Coding-free multi-language interface automatic testing method and system

Abstract

The invention discloses a coding-free multi-language interface automatic testing method and system, and belongs to the technical field of software testing, the realization of the method comprises a recursive traversal crawler algorithm, a crawler tool and a standardized language comparison model, and the method can realize quasi-user operation, traverse a tested product interface and capture internationalized interface element information of multi-language software; through a standardized language comparison model, whether the element information conforms to the context in various language environments and whether translation deviation exists are judged, and semantic accuracy is guaranteed; wherein the recursive traversal crawler algorithm execution logic comprises the following steps: logging in a tested system; opening a function menu; reading a current function title; and judging whether a new interface appears or not according to the function title, and if yes, automatically identifying interactive elements and static text information in the interface. According to the method, manual intervention can be reduced, the test efficiency is effectively improved, meanwhile, the test coverage rate is remarkably increased, and efficient and reliable technical guarantee is provided for internationalized quality control of products.

Description

Technical Field

[0001] This invention relates to the field of software testing technology, specifically to a no-code, multilingual interface automated testing method and system. Background Technology

[0002] In a globalized software development environment, product internationalization has become a core requirement for expanding into overseas markets. The accuracy and consistency of interface language translation directly impact user experience and brand image. Currently, traditional internationalization testing primarily relies on manual page-by-page verification or automated script-based testing, which faces the following technical bottlenecks:

[0003] (1) Manual testing is inefficient: Testers need to visually compare buttons, forms, pop-ups and other elements of multilingual interfaces one by one, which is time-consuming, labor-intensive and easy to miss. In particular, the test coverage of complex interfaces or multilingual versions (such as Chinese, English, French, Japanese and Korean) is insufficient, which makes it difficult to meet the iteration rhythm of agile development.

[0004] (2) Coding dependency leads to high technical threshold: Traditional automated testing requires writing a lot of code (such as Selenium scripts), which requires testers to have programming skills, and the code maintenance cost is high (the script needs to be modified synchronously when the interface elements change, making it difficult for small and medium-sized enterprises or non-technical teams to implement.

[0005] (3) Insufficient semantic verification capability: Existing tools mostly rely on string matching for translation verification, which cannot identify contextual differences (such as mistranslation of professional terms, deviation of colloquial expressions, etc.) and make it difficult to ensure the semantic accuracy of the translation. For example, when “submit order” is mistranslated as “Submit Order Form” (the correct translation should be “Submit Order”), traditional tools may not be able to locate it accurately.

[0006] (4) Incomplete multilingual adaptation coverage: For dynamically loaded content (such as asynchronously rendered table data, dynamic pop-ups) or niche languages ​​(such as Arabic, Russian and other right-to-left layout languages), existing solutions often miss detections due to insufficient parsing capabilities, thus limiting the comprehensiveness of internationalization testing. Summary of the Invention

[0007] The technical objective of this invention is to provide a no-code automated testing method and system for multilingual interfaces, which can greatly reduce manual intervention, effectively improve testing efficiency, and significantly increase test coverage, providing efficient and reliable technical support for international quality control of products; it is especially suitable for automated detection and translation quality verification of interface language information in multilingual software products.

[0008] The technical solution adopted by the present invention to solve its technical problem is:

[0009] A no-code automated testing method for multilingual interfaces is disclosed. The method includes a recursive traversal crawler algorithm, a crawler tool, and a standardized language comparison model. It simulates user operations, traverses the interface of the product under test, and captures internationalized interface element information of the multilingual software. Through the standardized language comparison model, it determines whether the element information conforms to the context in various language environments and whether there are translation biases, ensuring semantic accuracy. The execution logic of the recursive traversal crawler algorithm includes:

[0010] (1) Log in to the system under test. The default browser is Chrome. Currently, all mainstream browsers are supported. Select the language to be tested on the login screen.

[0011] (2) Open the function menu;

[0012] (3) After opening the function, read the current function title; determine whether a new interface appears based on the function title. If it does, automatically identify interactive elements (buttons, form controls, tabs, etc.) and static text (table headers, pop-up prompts) in the interface. Add or modify the information to be collected according to the product characteristics. Save the captured information to Excel. Save the current function title to memory so that the information of this function title interface will not be captured again in the future.

