Multi-language test method and related equipment

Abstract

The embodiment of the invention provides a multi-language test method and related equipment. According to the method, first, a first test case used for conducting multi-language testing on a first application is obtained, and the first test case comprises target test interface indication information written in a natural language and / or test content of a target test interface; and then running the first application according to the first test case to obtain an interface image of the target test interface. And finally, according to the first test case, performing multi-dimensional translation detection, including mistranslation, translation omission, text hyperbox, logic detection and the like, on the interface image of the target test interface, and then obtaining a first test result. Therefore, a tester can directly drive the automatic test of the multi-language application by writing the test case by using the natural language, so that the test threshold is greatly reduced, and the test efficiency is remarkably improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a multilingual testing method and related equipment. Background Technology

[0002] Multilingual applications are designed to meet the needs of users in different countries and regions, supporting multiple language versions. Multilingual testing refers to the testing of each language version of the application during its development to check for issues such as mistranslation, omissions, and text exceeding boundaries. However, current multilingual testing methods still have many shortcomings, such as requiring highly skilled testers, low testing efficiency, and high testing costs.

[0003] Therefore, how to conduct efficient and convenient multilingual testing of multilingual applications has become an urgent problem to be solved in this technical field. Summary of the Invention

[0004] This application provides a multilingual testing method and related equipment. The electronic device can run multilingual applications according to test cases, acquire interface images of the target test interface of the application, and then perform multi-dimensional translation detection on the interface images of the target test interface according to the test cases to generate test results. In this way, testers can directly drive automated testing of multilingual applications by writing test cases using natural language, significantly lowering the testing threshold while significantly improving testing efficiency.

[0005] In a first aspect, embodiments of this application provide a multilingual testing method applied to an electronic device. According to the multilingual testing method provided in the first aspect, the electronic device first acquires a first test case for multilingual testing of a first application. The first test case includes target test interface instruction information written in natural language and / or test content for the target test interface. Then, the electronic device runs the first application according to the first test case and acquires an interface image of the target test interface. After acquiring the interface image of the target test interface, the electronic device then performs translation detection on the interface image of the target test interface according to the first test case and obtains a first test result. The first test result may include one or more of the following translation problems in the target test interface: mistranslation, omission, text exceeding the frame, and logical problems between multiple translation objects. Finally, the electronic device displays the first test result.

[0006] For example, the first test case acquired by the electronic device could be "Open the Michelle character profile interface and check whether the text of the character name, character introduction, character skills, and character slogan in the interface is free of typos, ambiguity, and fluent sentences." This includes the target test interface instruction information "Michelle character profile interface" and the test content for the target test interface: "Check whether the text of the character name, character introduction, character skills, and character slogan in the interface is free of typos, ambiguity, and fluent sentences." Based on the first test case, the electronic device can run the first application, acquire the interface image of the Michelle character profile interface, then perform translation detection on the interface image of the Michelle character profile interface, and obtain the first test result.

[0007] By implementing the method provided in the first aspect, electronic devices can receive test cases written in natural language uploaded by testers, and then automatically run and test multilingual applications directly based on the received test cases. After the test is completed, the electronic device can also directly display the test results to the testers, thereby achieving efficient and convenient multilingual testing, significantly lowering the testing threshold while significantly improving testing efficiency.

[0008] In conjunction with the first aspect, in some embodiments, the method for an electronic device to run a first application according to a first test case and obtain an interface image of a target test interface may include the following steps: First, the electronic device runs the first application and obtains a first interface image of a first interface within the first application. Then, the electronic device uses a first model to identify the first test case and the first interface image, determining an operation instruction for controlling the first application to jump from the first interface to the target test interface. Next, the electronic device controls the first application to jump to the target test interface according to the determined operation instruction and obtains an interface image of the target test interface.

[0009] The first model can be a pre-trained natural language model capable of understanding test cases, recognizing interface images, and generating operation instructions to instruct an electronic device to run a first application and acquire interface images. By implementing the above embodiments, the natural language understanding and image recognition capabilities of the first model are utilized to convert test cases into interface operation instructions, then automatically navigate to the target test interface and capture the interface image of the target test interface, thereby improving the convenience of multilingual testing.

[0010] In conjunction with the first aspect, in some embodiments, the electronic device may first use a first model to identify a first test case and a first interface image, determine a first operation instruction, control a first application to jump from the first interface to a second interface according to the first operation instruction, and acquire a second interface image of the second interface. Then, the electronic device continues to use the first model to identify the first test case and the second interface image, determine a second operation instruction, and control the first application to jump from the second interface to a target test interface according to the second operation instruction.

[0011] See the attached diagram. Figures 7A-7C , Figures 7A-7C This example illustrates the process of displaying the target test interface after two interface transitions. Figure 7A An example is shown of the first interface, Figure 7B An example is shown of the second interface, Figure 7C An example of the target test interface is shown.

[0012] In some specific implementations, the second operation instruction can also include multiple operation instructions, and the process of jumping from the second interface to the target test interface can include more jump steps. For example, the electronic device controls the first application to jump from the second interface to the third interface according to the second operation instruction. Then, the electronic device continuously inputs the newly acquired interface image of the third interface into the first model, receives and executes feedback operation instructions (such as the third operation instruction, the fourth operation instruction, etc.), and simultaneously acquires subsequent interface images (such as the interface image of the fourth interface, the interface image of the fifth interface, etc.) until it jumps to the target test interface and acquires the interface image of the target test interface. That is to say, the electronic device can use the first model multiple times to parse the interface images of the first test case and the currently displayed application interface, generate multiple operation instructions, and control the first application to perform multiple interface jumps according to these multiple operation instructions, finally jumping to the target test interface and then acquiring the interface image of the target test interface.

[0013] In conjunction with the first aspect, in some embodiments, the electronic device can use the first model to identify target test interface indication information in the first test case, identify the first control in the first interface image according to the target test interface indication information, and generate a first operation instruction, wherein the first control is used to control the first application to jump from the first interface to the second interface, and the first operation instruction is an instruction for operating the first control.

[0014] By implementing the above embodiments, the natural language model's image understanding capabilities are used to identify controls in the interface image, and the natural language test cases are automatically converted into a series of interface navigation operations. This not only automates testing and improves convenience, but also adapts to dynamic changes in the application interface (such as layout adjustments and changes in control positions), thereby improving the reusability of test cases.

[0015] In conjunction with the first aspect, in some embodiments, the electronic device can run one or more language versions of the first application and acquire interface images of the target test interface for one or more language versions. Specifically, when the test case does not specify a target test language version, such as when the test case is "testing the Michelle character profile interface," the electronic device can sequentially acquire interface images of the target test interface for all language versions supported by the first application. When the test case specifies a target test language version, such as when the test case is "testing the French version of the Michelle character profile interface," the electronic device can acquire the interface image of the target test interface for the target language version (e.g., the French version) specified by the test case.

[0016] In conjunction with the first aspect, in some embodiments, the electronic device can utilize a second model to identify test content for the target test interface from the first test cases, and perform translation testing on the target test interface based on the test content. The second model is a pre-trained natural language model capable of understanding test cases and performing multilingual testing on the interface images based on the test cases.

[0017] For details, see Figure 9 The second model can perform multi-dimensional translation tests in parallel on the interface image of a target test interface, such as mistranslation tests, omission tests, out-of-frame tests, and logic tests. The test results from these multiple dimensions are then integrated to form the multilingual test results for the interface image. Specifically, the out-of-frame test detects whether text content exceeds the display area boundary, the mistranslation test verifies the accuracy of the translation, the omission test checks for untranslated text content, and the logic test detects the logical relationships between multiple translated texts in the interface image, ensuring the semantic, functional, and contextual completeness and rationality of the translated content. For example, in continuous dialogue scenarios (such as game NPC dialogues or teaching guidance processes), the model analyzes the logic between multiple dialogue texts to detect whether they are coherent and connected. The second model can also determine which dimensions of testing to perform on the interface image of the target test interface based on the first test case. For example, the test case is "Open the Michelle character profile interface and check whether the text of the character name, character introduction, character skills, and character slogan in the interface is free of typos, ambiguity, and fluent sentences". By parsing the test case, the second model can determine to perform mistranslation testing and semantic testing on the interface image of the test case.

[0018] In conjunction with the first aspect, in some embodiments, both the first model and the second model integrate a glossary of terms for the first application. The glossary of terms for the first application includes standard translations of the proper nouns of the first application in multiple language versions.

[0019] The first model, by integrating the terminology of the first application, enhances its understanding of the application's vocabulary, reduces ambiguity and misidentification, and thus more accurately identifies the text content of the application's interface images, generating more precise operation instructions. The second model, by integrating the terminology of the first application, also enhances its understanding of the application's vocabulary, reduces ambiguity and misidentification, and thus more accurately identifies the application's interface images, performing text semantic testing, translation testing, and so on.

[0020] In conjunction with the first aspect, in some embodiments, the first model and the second model are the same model, which may integrate a glossary of the first application; it has the capability of the first model, that is, it can generate operation instructions based on test cases and interface images, instructing the image acquisition tool to perform operations, thereby acquiring the target test image of the test cases; it also has the capability of the second model, that is, it can test the interface images based on test cases and obtain test results.

[0021] In conjunction with the first aspect, in some embodiments, the first model and the second model are deployed on a server, and the method provided in the first aspect further includes: the electronic device sending a first test case to the server; the electronic device receiving an operation instruction returned by the server; the electronic device sending an interface image of the target test interface to the server; and the electronic device receiving a first test result returned by the server.

