can be sorted based on the degree of matching between the codes and/or code annotations pre-stored by the server 300 corresponding to the multiple candidate format protocols and the codes and/or code annotations received by the server 300 and sent by the terminal device 100.

It should be understood that the above-listed several ways to implement the sorting of the candidate form protocols are only examples, and other ways that can implement the sorting of the candidate form protocols also belong to the protection scope of the present application, which is not specifically limited in the present application.

Optionally, when the server 300 sends multiple candidate form specifications to the terminal device 100 , it may also send the matching degrees or confidence levels corresponding to the multiple candidate form specifications to the terminal device 100 .

S306: The terminal device 100 displays multiple candidate form specifications to the user.

The multiple candidate form specifications displayed to the user by the terminal device 100 may be the multiple candidate form specifications after sorting. If the terminal device 100 also receives the matching degrees or confidence levels corresponding to the multiple candidate form specifications sent by the server 300, when the terminal device 100 displays the multiple candidate form specifications to the user, it may display the matching degrees or confidence levels corresponding to the multiple candidate form specifications to the user. In this way, it is convenient for the user to quickly and conveniently view the optimal candidate form specification from the multiple candidate form specifications, and optimize user experience.

S307: The terminal device 100 acquires a target format protocol selected by the user from multiple candidate format protocols.

S308: The terminal device 100 uses the target form protocol to verify the code to be verified.

From the above introduction to intermediate representation and structured data, it can be seen that the intermediate representation corresponding to the code is easier to analyze than the code itself, and the analysis results obtained are usually more accurate. Compared with unstructured data, structured data is easier to store, manage and process. Therefore, in the specific embodiment of this application, the formal protocol recommendation system shown in Figure 2 can also refer to the interactive schematic diagram shown in Figure 5 to implement the formal protocol recommendation:

S501: The terminal device 100 monitors the code writing operation performed by the user in real time.

S502: The terminal device 100 acquires the code and/or the comment of the code input by the user.

S503: The terminal device 100 sends the code and/or the comment of the code to the server 300 .

S504: The server 300 acquires an intermediate representation corresponding to the code and/or a structured annotation of the code.

S505: The server 300 obtains multiple candidate form specifications based on the intermediate representation corresponding to the code and/or the structured annotation of the code.

S506: The server 300 sends multiple candidate form protocols to the terminal device 100.

S507: The terminal device 100 displays multiple candidate form specifications to the user.

S508: The terminal device 100 acquires a target format protocol selected by the user from multiple candidate format protocols.

S509: The terminal device 100 uses the target form protocol to verify the code to be verified.

The specific implementation process of step S504 can be as follows: the server 300 can obtain the intermediate representation corresponding to the code from the front end of the compiler, and the form of the intermediate representation can be a three-address code, a control flow chart, or a static single assignment, etc., which are not specifically limited here; the server 300 can perform structural processing on the code comment, so as to obtain the structured comment of the code. Optionally, when the server 300 performs structural processing on the comments of the code, it can also clean the comments, such as deleting special symbols included in the comments, such as "*", "#", etc., or adjusting the order of the words included in the comments, etc., which are not specifically limited here.

It can be seen that steps S501, S502, S503, S506, S507, S508, and S509 are the same as steps S301, S302, S303, S305, S306, S307, and S308. For details, please refer to the relevant descriptions of steps S301, S302, S303, S305, S306, S307, and S308 above. , will not be repeated here.

The specific implementation process of step S505 is similar to the specific implementation process of the above-mentioned step S304, and reference may be made to the relevant description of the above-mentioned step S304, which will not be repeated here.

Optionally, after the terminal device 100 obtains the code and/or the annotation of the code input by the user, it may also directly obtain the intermediate representation corresponding to the code and/or the structured annotation of the code based on the code and/or the annotation of the code, and then send the intermediate representation corresponding to the code and/or the structured annotation of the code to the server 300.

It can be understood that, compared with the code itself, the intermediate representation corresponding to the code is easier to analyze, and the obtained analysis results are usually more accurate. Compared with the comment of the code (referring to the comment before the structured processing), the structured comment of the code is easier to process, and the processing result obtained is usually more accurate. This makes the multiple candidate form specifications recommended to users more accurate.