[0013] (4) Determine whether there is a pop-up or other window on the current interface. If there is, save the pop-up content to Excel and close the pop-up or other window to avoid pop-ups affecting subsequent element operations.

[0014] (5) Determine if there are still interactive elements on the current screen, iterate through and click on the interactive elements on the current screen, save the interactive elements on the current screen and determine if they have been clicked; if all interactive elements on the current screen have been operated on, close the current screen.

[0015] The script runs using a self-developed runtime platform, supporting containerized concurrent execution and improving script execution efficiency.

[0016] This innovative method employs browser crawling technology to deeply analyze the interface of the product under test, accurately capturing interface element information such as button names, form control names, table headers, tab labels, radio / selection label names, and pop-up content. It automatically saves interface element information from various language environments in the product under test to Excel. Then, using a standardized language comparison model, it accurately locates translation deviations and semantic errors during the product's internationalization process. This significantly reduces manual intervention, effectively improves testing efficiency, and achieves a substantial increase in test coverage, providing efficient and reliable technical support for the quality control of product internationalization.

[0017] Furthermore, the execution logic of the recursive traversal crawler algorithm also includes:

[0018] Each time an interactive element is operated, it is determined whether a new function title appears; if it appears, the recursive method of the crawler is entered, that is, the return logic (3) is returned, and the current function title is read; the interactive elements and static text in the interface are automatically identified and saved to Excel, and the current function title is saved to memory;

[0019] A pop-up window may appear when operating on an element. If a pop-up window appears, proceed to logic (4).

[0020] Furthermore, the recursive traversal crawler algorithm delves deeper into each of the interfaces that appear. When all interactive elements of the last interface have been operated on, the interface is closed, and the previous interface is returned to continue to judge and operate on the interactive elements that have not yet been clicked. The algorithm will exit only after all interactive elements of all interfaces have been operated on.

[0021] Furthermore, the recursive traversal crawler algorithm captures element information and saves it to Excel, including the following columns: function interface title (Simplified Chinese), function title location information, function interface element control location information, element control type, including language such as Simplified Chinese, English, Traditional Chinese, Arabic, etc., with each language in a separate column.

[0022] The above columns contain the content of the first sheet page; the content of the second sheet page is: element information in any two languages ​​as columns A and B, column C as the accuracy rate, and column D as correction suggestions (accuracy rate <95%).

[0023] Taking Chinese and English as an example, the interface element information captured in both Chinese and English is combined with a standardized language comparison model and prompts. The large model returns the accuracy rate of the Chinese-English translation and provides correction suggestions (accuracy rate <95%). The prompts from the large model are as follows: "Please analyze columns E and F in the Excel file, where column E is the Chinese name and column F is the corresponding English translation. Please generate a table where the first column is the original column E, the second column is the original column F, the third column is the percentage value of the accuracy rate of the large model in determining whether the English translation is accurate, and the fourth column provides the correct value for those with an accuracy rate below 95%."

[0024] Furthermore, the open function menu uses a table comparing Chinese menu names with menu names in other languages, so switching languages ​​does not require script modification. One script can adapt to multiple language environments, thereby improving script reusability.

[0025] Furthermore, the crawler tool used is a third-party crawler tool, including Selenium and DrissionPage crawlers.

[0026] Selenium is an open-source tool widely used for web automation testing and data collection. Its core function is to simulate user actions in a browser, enabling automated interaction and information extraction of web page elements. Key features include: multi-browser compatibility, supporting major browsers such as Chrome, Firefox, and Edge; automated operation capabilities, providing rich API interfaces to simulate user actions such as clicking, inputting, and selecting from dropdowns on web page elements; and support for obtaining web page source code, element attributes, and executing JavaScript scripts to meet complex interface interaction needs.

[0027] DrissionPage is a high-efficiency web automation tool developed in Python. Compared to Selenium, it offers greater flexibility and ease of use in certain scenarios, and supports all the Selenium features mentioned above. In this approach, DrissionPage is used as a supplementary tool to Selenium, specifically in the following two ways: First, DrissionPage can obtain the location information of elements in the interface, which can be the absolute path of XPath or the absolute path of CSS selectors, providing a basis for subsequent judgment on whether they are the same element in different language environments; Second, DrissionPage can search for elements across iframes without needing to switch in and out, and can operate on multiple tabs simultaneously without switching. The use of DrissionPage is particularly important when the product under test contains a large number of iframe elements.