[0022] In conjunction with the first aspect, in some embodiments, the first test case is either user-inputted or retrieved from a database. For example, the electronic device may display... Figure 2A The first user interface shown is used to receive first test cases uploaded by users. Uploading methods may include inputting information in the first user interface, uploading a test file containing the first test cases in the first user interface, selecting to retrieve test cases from a database through the first user interface, etc. For details, please refer to the relevant descriptions in the implementation details, including... Figure 2A ,right Figure 2A The introduction includes S501, etc.

[0023] In conjunction with the first aspect, in some embodiments, the electronic device can also determine and display the type of translation problem in the first test results, the severity level of the translation problem, suggestions for improvement, etc. For details, please refer to... Figure 2B And to Figure 2B The above embodiments not only display existing translation issues to testers, but also show the issue level and suggested modifications. This allows testers to more quickly and accurately locate problems and optimize the translation based on the suggestions, thereby reducing the time cost of manual screening and improving testing efficiency.

[0024] In conjunction with the first aspect, in some embodiments, the electronic device may also display controls for marking whether the first test result is correct. The electronic device can add the first test result to either an error result set or a correct test result set based on user actions applied to different controls. Then, the electronic device can, based on the user's selection, display only the test results from the error result set, only the test results from the correct test result set, or both the error result set and the correct test result set simultaneously. For details, please refer to [link to relevant documentation]. Figures 2D-2F And to Figures 2D-2F This section provides a related introduction. By labeling and categorizing test results, the ease with which testers can view the results can be effectively improved. Furthermore, by labeling the results, the testing capabilities of the second model can be further trained and adjusted based on the sets of incorrect and correct results, continuously improving the accuracy of the second model in testing interface images.

[0025] In a second aspect, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory; the processor executes the computer program to implement the method described in the first aspect and any possible implementation thereof.

[0026] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method as described in the first aspect and any possible implementation thereof.

[0027] Fourthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect and any possible implementation thereof.

[0028] Understandably, the electronic device provided in the second aspect, the computer storage medium provided in the third aspect, and the computer program product provided in the fourth aspect are all used to execute the method provided in this application. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of this application or the background art, the accompanying drawings used in the embodiments of this application or the background art will be described below.

[0030] Figure 1A and Figure 1B Examples show the application interface in two different languages; Figure 2A An example is shown of a first user interface for receiving test cases and starting tests; Figures 2B-2F Several examples of second user interfaces for displaying test results are shown; Figure 3 The overall flow of the multilingual testing method provided in the embodiments of this application is illustrated by way of example; Figure 4 The embodiments of this application exemplify several functional modules involved in the specific implementation of the multilingual testing method provided in the embodiments of this application; Figure 5 An exemplary method is provided for obtaining a first test case and, based on the first test case, obtaining an interface image of the target test interface; Figure 6 This example illustrates the specific steps of the image acquisition tool calling the first model to acquire the interface image of the target test interface; Figures 7A-7C This example illustrates the process of displaying the target test interface after two interface transitions; Figure 8 An example is shown of the process for performing multilingual testing on the interface images of the target test interface; Figure 9 An example illustrates the workflow for performing multilingual tests using the second model; Figure 10 An exemplary embodiment of this application provides a multilingual testing framework; Figure 11 An example is shown of a method for electronic devices and servers to work together; Figure 12 An electronic device provided in an embodiment of this application is illustrated by way of example; Figure 13 An exemplary embodiment of the server provided in this application is shown. Detailed Implementation

[0031] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be a limitation of this application.

[0032] Multilingual applications can support multiple language versions of the interface. For example, a multilingual application can support multiple language versions such as Chinese, English, German, and French. When a user selects a language version, the multilingual application will display the application interface in the language corresponding to that language version.

[0033] Figure 1A and Figure 1B Examples of application interfaces 10 in two different languages ​​are shown.

[0034] in, Figure 1A The Chinese version of the application interface 10 (hereinafter referred to as the Chinese interface). Figure 1BThe application interface 10 is in English (hereinafter referred to as the English interface).

[0035] like Figure 1A and Figure 1B As shown, the application interface 10 may include five text display areas: area 101, area 102, area 103, area 104, and area 105. Figure 1A In the Chinese interface shown, all text in the five text display areas of application interface 10 is displayed in Chinese, while... Figure 1B In the English interface shown, all text in the five text display areas of application interface 10 is displayed in English. The text in each display area has the same semantic meaning across different language versions. For example, area 101 displays the name of the game character shown on the right side of application interface 10, area 102 displays the game character's identity, area 103 displays the game character's introduction, and area 104 displays prompt text. Figure 1A In the Chinese interface shown, area 101 displays the Chinese name "Michelle," area 102 displays "A newly promoted, energetic investigator who loves games," area 103 displays the Chinese introduction "The energetic super player Michelle has arrived! As a rookie investigator in Opal, medals, cats, and games are her favorite things! She firmly believes that with the powerful cat by her side, one day she will be able to stand on her own and become a reliable hero like her father!", and area 104 displays "Character Skills." Figure 1B In the English interface shown, area 101 displays the English name "Michele", area 102 displays "Avibrant new detective who loves games", area 103 displays the English introduction "Here comes Michelle, the energetic super gamer! As a new Opal Investigator, her favorite things are medals, cats, and games! She firmly believes that with her powerful "Fire Kitty" by her side, she will one day be able to stand on her own and become a reliable hero just like her father!", and area 104 displays "CharacterSkills".

[0036] It is important to note that Figure 1A and Figure 1B This is only intended to illustrate the features of the unified application interface in different language versions and should not be construed as limiting the embodiments of this application.

[0037] However, the translation process for multilingual applications may introduce semantic problems such as mistranslation, omission, improper word order, and inclusion of culturally taboo words, which in turn affect the accuracy of the text in the application interface and reduce the user experience.

[0038] Furthermore, in multilingual applications, the text display area typically has a fixed size, while the length of text with the same meaning in different languages ​​often varies significantly. This can lead to display issues such as text exceeding the border, layout misalignment, and overlapping controls, negatively impacting the user experience. For example, Figure 1A and Figure 1B The text display area 105 in the application interface 10 shown is used to display the game character's declaration. For example, the game character's Chinese declaration is: "Let's work hard together, justice will never be absent!", and the English declaration is: "Let's work hard together, justice will never be absent!". It can be seen that the English declaration is longer than the Chinese declaration; therefore, in... Figure 1B In the English interface shown, the English declaration is likely to be displayed outside the text display area 105.

[0039] Therefore, before releasing a multilingual application, it is necessary to test the application interface of each language version to check for semantic or display issues, thus ensuring that the application interface can be displayed correctly in each language version after release. This testing process can be called multilingual testing.

[0040] Currently, multilingual testing methods mainly fall into two categories: manual testing and UI automation testing.

[0041] Manual testing refers to testers manually running various language versions of a multilingual application and then manually reviewing each application interface in each language version to check for semantic and display issues. This manual testing method relies on testers proficient in the target language and requires testers to invest a significant amount of time in repetitive operations and reviews, resulting in high testing costs and low testing efficiency, becoming a major obstacle in the rapid update and iteration process of applications.

[0042] UI automation testing refers to the method of simulating user actions (such as clicking, typing, and swiping) using testing tools to automatically test the application's interface and verify the expected results. Common testing tools include Selenium and Airtest. However, these tools require testers to write scripts to drive them, which has a high technical threshold and is difficult to apply on a large scale.

[0043] Therefore, this application provides a multilingual testing method. An electronic device can receive test cases written in natural language uploaded by testers, and then automatically run and test the multilingual application directly based on the received test cases. After the test is completed, the electronic device can directly display the test results to the testers, thereby achieving efficient and convenient multilingual testing.

[0044] A test case is a description of a testing task for a specific software product, embodying the test plan, methods, techniques, and strategies. Its content may include one or more of the following: test objectives, test environment, input data, test steps, expected results, etc., ultimately forming a document. Simply put, a test case is a detailed test guide used to explain the test tasks, test steps, expected results, etc.

[0045] In this embodiment, test cases can be written using natural language. Each test case describes the test steps for running the multilingual application under test. Natural language refers to languages ​​used by humans in daily life, such as Chinese and English, and is an important tool for exchanging ideas and transmitting information. For example, a test case might include the following: Step 1, switch to the English version; Step 2, open the character introduction interface of game character A and test the page; Step 3, open the skill introduction interface of skill B of game character A and test the page.

[0046] Test results can include detected problems and their corresponding problem levels, problem types, etc. Problem levels indicate the severity of the problem; for example, problem levels can be divided into three levels: severe, moderate, and minor. Problem types can include semantic problems, display problems, etc.

[0047] The following is combined with Figures 2A-2F This paper introduces the human-computer interaction method in the multilingual testing method provided in the embodiments of this application.

[0048] Figure 2A An example is shown of a first user interface for receiving test cases and starting tests.

[0049] See Figure 2AThe first user interface may include an area for uploading test cases 201, an area for selecting test cases 202, a start test control 203, and a view test results control 204.

[0050] The test case upload area 201 is used to receive test cases. The test case upload area 201 may include a file upload control 201A and an online writing control 201B. This application embodiment provides two methods for receiving test cases: one is to receive test case files uploaded by the user through the file upload control 201A, and the other is to receive test cases written online by the user through the online writing control 201B. For example, when an electronic device detects an action acting on… Figure 2A When the user operates the file upload control 201A in the first user interface shown, the electronic device can display a window for selecting a local file, and then retrieve the test case file selected by the user from the local storage path. For example, when the electronic device detects an action on... Figure 2A When the user operates the online writing control 201B in the first user interface shown, the electronic device can display a window for writing test cases to receive test cases written online by the user. For example... Figure 2A As shown, the file upload control 201A can also display the prompt text "Click here to upload a file", and the online writing control 201B can also display the prompt text "Click here to write test cases online".

