Automatic test script generation method and device and related equipment
By parsing interface elements and combining them with large language model training, meta-functional operation groups are generated, solving the problem of accurate matching of automated test scripts in existing technologies and realizing the generation of high-quality automated test scripts.
Patent Information
- Application Number
- CN202511024643.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-24
- Publication Date
- 2025-11-07
AI Technical Summary
Existing technologies cannot accurately match interface elements when generating automated test scripts, resulting in script execution errors, poor compatibility, and a lack of structured modeling for complex page operations, leading to low accuracy and reusability of the generated scripts.
By analyzing the interface elements of the software under test, a meta-functional operation group is generated. The base large language model is then trained and fine-tuned. Combined with historical test datasets and preprocessing algorithms, the target test script is generated to improve business adaptability.
It enables precise generation of automated test scripts, improving script accuracy and reusability, and meeting the needs of high-quality automated testing.
Smart Images
Figure CN120909932A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] Embodiments of the present application relate to the field of automatic testing technology, and in particular to an automatic test script generation method and device and related equipment. BACKGROUND
[0002] In the business scenario of generating automatic test scripts by large models, the interface elements of the software under test have diversity and complexity, and the layout structure, interaction logic and element attributes of different pages differ significantly. The automatic scripts directly generated by large models often cannot accurately match the interface elements, resulting in script execution errors and poor compatibility, and a large amount of manpower is required for debugging and modification.
[0003] The prior art mainly relies on large models to directly map test cases described in natural language to generate scripts, but does not consider the internal association between interface elements and business logic, and lacks structured decomposition of complex page operations. Since the positioning method (such as XPath, CSS selector) of the interface element, the operation type (click, input, swipe) and the business context are closely related, simply relying on text generation cannot capture the semantic differences of elements in different business scenarios, resulting in low accuracy and reusability of the generated scripts, and failing to meet the needs of high-quality automatic testing. Therefore, a method is needed that can structure the modeling of page operations in combination with business logic and use large models to achieve accurate script generation. SUMMARY
[0004] Embodiments of the present application provide an automatic test script generation method, device and related equipment, aiming to solve the technical problem of complex cross-page table recognition and splicing effect intersection in the prior art.
[0005] In a first aspect, embodiments of the present application provide an automatic test script generation method, which includes:
[0006] Parsing the interface elements of the current software under test, combining and arranging the interface elements according to the test case flow of the current software under test, and generating a plurality of meta-function operation groups;
[0007] Obtaining a historical test data set of the current software under test, and parsing the historical test data set to obtain description texts of historical test cases and corresponding meta-function operation group sequences, and constructing a training data set by taking the description texts as sample description texts and taking the meta-function operation group sequence corresponding to each description text as a label sequence, the meta-function group sequence at least including one meta-function operation group;
[0008] Obtaining a base large language model, and training the base large language model with the training data set to obtain a fine-tuned large language model;
[0009] obtain a target description text of a current test case, input the target description text into the large language model for business recognition, and output a corresponding target meta-function group sequence;
[0010] arrange based on the target meta-function group sequence and corresponding interface elements to generate a target test script.
[0011] In a second aspect, an embodiment of the present application provides an automatic test script generation device, which comprises:
[0012] The parsing module is configured to parse interface elements of a current software under test, combine and arrange the interface elements according to a test case process of the current software under test, and generate a plurality of meta-function operation groups.
[0013] The sample construction module is configured to obtain a historical test data set of the current software under test, parse the historical test data set, obtain description texts of historical test cases and corresponding meta-function operation group sequences, use the description texts as sample description texts, use the meta-function operation group sequences corresponding to each description text as label sequences, and construct a training data set, wherein the meta-function group sequence comprises at least one meta-function operation group.
[0014] The training module is configured to obtain a base large language model, train the base large language model based on the training data set, and obtain a fine-tuned large language model.
[0015] The recognition module is configured to obtain a target description text of a current test case, input the target description text into the large language model for business recognition, and output a corresponding target meta-function group sequence.
[0016] The arrangement module is configured to arrange based on the target meta-function group sequence and corresponding interface elements to generate a target test script.
[0017] In a third aspect, an embodiment of the present application further provides a computer device, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the automatic test script generation method of the first aspect when executing the computer program.
[0018] In a fourth aspect, an embodiment of the present application further provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program causes the processor to execute the automatic test script generation method of the first aspect when executed by the processor.
