Cross-platform software compatibility testing and optimization methods

Abstract

The invention discloses a cross-platform software compatibility test and optimization method, belongs to the technical field of software testing, and aims to solve the difficult problem of existing cross-platform software compatibility test; firstly, the software is installed on different test platforms, and function, interface and performance tests are performed; the function test simulates user operation, calculates average response delay time difference and success rate, and marks function abnormal points; the interface test divides detection points, analyzes color, clarity and coordinate values, and marks interface abnormal points; the performance test monitors the number of flashbacks and completion time, and marks performance abnormal points; then, the abnormal points are sent to a positioning display module, and different marking circles and colors are set according to the abnormal type and severity in a problem display model, and priority sorting is performed; maintenance personnel can optimize the software according to the data in the problem display model, so as to quickly locate the root cause of the problem, improve development efficiency, and ensure the quality and stability of the software on different platforms.

Description

Technical Field

[0001] The present invention belongs to the technical field of software testing, and in particular relates to a cross-platform software compatibility testing and optimization method. Background Art

[0002] In the field of software testing technology, cross-platform software compatibility testing and optimization are key links to ensure the normal operation of software on different operating systems, hardware environments, browsers and other platforms; by adopting various testing methods and optimization strategies, the quality and stability of the software can be improved to meet the user's usage needs on different platforms.

[0003] Existing cross-platform software faces many challenges in terms of compatibility. When compatibility issues arise, the existing testing methods often make it difficult to quickly and accurately locate the root cause of the problem, which results in a lot of time and effort required for troubleshooting. At the same time, when problems are found, it is also difficult to prioritize the problem handling, which greatly reduces development efficiency. For this reason, we propose a cross-platform software compatibility testing and optimization method. Summary of the invention

[0004] The purpose of the present invention is to provide a cross-platform software compatibility testing and optimization method to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a cross-platform software compatibility testing and optimization method, comprising the following steps:

[0006] Step 1: Install the software to be tested on different test platforms and conduct compatibility testing on the test platforms; compatibility testing includes: functional testing, interface testing, and performance testing;

[0007] Step 2: When performing functional testing, when the average response delay time difference of the test point execution operation is greater than the preset threshold, or the success rate of the test point execution operation is less than the preset threshold, it is marked as a functional abnormal test point, and the functional abnormal test point is sent to the problem display model;

[0008] Step 3: When performing interface testing, the interface of the test point on the test platform is divided into several detection points. The color value, clarity value and coordinate value of the detection point are comprehensively analyzed to obtain the interface deviation evaluation value of the test point. If the interface deviation evaluation value of the test point is greater than the preset threshold, the test point is marked as an interface abnormality test point, and the interface abnormality test point is sent to the problem display model;

[0009] Step 4: When performing performance testing, monitor the number of flashbacks of the test platform during the execution of the test point and the duration of the execution operation. If the number of flashbacks of the test point is greater than a preset threshold or the duration of the execution operation of the test point is greater than a preset threshold, mark the test point as a performance abnormality test point and send the performance abnormality test point to the problem display model.

[0010] Step 5: Display the functional abnormality test points, interface abnormality test points and performance abnormality test points in the problem display model, and scan them in real time to locate the abnormal test points where software testing problems occur, and prioritize the abnormal test points.

[0011] Preferably, the specific process of the compatibility test in step 1 is:

[0012] The test platform includes: different operating systems, hardware environments and browsers; when conducting compatibility tests on different test platforms, develop matching test cases. Test cases refer to a collection of specific test points in terms of software function implementation, interface presentation and performance in cross-platform usage scenarios; each test case contains several test points, and each test point corresponds to an execution operation behavior.

[0013] Preferably, the specific process of the functional test in step 2 is:

[0014] Start the software on different test platforms and simulate the user's operation behaviors according to the execution operations contained in the test cases. The operation behaviors include: inputting data, clicking buttons, and selecting menu options. For each test platform, multiple repetitive tests are performed when the corresponding test cases are tested;

[0015] For each test point, obtain the execution operation time of the test point and the time for the test platform to respond to the corresponding operation, and subtract them to obtain the response delay time difference of the test point execution operation, sum and average the response delay time differences of the test point execution operations for multiple repetitive tests, and obtain the average response delay time difference of the test point execution operations; at the same time, obtain the number of successful execution operations of the test point during multiple repetitive tests, and obtain the success rate of the test point execution operations by dividing the number of successful execution operations of the test point by the total number of repetitive tests; preset each test point execution operation has a corresponding response delay time difference threshold and execution operation success rate threshold, compare the average response delay time difference of the test point execution operation with the corresponding preset threshold, and compare the success rate of the test point execution operation with the corresponding preset operation success rate threshold, when the average response delay time difference of the test point execution operation is greater than the corresponding preset threshold, or the success rate of the test point execution operation is less than the corresponding preset threshold, mark the test point as a functionally abnormal test point, and display the functionally abnormal test point by sending it to the problem display model.