[0028] Furthermore, the standardized language comparison model selects a corresponding large model for the language currently being tested;

[0029] A private model is deployed locally, combined with the RAG knowledge base, to build a knowledge base for specialized or industry-specific terms appearing in the product. Product documents are saved to a vector database, forming a vertical model for the industry domain based on the large model, which is used to check for semantic biases and translation errors.

[0030] By leveraging browser crawling technology to capture interface elements comprehensively and combining it with a standardized language comparison model, this approach overcomes the dual limitations of manual work and coding reliance, significantly improving testing efficiency and semantic verification accuracy. It provides a new technical path for the international quality control of multilingual software.

[0031] This invention also claims protection for a no-code multilingual interface automated testing system, including a recursive traversal crawler algorithm module, a crawler tool, and a standardized language comparison model;

[0032] This system achieves automated testing of multilingual interfaces without coding using the methods described above.

[0033] The present invention also claims a coding-free multilingual interface automated testing device, comprising: at least one memory and at least one processor;

[0034] The at least one memory is used to store a machine-readable program;

[0035] The at least one processor is used to call the machine-readable program to implement the above method.

[0036] The present invention also claims a computer-readable medium storing computer instructions that, when executed by a processor, enable the implementation of the above-described method.

[0037] Compared with the prior art, the no-coding multilingual interface automated testing method and system of the present invention has the following advantages:

[0038] This invention achieves a technological breakthrough in the field of internationalization testing by combining browser crawling technology with a standardized language comparison model.

[0039] In terms of efficiency, automated processes replace over 80% of manual operations. Compared to traditional solutions, a single round of internationalization testing involves at least 5,000 functions, saving half an hour of manual work per function. Multiple rounds of internationalization testing are typically conducted with product updates. This method saves at least several thousand hours compared to traditional solutions, significantly reducing labor costs.

[0040] In terms of quality, by combining a multi-dimensional standardized language model, semantic deviations are accurately located, the error detection rate of translation is raised to a new level, and bold attempts are made for dynamically loaded content and right-to-left language for the first time.

[0041] In terms of technology popularization, the no-code design enables non-technical personnel to directly perform tests, promoting the transformation of international testing from "professional technical positions" to "participation of all personnel", especially reducing the technical entry threshold for non-technical teams by more than 70%.

[0042] At the engineering level, standardized Excel reports and defect management interfaces enable test results to directly drive the R&D process, forming a quality closed loop of "detection-repair-verification" and helping enterprises build a high-standard international management and control system. Attached Figure Description

[0043] Figure 1 This is a flowchart illustrating the no-coding multilingual interface automated testing method provided in this embodiment of the invention;

[0044] Figure 2 This is an example image showing how the execution result of this invention generates an Excel spreadsheet for each function menu.

[0045] Figure 3 This is an example image of the large model determination result sheet in Excel provided in an embodiment of the present invention. Detailed Implementation

[0046] The present invention will be further described below with reference to specific embodiments.

[0047] This invention provides a no-code automated testing method for multilingual interfaces. The method includes a recursive traversal crawler algorithm, a crawler tool, and a standardized language comparison model. It can simulate user operations, traverse the interface of the product under test, and capture internationalized interface element information of the multilingual software. Then, through the standardized language comparison model, it determines whether the element information conforms to the context in various language environments and whether there are translation errors, ensuring semantic accuracy. The specific implementation is as follows:

[0048] 1. This method is based on a self-developed recursive traversal crawler algorithm, such as... Figure 1 As shown, the execution logic of the recursive traversal crawler algorithm includes:

[0049] (1) The system under test is logged in using the Chrome browser by default, which is supported by all major browsers. Select the language to be tested on the login screen.

[0050] (2) Open the function menu. Here, a table comparing Chinese menu names with menu names in other languages ​​is used. Switching languages ​​does not require modifying the script. One script can adapt to multiple language environments, improving script reusability.

[0051] (3) After opening the function, read the current function title. Determine if a new interface appears based on the function title. If so, automatically identify interactive elements (buttons, form controls, tabs, etc.) and static text (table headers, pop-up prompts) in the interface. The information to be collected can be customized, added, or modified according to product characteristics. The captured information is saved to Excel. The current function title is saved in memory so that information is not captured again from this function title interface in the future.