In a specific embodiment of the present application, the terminal device 100 can also trigger the execution of step S301 or S501 through a keyword. Specifically, the terminal device 100 can monitor whether the user has input a keyword (such as requires or prey, etc.), and when it is detected that the user has input a keyword, execute step S301 or S501. Wherein, the position of the keyword input by the user is the position where the user expects to start the form specification recommendation, so that the device (terminal device 100 shown in FIG. 1 ) or system (the form specification recommendation system shown in FIG. 2 ) applying the form specification recommendation method provided by this application can only monitor the code writing operation performed by the user after inputting the keyword, and not monitor the code writing operation performed by the user before inputting the keyword, which can optimize user experience and save computing resources.

More specifically, the terminal device 100 can monitor the content input by the user in real time, and then match the content input by the user with multiple keywords in the keyword database to determine whether there is a keyword matching the content input by the user in the keyword database.

Optionally, the terminal device 100 may also receive a time range input by the user. This time range is the time range in which the user will perform code writing operations and is also the time when the user expects to recommend form specifications. In this way, the device or system that applies the form specification recommendation method provided by this application can only monitor the code writing operations performed by the user within this time range, and perform form specification recommendations based on the code writing operations within this time range. It does not monitor the code writing operations performed by the user outside the time range, which can optimize user experience and save computing resources.

In a specific implementation, before the server 300 sends the multiple candidate protocols to the terminal device 100, or before the terminal device 100 displays the multiple candidate protocols to the user, it may further screen the multiple candidate protocols to filter out some obviously worthless protocols, for example, filter out the protocols with obvious errors (such as grammatical errors or semantic errors, etc.), so that the multiple candidate protocols finally recommended to the user can be more accurate.

In summary, the formal specification recommendation system and formal specification recommendation method provided by this application can monitor the code writing operation performed by the user in real time, obtain the code and/or code comments entered by the user, and then automatically obtain multiple candidate formal specifications according to the code and/or code comments, and finally recommend the obtained multiple candidate formal specifications to the user. , and use the selected target formal specification to verify the code to be verified, the user does not need to write the entire formal specification, which can save the user's energy and time, and can improve the efficiency of formal specification acquisition and computer system development.

In order to facilitate the understanding of the beneficial effects of the solution provided by the embodiment of the present application, some exemplary graphical user interfaces during the above steps S301-S308 and steps S501-S509 are introduced below. Please refer to FIG. 6 . FIG. 6 is an exemplary graphical user interface 600 provided by the present application. It can be seen from FIG. 6 that the interface 600 may include a code input area 610 .

Among them, the code input area 610 can provide the user with an interface for inputting codes and code comments. The user can input keywords and perform code writing operations in this area. For example, "requires" shown in FIG.

Optionally, the interface 600 can also provide the user with a "Add Code" control 620, as shown in FIG.

After the device or system applying the formal protocol recommendation method provided by this application detects that the user has input the keyword "requires", it can execute the above steps S301-S304 or execute the above-mentioned steps S501-S505 to obtain multiple candidate formal protocols, sort the multiple candidate formal protocols, and then display the sorting results to the user. The protocol recommendation pop-up window 630 shown in FIG. c, may be the sorted candidate formal specifications recommended to the user by a device or a system applying the formal specification recommendation method provided in this application. If the user determines that there is a candidate format specification that meets his needs in the pop-up window 630, he can click on the format specification, such as clicking on the format specification d, and then click on the "OK" control 6301 displayed in the pop-up window 630 to determine the clicked candidate format specification as the target format specification.

It should be understood that FIG. 6 is used for illustration, and is not specifically limited in the present application.

According to the interface 600 shown in FIG. 6 , the method for recommending formal specifications provided by this application, for the user, after inputting code and/or code comments on the interface shown in interface 600, can obtain multiple sorted candidate formal specifications on the display interface shown in interface 600, and the sorting results can provide good reference for users. .

It should be understood that the protocol recommendation method provided by the present application shown in FIG. 3 and FIG. 5 can be applied to the terminal device 100 shown in FIG. 1 in addition to the protocol recommendation system shown in FIG. 2 , that is, the steps performed by the server 300 in the embodiment shown in FIG. 3 and FIG. and using the selected target formal specification to verify the code to be verified, the user does not need to write the entire formal specification, which can save the user's energy and time, and can improve the efficiency of formal specification acquisition and computer system development.

Next, the form specification recommendation device provided by this application is introduced. This device can be applied to the terminal device 100 shown in FIG. 1, or to the form specification recommendation system shown in FIG. 2, and can be applied to the server 300 shown in FIG. It should be understood that the unit modules inside the formal specification recommendation device can also be divided into multiple types. Each module can be a software module, or a hardware module, or partly a software module and partly a hardware module, which is not limited in this application. When the form specification recommendation device is applied to the server 300 shown in FIG. 2, and the form specification recommendation device includes a plurality of unit modules, each module in the plurality of unit modules can be deployed on the same server 300, or can be deployed on different servers 300, which is not specifically limited in this application.