[0051] In some implementations... Figure 2A The online editing control 201B in the first user interface shown can also be used to edit historically received test cases. For example, when the electronic device detects a user action on the online editing control 201B in the first user interface, the electronic device can display a window for selecting historically uploaded test cases, and then display the content of the selected test case in the editing area.

[0052] The test case selection area 202 can be used to display the identifiers of all available test cases (including those uploaded via file upload control 201A and those written online via online writing control 201B), and receive user selections of one or more test case identifiers to determine the set of test cases to be executed. See also Figure 2AThe test case selection area 202 in the first user interface shown can display multiple test case numbers, and each test case number can correspond to a selection control (such as a checkbox). The electronic device can determine whether to add the corresponding test case to the test case set by detecting user actions on each selection control. For example, after the electronic device detects actions on the checkboxes corresponding to test cases 10000, 10001, and 10002, it can determine to add test cases 10000, 10001, and 10002 to the test case set to be executed and display the checkboxes corresponding to these test cases as selected, while the remaining test cases 10003, 10004, and 10005 remain unselected.

[0053] The Start Test control 203 can be used to trigger the execution of test cases selected by testers in the test case selection area 202. That is, when the electronic device detects a user action on the Start Test control 203, it begins to execute the selected test cases.

[0054] The test result viewing control 204 can be used to trigger a jump to a second user interface for displaying the test results. That is, when the electronic device detects a user action on the test result viewing control 204, it displays the test results.

[0055] Testers can upload or write test cases in the test case upload area 201, and select the test cases to be executed in the test case selection area 202. They can then trigger the execution of the selected test cases using the start test control 203. After the test is completed, testers can use the view test results control 204 to navigate to a second user interface that displays the test results.

[0056] It is important to note that Figure 2A This is merely an exemplary description of a first user interface used to initiate testing. In other embodiments, the first user interface used to initiate testing may also be other forms of user interfaces with the above-described functions.

[0057] In some implementations, the electronic device can directly start executing the received test cases after receiving them through the upload test case area 201, without the tester needing to select test cases in the select test case area 202 or click the start test control 203.

[0058] In some implementations, after the electronic device completes the test, it can automatically jump to the second user interface to display the test results, without requiring the tester to click the test result viewing control 204.

[0059] Figure 2B An example of a second user interface for displaying test results is shown.

[0060] The second user interface can be used to display all or part of the historical test results. Historical test results refer to the results of tests completed before the results are displayed through the second user interface.

[0061] like Figure 2B As shown, the second user interface may include a display area 206 for displaying test results.

[0062] In some implementations, the electronic device can display test results in the form of cards within the display area 206. Each card is a rectangular visual container component used to encapsulate and display information from a single test result. The electronic device can display multiple test results using multiple cards within the display area 206. Each card can correspond to one test result. Within a single card, information such as the corresponding problem interface image, problem description, problem type, problem level, and suggested modifications can be displayed using images, text, and other formats, providing testers with a highly visual and information-rich way to present test results. Testers can also interact with the cards using gestures such as clicking and swiping to further view the corresponding problem or perform operations such as deletion or marking.

[0063] like Figure 2B As shown, display area 206 can display cards 206A and 206B. Cards 206A and 206B are used to display two different test results.

[0064] Card 206A may include two areas: an image display area 206A1 on the left and a problem description area 206A2 on the right. The image display area 206A1 displays an image of the interface containing the problem, while the problem description area 206A2 displays the problem content. Specifically, the problem content may include one or more of the following: the location of the problem, the problem text, the problem type, the problem description, suggested modifications, etc. Figure 2BAs shown in the problem description area 206A2, there are two problems, Problem 1 and Problem 2. Problem 1 includes the following: "Location: The 'CharacterDescription(#044)' section at the top of the interface. Problem text: Character Description(#044). Problem type: Display problem. Problem description: This should be a description of the character's personality, but it only displays the placeholder 'Character Description(#044)', without showing the actual description, making it impossible for users to understand the character's characteristics." Problem 2 includes the following: "Location: The dialog box next to the portrait on the right side of the interface. Problem text: Let's work hard together, justice will never be absent! Problem type: Display problem. Problem description: The text here exceeds the display area, causing an overflow issue. It is recommended to expand the display area." Similarly, card 206B also includes two areas: an image display area 206B1 on the left and a problem description area 206B2 on the right. The image display area 206B1 displays an image of the interface containing the problem, while the problem description area 206B2 displays the problem content. Specifically, the problem content may include one or more of the following: the location of the problem, the problem text, the problem type, the problem description, suggested modifications, etc. Figure 2BAs shown, problem description area 206B2 displays one problem, namely Problem 1. Problem 1 includes the following: Location: Text on the right side of the interface. Problem text: Michelle throws a small turret that can attach to building surfaces. The turret will automatically attack enemy heroes within a certain range and slow them down. Problem type: Semantic problem. Problem description: There is a grammatical error. Michelle is a third-person singular, and the verb following her should be in the third-person singular form "throws". Additionally, it is suggested that "throws" be replaced with the more appropriate verb "deploys" for "throwing turrets". The corrected text should be: Michelle deploys a small turret that can attach to building surfaces. The turret will automatically attack enemy heroes within a certain range and slow them down.

[0065] Figure 2C An example is shown of another second user interface for displaying test results.

[0066] See Figure 2C In some embodiments, the second user interface may further include a filtering area 205 for filtering test results. Figure 2C In the middle, display area 206 can be used to display the test results filtered by filter area 205.

[0067] like Figure 2C As shown, the filtering area 205 may include multiple filtering controls, such as language version filtering control 205A, issue level filtering control 205B, issue type filtering control 205C, and test case filtering control 205D, etc.

[0068] The language version filtering control 205A may include multiple selection controls. These selection controls correspond to multiple language versions of the multilingual application. The electronic device can determine whether to select the corresponding language version by detecting user actions on each selection control in the language version filtering control 205A, and then filter the test results corresponding to the selected language version from all test results. Figure 2C As shown, Figure 2CThe language version selection control 205A includes three selection controls, corresponding to Chinese, English, and French language versions respectively. When the electronic device detects that the selection control corresponding to the English version is being used, it determines to filter the test results corresponding to the English version.

[0069] Similarly, other filter controls, such as the issue level filter control 205B, the issue type filter control 205C, and the test case filter control 205D, can also include multiple selection controls. Specifically, the multiple selection controls in the issue level filter control 205B can each correspond to a different issue level, such as severe, moderate, or minor; the multiple selection controls in the issue type filter control 205C can each correspond to a different issue type, such as semantic issues or display issues; and the multiple selection controls in the test case filter control 205D can each correspond to a different test case, such as 10000, 10001, 10002, or 10003. The electronic device can then determine the filter items by detecting user actions performed on each of these selection controls.

[0070] The filtering area 205 may also include a query control 205E. After the electronic device detects the user operation on the query control 205E, it can determine which test results need to be filtered from the historical test results based on the detection results of each filtering control in the filtering area 205, and then display the filtered test results in the display area 206.

[0071] Figure 2C The filter controls shown are merely an exemplary illustration of the filter area 205. In some implementations, the filter area 205 may include more or fewer filter controls for filtering from other dimensions, such as filtering by test time, filtering according to the application scenario, etc.

[0072] Figure 2D An example is shown of yet another second user interface for displaying test results.

[0073] like Figure 2D As shown, the display area 206 may also include controls for classifying test results according to their correctness.

[0074] For example, card 206A may also include controls 206A3 and 206A4. Control 206A3 can be used to mark the test results of the interface image corresponding to card 206A as "Analysis Error" and add it to the error result collection. The error result collection can be used to track defects that need to be fixed, generate error reports, or assign them to developers. Control 206A4 is used to mark the test results of the interface image corresponding to card 206A as "Confirmed Problem" and add it to the correct test result collection. The correct test result collection can be used to record verified correct translations or interface displays as a benchmark or reference for subsequent testing. Testers can use these two controls to quickly classify and filter a large number of automated test results with one click, categorizing the test results.

[0075] Card 206B is similar to card 206A, so it will not be described in detail.

[0076] The display area 206 may also include a correct analysis control 207 and an abnormal analysis control 208. When the electronic device detects that the correct analysis control 207 is selected, it can display only the test results marked as "confirmed to be a problem" and added to the correct results set in the display area 206. When the electronic device detects that the abnormal analysis control 208 is selected, it can display only the test results marked as "analysis error" and added to the error results set in the display area 206.

[0077] For example, suppose the test result corresponding to card 206A is marked as "Analysis error" by the user via control 206A3 and added to the error result set, and the test result corresponding to card 206B is marked as "Confirmed problem" by the user via control 206B4 and added to the correct result set. Then, as... Figure 2E As shown, after the electronic device detects that the analysis control 207 has been activated, it can display the test result corresponding to card 206B, but will not display the test result corresponding to card 206A. For example... Figure 2F As shown, after the electronic device detects the action on the analysis anomaly control 208, it can display only the test results corresponding to card 206A and not the test results corresponding to card 206B. Based on the first and second user interfaces shown above, the electronic device can first receive test cases uploaded by users through the first user interface, determine the test cases to be executed, and start the test. Then, it can directly display the test results to the testers through the second user interface. The testers can filter the test results according to language version, problem type, problem level, test cases, etc., and display test results that meet the testers' needs. The whole testing process is both efficient and convenient.