[0016] Preferably, the process of judging whether the test point execution operation is successful is as follows:

[0017] The response delay time difference of the test point execution operation is less than the corresponding preset test point execution operation response delay time difference threshold, and after the test point executes the operation, the result of the operation is consistent with the expected result of the operation; the consistency between the result of the operation and the expected result of the operation includes: consistency in functional integrity, consistency in data accuracy, and consistency in the effectiveness of the error handling mechanism.

[0018] Preferably, the specific process of the interface test in step 3 is:

[0019] When the test platform performs interface testing through test cases, for each test point, the interface where the test point is tested on the test platform is divided into a number of test points i, i=1, 2, ..., n; the color value, clarity value, and coordinate value of each test point in the test interface are recorded; each test point is preset to have a corresponding standard color value, standard clarity value, and standard coordinate value; the color value, clarity value, and coordinate value of the test point are compared with the preset standard color value, standard clarity value, and standard coordinate value of the corresponding test point to obtain the color deviation value SP of the test point. i , clarity deviation value FP i And the coordinate deviation value ZP i ; Using the formula: , obtain the test point interface deviation evaluation value MP, where a1, a2, and a3 are preset weight coefficients, and each test point interface deviation evaluation value is preset to correspond to a deviation evaluation threshold;

[0020] For each test point, the test point interface deviation evaluation value MP is compared with the corresponding preset deviation evaluation threshold. If MP is greater than the corresponding preset deviation evaluation threshold, the test point is marked as an interface abnormality test point, and the interface abnormality test point is sent to the problem display model for display.

[0021] The analysis process of the color deviation value, clarity deviation value and coordinate deviation value of the detection point is as follows:

[0022] Color deviation value: By comparing the RGB values ​​of each pixel constituting the detection point with the RGB values ​​of each pixel corresponding to the preset standard color value, the RGB deviation value of each pixel constituting the detection point is obtained, and by summing the RGB deviation values ​​of each pixel constituting the detection point, the color deviation value SP of the detection point is obtained. i ;

[0023] Clarity deviation value: Convert the detection point image and the image corresponding to the preset standard clarity value into grayscale images, apply the Canny algorithm to perform edge detection on the detection point image and the corresponding preset image, and then obtain the edge image of the detection point and the edge image of the corresponding preset image. Calculate the sum of the gradient values ​​of all pixels in the edge image of the detection point and the edge image of the corresponding preset image, and obtain the total gradient G1 of the detection point image. i The sum of the gradients with the preset image G2 i ; Using the formula: FP i =|G1 i -G2 i | / G2 i , get the clarity deviation value FP of the detection point i ;

[0024] Coordinate deviation value: Get the coordinates of the center point of the detection point and the coordinates of each edge vertex, and mark the coordinates of the center point and each edge vertex as feature point coordinates (X1 e , Y1 e ), where e=1, 2, ..., g; g represents the number of the feature point, and the preset standard coordinate value corresponds to the feature point coordinate (X2 e , Y2 e ), using the formula: , get the coordinate deviation value ZP of the detection point i .

[0025] Preferably, the specific process of the performance test in step 4 is:

[0026] For each test point, a threshold for the number of operations to be performed is preset, and each threshold for the number of operations to be performed for each test point has a threshold for the number of flashbacks and a threshold for the length of time it takes to complete the operation.

[0027] For each test point, monitor the number of times the test platform crashes and the time it takes to complete the operation during each test of the corresponding operation quantity threshold. Sum the number of crashes and the time it takes to complete the operation for all the operations during the performance test, and take the average as the number of crashes and the time it takes to complete the operation for the test point during the performance test.

[0028] The number of flashbacks and the time to complete the operation of the test point are compared with the corresponding preset flashback number threshold; at the same time, the time to complete the execution of the operation is compared with the corresponding preset execution of the operation completion time threshold;

[0029] If the number of test point flashbacks is greater than the corresponding preset flashback threshold or the test point execution operation completion time is greater than the corresponding preset execution operation completion time threshold, the test point will be marked as a performance abnormality test point, and the performance abnormality test point will be sent to the problem display model.