[0052] (4) Determine whether there is a pop-up or other window on the current interface. If there is, save the pop-up content to Excel and close the pop-up or other window to avoid pop-ups affecting subsequent element operations.

[0053] (5) Check if there are any interactive elements on the current screen. Iterate through and click on the interactive elements on the current screen, and save whether each interactive element has been clicked. If all interactive elements on the current screen have been interacted with, close the current screen.

[0054] (6) Each time an interactive element is operated, it is determined whether a new function title appears. If it appears, the recursive method of the crawler is entered, i.e., logic (3), to read the current function title; the interactive elements and static text in the interface are automatically identified and saved to Excel, and the current function title is saved to memory.

[0055] (7) A pop-up window may appear when operating on an element. If a pop-up window appears, proceed to logic (4).

[0056] (8) This recursive traversal algorithm goes deeper and deeper according to the interface that appears. When all interactive elements of the last interface have been operated, the interface will be closed and the previous interface will be returned to continue to judge the interactive elements that have not yet been clicked and continue to operate until all interactive elements of all interfaces have been operated, and then the algorithm will exit.

[0057] (9) The captured element information is saved to Excel, mainly including the following columns: function interface title (Simplified Chinese), function title location information, function interface element control location information, element control type, and a separate column for each language such as Simplified Chinese, English, Traditional Chinese, and Arabic. Figure 2 As shown.

[0058] Taking Chinese and English as an example, the interface element information captured in both Chinese and English is combined with a standardized language comparison model and prompts. The large model returns the accuracy rate of the Chinese-English translation and provides correction suggestions (accuracy rate <95%). The prompts from the large model are as follows: "Please analyze columns E and F in the Excel file, where column E is the Chinese name and column F is the corresponding English translation. Please generate a table where the first column is the original column E, the second column is the original column F, the third column is the percentage value of the accuracy rate of the large model in determining whether the English translation is accurate, and the fourth column provides the correct value for those with an accuracy rate below 95%."

[0059] Output result:

[0060] The core of the output report is an Excel file. The first sheet contains the execution logic (9th) line, the second sheet contains element information in any two languages ​​as columns A and B, column C is the accuracy rate, and column D is the correction suggestion (accuracy <95%). Figure 3 As shown.

[0061] 2. Third-party web crawling tools are used, including Selenium and DrissionPage.

[0062] Selenium is an open-source tool widely used for web automation testing and data collection. Its core function is to simulate user actions in a browser, enabling automated interaction and information extraction of web page elements. Key features include: multi-browser compatibility, supporting major browsers such as Chrome, Firefox, and Edge; automated operation capabilities, providing rich API interfaces to simulate user actions such as clicking, inputting, and selecting from dropdowns on web page elements; and support for obtaining web page source code, element attributes, and executing JavaScript scripts to meet complex interface interaction needs.

[0063] DrissionPage is a high-efficiency web automation tool developed in Python. Compared to Selenium, it offers greater flexibility and ease of use in certain scenarios, and supports all the Selenium features mentioned above. In this approach, DrissionPage is used as a supplementary tool to Selenium, specifically in the following two ways: First, DrissionPage can obtain the location information of elements in the interface, which can be the absolute path of XPath or the absolute path of CSS selectors, providing a basis for subsequent judgment on whether they are the same element in different language environments; Second, DrissionPage can search for elements across iframes without needing to switch in and out, and can operate on multiple tabs simultaneously without switching. The use of DrissionPage is particularly important when the product under test contains a large number of iframe elements.

[0064] 3. The standardized language comparison model:

[0065] Most mainstream models support major global languages, but different models perform differently with different languages. Therefore, it's necessary to select a high-performing model for the language being tested. We recommend deploying a private model locally, combining it with the RAG knowledge base, and building your own knowledge base for specialized or industry-specific terms appearing in the product. Store product documents in a vector database, and build your own industry-specific vertical model based on the large model. This is crucial for checking semantic biases and translation errors.

[0066] 4. The script runs using a self-developed runtime platform, supporting containerized concurrent execution and improving script execution efficiency.

[0067] A specific application example of this method can be found in the reference section. Figure 2 and Figure 3 As shown:

[0068] Configure the test environment information and run the Python recursive traversal crawler algorithm program. A self-developed script execution platform supports headless browser execution and concurrent execution in Docker containers, improving operational efficiency.