[0078] The following describes the implementation method of the multilingual testing method provided in this application embodiment. In the multilingual testing method provided in this application embodiment, the electronic device first obtains test cases, then runs the multilingual application to be tested according to the obtained test cases and obtains the interface images to be tested. Next, the electronic device can perform translation tests on these interface images to detect whether there are semantic problems and / or display problems. Finally, after the test is completed, the electronic device can display the test results to the user.

[0079] Figure 3 The overall flow of the multilingual testing method provided in this application embodiment is illustrated by way of example, and may include, but is not limited to, the following steps: S301. Obtain a first test case for multilingual testing of the first application. The first test case includes target test interface instruction information written in natural language and / or test content of the target test interface.

[0080] Specifically, electronic devices can... Figure 2A The first user interface shown retrieves the first test case; details can be found in the [link / view]. Figure 2A And to Figure 2A The relevant information will not be repeated here.

[0081] In this embodiment, the first test case may be written in natural language, and may include a description of the test interface, the tests to be performed on each test interface, and the expected results, etc. In some embodiments, the test cases may adopt structured data formats, such as JSON, XML, etc. In other embodiments, the test cases may adopt unstructured plain text. For example, the first test case may be: "Open the Michelle character profile interface and check whether the text of the character name, character introduction, character skills, and character slogan in the interface is free of typos, ambiguity, and fluent."

[0082] In some implementations, test cases may also be represented in tabular format.

[0083] Table 1 provides an example of three test cases, including test case 1, test case 2, and test case 3. Table 1

[0084] Referring to Table 1, each test case can include a description of the test interface, a description of the test content, and a description of the expected result, etc. For example, in test case 1, the test interface is described as "game character display interface," the test content is described as "checking the names of each character," and the expected result is described as "text without typos, ambiguity, and fluent sentences." In test case 2, the test interface is described as "Michelle's character introduction interface," the test content is described as "checking the character name, character introduction, character skills, and character slogan," and the expected result is described as "text without typos, ambiguity, and fluent sentences." In test case 3, the test interface is described as "Michelle's skill 'Meow Meow Turret' introduction interface," the test content is described as "checking the skill name and skill introduction," and the expected result is described as "text without typos, ambiguity, fluent sentences, and no exceeding the frame."

[0085] Table 1 is only used as an example to illustrate the content of the test cases, and the embodiments of this application do not limit the data format of the test cases. In other embodiments, test cases can be written using various data formats.

[0086] In some implementations, test cases may include more detailed descriptions. On one hand, test cases may include detailed operational descriptions of obtaining the target test interface, such as "Open the character panel in the main game interface, find Michelle in the character panel, and click on Michelle's avatar to open Michelle's character profile interface." On the other hand, test cases may also include detailed descriptions of the test content, such as "Check if the character name is Michele, check if the character slogan exceeds the frame, check if the character introduction is ambiguous," and so on.

[0087] In other implementations, test cases may include more concise descriptions, such as "Testing the Michelle character profile interface".

[0088] S302. Run the first application according to the first test case and obtain the interface image of the target test interface.

[0089] The first application is the multilingual application to be tested, which can be a mobile application, web application, desktop software, or other software product that needs to support multiple language interfaces. Specifically, the electronic device can control the first application to jump to the target test interface described in the first test case, and then acquire the interface image of the target test interface. The method of controlling the first application to jump to the interface will be described in detail below, and will not be elaborated here.

[0090] S303. Perform translation detection on the interface image of the target test interface according to the first test case to obtain the first test result. The first test result includes one or more of the following types of translation problems in the target test interface: mistranslation, omission, text exceeding the frame, and logical problems between multiple translation objects.

[0091] Specifically, in the embodiments of this application, the electronic device can utilize a natural language model to recognize interface images and conduct multi-dimensional tests on the interface images, including but not limited to mistranslation detection, omission detection, text over-boundary detection, logic detection, etc., to detect possible translation problems in the interface images and obtain a first test result. Specific testing methods will be described in detail below and will not be elaborated here.

[0092] S304, Display the first test result.

[0093] Specifically, electronic devices can... Figures 2B-2C The second user interface shown displays the first test results; details can be found by viewing [link / reference]. Figures 2B-2C And to Figures 2B-2C The relevant information will not be repeated here.

[0094] See below. Figures 4-9 This paper introduces the specific implementation method of the multilingual testing method provided in the embodiments of this application.

[0095] Figure 4 The following examples illustrate several functional modules involved in the specific implementation of the multilingual testing method provided in this application. For instance... Figure 4 As shown, the specific implementation of the multilingual testing method provided in this application involves multiple functional modules, including but not limited to a testing tool, a first application, an image acquisition tool, and a test execution tool. The interaction between these functional modules may include: the testing tool receiving a first test case uploaded by a user, and then sending the first test case to the image acquisition tool. The image acquisition tool runs the first application according to the first test case and acquires the interface image of the target test interface. Then, the image acquisition tool can pass the interface image of the target test interface to the test execution tool. The test execution tool performs multilingual testing on the interface image of the target test interface and obtains a first test result. Finally, the testing tool displays the first test result.

[0096] Among these, testing tools can be used to provide users with interactive user interfaces, such as... Figure 2A The first user interface shown and Figure 2B The second user interface is shown. The first application can be the multilingual application to be tested described above.

[0097] Image acquisition tools may include intelligent agent applications. An intelligent agent is a proxy capable of perceiving its environment and taking actions to achieve a specific goal. It can be software, hardware, or a system, possessing autonomy, adaptability, and interactivity, and can proactively complete a series of tasks. In the embodiments of this application, the intelligent agent has the ability to schedule and run other applications in the system, and can be used to coordinate testing processes, schedule testing tasks, etc.

[0098] In some implementations, these functional modules can be entirely deployed on the terminal electronic device, such as a computer, mobile phone, or tablet. In other implementations, these functional modules can be partially deployed on the terminal electronic device and partially deployed on a server, including a cloud server and a local server. For example, one possible implementation is that the testing tool, the first application, and the image acquisition tool are deployed on the terminal electronic device, while the test execution tool is deployed on the server.

[0099] Figure 4 The multiple functional modules shown are merely one implementation of this application's embodiments. In actual applications, more or fewer modules may be included, and this is not a limitation. For example, in some implementations, the testing tool may also be part of an image acquisition tool. That is, the image acquisition tool may also provide an interactive user interface for receiving test cases, displaying test results, and so on. Figure 5 An exemplary method is illustrated for obtaining a first test case and, based on the first test case, obtaining the interface image of the target test interface. Figure 5 The steps shown include steps S501-S504 for obtaining the first test case, and steps S505-S507 for obtaining the interface image of the target test interface. These are elaborated below: S501, The testing tool receives the first test case uploaded by the user through the user interface.

[0100] For example, testing tools can be used Figure 2A The first user interface shown receives the first test case uploaded by the user; details can be viewed here. Figure 2A The first user interface shown and the previous introduction to the first user interface will not be repeated here.

[0101] S502, The testing tool stores the first test case uploaded by the user in the database.

[0102] In this embodiment, step S502 is optional. That is, the testing tool can process the received test cases in two ways: a first processing method and a second processing method.

[0103] One approach is for the testing tool to receive test cases uploaded by users and then store them in a database. This allows the stored test cases to be reused, eliminating the need for users to write or upload test cases again in subsequent tests, thus improving testing convenience.

[0104] The second approach is for the testing tool to receive the test cases uploaded by the user, but instead of storing the test cases in the database, directly execute the test cases.

[0105] In some implementations, the testing tool may provide users with a method to choose whether to store test cases in a database. Specifically, on the user interface of the testing tool (e.g., ... Figure 2A The first user interface shown may include a control for selecting whether to store test cases in the database. The testing tool can then determine whether to store user-uploaded test cases in the database by detecting user actions on this control. This allows users to selectively store reusable standard test cases in the database while leaving temporary test cases for ad-hoc testing unstored, thus improving testing convenience while efficiently conserving database storage space.

[0106] The database can be a cloud database or a local database, and in this embodiment, it can be used to store test cases, interface images, and test results.

[0107] S503, Start Test.

[0108] This application provides two triggering methods for starting tests, namely a first triggering method and a second triggering method, to adapt to different test scenarios and requirements.

[0109] Specifically, the first trigger method can be that the testing tool triggers the test through a user interface (such as...). Figure 2A After receiving the first test case uploaded by the user (as shown in the first user interface), the testing tool immediately starts executing the first test case. Specifically, the testing tool can directly send the received first test case to the image acquisition tool, instructing the image acquisition tool to start the test.

[0110] The second triggering method is that after the testing tool detects a user action that triggers the start of the test, it sends a command to the image acquisition tool to start the test, instructing the image acquisition tool to initiate the test. For example, the testing tool can detect a user action that triggers the start of the test. Figure 2A After the user operates the start test control 203 in the first user interface shown, a start command is sent to the image acquisition tool, instructing the image acquisition tool to start the test.

[0111] S504, The image acquisition tool obtains the first test case.

[0112] In some implementations, the image acquisition tool obtains the first test case from the testing tool. Specifically, the testing tool receives the first test case uploaded by the user and directly sends the received first test case to the image acquisition tool.

[0113] In some implementations, the image acquisition tool retrieves a first test case from a database. Specifically, the start command sent by the testing tool to the image acquisition tool may include an identifier of the first test case to be executed, such as the first test case number, name, ID, etc. Then, the image acquisition tool can query and retrieve the first test case from the database based on the identifier of the first test case. In some implementations, the start command may also include identifiers of multiple test cases, and the image acquisition tool retrieves multiple test cases from the database based on the start command. For example, after a user selects multiple test cases, such as test case 10000, test case 10001, and test case 10002, through the test case selection area 202 of the first user interface, and clicks the start test control 203, the start command sent by the testing tool to the image acquisition tool may include the identifiers of test case 10000, test case 10001, and test case 10002.