[0030] Preferably, in step 5, the process of positioning and processing the function abnormality test points, interface abnormality test points and performance abnormality test points displayed in the problem display model is as follows:

[0031] The software icon for cross-platform testing is placed at the center of the problem display model. Three coordinate axes extend outward from the software icon, namely the function test direction axis, the interface test direction axis, and the performance test direction axis, and the angle between each direction axis is 120°;

[0032] On the function test direction axis, the interface test direction axis and the performance test direction axis, different sizes of marking circles and colors are set for marking according to the types and severity of the function abnormality test points, the interface abnormality test points and the performance abnormality test points, and at the same time, the functional problems are prioritized on the function test direction axis according to the size of the marking circle. The larger the marking circle, the closer the position on the function test direction axis is to the software icon, and the higher the priority of the functional problem handling;

[0033] When processing the marked circle in the problem display model, the problem display model is scanned in real time. When a circle is marked in the problem display model, the direction axis, color and size corresponding to the marked circle are retrieved to obtain the test points corresponding to the marked circle, the abnormal types of the test points and the abnormal evaluation value data corresponding to the abnormal types of the test points. The test points corresponding to the marked circle, the abnormal types of the test points and the abnormal evaluation value data corresponding to the abnormal types of the test points are compressed into a data packet and sent to the maintenance personnel. After receiving the data packet, the maintenance personnel optimize the cross-platform test software according to the data in the data packet.

[0034] When a single marked circle appears in the problem display model, the data corresponding to the marked circle is directly sent to the maintenance personnel;

[0035] When multiple marked circles appear on the same direction axis in the problem display model, the test points corresponding to the marked circle with the largest radius, the abnormal types of the test points, and the abnormal evaluation value data corresponding to the abnormal types of the test points are compressed into a data packet and sent to the maintenance personnel; the maintenance personnel optimize the cross-platform test software according to the data in the data packet according to the order in which the data packets are sent;

[0036] When marked circles appear on the function test direction axis, the interface test direction axis, and the performance test direction axis in the problem display model, the priority processing method is as follows:

[0037] A proportional normalization adjustment coefficient is preset for each of the preset function test direction axis, interface test direction axis and mark circle on the performance test. The radius of the mark circle on each direction axis is multiplied by the corresponding normalization adjustment coefficient to obtain the radius of the mark circle on each direction axis after normalization adjustment. The priority is determined by comparing the radii of the mark circles on each direction axis after normalization adjustment, that is, the larger the radius, the higher the priority of problem handling. The problem handling priorities are sorted according to the size of the priority, and the test points corresponding to the mark circle with the highest priority, the abnormal types of the test points and the abnormal evaluation value data corresponding to the abnormal types of the test points are compressed into a data packet and sent to the maintenance personnel. The maintenance personnel optimize the cross-platform test software according to the data in the data packets according to the order in which the data packets are sent.

[0038] Preferably, the specific process of analyzing the corresponding color and the size of the marked circle for each functional abnormality test point is as follows:

[0039] A color set of function abnormality test key points is preset, wherein each function abnormality test key point in the color set corresponds to a color, and for each function abnormality test key point, the corresponding color is obtained by substituting the function abnormality test key point into the preset function abnormality test key point color set;

[0040] For each function abnormality test point, obtain the difference between the average response delay time difference of its execution operation and the corresponding preset threshold value, and mark it as the function abnormality test point response deviation value XP. At the same time, obtain the success rate CG of the execution operation of the abnormal test point, and use the formula CXP= (XP×b1-CG×b2)×r to obtain the function abnormality evaluation value CXP, where b1 and b2 are preset weight coefficients, and r is the function adjustment coefficient. Different function abnormality test points have different r values.

[0041] A number of functional abnormality evaluation value intervals are preset, each functional abnormality evaluation value interval corresponds to a marking circle radius, and for each functional abnormality test point, its corresponding functional abnormality evaluation value is substituted into the preset number of functional abnormality evaluation value intervals for matching, and its corresponding marking circle radius is output.