[0019] The embodiment of the present application provides a kind of automatic test script generation method, device and related equipment.The method parses the interface element of current software under test, and the interface element is combined according to the test case process of current software under test, generates several meta function operation groups;Current software under test historical test data set is obtained, and historical test data set is parsed, the description text of historical test case and corresponding meta function operation group sequence are obtained, and the description text is used as sample description text, and the meta function operation group sequence corresponding to each description text is used as label sequence;Base large language model is obtained, and base large language model is trained based on sample description text, and the generation sequence of sample description text is output;The sequence loss of generation sequence and label sequence is calculated based on the preset loss function, and the model parameter of base large language model is optimized according to sequence loss, and the fine-tuned large language model is obtained;The target description text of current test case is obtained, and the target description text is input into large language model for business identification, and the corresponding target meta function group sequence is output;Target meta function group sequence and corresponding interface element are arranged to generate target test script.The method realizes the generation of automatic script, and improves the business adaptability of the generated script. BRIEF DESCRIPTION OF DRAWINGS
[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.
[0021] Figure 1 The flowchart of an embodiment of the automatic test script generation method provided by the embodiment of the present application is shown.
[0022] Figure 2 For Figure 1 The subflowchart of an embodiment of step S130 in the method is shown.
[0023] Figure 3 The schematic block diagram of the automatic test script generation device provided by the embodiment of the present application is shown. DETAILED DESCRIPTION
[0024] The technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.
[0025] It should be understood that the terms "comprises" and "comprising," when used in this specification and the following claims, indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.
[0026] It should also be understood that the terms used in the specification and the appended claims are intended to describe particular embodiments and do not intend to limit the present application. As used in the specification and the appended claims, the singular forms "a," "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
[0027] It should further be understood that the term "and / or" used in the specification and the appended claims means one or more of the associated listed items as well as all possible combinations of the items and includes the combinations.
[0028] Reference is made to Figure 1 The flowchart of the method for generating an automated test script provided by the embodiment of the present application includes steps S110-S150.
[0029] Step S110, parse the interface elements of the current software under test, combine and arrange the interface elements according to the test case flow of the current software under test, and generate a plurality of meta-function operation groups;
[0030] In the embodiment, the interface elements of the current software under test are parsed through page analysis to obtain a plurality of interface elements, including element positioning information (XPath, CSS selector, element ID), operation type (click, input, swipe, test), element attribute (text content, visibility, disabled state), and business correlation parameter (such as the field name corresponding to the input box, the option value of the drop-down list). The interface elements are combined according to the test case flow of the software under test to obtain an element set of each test case flow. Finally, the interface elements in the corresponding element set are arranged according to the test logic corresponding to each test case flow to obtain a meta-function operation group of each test case flow.
[0031] For example, the interface elements of the test case flow "login page" include three sub-operations "username input box input", "password input box input", and "login button click", and the corresponding interface elements are:
[0032] Username input: <input type＝"text”id＝"username”name＝"username”placeholder＝"请输入用户名”> ;
[0033] Password input: <input type＝"password”id＝"password”name＝"password”placeholder＝"请输入密码”> ;
[0034] Login button:<button type＝"submit”id＝"loginbtn”>Log in< / button> .
[0035] Step S120, obtain the historical test data set of the current software under test, and parse the historical test data set to obtain the description text of the historical test case and the corresponding meta-function operation group sequence, and take the description text as the sample description text, and take the meta-function operation group sequence corresponding to each description text as the label sequence, and construct the training data set, wherein the meta-function group sequence at least includes one meta-function operation group;
[0036] Step S130, obtain the base large language model, and train the base large language model with the training data set to obtain the fine-tuned large language model;
[0037] In the embodiment, the historical test data set of the historical test scheme, test scene and expected result of the current software under test is called through the local database. The description text of the historical test case and the corresponding meta-function operation group sequence are disassembled from the historical test data set. Since some test cases need multiple meta-function operation groups to complete, the meta-function operation group sequence is used for representation. For example, the meta-function operation group sequence corresponding to the test case "verify user login function" includes
login meta-function operation group
verification meta-function operation group
[0038] Further, for example, the test case of shopping cart settlement includes the meta-function operation groups of "product addition -- settlement process -- payment verification -- address verification" in the corresponding meta-function operation group sequence, which is as follows:
[0039] 1. AddToCartGroup (product addition group)
[0040] Operation process: Enter the product detail page (page URL: / product / 123), locate the product quantity input box (element location: XPath = / / input[@id='quantity']), input quantity "2" (operation type: input, business parameter: quantity = 2); click the "Add to Cart" button (element location: CSS = #add-to-cart-btn, operation type: click).