[0114] In summary, the image acquisition tool's acquisition of the first test case can include the following two scenarios: Scenario 1: The testing tool receives test cases uploaded by the user and directly sends the received test cases to the image acquisition tool.

[0115] Scenario 2: After receiving test cases uploaded by the user, the testing tool first stores the test cases in the database. Then, when the testing tool detects a user action that triggers the start of the test, it notifies the image acquisition tool to retrieve the test cases from the database.

[0116] S505, The image acquisition tool controls the startup of the first application and acquires the first interface image of the first interface of the first application.

[0117] In some implementations, the image acquisition tool can launch a first application by simulating user actions. For example, the image acquisition tool can simulate a user clicking the desktop icon corresponding to the first application to launch the first application. Specifically, both the image acquisition tool and the first application can be deployed on an electronic device, which also has a first operating system installed. The first operating system provides a first interface. The image acquisition tool can use the first interface provided by the first operating system to simulate a series of input device events, including but not limited to mouse events (such as movement and click) or touch events (such as touch and swipe gestures). After these events are recognized by the first operating system, they are distributed and processed as if they were real user actions.

[0118] In some embodiments, both the first application and the image acquisition tool are deployed on an electronic device. The electronic device also has a first operating system installed, which includes a second interface. The first operating system can receive control commands sent by the image acquisition tool through the second interface and perform corresponding operations on the first application according to the received control commands. That is, the image acquisition tool can send control commands to the first operating system through the second interface, instructing the first operating system to launch the first application. The control commands sent by the image acquisition tool to the first operating system through the second interface may include the identifier of the first application, such as the application package name of the first application.

[0119] In this embodiment, the first interface is the first interface displayed by the first application after the image acquisition tool controls its launch. Specifically, if the image acquisition tool controls the first application to launch as a cold start (i.e., launch from a completely closed state), the first interface can be the initial interface of the first application (e.g., a game lobby, a shopping application homepage, etc.). If the image acquisition tool controls the first application to launch as a warm start (i.e., switch from a background running state to the foreground), the first interface is the interface that the application was previously displaying (e.g., a game task introduction interface, a product details page, etc.).

[0120] In some implementations, the first interface image of the first interface can be a screenshot of the first interface, or it can be the raw bitmap data or vector graphics data output by the graphics rendering pipeline, or it can be other image data that can reflect the content of the interface.

[0121] S506. The image acquisition tool, based on the first test case and the first interface image of the first interface, calls the first model to determine the target test interface of the first application and acquires the interface image of the target test interface.

[0122] In this embodiment, the first model can be a pre-trained natural language model capable of understanding test cases, recognizing interface images, and generating operation instructions for instructing an image acquisition tool to run the first application and acquire interface images. Here, a natural language model refers to an artificial intelligence model capable of natural language understanding, generation, and analysis.

[0123] Specifically, in S506, the first model can identify the first test case and the first interface image, determine the operation instructions for jumping from the first interface to the target test interface, and then the image acquisition tool controls the first application to jump to the target test interface according to the operation instructions and acquires the interface image of the target test interface. Figure 6 The following is an example illustrating the specific steps of the image acquisition tool calling the first model to acquire the interface image of the target test interface, including S601-S605: S601, The image acquisition tool inputs the first test case and the first interface image of the first interface into the first model.

[0124] S602, The first model determines the jump path 1 from the first interface to the target test interface based on the first test case.

[0125] S603, The first model sends the operation command corresponding to jump path 1 to the image acquisition tool.

[0126] After receiving the first test case and the interface image of the first interface, the first model can understand the semantics of the test case through its natural language understanding capabilities. It can identify the text content, UI element layout, control meanings, and relative positions of the controls in the interface image of the first interface. Based on the semantic understanding of the first test case and the recognition results of the interface image of the first application, it determines the jump path 1 from the first interface to the target test interface and generates operation instructions to control the first application to jump to the target test interface. Jump path 1 can be specifically implemented through controls in the interface. In other words, the first model identifies the target test interface indication information from the first test case, then identifies the target control that can control the jump to the target test interface from the interface image of the first interface based on the target test interface indication information, and finally generates operation instructions based on the target control.

[0127] In this embodiment, the operation instruction can be a structured action command, which may include an action and its corresponding parameters, used to instruct the image acquisition tool to perform the action to achieve the corresponding interface jump. For example, the operation instruction can be in JSON format.

[0128] The first test case is: "Open the Michelle character profile interface and check whether the text of the character name, character introduction, character skills, and character slogan in the interface is free of typos, ambiguities, and fluent sentences." The interface image of the first interface is: Figure 7A Taking the interface image shown as an example, the first model determines the operation goal as "open the Michelle character profile interface" by learning the semantics of the first test case. Then, the first model identifies... Figure 7A The interface image shown indicates that the control 701 on the lower left of the interface image can be used to open Michelle's character profile. Then, the first model can identify the coordinates of the control 701 in the first interface, such as [x1, y1], and generate operation instructions based on the coordinates of the control 701 in the first interface, such as {"action":"tap", "coordinates": [x1, y1]}, where the action parameter can be used to indicate the action type, i.e., click, and the coordinates parameter can be used to indicate the position of the action, i.e., the position with coordinates [x1, y1] in the first interface.

[0129] In some implementations, the operation instructions can also correspond to more complex interactions, such as {"action": "input", "coordinates": [x2, y2], "text": "test_user"}, which instructs the user to enter text in the input box at the specified coordinates; {"action": "long_press", "coordinates": [x, y], "duration": 500}, which instructs the user to press and hold at the specified coordinates for 500 milliseconds, and so on.

[0130] In some implementations, the first model may integrate a glossary of the first application. This glossary refers to a collection of key terms within the first application, including proper nouns, specific concepts, user interface element identifiers, functional module names, and their standard translations in a multilingual environment. By integrating this glossary, the first model can enhance its understanding of the application's vocabulary, reduce ambiguity and misidentification, and thus more accurately identify the text content of the application's interface images, thereby generating more precise operational instructions.

[0131] S604, the image acquisition tool controls the first application to jump to the target test interface according to the operation command corresponding to jump path 1.

[0132] S605. The image acquisition tool acquires the interface image of the target test interface.

[0133] The method by which the image acquisition tool controls the first application to jump to the target test interface is similar to the method in S505 where the image acquisition tool controls the launch of the first application. In some embodiments, the image acquisition tool can simulate corresponding input device events, such as simulating mouse clicks or gesture touch events, through a first interface provided by the first operating system, based on operation instructions, triggering the interface jump. Taking the operation instruction {"action": "tap", "coordinates": [x1,y1]} as an example, the image acquisition tool parses the operation instruction and converts it into a simulated device input event for the first application, thereby implementing the operation of clicking the position with coordinates [x1, y1] on the first interface as indicated by the operation instruction.

[0134] S507, The image acquisition tool stores the interface image of the target test interface into the database.

[0135] In this embodiment, a relationship can be established between interface images stored in the database and their corresponding test cases. For example, when storing an interface image in the database, a set of structured metadata can be attached to it. This metadata includes at least one field for recording the unique identifier of its corresponding test case. By establishing the relationship between the interface image and the test case, the interface image corresponding to a test case can be directly queried from the database based on this relationship.

[0136] Table 2 provides an example of the association between test cases in the database and interface images. Table 2

[0137] Referring to Table 2, the database stores a first relationship table, which records the association between test cases and interface images. When it is necessary to retrieve a test case and its corresponding target test interface image from the database, the first relationship table can be queried to determine which interface images in the database correspond to the target test interface image of that test case. As shown in Table 2, a unique identifier can be used to establish an association between a test case and an interface image in the first relationship table. For example, the unique identifier of the first test case can be 0x001, which indicates that the interface image corresponding to the first test case in the database includes interface image 1. Similarly, the unique identifier of the second test case can be 0x002, which indicates that the interface image corresponding to the second test case in the database includes interface image 2; the unique identifier of the third test case can be 0x003, which indicates that the interface image corresponding to the third test case in the database includes interface image 3; the unique identifier of the fourth test case can be 0x004, which indicates that the interface image corresponding to the fourth test case in the database includes interface image 4, and so on.

[0138] It should be noted that in some implementations, there can be multiple interface images corresponding to a single test case. When a test case corresponds to multiple interface images, all of these interface images are associated with that test case.

[0139] This application does not limit the method for establishing the association between interface images and test cases. In some implementations, it can also be achieved through various other methods, such as packaging the test cases and their corresponding interface images into a file for storage, etc.

[0140] In some implementations, the jump path 1 from the first interface to the target test interface may involve multiple interface jumps. In this case, the first model can generate multiple operation instructions in sequence. The image acquisition tool controls the first application to perform multiple interface jumps in sequence according to these multiple operation instructions generated in sequence, so as to jump from the first interface to the target test interface.

[0141] For example, when jump path 1 includes two interface jumps, the following steps may be included: Step 1: The image acquisition tool inputs the first test case and the first interface image into the first model.

[0142] Step 2: The first model identifies the first test case and the first interface image, determines the first control, and generates the first operation instruction based on the first control.

[0143] Step 3: The image acquisition tool controls the first application to jump from the first interface to the second interface according to the first operation instruction, and acquires the second interface image of the second interface.

[0144] Step 4: The image acquisition tool inputs the second interface image into the first model.

[0145] Step 5: The first model identifies the first test case and the second interface image, determines the second control, and generates a second operation instruction based on the second control.