Referring to FIG. 7 , FIG. 7 is a schematic structural diagram of an apparatus 700 for form protocol recommendation in this application. The apparatus 700 includes: an acquisition module 710 and a protocol recommendation module 720 .

Obtaining module 710, configured to obtain codes and/or comments of codes input by the user.

The protocol recommending module 720 is configured to obtain multiple candidate formal specifications based on the code and/or code comments input by the user, and display the multiple candidate formal specifications to the user.

In a possible implementation, the protocol recommendation module 720 can specifically obtain multiple candidate formal specifications based on the code and/or code comments input by the user in the following manner: identify the code and/or code comments as input to the formal specification generation model, and use the formal specification output by the formal specification generation model as a candidate formal specification.

In a possible implementation, the protocol recommendation module 720 can specifically obtain multiple candidate formal specifications based on user-input code and/or code annotations in the following manner: first, obtain the matching degree between the code and/or code annotations and multiple rules included in the rule base, and then use the formal specification corresponding to the rule whose matching degree is greater than the matching degree threshold among the multiple rules as a candidate formal specification.

In a possible implementation manner, the protocol recommending module 720 can specifically obtain multiple candidate formal specifications based on user-input codes and/or code annotations in the following manner: generate multiple candidate formal specifications based on the intermediate representation corresponding to the codes and/or code structured annotations, wherein the above-mentioned structured annotations of codes are obtained by performing structural processing on the code annotations.

In a possible implementation manner, the acquiring module 710 may specifically acquire the code and/or comment of the code input by the user in the following manner: triggering acquisition of the code and/or comment of the code input by the user through keywords.

In a possible implementation manner, the acquisition module 710 can specifically implement the acquisition of the code and/or code annotations entered by the user through keywords in the following manner: monitor whether the user has entered a keyword, and obtain the code and/or code annotations entered by the user when it is detected that the user has entered the keyword. The position of the above-mentioned keywords is the position where the user expects to start the form specification recommendation, so that the device or system that applies the form specification recommendation method provided by this application can only monitor the code writing operation performed by the user after inputting the keyword, and not monitor the code writing operation performed by the user before inputting the keyword, which can optimize user experience and save computing resources.

In a possible implementation, the acquisition module 710 can also obtain the code and/or code comments input by the user in the following manner: the time range for receiving user input, the time range is the time range for the user to write codes, and the time when the user expects to recommend formal specifications, so that the device or system that applies the formal specification recommendation method provided by this application can only monitor the code writing operations performed by the user within this time range, and perform formal specification recommendations based on the code writing operations within this time range, and not monitor the codes performed by users outside this time range Writing operations can optimize user experience and save computing resources.

In a possible implementation, the protocol recommendation module 720, when displaying multiple candidate protocols to the user, may display the sorted multiple candidate protocols to the user, wherein among the multiple candidate protocols, the candidate protocol that best matches the code and/or code comments is ranked first, so that the user can conveniently and quickly view the optimal candidate protocol from the multiple candidate protocols.

In a possible implementation, as shown in FIG. 7 , the device 700 further includes a verification module 730. When the acquisition module 710 is also used to obtain the target form specification selected by the user from multiple candidate form specifications, the verification module 730 is used to use the target form specification to verify the code to be verified. Specifically, for the implementation of various operations performed by the formal specification recommending apparatus 700, reference may be made to the descriptions in the related content in the above-mentioned formal specification recommending method embodiment, and for the sake of brevity, details are not repeated here.

To sum up, the formal specification recommendation device provided by this application (the device 700 shown in Figure 7 ) can automatically generate multiple candidate formal specifications according to the code and/or code annotations by obtaining the code and/or code comments input by the user, and finally recommend the generated multiple candidate formal specifications to the user. The multiple candidate formal specifications can provide users with good references. The target form specification is verified by the code to be verified, and the user does not need to write the entire form specification, which can save the user's energy and time, and can improve the efficiency of obtaining the form specification and the development efficiency of the computer system.

Referring to FIG. 8, FIG. 8 is a schematic structural diagram of a computing device 800 provided by the present application. As shown in FIG.

The processor 810 can read the program codes (including instructions) stored in the memory 820, and execute the program codes stored in the memory 820, so that the computing device 800 executes the steps in the formal specification recommendation method provided by the above method embodiments, or makes the computing device 800 deploy the formal specification recommendation apparatus 700.