[0041] Test purpose: Simulate the core operation of users selecting and adding products to the shopping cart, providing test data for subsequent settlement processes.
[0042] 2. CheckoutProcessGroup (Checkout Process Group)
[0043] Operation process: Jump to the shopping cart page ( / cart), click the "Go to Settlement" button (element location: class = checkout-btn, operation type: click); after entering the settlement page, locate the default payment method single selection box (element location: id = wechat-pay, operation type: verify selected state); if the user is not logged in, automatically fill in the login information (username input box: id = username, password input box: id = password, operation type: input, business parameter: username = test_user, password = 123456) and click login (operation type: click).
[0044] Test purpose: Cover the complete process from shopping cart to settlement, including login state verification and payment method default value verification.
[0045] 3. PaymentAssertGroup (Payment Verification Group)
[0046] Operation process: Locate the product total price element (XPath = / / span[@class='total-price'], expected value: calculated according to product unit price x quantity, such as unit price 100 yuan x 2 pieces = 200 yuan); locate the settlement amount element (id = final-amount, operation type: numerical comparison), test rule: settlement amount == product total price + shipping fee (if any) (assuming shipping fee is 0 here, i.e. verify that the settlement amount is equal to 200 yuan).
[0047] Technical implementation: Extract the page amount value using regular expressions, and use the Python verification statement assert final_amount == total_price for verification.
[0048] 4. AddressVerifyGroup (Address Verification Group)
[0049] Operation process: locate the delivery address display area (class = address-info), extract the text content (operation type: text acquisition), and perform string matching with the user account binding address (obtained from the test data center, such as "No. 123, XX Street, Chaoyang District, Beijing") (verification rule: the settlement address contains province, city and district information and is consistent with the account address).
[0050] As shown in Figure 2 In an embodiment, step S130 includes:
[0051] Step S210, training the base large language model based on the sample description text of the sample training set, outputting the generation sequence of the sample description text, the generation sequence being the meta-function operation group sequence generated by the base large language model based on the sample description text;
[0052] Step S220, calculating the sequence loss of the generation sequence and the sample description text pair corresponding to the label sequence based on the preset loss function, and optimizing the model parameters of the base large language model according to the sequence loss, to obtain the large language model.
[0053] In this embodiment, the description text of the historical test case (such as "user logs into the system") is converted into an input format acceptable to the model (such as a token sequence). The label sequence (i.e., the corresponding meta-function operation group sequence, such as "enter username → enter password → click login") is encoded to form the training target.
[0054] The base large language model is trained based on the training data set, and the base large language model outputs a generation sequence based on the sample description text. A preset loss function is used to calculate the sequence loss between the generation sequence and the label sequence corresponding to the sample description text, and the model parameters of the base large language model are optimized based on the sequence loss, to improve the matching accuracy between the generation sequence output by the large language model and the description text of the test case. Wherein, the loss function L is as follows:
[0055]
[0056] Wherein, N represents the total number of samples, n represents the nth sample, y n represents the probability of the sequence label, represents the probability of the generated label.
[0057] After multiple rounds of training, the large language model can accurately generate a meta-function operation group sequence, and the loss value is significantly reduced.
[0058] Step S140, obtaining the target description text of the current test case, inputting the target description text into the large language model for business recognition, and outputting the corresponding target meta-function group sequence;
[0059] Step S150, based on the target meta-function group sequence and the corresponding interface element, the target test script is generated.
[0060] In this embodiment, the target description text of the current test case is obtained, and the target description text is preprocessed, including word segmentation, stop word removal, and standardized expression. The preprocessed target description text is converted into a token sequence and input into the fine-tuned large language model. The model analyzes the text semantics through the attention mechanism and identifies the key business intent (such as "search", "login", "payment"). Based on the language patterns learned in the pre-training stage and the optimized parameters in the fine-tuning stage, the corresponding meta-function operation group sequence is generated. The generated meta-function operation group sequence is checked to ensure the logical coherence of the meta-function operation group (such as "payment" operation needs to be completed before "adding goods to the shopping cart"). If the generated meta-function operation group sequence has ambiguity, it is corrected through preset rules or manual intervention.