[0146] Step 6: The image acquisition tool controls the first application to jump from the second interface to the target test interface according to the second operation command.

[0147] Figures 7A-7C This example illustrates the process of displaying the target test interface after two interface transitions. Figure 7A An example is shown of the first interface, Figure 7B An example is shown of the second interface, Figure 7C An example of the target test interface is shown.

[0148] Taking the first test case as "Open Michelle's skill 'Meow Meow Turret' introduction interface and check whether the skill name and skill description are free of typos, ambiguities, fluent sentences, and no exceeding the frame," after the image acquisition tool controls the launch of the first application, the first interface of the first application is... Figure 7A The interface shown is as follows. After the image acquisition tool acquires the first interface image, it inputs it along with the first test case into the first model. The first model parses the first interface image and the first test case, identifies control 701 from the first interface image, and generates a first operation instruction based on control 701, instructing the image acquisition tool to simulate clicking control 701 to jump to and display the second interface, i.e. Figure 7B The image acquisition tool displays the second interface image and then acquires the second interface image. The second interface is not the target test interface described in the first test case. Therefore, the image acquisition tool continues to input the second interface image into the first model. The first model continues to identify control 702 from the second interface image according to the first test case and generates a second operation instruction, instructing the image acquisition tool to simulate clicking control 702, causing the interface of the first application to jump to the target test interface described in the first test case. Figure 7C The 'Meow Meow Turret' introduction interface is shown.

[0149] The above is an introduction to jump path 1, which includes two interface jumps. In some embodiments, jump path 1 may include more interface jumps. In this case, the image acquisition tool will continuously input the newly acquired interface image into the first model and execute the feedback operation instructions (such as the third operation instruction, the fourth operation instruction, etc.), while acquiring subsequent interface images (the third interface image, the fourth interface image, etc.) until it jumps to the target test interface and acquires the interface image of the target test interface.

[0150] In some implementations, a single test case can be used to obtain an image of the interface to be tested. For example, the test case "Open Michelle's character profile interface and check whether the text of the character name, character introduction, character skills, and character slogan in the interface is free of typos, ambiguity, and fluent sentences" in the example above is used to obtain an image of Michelle's character profile interface, and the test case "Open Michelle's skill 'Meow Meow Turret' introduction interface and check whether the skill name and skill introduction are free of typos, ambiguity, fluent sentences, and do not exceed the frame" is used to obtain an image of Michelle's skill 'Meow Meow Turret' introduction interface.

[0151] In some implementations, a single test case can be used to acquire interface images of multiple target test interfaces. For example, a test case for acquiring multiple target test interfaces could be: "First, open the Michelle character profile interface, then sequentially open the Michelle skill introduction interfaces, and check whether the text in these interfaces is free of typos, ambiguities, overflows, and is grammatically correct." Based on the content of this test case, the first model can sequentially send multiple operation commands to the image acquisition tool to control the interface transitions and acquire the interface images of the Michelle character profile interface and multiple skill introduction interfaces as the interface images to be tested.

[0152] In some implementations, the image acquisition tool can acquire multiple test cases at once, and then execute these test cases sequentially to acquire the interface image of the target test interface. For example, based on the start command sent by the test tool, the image acquisition tool retrieves a test file from the database, which contains multiple test cases, and then the image acquisition tool can execute the multiple test cases in the test file sequentially.

[0153] In some implementations, the image acquisition tool may only store the interface image of the target test interface in the database, while interface images that do not need to be tested may not be stored in the database. The operation instructions sent by the first model to the image acquisition tool may include first indication information, used to indicate whether to store the interface image acquired in this operation in the database.

[0154] In some implementations, the image acquisition tool can control one or more language versions of the first application to run and acquire interface images of the target test interface in one or more language versions. Specifically, when the test case does not specify a target test language version, such as when the test case is "testing the Michelle character profile interface," the image acquisition tool can sequentially acquire interface images of the target test interface in all language versions supported by the first application. When the test case specifies a target test language version, such as when the test case is "testing the French version of the Michelle character profile interface," the image acquisition tool can acquire the interface image of the target test interface in the target language version (e.g., the French version) specified by the test case.

[0155] The above is an introduction to the method of obtaining the interface image of the target test interface based on the first test case. By utilizing the understanding ability of the natural language model to identify the controls in the interface image, the natural language test case is automatically converted into a series of interface jump operations. This not only automates the test and improves convenience, but also adapts to dynamic changes in the UI (such as layout adjustment and control position change), thereby improving the reusability of the test case.

[0156] In this embodiment of the application, after obtaining the interface image of the target test interface, multilingual testing can be further performed on the interface image of the target test interface. Figure 8 The following is an example of the process for performing multilingual testing on the interface images of the target test interface, which will be elaborated below: S801, The test execution tool retrieves the first test case and the interface image of the target test interface corresponding to the first test case from the database.

[0157] Specifically, the test execution tool can retrieve the interface image of the target test interface corresponding to the first test case based on the association between the first test case and the interface images stored in the database. For example, the test execution tool can query the database based on the unique identifier of test case A and retrieve test case A and its corresponding interface image 1 of the target test interface (e.g., ...). Figure 7C The interface image shown.

[0158] In some implementations, after the image acquisition tool obtains the interface image of the target test interface, it can send instruction information to the test execution tool, instructing the test execution tool to retrieve the first test case and the interface image of the target test interface of the first test case from the database. The instruction information may include a unique identifier for the first test case, enabling the test execution tool to determine the first test case and the corresponding interface image of the target test interface. For example, the instruction information may include the unique identifier 0x001 of the first test case. Based on 0x001, a lookup of the first relationship table shown in Table 2 determines that the interface image of the target test interface corresponding to the first test case includes interface image 1.

[0159] S802, The test execution tool inputs the first test case and the interface image of the target test interface corresponding to the first test case into the second model.

[0160] The second model is a pre-trained natural language model capable of understanding test cases and performing multilingual testing on interface images based on those test cases. The second model can integrate the terminology of the first application. By integrating this terminology, the second model enhances its understanding of the first application's vocabulary, reduces ambiguity and misidentification, and thus more accurately identifies the interface images of the first application before performing text semantic testing, translation testing, and so on.

[0161] S803. The second model detects one or more of the following translation problems in the interface image of the target test interface based on the first test case: mistranslation, omission, text exceeding the frame, and logical problems, and obtains the first multilingual test result.

[0162] Figure 9 An example is shown illustrating the workflow for performing multilingual tests using the second model.

[0163] like Figure 9 As shown, the second model can perform multi-dimensional translation tests on the interface image of a target test interface in parallel, such as mistranslation test, omission test, over-frame test, logic test, etc., and then integrate the test results of multiple dimensions to form the multilingual test results of this interface image.

[0164] Among them, the over-frame test is used to detect whether the text content exceeds the display area boundary, the mistranslation test is used to verify the accuracy of the translation, and the omission test is used to check whether there is any untranslated text content.

[0165] Logic testing is used to detect the logical relationships between multiple translated texts in an interface image, ensuring the semantic, functional, and contextual completeness and rationality of the translated content. For example, in continuous dialogue scenarios (such as game NPC dialogues, tutorial processes, etc.), it analyzes the logic between multiple dialogue texts to check whether they are coherent and connected. Another example is scene conformity testing, checking whether the text matches the visual and functional scenes displayed on the interface. Yet another example is for texts with functional attribute descriptions, checking whether the descriptions are consistent with the actual parameters and operation results (e.g., whether the damage value in a skill description matches the actual change in health after being damaged, etc.).

[0166] To achieve the aforementioned multi-dimensional detection, the second model can be pre-trained using pre-designed training samples. These training samples can include positive training samples, which are error-free interface images used to train the second model to learn correct translations. They can also include negative training samples, which can include interface images containing translation errors and expected detection results, used to train the second model to correctly identify translation problems from the interface images. Specifically, different training samples can be set for each dimension of detection. For example, for logic detection, interface images containing dialogue logic errors, text mismatches with the scene, and incorrect function descriptions can be used as training samples.

[0167] The multi-dimensional detection workflow shown above is only an exemplary introduction. In other embodiments, the second model may also include more detailed or broader multi-dimensional detection workflows, such as performing semantic tests to analyze the fluency and clarity of sentences, and performing terminology tests to detect whether the translation of professional terms is accurate.

[0168] Specifically, in this embodiment, the second model can determine which dimensions of the interface image of the target test interface to test based on the first test case. In other words, the test case can serve as a prompt to guide the second model to execute the test.

[0169] In some implementations, test cases include a description of the test content. The second model can determine which dimensions of testing to perform on the interface image corresponding to the test case based on the description of the test content. For example, if the test case is "Open the Michelle character profile interface and check whether the text of the character name, character introduction, character skills, and character slogan in the interface is free of typos, ambiguity, and fluent sentences," by parsing the test case, the second model can determine whether to perform mistranslation testing and semantic testing on the interface image of the test case.

[0170] In other implementations, the test cases do not include a description of the test content. In this case, the second model can perform tests on all dimensions of the interface image corresponding to the test case. For example, when the test case is "Test the Michelle character profile interface", the second model can perform tests on all dimensions of the interface image, including the aforementioned overframe test, mistranslation test, omission test, logic test, semantic test, terminology test, etc.

[0171] In some implementations, the second model and the first model can be the same model, which can integrate the terminology of the first application; it has the capability of the first model, that is, it can generate operation instructions based on test cases and interface images, instructing the image acquisition tool to perform operations, thereby acquiring the target test image of the test cases; it also has the capability of the second model, that is, it can test the interface images based on test cases and obtain test results.

[0172] S804, the second model sends the first multilingual test result to the test execution tool.