The processor 810 may have multiple specific implementation forms, such as a central processing unit (central processing unit, CPU), or a combination of a CPU and a hardware chip. The aforementioned hardware chip may be an application-specific integrated circuit (application-specific integrated circuit, ASIC), a programmable logic device (programmable logic device, PLD) or a combination thereof. The above-mentioned PLD may be a complex programmable logic device (complex programmable logic device, CPLD), a field-programmable gate array (field-programmable gate array, FPGA), a general array logic (generic array logic, GAL) or any combination thereof. Processor 810 executes various types of digitally stored instructions, such as software or firmware programs stored in memory 820, which enable computing device 800 to provide various services.

The memory 820 is used to store program codes, which are controlled and executed by the processor 810, so as to execute the processing steps in any one of the above-mentioned embodiments in FIG. 3 and FIG. 5 . The program code may include one or more software modules, and the one or more software modules may be the software modules provided in the embodiment of FIG. 7 , such as an acquisition module 710 , a protocol recommendation module 720 and a verification module 730 .

Memory 820 can comprise volatile memory (volatile memory), such as random access memory (random access memory, RAM); Memory 820 can also comprise nonvolatile memory (non-volatile memory), such as read-only memory (read-only memory, ROM), flash memory (flash memory), hard disk (hard disk drive, HDD) or solid-state drive (solid-state drive) ive, SSD); the memory 820 may also include a combination of the above types.

The communication interface 830 can be a wired interface (such as an Ethernet interface, an optical fiber interface, other types of interfaces (such as an infiniBand interface)) or a wireless interface (such as a cellular network interface or using a wireless local area network interface) for communicating with other computing devices or devices. The communication interface 830 can adopt a protocol family above the transmission control protocol/internet protocol (transmission control protocol/internet protocol, TCP/IP), for example, a remote function call (remote function call, RFC) protocol, a simple object access protocol (simple object access protocol, SOAP) protocol, a simple network management protocol (simple network management protocol, SNMP) protocol , common object request broker architecture (common object request broker architecture, CORBA) protocol and distributed protocol, etc.

The bus 840 can be a peripheral component interconnect express (PCIe) bus, or an extended industry standard architecture (EISA) bus, a unified bus (Ubus or UB), a computer fast link (compute express link, CXL), or a cache coherent interconnect for accelerator s, CCIX) etc. The bus 840 can be divided into an address bus, a data bus, a control bus, and the like. In addition to the data bus, the bus 840 may also include a power bus, a control bus, and a status signal bus. However, for clarity of illustration, the various buses are labeled as bus 840 in the figure. For ease of representation, only one thick line is used in FIG. 8 , but it does not mean that there is only one bus or one type of bus.

The above-mentioned computing device 800 is used to execute the form specification recommendation method provided by this application as shown in FIG. 3 or FIG. 5 , which belongs to the same concept as the above-mentioned method embodiment. For the specific implementation process, refer to the above-mentioned method embodiment for details, and will not be repeated here. It should be understood that computing device 800 is only an example provided by the embodiment of the present application, and computing device 800 may have more or fewer components than those shown in FIG. 8 , may combine two or more components, or may have different configurations of components.

The present application also provides a computer-readable storage medium, and the computer-readable storage medium stores instructions, and when the instructions are executed, some or all steps of the formal specification recommendation method described in the above-mentioned embodiments can be implemented.

The present application also provides a computer program product. When the computer program product is read and executed by a computer, it can realize some or all steps of the formal specification recommendation method described in the above method embodiments.

In the above-mentioned embodiments, the descriptions of each embodiment have their own emphases, and for parts not described in detail in a certain embodiment, reference may be made to relevant descriptions of other embodiments.

In the above-mentioned embodiments, all or part may be implemented by software, hardware or any combination thereof. When implemented using software, it may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the processes or functions according to the embodiments of the present application will be generated in whole or in part. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line) or wireless (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that can be accessed by a computer, or a data storage device such as a server or a data center integrated with one or more available media. The available medium may be a magnetic medium (such as a floppy disk, a hard disk, or a magnetic tape), an optical medium, or a semiconductor medium.

The foregoing is only a specific implementation manner of the present application. Those skilled in the art may conceive changes or substitutions based on the specific implementation methods provided in this application, and all of them shall fall within the protection scope of this application.