[0061] Further, according to the operation type (such as "click", "input") and business parameters (such as "username", "password") in the meta-function operation group, the corresponding interface elements are matched. According to the interface element information (such as XPath, CSS selector), the mapping relationship between the meta-function operation group and the actual interface element is established. According to the order of the meta-function operation group sequence, the execution flow of the test steps is generated. The business parameters (such as test data, environment configuration) are dynamically injected into the operation steps. Test logic is inserted after key operations to verify the operation results (such as "check if the user avatar is displayed after successful login"). According to the syntax specifications of the target test framework (such as Selenium, Pytest, JMeter), executable test script code is generated. Support for multi-language output (such as Python, Java, JavaScript) adapts to different technology stack test environments.
[0062] Further, the generated automated script is run in the software environment under test, and the running log and test results are captured in real time. Through the abnormality detection algorithm (such as the statistical-based anomaly detection method), the running log is deeply analyzed, and the element positioning failure, verification error and other abnormal situations are accurately identified and classified. When an abnormality is detected, the system automatically records detailed abnormal information (such as abnormal type, abnormal step and interface element information, etc.), providing data support for subsequent script debugging and optimization.
[0063] This method parses the interface elements of the software under test and combines them according to the test case flow to generate several meta-functional operation groups. It then acquires the historical test dataset of the software under test, parses it to obtain the description text of historical test cases and the corresponding meta-functional operation group sequences, using the description text as sample description text and the meta-functional operation group sequence corresponding to each description text as a label sequence. A base language model is acquired and trained based on the sample description text, outputting a generated sequence of sample description text. The sequence loss of the generated sequence and label sequence is calculated based on a preset loss function, and the model parameters of the base language model are optimized according to the sequence loss to obtain a fine-tuned language model. Finally, the target description text of the current test case is acquired, input into the language model for business recognition, and the corresponding target meta-functional group sequence is output. Based on the target meta-functional group sequence and the corresponding interface elements, the target test script is generated. This method automates script generation and improves the business adaptability of the generated script.
[0064] This invention also provides an automated test script generation apparatus, which is used to execute any embodiment of the aforementioned automated test script generation method. Specifically, please refer to... Figure 3 , Figure 3 This is a schematic block diagram of an automated test script generation device provided in an embodiment of the present invention. The automated test script generation device 100 can be configured in a server.
[0065] like Figure 3 As shown, the automated test script generation device 100 includes a parsing module 110, a sample construction module 120, a training module 130, a recognition module 140, and an arrangement module 150.
[0066] The parsing module 110 is used to parse the interface elements of the current software under test, combine and arrange the interface elements according to the test case flow of the current software under test, and generate several meta-function operation groups.
[0067] The sample construction module 120 is used to obtain the historical test dataset of the current software under test, and parse the historical test dataset to obtain the description text of the historical test cases and the corresponding meta-function operation group sequence. The description text is used as the sample description text, and the meta-function operation group sequence corresponding to each description text is used as the label sequence to construct the training dataset. The meta-function group sequence includes at least one meta-function operation group.
[0068] Training module 130 is used to acquire the base large language model and train the base large language model with the training dataset to obtain a fine-tuned large language model.
[0069] The recognition module 140 is configured to obtain a target description text of a current test case, input the target description text into the large language model to perform business recognition, and output a corresponding target meta-function group sequence.
[0070] The arrangement module 150 is configured to arrange based on the target meta-function group sequence and a corresponding interface element, and generate a target test script.
[0071] In an embodiment, the parsing module 110 includes:
[0072] The recognition unit is configured to determine an element set of each test case process according to a test case process of the software under test, the element set including at least one interface element;
[0073] The arrangement unit is configured to arrange the interface elements in the corresponding element set according to the test logic corresponding to each test case process, to obtain a meta-function operation group of each test business process.
[0074] In an embodiment, the training module 130 includes:
[0075] The prediction unit is configured to train the base large language model based on a sample description text of the sample training set, and output a generation sequence of the sample description text, the generation sequence being a meta-function operation group sequence generated by the base large language model based on the sample description text.
[0076] The optimization unit is configured to calculate a sequence loss of the generation sequence and a corresponding label sequence of the sample description text based on a preset loss function, and optimize model parameters of the base large language model according to the sequence loss, to obtain the large language model.
[0077] The embodiment of the present application also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the automatic test script generation method as described above when executing the computer program.
[0078] In another embodiment of the present application, a computer readable storage medium is provided. The computer readable storage medium can be a non-volatile computer readable storage medium. The computer readable storage medium stores a computer program, which, when executed by a processor, causes the processor to execute the automatic test script generation method as described above.
[0079] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the devices, apparatuses and units described above can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here. Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, each example has been described in the foregoing description in terms of its general functionality, without regard to its specific combination of hardware and software. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.