[0173] In this embodiment, the multilingual test results output by the second model may include various semantic problems (such as mistranslation, omission, ambiguity, logical problems between multiple translation objects, etc.) and display problems (such as over-frame display, omission display, etc.) detected from the interface image, as well as specific descriptions of each problem (including the location of the problem in the interface, problem description, etc.), modification suggestions, problem severity level, etc.

[0174] S805, the test execution tool stores the first multilingual test results in the database.

[0175] In this embodiment, test results stored in the database can be associated with their corresponding test cases. The method for establishing this association is similar to the method for associating interface images with suggested test cases, and will not be elaborated here.

[0176] Long-term storage is achieved by storing test results in a database for later review.

[0177] S806, The testing tool retrieves the first multilingual test results from the database.

[0178] In some embodiments, the database or test execution tool can send the first multilingual test result to the test tool immediately after the test is completed.

[0179] In other embodiments, the testing tool may send a request to the database based on user actions to obtain the first multilingual test results.

[0180] For example, testing tools can be used via... Figure 2AThe first user interface shown detects a user action on the test result viewing control 204 after it has been activated, or through... Figure 2C After the second user interface shown detects the user action on the query control 205E, it sends a request to the database to obtain the first multilingual test results.

[0181] S807, the test tool displays the first multilingual test results.

[0182] Specifically, testing tools can be used through Figures 2B-2F The second user interface shown displays the results of the first multilingual test. The method for displaying the first multilingual test results using the testing tool can be viewed here. Figures 2B-2F The second user interface is shown, along with an introduction to it.

[0183] For example, based on the test case "Test the Michelle character profile interface", the test result for the Michelle character profile interface can be: Figure 2B The content displayed in card 206A, according to the test case "Testing the introduction interface of Michelle's skill 'Meow Meow Turret'", the test result of the introduction interface of Michelle's skill 'Meow Meow Turret' can be: Figure 2B The content displayed in card 206B.

[0184] like Figure 2B As shown, the second model detected two display issues in Michelle's character profile interface. Issue 1 indicates that the "Character Description (#044)" section at the top of the interface is missing, pointing out that this should be a description of the character's personality, but only the placeholder "Character Description (#044)" is displayed, without showing the actual description content, making it impossible for users to understand the character's characteristics. Issue 2 indicates that the text in the dialog box next to the portrait on the right side of the interface is displayed beyond the frame, and suggests adjusting the size of the dialog box. The second model detected a semantic issue in Michelle's skill 'Meow Meow Turret' description interface, indicating a semantic error in the text on the right side of the interface, and suggests changing "throw" to "deploys".

[0185] As described above regarding the second user interface, in some embodiments, the second user interface may include controls for marking test results. Through these controls, users can view the test results and determine their accuracy. If the test result is inaccurate, the control can mark it as "Analysis Error" and add it to the error result set. If the test result is accurate, the control can mark it as "Confirmed Problem" and add it to the correct result set. By marking the results, the testing capabilities of the second model can be further trained and adjusted based on the error result set and the correct result set, continuously improving the accuracy of the second model in testing interface images. Specifically, after detecting a user action on the "Analysis Error" control, the testing tool can also prompt the user to select or input the error reason and expected test result, and add the collected error reason, expected test result, and error result set to the error result set. Then, the error result set can be used to further train the second model, improving its testing accuracy.

[0186] The above describes the implementation steps of the multilingual testing method provided in this application embodiment. First, the image acquisition tool and the first model work together to convert natural language test cases into precise interface operation instructions, automatically navigating to the target interface and capturing images. Subsequently, the second model analyzes the interface images, intelligently detecting multi-dimensional issues such as translation accuracy, text semantics, and layout display. The entire process achieves end-to-end integration of test case management, automated execution, intelligent analysis, and result traceability, significantly reducing the testing threshold while significantly improving testing efficiency.

[0187] Figure 10 An exemplary embodiment of this application provides a multilingual testing framework.

[0188] like Figure 10 As shown, the multilingual testing framework 100 may include a user interaction module 101, a data management module 102, an image acquisition module 103, and a testing module 104.

[0189] The user interaction module 101 may include testing tools to provide users with an interactive user interface, such as a first user interface and a second user interface. The first user interface receives test cases uploaded by the user and stores them in a database. It can also retrieve historically uploaded test cases from the database through the first user interface. The second user interface displays test results to the user and provides relevant controls, such as result labeling controls and filtering controls, so that the user can perform subsequent processing based on the test results.

[0190] The data management module 102 may include a database for storing test cases received through the user interaction module 101, target test interface images received through the image acquisition module 103, and test results obtained through the test module 104. It manages these test cases, target test interface images, and test results, and establishes the association between test cases, target test interface images, and test results for easy subsequent querying.

[0191] The image acquisition module 103 may include a first model and an image acquisition tool, used to control the execution of a first application and acquire the target test interface of the first application. Specifically, the first model in the image acquisition module 103 can identify target controls from the interface image of the current interface of the first application that has been acquired, based on test cases, and generate operation instructions based on the target controls. Then, the image acquisition tool controls the interface of the first application to jump to a new interface according to the operation instructions. The above steps are then repeated until the target test interface is reached and the image of the target test interface is acquired.

[0192] The testing module 104 may include a second model and a test execution tool, used to acquire interface images from the data management module 102, test the interface images, and store the test results in the data management module 102. Specifically, the test execution tool in the testing module 104 can obtain test cases and interface images from the database of the data management module 102, and then input the test cases and interface images into the second model. The second model tests the interface images according to the description of the test cases and generates test results. Finally, the test execution tool stores the test results output by the second model into the database of the data management module 102.

[0193] Figure 10 The multilingual testing framework 100 shown is merely one implementation of an embodiment of this application. In practical applications, the multilingual testing framework 100 may include more or fewer modules, which is not limited here.

[0194] This application also provides an electronic device 200 and a server 300. In some embodiments, multiple modules of the multilingual testing framework 100 can be deployed entirely on the electronic device 200, and the electronic device 200 executes all steps of the multilingual testing method provided in this application. In other embodiments, multiple modules of the multilingual testing framework 100 can be deployed entirely on the server 300, and the server 300 executes all steps of the multilingual testing method provided in this application. In still other embodiments, several modules of the multilingual testing framework 100 can be partially deployed on the electronic device 200 and partially deployed on the server 300, with the electronic device 200 and the server 300 working collaboratively to jointly execute the multilingual testing method provided in this application.

[0195] Figure 11 An example is shown of a method for electronic devices and servers to work together.

[0196] like Figure 11 As shown, the electronic device 200 may include testing tools, a first application, and an image acquisition tool, while the server 300 may include a first model, a database, a test execution tool, and a second model. The electronic device 200 interacts with the server 300 to jointly implement the multilingual testing method provided in this application embodiment.

[0197] For the process of electronic device 200 and server 300 jointly implementing the multilingual testing method provided in the embodiments of this application, please refer to... Figure 11 The multiple steps involved, as well as the content of the method embodiments described above, will not be repeated here.

[0198] Figure 12 An electronic device provided by an embodiment of this application is illustrated by way of example.

[0199] like Figure 12 As shown, the electronic device 200 may include: input / output modules (including an audio input / output module 218, a key input module 216, and a display 220, etc.), a communication interface 201, a user interface 202, one or more processors 204, a transmitter 206, a receiver 208, a coupler 210, an antenna 214, and a memory 212. These components can be connected via a bus or other means. Figure 12 Taking a bus connection as an example: The communication interface 201 can be used by the electronic device 200 to communicate with other communication devices, such as base stations. Specifically, the communication interface 201 may include one or more of the following: Global System for Mobile Communication (GSM) (2G) communication interface, Wideband Code Division Multiple Access (WCDMA) (3G) communication interface, and Long Term Evolution (LTE) (4G) communication interface, or it may be a 4.5G, 5G, or future New Radio (NR) communication interface. Not limited to wireless communication interfaces, the electronic device 200 may also be configured with a wired communication interface 201, such as a Local Access Network (LAN) interface.

[0200] Antenna 214 can be used to convert electromagnetic energy in a transmission line into electromagnetic waves in free space, or to convert electromagnetic waves in free space into electromagnetic energy in a transmission line. Coupler 210 is used to split the mobile communication signal received by antenna 214 into multiple paths and distribute them to multiple receivers 208.

[0201] Transmitter 206 can be used to transmit signals output by processor 204, and receiver 208 can be used to receive mobile communication signals received by antenna 214. In some embodiments of this application, transmitter 206 and receiver 208 can be considered as a wireless modem. In electronic device 200, the number of transmitter 206 and receiver 208 can be one or more.

[0202] Apart from Figure 12 The transmitter 206 and receiver 208 shown may also include other communication components, such as a GPS module, a Bluetooth module, or a Wi-Fi module. Beyond the wireless communication signals described above, the electronic device 200 may also support other wireless communication signals, such as satellite signals, shortwave signals, etc. In addition to wireless communication, the electronic device 200 may also be configured with a wired network interface (such as a LAN interface) to support wired communication.

[0203] The input / output module is used to realize the interaction between the electronic device 200 and the user / external environment, and may mainly include an audio input / output module 218, a key input module 216, and a display 220. Specifically, the input / output module may also include a camera, a touch screen, and sensors, etc. All input / output modules communicate with the processor 204 through the user interface 202.