[0042] Preferably, the specific process of analyzing the corresponding color and the size of the marking circle for each interface abnormality test point is as follows:

[0043] A preset interface abnormality test key point color set, wherein each interface abnormality test key point in the interface abnormality test key point color set corresponds to a color, and for each interface abnormality test key point, the corresponding color is obtained by substituting the interface abnormality test key point into the preset interface abnormality test key point color set;

[0044] For each interface abnormality test point, obtain the difference between the detection point deviation evaluation value and the preset deviation evaluation threshold of the corresponding detection point, and mark it as the interface test deviation value ZP. Use the formula: ZMP=ZP×z to obtain the interface abnormality evaluation value ZMP, where z is the interface adjustment coefficient, and the z value corresponding to different interface abnormality test points is different;

[0045] Several interface anomaly evaluation value intervals are preset, each interface anomaly evaluation value interval corresponds to a marking circle radius, and for each interface anomaly test point, its corresponding interface anomaly evaluation value is substituted into several preset interface anomaly evaluation value intervals for matching, and its corresponding marking circle radius is output.

[0046] Preferably, the specific process of analyzing the corresponding color and the size of the marked circle for each performance abnormality test point is as follows:

[0047] A preset performance abnormality test key point color set, wherein each performance abnormality test key point in the performance abnormality test key point color set corresponds to a color, and for each performance abnormality test key point, the corresponding color is obtained by substituting the performance abnormality test key point into the preset performance abnormality test key point color set;

[0048] For each performance abnormality test point, obtain the difference between the number of flashbacks greater than the corresponding preset flashback threshold, and mark it as the abnormal performance test point flashback deviation value TP. At the same time, obtain the difference between the execution operation completion time greater than the corresponding preset execution operation completion time threshold, and mark it as the abnormal performance test point operation time deviation value WP. Use the formula WTP=(WP×c1+TP×c2)×w to obtain the performance abnormality evaluation value WTP, where c1 and c2 are preset weight coefficients, and w is the performance adjustment coefficient. Different performance abnormality test points have different w values.

[0049] Several performance anomaly evaluation value intervals are preset, each performance anomaly evaluation value interval corresponds to a marking circle radius, and for each performance anomaly test point, its corresponding performance anomaly evaluation value is substituted into several preset interface anomaly evaluation value intervals for matching, and its corresponding marking circle radius is output.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] (1) This cross-platform software compatibility testing and optimization method conducts compatibility testing on various test platforms such as different operating systems, hardware environments and browsers, comprehensively covering various environments in which the software may run, ensuring that the software can run normally on different platforms. The test cases include functional testing, interface testing and performance testing, evaluating the software from multiple aspects to ensure the comprehensiveness and completeness of the test.

[0052] (2) The cross-platform software compatibility testing and optimization method uses a problem display model with the software icon as the center to extend outwardly into a function test direction axis, an interface test direction axis, and a performance test direction axis. According to the type and severity of the function abnormality test points, the interface abnormality test points, and the performance abnormality test points, marking circles of different radii and colors are set on the corresponding direction axes for marking, so that the problem can be seen at a glance and the test status of the software in different aspects is intuitively displayed. By performing real-time scanning on the problem display model, when a marked circle is found, the test point, abnormality type, and abnormal evaluation value data corresponding to the marked circle can be quickly retrieved and compressed into a data packet and sent to the maintenance personnel, thereby improving the accuracy and efficiency of problem location.

[0053] (3) The cross-platform software compatibility testing and optimization method reflects the severity of the problem through the size of the marked circle. The closer the position is to the software icon, the higher the severity and the higher the priority. When multiple marked circles appear on the same direction axis in the problem display model, the data corresponding to the marked circle with the largest radius are sent to the maintenance personnel in turn to ensure that the most serious problem is handled first; when marked circles appear on the function test direction axis, the interface test direction axis and the performance test direction axis in the problem display model, the radius of the marked circles on each direction axis is normalized by a preset proportional normalization adjustment coefficient, and then the priority is determined by comparing the radius size, thereby ensuring the rationality and orderliness of problem handling, orderly solving various compatibility problems that occur in cross-platform software testing, reducing development delays caused by compatibility problems, accelerating the software development cycle, and improving development efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a flowchart of the present invention;

[0055] Figure 2 A schematic diagram showing a model for the problem of the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example

[0057] See also Figure 1 - Figure 2 , the present invention provides a cross-platform software compatibility testing and optimization method, including an installation module, a function testing module, an interface testing module, a performance testing module, and a problem display module;

[0058] The installation module will reasonably install the software for cross-platform testing on different test platforms. The test platforms include but are not limited to: different operating systems, hardware environments and browsers; and perform compatibility testing on the test platforms to detect whether the software can run normally on different platforms; compatibility testing includes: functional testing, interface testing, and performance testing; the specific process is:

[0059] In order to conduct compatibility tests on different test platforms, matching test cases are developed; test cases refer to a collection of specific test points in terms of software function implementation, interface presentation, and performance in cross-platform usage scenarios, which comprehensively evaluate the compatibility of the software on different platforms; whether it is the accuracy of functional operation, the display effect of interface elements or the compliance of performance indicators, etc., they are all tested through these carefully designed test cases.