[0080] In several embodiments provided by the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely schematic, for example, the division of the units is merely logical functional division, and actual implementation can have another division, or units with the same function can be combined into one unit, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, or can be electrical, mechanical or other form of connection.
[0081] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, i.e. they can be located in one place or distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiments of the present application.
[0082] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be realized in the form of hardware or software functional unit.
[0083] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a storage medium. Based on such understanding, the technical solutions of the present application essentially or the part of the prior art that contributes to the present application, or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in various embodiments of the present application. The aforementioned storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a magnetic disk or an optical disk, and various media that can store program codes.
[0084] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present application, and these modifications or replacements should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.
Claims
1. An automated test script generation method, characterized by, The method comprises the following steps: parsing interface elements of the current software under test, combining and arranging the interface elements according to the test case flow of the current software under test, and generating a plurality of meta-function operation groups; obtaining a historical test data set of the current software under test, parsing the historical test data set, obtaining a description text of a historical test case and a corresponding meta-function operation group sequence, taking the description text as a sample description text, and taking the meta-function operation group sequence corresponding to each description text as a label sequence to construct a training data set, wherein the meta-function group sequence at least includes one meta-function operation group; obtaining a base large language model, and training the base large language model with the training data set to obtain a fine-tuned large language model; obtaining a target description text of a current test case, inputting the target description text into the large language model for business identification, and outputting a corresponding target meta-function group sequence; arranging based on the target meta-function group sequence and the corresponding interface elements to generate a target test script.
2. The automated test script generation method of claim 1, wherein, The interface elements include element positioning information, operation type, element attribute and business association parameter.
3. The method of claim 1, wherein, The combining and arranging of the interface elements according to the test case flow of the current software under test comprises: determining an element set of each test case flow according to the test case flow of the software under test, wherein the element set at least includes one interface element; arranging the interface elements in the corresponding element set according to the test logic corresponding to each test case flow to obtain the meta-function operation group of each test business flow.
4. The method of claim 1, wherein, The training of the base large language model with the training data set to obtain a fine-tuned large language model comprises: training the base large language model based on the sample description text of the sample training set, outputting the generation sequence of the sample description text, and the generation sequence is the meta-function operation group sequence generated by the base large language model based on the sample description text; calculating the sequence loss of the generation sequence and the label sequence corresponding to the sample description text based on the preset loss function, and optimizing the model parameters of the base large language model according to the sequence loss to obtain the large language model.
5. The method of claim 1, wherein, After the arranging based on the target meta-function group sequence and the corresponding interface elements to generate a target test script, the method comprises the following steps: running the target test script and generating a running log to record the corresponding test result.
6. The method of claim 4, wherein, After executing the target test script and monitoring the running log and the test result in real time, the method comprises the following steps: analyzing the running log by an abnormality detection algorithm to determine whether the running log contains abnormal information; if there is abnormal information, recording the abnormal data corresponding to the abnormal information, wherein the abnormal data at least includes abnormal type, abnormal occurrence step and interface element information.
7. An automated test script generation apparatus characterized by comprising: The method comprises the following steps: a parsing module for parsing interface elements of the current software under test, combining and arranging the interface elements according to the test case flow of the current software under test, and generating a plurality of meta-function operation groups; The sample construction module is configured to obtain a historical test data set of a current software under test, parse the historical test data set to obtain description texts of historical test cases and corresponding meta-function operation group sequences, and construct a training data set by taking the description texts as sample description texts and taking the meta-function operation group sequences corresponding to each description text as label sequences, wherein the meta-function group sequence at least includes one meta-function operation group; The training module is configured to obtain a base large language model, and train the base large language model by using the training data set to obtain a fine-tuned large language model; The recognition module is configured to obtain a target description text of a current test case, input the target description text into the large language model for business recognition, and output a corresponding target meta-function group sequence; The arrangement module is configured to arrange based on the target meta-function group sequence and corresponding interface elements to generate a target test script.
8. The automated test script generation apparatus of claim 6, wherein, The parsing module comprises: An identification unit configured to determine an element set of each test case process according to a test case process of the software under test, wherein the element set at least includes one interface element; An arrangement unit configured to arrange the interface elements in the corresponding element set according to the test logic corresponding to each test case process to obtain a meta-function operation group of each test business process. 9.A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the automatic test script generation method according to any one of claims 1 to 6 when executing the computer program.
10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, which, when executed by a processor, causes the processor to perform the automatic test script generation method according to any one of claims 1 to 6.