[0204] Memory 212 is coupled to processor 204 and is used to store various software programs and / or multiple sets of instructions. Specifically, memory 212 may include high-speed random access memory and may also include non-volatile memory, such as one or more disk storage devices, flash memory devices, or other non-volatile solid-state storage devices. Memory 212 may store an operating system (hereinafter referred to as the system), such as embedded operating systems like Android, iOS, Windows, or Linux. Memory 212 may also store a network communication program, which can be used to communicate with one or more additional devices, one or more terminal devices, or one or more network devices. Memory 212 may also store a user interface program, which can realistically display the content of the application through a graphical user interface and receive user control operations on the application through input controls such as menus, dialog boxes, and buttons.

[0205] In some embodiments of this application, memory 212 may be used to store the implementation program of the multilingual testing method provided in one or more embodiments of this application on the electronic device 200 side.

[0206] The processor 204 can be used to read and execute computer-readable instructions. Specifically, the processor 204 can be used to call a program stored in the memory 212, such as the implementation program of the multilingual testing method provided in one or more embodiments of this application on the electronic device 200 side, and execute the instructions contained in the program.

[0207] It needs to be explained that, Figure 12 The electronic device 200 shown is merely one implementation of the embodiments of this application. In actual applications, the electronic device 200 may include more or fewer components, which is not limited here.

[0208] Figure 13 An exemplary embodiment of the server provided in this application is shown.

[0209] like Figure 13 As shown, server 300 may include: input / output modules (including audio input / output module 318, key input module 316, and display 320, etc.), communication interface 301, user interface 302, one or more processors 304, transmitter 306, receiver 308, coupler 310, antenna 314, and memory 312. These components can be connected via a bus or other means. Figure 13 Taking a bus connection as an example: The communication interface 301 can be used by the server 300 to communicate with other communication devices, such as base stations. Specifically, the communication interface 301 may include one or more of the following: Global System for Mobile Communication (GSM) (2G) communication interface, Wideband Code Division Multiple Access (WCDMA) (3G) communication interface, and Long Term Evolution (LTE) (4G) communication interface, or a 4.5G, 5G, or future New Radio (NR) communication interface. Not limited to wireless communication interfaces, the server 300 may also be configured with a wired communication interface 301, such as a Local Access Network (LAN) interface.

[0210] Antenna 314 can be used to convert electromagnetic energy in a transmission line into electromagnetic waves in free space, or to convert electromagnetic waves in free space into electromagnetic energy in a transmission line. Coupler 310 is used to split the mobile communication signal received by antenna 314 into multiple paths and distribute them to multiple receivers 308.

[0211] Transmitter 306 can be used to transmit signals output by processor 304, and receiver 308 can be used to receive mobile communication signals received by antenna 314. In some embodiments of this application, transmitter 306 and receiver 308 can be considered as a wireless modem. In server 300, the number of transmitter 306 and receiver 308 can be one or more.

[0212] Apart from Figure 13 The transmitter 306 and receiver 308 shown may be accompanied by other communication components in the server 300, such as a GPS module, a Bluetooth module, or a Wi-Fi module. Not limited to the wireless communication signals described above, the server 300 may also support other wireless communication signals, such as satellite signals, shortwave signals, etc. In addition to wireless communication, the server 300 may also be configured with a wired network interface (such as a LAN interface) to support wired communication.

[0213] The input / output module is used to realize the interaction between the server 300 and the user / external environment, and may mainly include an audio input / output module 318, a key input module 316, and a display 320. Specifically, the input / output module may also include a camera, a touch screen, and sensors, etc. All input / output modules communicate with the processor 304 through the user interface 302.

[0214] Memory 312 is coupled to processor 304 and is used to store various software programs and / or multiple sets of instructions. Specifically, memory 312 may include high-speed random access memory and may also include non-volatile memory, such as one or more disk storage devices, flash memory devices, or other non-volatile solid-state storage devices. Memory 312 may store an operating system (hereinafter referred to as the system), such as embedded operating systems like Android, iOS, Windows, or Linux. Memory 312 may also store a network communication program, which can be used to communicate with one or more additional devices, one or more terminal devices, or one or more network devices. Memory 312 may also store a user interface program, which can realistically display the content of the application through a graphical user interface and receive user control operations on the application through input controls such as menus, dialog boxes, and buttons.

[0215] In some embodiments of this application, memory 312 may be used to store the implementation program of the multilingual testing method provided in one or more embodiments of this application on the server 300 side.

[0216] Processor 304 can be used to read and execute computer-readable instructions. Specifically, processor 304 can be used to invoke a program stored in memory 312, such as the implementation program of the multilingual testing method provided in one or more embodiments of this application on the server 300 side, and execute the instructions contained in the program.

[0217] It needs to be explained that, Figure 13 The server 300 shown is merely one implementation of the embodiments of this application. In actual applications, the server 300 may include more or fewer components, which is not limited here.

[0218] This application also provides a computer program product comprising one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or data center that integrates one or more available media. For example, available media may include magnetic media (e.g., floppy disk, hard disk, or magnetic tape), optical media (e.g., digital versatile disc (DVD)), or semiconductor media (e.g., solid-state disk (SSD)).

[0219] This application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program or instructions. When the computer program or instructions are executed by a processor, they implement the methods described above. The method flow described in the above embodiments can be implemented wholly or partially by software, hardware, firmware, or any combination thereof. If implemented in software, the functionality can be stored as one or more instructions or code on or transmitted over the computer-readable medium. The computer-readable medium can include computer storage media and communication media, and can also include any medium that can transfer a computer program from one place to another. The storage medium can be any target medium accessible by a computer.

[0220] In some implementations, computer-readable media may include compact disc read-only memory (CD-ROM), RAM, ROM, EEPROM, or other optical disc storage; computer-readable media may include disk storage or other disk storage devices. Furthermore, any connecting cable may also be appropriately referred to as computer-readable media. For example, if software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of media. As used herein, disks and optical discs include optical discs (CD), laser discs, optical discs, DVDs, floppy disks, and Blu-ray discs, where disks typically reproduce data magnetically, while optical discs optically reproduce data using lasers. Combinations of the above should also be included within the scope of computer-readable media.

[0221] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0222] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially 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 a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0223] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

[0224] The above detailed embodiments further illustrate the purpose, technical solution, and beneficial effects of the embodiments of this application. It should be understood that the above are merely specific embodiments of the embodiments of this application and are not intended to limit the protection scope of the embodiments of this application. Any modifications, equivalent substitutions, improvements, etc., made on the basis of the technical solutions of the embodiments of this application should be included within the protection scope of the embodiments of this application.

Claims

1. A multilingual testing method, characterized in that, Applied to electronic devices, the method includes: Obtain a first test case for multilingual testing of the first application, the first test case including target test interface instruction information written in natural language and / or test content of the target test interface; Run the first application according to the first test case and obtain the interface image of the target test interface; The interface image of the target test interface is translated and detected according to the first test case to obtain a first test result; the first test result includes one or more of the following translation problems in the target test interface: mistranslation, omission, text exceeding the frame, and logical problems between multiple translation objects; The results of the first test are displayed.

2. The method as described in claim 1, characterized in that, The step of running the first application according to the first test case and obtaining the interface image of the target test interface specifically includes: Run the first application and obtain the first interface image of the first interface in the first application; The first model is used to identify the first test case and the first interface image to determine the operation instructions used to control the first application to jump from the first interface to the target test interface; The operation command controls the first application to jump to the target test interface; Obtain the interface image of the target test interface.

3. The method as described in claim 2, characterized in that, The step of using a first model to identify the first test case and the first interface image, determining the operation command for jumping from the first interface to the target test interface, and controlling the first application to jump to the target test interface according to the operation command specifically includes: The first model is used to identify the first test case and the first interface image to determine the first operation instruction; According to the first operation instruction, control the first application to jump from the first interface to the second interface, and obtain the second interface image of the second interface; The first model is used to identify the first test case and the second interface image to determine the second operation instruction; The first application is controlled to jump from the second interface to the target test interface according to the second operation instruction.

4. The method as described in claim 3, characterized in that, The step of using the first model to identify the first test case and the first interface image to determine the first operation instruction specifically includes: The first model is used to identify the target test interface indication information in the first test case, and the first control in the first interface image is identified according to the target test interface indication information. The first control is used to control the first application to jump from the first interface to the second interface. Generate a first operation instruction, which is an instruction used to operate on the first control.

5. The method according to any one of claims 1-4, characterized in that, The step of running the first application and acquiring the interface image of the target test interface specifically includes: Run one or more language versions of the first application and obtain the interface image of the target test interface for the one or more language versions.

6. The method according to any one of claims 1-5, characterized in that, The step of translating and detecting the interface image of the target test interface according to the first test case specifically includes: The second model is used to identify test content for the target test interface from the first test case, and the target test interface is translated and tested according to the test content.

7. The method as described in claim 6, characterized in that, Both the first model and the second model integrate a glossary of terms from the first application, which includes standard translations of proper nouns from the first application in multiple language versions.

8. The method as described in any one of claims 6 or 7, characterized in that, The first model and the second model are the same model.

9. The method according to any one of claims 6-8, characterized in that, The first model and the second model are deployed on a server, and the method further includes: The electronic device sends the first test case to the server; The electronic device receives the operation instructions returned by the server; The electronic device sends the interface image of the target test interface to the server; The electronic device receives the first test result returned by the server.

10. The method according to any one of claims 1-9, characterized in that, The first test case is either user input or retrieved from a database.

11. The method according to any one of claims 1-10, characterized in that, The method further includes: determining and displaying the type of translation problem, the severity level of the translation problem, and modification suggestions for the translation problem in the first test results.

12. The method according to any one of claims 1-11, characterized in that, The method further includes displaying a control for marking whether the first test result is correct.

13. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method according to any one of claims 1-12.

14. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the steps of the method described in any one of claims 1-12.

15. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the steps of the method according to any one of claims 1-12.

Images