[0069] The result is an Excel spreadsheet generated for each function menu item, with the following content: Figure 2 As shown.

[0070] The content of the large model determination results sheet in Excel is as follows: Figure 3 As shown.

[0071] This method overcomes the technical shortcomings of existing internationalization testing, such as high manual costs, strong coding dependencies, and insufficient semantic verification. It provides a coding-free, fully automated, and high-precision internationalization testing method, specifically achieving the following technical goals:

[0072] Eliminate reliance on manual intervention and improve testing efficiency—by automatically capturing multilingual information (including all elements such as buttons, forms, and pop-ups) from the interface of the product under test, the inefficient traditional manual page-by-page verification mode is replaced, thereby improving testing efficiency and meeting the testing needs of high-frequency iterations in agile development.

[0073] Lowering the technical barrier and enabling universal testing—adopting a no-code design eliminates the need for testers to have programming skills or write automation scripts. By using browser crawling technology and comparing standardized language models, non-technical personnel (such as product managers and operations personnel) can quickly execute international tests, expanding the applicability of the testing subject.

[0074] Accurately pinpoint translation deviations and ensure semantic accuracy—by leveraging a multi-dimensional standardized language comparison model, not only is string consistency verified, but also contextual differences are analyzed in depth (such as the accuracy of professional terminology and the adaptability of language habits). For example, it can identify whether “user agreement” is mistranslated as “User Agreement Document” (a redundant translation), ensuring that the translation results conform to the semantic habits and industry standards of the target language.

[0075] Comprehensive coverage of multilingual scenarios, improving test coverage—supports automated parsing of multilingual interfaces such as Chinese, English, Japanese, and Arabic, solving the problem of missed detections in special scenarios by traditional solutions.

[0076] Establish a standardized quality control system – by automatically synchronizing the captured language information to an Excel document and generating a visual comparison report, a reusable international testing standard process is formed, providing the product development team with an intuitive basis for translation quality assessment and helping to build a standardized multilingual adaptation management system.

[0077] This invention also provides a no-code multilingual interface automated testing system, including a recursive traversal crawler algorithm module, a crawler tool, and a standardized language comparison model; the system achieves no-code multilingual interface automated testing through the no-code multilingual interface automated testing method described in the above embodiments.

[0078] The execution logic of the recursive traversal crawler algorithm module includes:

[0079] (1) The system under test is logged in using the Chrome browser by default, which is supported by all major browsers. Select the language to be tested on the login screen.

[0080] (2) Open the function menu. Here, a table comparing Chinese menu names with menu names in other languages ​​is used. Switching languages ​​does not require modifying the script. One script can adapt to multiple language environments, improving script reusability.

[0081] (3) After opening the function, read the current function title. Determine if a new interface appears based on the function title. If so, automatically identify interactive elements (buttons, form controls, tabs, etc.) and static text (table headers, pop-up prompts) in the interface. The information to be collected can be customized, added, or modified according to product characteristics. The captured information is saved to Excel. The current function title is saved in memory so that information is not captured again from this function title interface in the future.

[0082] (4) Determine whether there is a pop-up or other window on the current interface. If there is, save the pop-up content to Excel and close the pop-up or other window to avoid pop-ups affecting subsequent element operations.

[0083] (5) Check if there are any interactive elements on the current screen. Iterate through and click on the interactive elements on the current screen, and save whether each interactive element has been clicked. If all interactive elements on the current screen have been interacted with, close the current screen.

[0084] (6) Each time an interactive element is operated, it is determined whether a new function title appears. If it appears, the recursive method of the crawler is entered, i.e., logic (3), to read the current function title; the interactive elements and static text in the interface are automatically identified and saved to Excel, and the current function title is saved to memory.

[0085] (7) A pop-up window may appear when operating on an element. If a pop-up window appears, proceed to logic (4).

[0086] (8) This recursive traversal algorithm goes deeper and deeper according to the interface that appears. When all interactive elements of the last interface have been operated, the interface will be closed and the previous interface will be returned to continue to judge the interactive elements that have not yet been clicked and continue to operate until all interactive elements of all interfaces have been operated, and then the algorithm will exit.

[0087] (9) The captured element information is saved to Excel, which mainly includes the following columns: function interface title (Simplified Chinese), function title location information, function interface element control location information, element control type, and a separate column for each language such as Simplified Chinese, English, Traditional Chinese, and Arabic.