[0060] The functional test module performs functional testing on different test platforms according to pre-defined functional test cases. The specific process is as follows:

[0061] Start the software on different test platforms and simulate the user's operation behaviors according to the execution operation behaviors contained in the test cases. The operation behaviors include but are not limited to: inputting data, clicking buttons, and selecting menu options. Each test case contains several test points, and each test point corresponds to an execution operation behavior. For each test platform, multiple repetitive tests are performed when testing the corresponding test case;

[0062] For each test point, the execution operation time of the test point is subtracted from the time the test platform responds to the corresponding operation, and the response delay time difference of the test point execution operation is obtained. The response delay time differences of the test point execution operations of multiple repetitive tests are summed and averaged to obtain the average response delay time difference of the test point execution operations. At the same time, the number of successful execution operations of the test point in multiple repetitive tests is obtained, and the test delay time difference is obtained by dividing the number of successful execution operations of the test point by the total number of repetitive tests. The success rate of the test point execution operation; each test point execution operation is preset to have a corresponding response delay time difference threshold and an execution operation success rate threshold, the average response delay time difference of the test point execution operation is compared with the corresponding preset threshold, and the success rate of the test point execution operation is compared with the corresponding preset operation success rate threshold; when the average response delay time difference of the test point execution operation is greater than the corresponding preset threshold, or the success rate of the test point execution operation is less than the corresponding preset threshold, the test point is marked as a functional abnormal test point, and the functional abnormal test point is sent to the positioning display module;

[0063] The judgment process for successful execution of the test point operation is as follows: the response delay time difference of the test point execution operation is less than the corresponding preset test point execution operation response delay time difference threshold, and after the test point execution operation, the result of the operation is consistent with the expected result of the operation; the consistency between the result of the operation and the expected result of the operation includes: consistency in functional integrity, consistency in data accuracy, and consistency in the effectiveness of the error handling mechanism.

[0064] The interface testing module performs interface testing on different test platforms according to pre-designed interface test cases. The specific process is as follows:

[0065] When the test platform is testing the corresponding test case, for each test point, the test point is divided into a plurality of test points i, i=1, 2, ..., n, by dividing the interface where the test point is tested on the test platform into a plurality of test points i, i=1, 2, ..., n; the color value, clarity value, and coordinate value of each test point in the test interface are recorded; each test point is preset to have a corresponding standard color value, standard clarity value, and standard coordinate value; the color value, clarity value, and coordinate value of the test point are compared with the preset standard color value, standard clarity value, and standard coordinate value of the corresponding test point, and the color deviation value SP of the test point is obtained. i , clarity deviation value FP i And the coordinate deviation value ZP i ; Using the formula: , obtain the test point interface deviation evaluation value MP, where a1, a2, and a3 are preset weight coefficients, and each test point interface deviation evaluation value is preset to correspond to a deviation evaluation threshold;

[0066] For each test point, the test point interface deviation evaluation value MP is compared with the corresponding preset deviation evaluation threshold. If MP is greater than the corresponding preset deviation evaluation threshold, the test point is marked as an interface abnormality test point, and the interface abnormality test point is sent to the positioning display module;

[0067] Color deviation value SP of the detection point i The analysis process is as follows: by comparing the RGB values ​​of each pixel constituting the detection point with the RGB values ​​of each pixel corresponding to the preset standard color value, the RGB deviation value of each pixel constituting the detection point is obtained; by summing the RGB deviation values ​​of each pixel constituting the detection point, the color deviation value SP of the detection point is obtained. i ;