Claims (19)

1. A form specification recommendation method, characterized in that the method includes:

   Get user-entered code and/or code comments;

   Based on the code and/or annotations of the code, a plurality of candidate formal specifications is obtained, and the plurality of candidate formal specifications are displayed to the user.
2. The method according to claim 1, wherein said code and/or code annotations are used to obtain a plurality of candidate form specifications, including:

   Inputting the code and/or comments of the code into a formal specification generating model for identification, and using the formal specification output by the formal specification generating model as a candidate formal specification.
3. The method according to claim 1, characterized in that, said based on said code and/or code annotations, obtaining a candidate form specification, comprising:

   Based on the matching degree between the code and/or the comment of the code and the multiple rules included in the rule base, the formal specification corresponding to the rule whose matching degree is greater than the matching degree threshold among the multiple rules is used as a candidate formal specification.
4. The method according to claim 1, wherein said code and/or code annotations are used to obtain a plurality of candidate form specifications, including:

   Multiple candidate form specifications are obtained based on the intermediate representation corresponding to the code and/or the structural annotation of the code, wherein the structural annotation of the code is obtained by performing structural processing on the annotation of the code.
5. The method according to any one of claims 1 to 4, characterized in that said acquiring user-input codes and/or code comments includes:

   Key words are used to trigger acquisition of the code and/or annotations of the code input by the user.
6. The method according to claim 5, characterized in that, the acquisition of the code input by the user and/or code annotations triggered by keywords includes:

   When the user inputs a keyword, acquire the code and/or the comment of the code input by the user.
7. The method according to any one of claims 1 to 6, wherein the displaying the plurality of candidate form specifications to the user comprises:

   Displaying the sorted plurality of candidate formal specifications to the user, wherein, among the plurality of candidate formal specifications, the candidate formal specifications that best match the code and/or the annotations of the code are ranked first.
8. The method according to any one of claims 1 to 7, further comprising:

   acquiring a target form specification selected by the user from the plurality of candidate form specifications;

   The code to be verified is verified using the target form specification.
9. A form specification recommendation device, characterized in that the device comprises:

   Get module, used to get user input code and/or code comments;

   The protocol recommendation module is used to obtain multiple candidate form specifications based on the code and/or code annotations, and display the multiple candidate form specifications to the user.
10. The device according to claim 9, wherein the protocol recommendation module is specifically used for:

    Inputting the code and/or comments of the code into a formal specification generating model for identification, and using the formal specification output by the formal specification generating model as a candidate formal specification.
11. The device according to claim 9, wherein the protocol recommendation module is specifically used for:

    Based on the matching degree between the code and/or the comment of the code and the multiple rules included in the rule base, the formal specification corresponding to the rule whose matching degree is greater than the matching degree threshold among the multiple rules is used as a candidate formal specification.
12. The device according to claim 9, wherein the protocol recommendation module is specifically used for:

    Multiple candidate form specifications are obtained based on the intermediate representation corresponding to the code and/or the structural annotation of the code, wherein the structural annotation of the code is obtained by performing structural processing on the annotation of the code.
13. The device according to any one of claims 9 to 12, wherein the acquisition module is specifically used for:

    Key words are used to trigger acquisition of the code and/or annotations of the code input by the user.
14. The device according to claim 13, wherein the acquiring module is specifically used for:

    In the case that the user has input a keyword, the code and/or the comment of the code input by the user are acquired.
15. The device according to any one of claims 9 to 14, wherein the protocol recommendation module is specifically used for:

    Displaying the sorted plurality of candidate formal specifications to the user, wherein, among the plurality of candidate formal specifications, the candidate formal specifications that best match the code and/or the annotations of the code are ranked first.
16. The device according to any one of claims 9 to 15, wherein the device further comprises a verification module;

    The acquiring module is further configured to acquire a target form specification selected by the user from the plurality of candidate form specifications;

    The verification module is configured to use the target form specification to verify the code to be verified.
17. A computing device, characterized in that the computing device includes a processor and a memory; the processor is configured to execute instructions stored in the memory, so that the computing device implements the method according to any one of claims 1 to 8.
18. A form specification recommendation system, characterized in that the system includes a terminal device and a server;

    The terminal device is used to obtain the code input by the user and/or the comment of the code;

    The terminal device is configured to send the code and/or comments of the code to the server;

    The server is configured to obtain multiple candidate form specifications according to the code and/or code comments, and send the multiple candidate form specifications to the terminal device;

    The terminal device is configured to display the plurality of candidate form specifications to the user.
19. A computer-readable storage medium, wherein the computer-readable storage medium stores instructions, and the instructions are used to implement the method according to any one of claims 1 to 8.

Images