[0088] Taking Chinese and English as an example, the interface element information captured in both Chinese and English is combined with a standardized language comparison model and prompts. The large model returns the accuracy rate of the Chinese-English translation and provides correction suggestions (accuracy rate <95%). The prompts from the large model are as follows: "Please analyze columns E and F in the Excel file, where column E is the Chinese name and column F is the corresponding English translation. Please generate a table where the first column is the original column E, the second column is the original column F, the third column is the percentage value of the accuracy rate of the large model in determining whether the English translation is accurate, and the fourth column provides the correct value for those with an accuracy rate below 95%."

[0089] Output result:

[0090] The core of the output report is an Excel file. The first sheet contains the execution logic (9), the second sheet contains element information in any two languages ​​as columns A and B, column C is the accuracy rate, and column D is the correction suggestion (accuracy <95%).

[0091] The crawler tool used is a third-party crawler tool, including Selenium and DrissionPage.

[0092] Selenium is an open-source tool widely used for web automation testing and data collection. Its core function is to simulate user actions in a browser, enabling automated interaction and information extraction of web page elements. Key features include: multi-browser compatibility, supporting major browsers such as Chrome, Firefox, and Edge; automated operation capabilities, providing rich API interfaces to simulate user actions such as clicking, inputting, and selecting from dropdowns on web page elements; and support for obtaining web page source code, element attributes, and executing JavaScript scripts to meet complex interface interaction needs.

[0093] DrissionPage is a high-efficiency web automation tool developed in Python. Compared to Selenium, it offers greater flexibility and ease of use in certain scenarios, and supports all the Selenium features mentioned above. In this approach, DrissionPage is used as a supplementary tool to Selenium, specifically in the following two ways: First, DrissionPage can obtain the location information of elements in the interface, which can be the absolute path of XPath or the absolute path of CSS selectors, providing a basis for subsequent judgment on whether they are the same element in different language environments; Second, DrissionPage can search for elements across iframes without needing to switch in and out, and can operate on multiple tabs simultaneously without switching. The use of DrissionPage is particularly important when the product under test contains a large number of iframe elements.

[0094] The standardized language comparison model:

[0095] Most mainstream models support major global languages, but different models perform differently with different languages. Therefore, it's necessary to select a high-performing model for the language being tested. We recommend deploying a private model locally, combining it with the RAG knowledge base, and building your own knowledge base for specialized or industry-specific terms appearing in the product. Store product documents in a vector database, and build your own industry-specific vertical model based on the large model. This is crucial for checking semantic biases and translation errors.

[0096] The script runs using a self-developed runtime platform, supporting containerized concurrent execution and improving script execution efficiency.

[0097] This invention also provides an automated testing device for a code-free multilingual interface, comprising: at least one memory and at least one processor;

[0098] The at least one memory is used to store a machine-readable program;

[0099] The at least one processor is used to call the machine-readable program to implement the code-free multilingual interface automated testing method described in the above embodiments.

[0100] This invention also provides a computer-readable medium storing computer instructions. When executed by a processor, the computer instructions cause the processor to perform the code-free multilingual interface automated testing method described in the above embodiments. Specifically, a system or apparatus equipped with a storage medium storing software program code that implements the functions of any of the embodiments described above can be provided, and the computer (or CPU or MPU) of the system or apparatus can read and execute the program code stored in the storage medium.

[0101] In this case, the program code read from the storage medium can itself implement the function of any of the above embodiments, and therefore the program code and the storage medium storing the program code constitute part of the present invention.

[0102] Examples of storage media used to provide program code include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, program code can be downloaded from a server computer via a communication network.

[0103] Furthermore, it should be clear that not only can the program code read by the computer be executed, but also the operating system or other components operating on the computer can be instructed based on the program code to perform some or all of the actual operations, thereby realizing the function of any of the embodiments described above.

[0104] Furthermore, it is understood that the program code read from the storage medium is written to the memory set in the expansion board inserted into the computer or to the memory set in the expansion unit connected to the computer. Then, based on the instructions of the program code, the CPU or other components installed on the expansion board or expansion unit execute some and all of the actual operations, thereby realizing the function of any of the embodiments described above.