[0068] The sharpness deviation value FP of the detection point i The analysis process is as follows: the detection point image and the image corresponding to the preset standard definition value are converted into grayscale images, and the Canny algorithm is used to perform edge detection on the detection point image and the corresponding preset image, and then the edge image of the detection point and the edge image of the corresponding preset image are obtained, and the sum of the gradient values ​​of all pixels in the edge image of the detection point and the edge image of the corresponding preset image are calculated respectively to obtain the detection point image gradient sum G1 i The sum of the gradients with the preset image G2 i ; Using the formula: FP i =|G1 i -G2 i | / G2i , get the clarity deviation value FP of the detection point i ;

[0069] Coordinate deviation value ZP of the detection point i The analysis process is as follows: obtain the coordinates of the center point of the detection point and the coordinates of each edge vertex, and mark the coordinates of the center point and the coordinates of each edge vertex as feature point coordinates (X1 e , Y1 e ), where e=1, 2, ..., g; g represents the number of the feature point, and the preset standard coordinate value corresponds to the feature point coordinate (X2 e , Y2 e ), using the formula: , get the coordinate deviation value ZP of the detection point i .

[0070] The performance test module performs performance tests on different test platforms according to pre-designed performance test cases. For each test point, the number of flashbacks and the time it takes to complete the execution of the test platform during the execution of the test point are monitored. The specific process is as follows:

[0071] For each test point, a threshold for the number of operations to be performed is preset, and each threshold for the number of operations to be performed for each test point has a threshold for the number of flashbacks and a threshold for the length of time it takes to complete the operation.

[0072] For each test point, monitor the number of times the test platform crashes and the time it takes to complete the operation during each test of the corresponding operation quantity threshold. Sum the number of crashes and the time it takes to complete the operation for all the operations during the performance test, and take the average as the number of crashes and the time it takes to complete the operation for the test point during the performance test.

[0073] The number of flashbacks and the time to complete the operation of the test point are compared with the corresponding preset flashback number threshold; at the same time, the time to complete the execution of the operation is compared with the corresponding preset execution of the operation completion time threshold;

[0074] If the number of test point flashbacks is greater than the corresponding preset flashback threshold or the test point execution operation completion time is greater than the corresponding preset execution operation completion time threshold, the test point will be marked as a performance abnormality test point and the performance abnormality test point will be sent to the positioning display module.

[0075] The positioning display module receives the key points of function abnormality test, interface abnormality test and performance abnormality test, and displays them in the problem display model. By scanning the problem display model in real time, the test points where software test problems occur can be obtained in time. The specific process is as follows:

[0076] Set up a problem display model, such as Figure 2 ,The software icon for cross-platform testing is placed at the center of the problem display model. Three coordinate axes extend outward from the software icon, namely the function test direction axis, the interface test direction axis and the performance test direction axis, and the angle between each direction axis is 120°;

[0077] On the function test direction axis, mark circles of different sizes and colors are set according to the type and severity of the function abnormality test points. At the same time, the function problems are prioritized on the function test direction axis according to the size of the mark circle. The larger the mark circle, the closer it is to the software icon on the function test direction axis, and the higher the priority of the function problem handling;

[0078] The specific process of analyzing each abnormal function test point, its corresponding color and the size of the marking circle is as follows:

[0079] A color set of function abnormality test key points is preset, wherein each function abnormality test key point in the color set corresponds to a color, and for each function abnormality test key point, the corresponding color is obtained by substituting the function abnormality test key point into the preset function abnormality test key point color set;

[0080] For each function abnormality test point, obtain the difference between the average response delay time difference of its execution operation and the corresponding preset threshold value, and mark it as the function abnormality test point response deviation value XP, and at the same time obtain the success rate CG of the execution operation of the abnormal test point, and use the formula CXP= (XP×b1-CG×b2)×r to obtain the function abnormality evaluation value CXP, where b1 and b2 are preset weight coefficients, and r is the function adjustment coefficient. The r value corresponding to different function abnormality test points is different. The role of r is to unify the function abnormality evaluation value corresponding to each function abnormality test point to the same scale for comparison; the larger the function abnormality evaluation value, the more serious the abnormality of the function test performed by the corresponding test point;

[0081] Preset a number of abnormal function evaluation value intervals, each of which corresponds to a marking circle radius. For each abnormal function test point, substitute its corresponding abnormal function evaluation value into the preset abnormal function evaluation value intervals for matching, and output its corresponding marking circle radius;

[0082] On the interface test direction axis, mark circles of different sizes and colors are set according to the type and severity of the interface abnormality test points for marking. At the same time, the interface problems are prioritized on the interface test direction axis according to the size of the mark circle. The larger the mark circle, the closer it is to the software icon on the interface test direction axis, and the higher the priority of the interface problem handling;