[0105] The present invention has been shown and described in detail above with reference to the accompanying drawings and preferred embodiments. However, the present invention is not limited to these disclosed embodiments. Based on the above embodiments, those skilled in the art will know that more embodiments of the present invention can be obtained by combining the code review methods in the different embodiments. These embodiments are also within the protection scope of the present invention.

Claims

1. A no-code, multilingual interface automated testing method, characterized in that, The implementation of this method includes a recursive traversal crawler algorithm, crawler tools, and a standardized language comparison model. It can simulate user operations, traverse the interface of the product under test, and capture internationalized interface element information of multilingual software. Through the standardized language comparison model, it can determine whether the element information is consistent with the context in various language environments and whether there are translation biases, thus ensuring semantic accuracy. The execution logic of the recursive traversal crawler algorithm includes: (1) Log in to the system under test and select the language to be tested on the login screen; (2) Open the function menu; (3) After opening the function, read the current function title; determine whether a new interface appears based on the function title. If it appears, automatically identify the interactive elements and static text information in the interface, add or modify the information to be collected according to the product characteristics, and save the captured information to Excel; save the current function title to memory so that the information of this function title interface will not be captured again in the future. (4) Determine whether there is a pop-up or other window on the current interface. If there is, save the pop-up content to Excel and close the pop-up or other window to avoid pop-ups affecting subsequent element operations. (5) Determine if there are still interactive elements on the current screen, iterate through and click on the interactive elements on the current screen, save the interactive elements on the current screen and determine if they have been clicked; if all interactive elements on the current screen have been operated on, close the current screen. The script supports containerized concurrent execution.

2. The automated testing method for a code-free multilingual interface according to claim 1, characterized in that, The execution logic of the recursive traversal crawler algorithm also includes: Each time an interactive element is operated, it is determined whether a new function title appears; if it appears, the recursive method of the crawler is entered, that is, the return logic (3) is returned, and the current function title is read; the interactive elements and static text in the interface are automatically identified and saved to Excel, and the current function title is saved to memory; A pop-up window may appear when operating on an element. If a pop-up window appears, proceed to logic (4).

3. The automated testing method for a code-free multilingual interface according to claim 1, characterized in that, The recursive traversal crawler algorithm delves deeper into each interface as it appears. Once all interactive elements on the last interface have been interacted with, the algorithm closes the interface, returns to the previous interface, and continues to check and interact with any unclicked interactive elements until all interactive elements on all interfaces have been interacted with, at which point the algorithm exits.

4. The automated testing method for a code-free multilingual interface according to claim 1, characterized in that, The recursive traversal crawler algorithm captures element information and saves it to Excel, including the following columns: function interface title, function title location information, function interface element control location information, element control type, including the language of Simplified Chinese, English, Traditional Chinese, and Arabic, with each language in a separate column. The above columns represent the content of the first sheet; the second sheet contains: element information in any two languages ​​as columns A and B, accuracy rate as column C, and correction suggestions as column D.

5. The automated testing method for a code-free multilingual interface according to claim 1, characterized in that, The open function menu uses a table comparing Chinese menu names with menu names in other languages. When switching languages, no script modification is required. A single script can be used to adapt to multiple language environments, thereby improving script reusability.

6. The automated testing method for a code-free multilingual interface according to claim 1, characterized in that, The crawler tool used is a third-party crawler tool, including Selenium and DrissionPage.

7. The automated testing method for a code-free multilingual interface according to claim 1, characterized in that, The standardized language comparison model selects the corresponding large model for the language currently being tested. A private model is deployed locally, combined with the RAG knowledge base, to build a knowledge base for specialized or industry-specific terms appearing in the product. Product documents are saved to a vector database, forming a vertical model for the industry domain based on the large model, which is used to check for semantic biases and translation errors.

8. A no-code, multilingual interface automated testing system, characterized in that, Includes a recursive traversal crawler algorithm module, crawler tools, and a standardized language comparison model; The system achieves automated testing of multilingual interfaces without coding through the method described in any one of claims 1 to 7.

9. A code-free, multilingual interface automated testing device, characterized in that, include: At least one memory and at least one processor; The at least one memory is used to store a machine-readable program; The at least one processor is configured to invoke the machine-readable program to implement the method according to any one of claims 1 to 7.

10. A computer-readable medium, characterized in that, The computer-readable medium stores computer instructions that, when executed by a processor, enable the implementation of the method described in any one of claims 1 to 7.

Images