[0083] The specific process of analyzing the corresponding color and the size of the marking circle for each interface abnormality test point is as follows:

[0084] A preset interface abnormality test key point color set, wherein each interface abnormality test key point in the interface abnormality test key point color set corresponds to a color, and for each interface abnormality test key point, the corresponding color is obtained by substituting the interface abnormality test key point into the preset interface abnormality test key point color set;

[0085] For each interface abnormality test point, obtain the difference between the deviation evaluation value of the detection point and the preset deviation evaluation threshold of the corresponding detection point, and mark it as the interface test deviation value ZP. Use the formula: ZMP=ZP×z to obtain the interface abnormality evaluation value ZMP, where z is the interface adjustment coefficient, and the z value corresponding to different interface abnormality test points is different. The role of z is to unify the interface abnormality evaluation value corresponding to each interface abnormality test point to the same scale for comparison; the larger the interface abnormality evaluation value, the more serious the abnormality of the interface test conducted by the corresponding test point;

[0086] Preset several interface anomaly evaluation value intervals, each interface anomaly evaluation value interval corresponds to a marking circle radius, and for each interface anomaly test point, substitute its corresponding interface anomaly evaluation value into several preset interface anomaly evaluation value intervals for matching, and output its corresponding marking circle radius;

[0087] On the performance test direction axis, mark circles of different sizes and colors are set according to the type and severity of the performance abnormality test points. At the same time, the performance issues are prioritized on the performance test direction axis according to the size of the mark circle. The larger the mark circle, the closer it is to the software icon on the performance test direction axis, and the higher the priority of performance issue handling.

[0088] The specific process of analyzing each performance abnormality test point, its corresponding color and the size of the marked circle is as follows:

[0089] A preset performance abnormality test key point color set, wherein each performance abnormality test key point in the performance abnormality test key point color set corresponds to a color, and for each performance abnormality test key point, the corresponding color is obtained by substituting the performance abnormality test key point into the preset performance abnormality test key point color set;

[0090] For each performance abnormality test point, obtain the difference between the number of flashbacks greater than the corresponding preset flashback threshold, and mark it as the abnormal performance test point flashback deviation value TP. At the same time, obtain the difference between the execution operation completion time greater than the corresponding preset execution operation completion time threshold, and mark it as the abnormal performance test point operation time deviation value WP. Use the formula WTP=(WP×c1+TP×c2)×w to obtain the performance abnormality evaluation value WTP, where c1 and c2 are preset weight coefficients, and w is the performance adjustment coefficient. The w values ​​corresponding to different performance abnormality test points are different. The role of w is to unify the performance abnormality evaluation values ​​corresponding to each performance abnormality test point to the same scale for comparison; the larger the performance abnormality evaluation value, the more serious the performance test abnormality of the corresponding test point;

[0091] Preset several performance anomaly evaluation value intervals, each performance anomaly evaluation value interval corresponds to a marking circle radius, and for each performance anomaly test point, substitute its corresponding performance anomaly evaluation value into several preset interface anomaly evaluation value intervals for matching, and output its corresponding marking circle radius;

[0092] When processing the marked circle in the problem display model, the problem display model is scanned in real time. When a circle is marked in the problem display model, the direction axis, color and size corresponding to the marked circle are retrieved to obtain the test points corresponding to the marked circle, the abnormal types of the test points and the abnormal evaluation value data corresponding to the abnormal types of the test points. The test points corresponding to the marked circle, the abnormal types of the test points and the abnormal evaluation value data corresponding to the abnormal types of the test points are compressed into a data packet and sent to the maintenance personnel. After receiving the data packet, the maintenance personnel optimize the cross-platform test software according to the data in the data packet.

[0093] When a single marked circle appears in the problem display model, the data corresponding to the marked circle is directly sent to the maintenance personnel;

[0094] When multiple marked circles appear on the same direction axis in the problem display model, the test points corresponding to the marked circle with the largest radius, the abnormal types of the test points, and the abnormal evaluation value data corresponding to the abnormal types of the test points are compressed into a data packet and sent to the maintenance personnel; the maintenance personnel optimize the cross-platform test software according to the data in the data packet according to the order in which the data packets are sent;

[0095] When marked circles appear on the function test direction axis, the interface test direction axis, and the performance test direction axis in the problem display model, the priority processing method is as follows:

[0096] A proportional normalization adjustment coefficient is preset for each of the marking circles on the preset function test direction axis, the interface test direction axis and the performance test. The radius of the marking circle on each direction axis is multiplied by the corresponding normalization adjustment coefficient to obtain the radius of the marking circle on each direction axis after normalization adjustment. The purpose of presetting a proportional normalization adjustment coefficient is to enable the marking circles on different direction axes to be compared at the same scale so as to more accurately determine the priority of the problem. The priority is determined by comparing the radius of the marking circles on each direction axis after normalization adjustment, that is, the larger the radius, the higher the priority of the problem handling. The problem handling priority is sorted according to the size of the priority, and the test points corresponding to the marking circle with the highest priority, the abnormal types of the test points and the abnormal evaluation value data corresponding to the abnormal types of the test points are compressed into a data packet and sent to the maintenance personnel. The maintenance personnel optimize the cross-platform tested software according to the data in the data packet according to the order in which the data packets are sent. This method facilitates the orderly resolution of various compatibility issues arising from cross-platform testing of the software.

[0097] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A cross-platform software compatibility testing and optimization method, characterized in that: The following steps are involved: Step 1: Install the software to be tested on different test platforms and conduct compatibility testing on the test platforms; compatibility testing includes: functional testing, interface testing, and performance testing; Step 2: When performing functional testing, when the average response delay time difference of the test point execution operation is greater than the preset threshold, or the success rate of the test point execution operation is less than the preset threshold, it is marked as a functional abnormal test point, and the functional abnormal test point is sent to the problem display model; Step 3: When performing interface testing, the interface of the test point on the test platform is divided into several detection points. The color value, clarity value and coordinate value of the detection point are comprehensively analyzed to obtain the interface deviation evaluation value of the test point. If the interface deviation evaluation value of the test point is greater than the preset threshold, the test point is marked as an interface abnormality test point, and the interface abnormality test point is sent to the problem display model; The specific process of interface testing in step 3 is as follows: When the test platform performs interface testing through test cases, for each test point, the interface where the test point is tested on the test platform is divided into a number of test points i, i=1, 2, ..., n; the color value, clarity value, and coordinate value of each test point in the test interface are recorded; each test point is preset to have a corresponding standard color value, standard clarity value, and standard coordinate value; the color value, clarity value, and coordinate value of the test point are compared with the preset standard color value, standard clarity value, and standard coordinate value of the corresponding test point to obtain the color deviation value SP of the test point. i , clarity deviation value FP i And the coordinate deviation value ZP i ; Using the formula: , obtain the test point interface deviation evaluation value MP, where a1, a2, and a3 are preset weight coefficients, and each test point interface deviation evaluation value is preset to correspond to a deviation evaluation threshold; For each test point, the test point interface deviation evaluation value MP is compared with the corresponding preset deviation evaluation threshold. If MP is greater than the corresponding preset deviation evaluation threshold, the test point is marked as an interface abnormality test point, and the interface abnormality test point is sent to the problem display model for display; The analysis process of the color deviation value, clarity deviation value and coordinate deviation value of the detection point is as follows: Color deviation value: By comparing the RGB values ​​of each pixel constituting the detection point with the RGB values ​​of each pixel corresponding to the preset standard color value, the RGB deviation value of each pixel constituting the detection point is obtained, and by summing the RGB deviation values ​​of each pixel constituting the detection point, the color deviation value SP of the detection point is obtained. i ; Clarity deviation value: Convert the detection point image and the image corresponding to the preset standard clarity value into grayscale images, apply the Canny algorithm to perform edge detection on the detection point image and the corresponding preset image, and then obtain the edge image of the detection point and the edge image of the corresponding preset image. Calculate the sum of the gradient values ​​of all pixels in the edge image of the detection point and the edge image of the corresponding preset image, and obtain the total gradient G1 of the detection point image. i The sum of the gradients with the preset image G2 i ; Using the formula: FP i =|G1 i -G2 i | / G2 i , get the definition deviation value FP of the detection point i ; Coordinate deviation value: Get the coordinates of the center point of the detection point and the coordinates of each edge vertex, and mark the coordinates of the center point and each edge vertex as feature point coordinates (X1 e , Y1 e ), where e=1, 2, ..., g; g represents the number of the feature point, and the preset standard coordinate value corresponds to the feature point coordinate (X2 e , Y2 e ), using the formula: , get the coordinate deviation value ZP of the detection point i ; Step 4: When performing performance testing, monitor the number of flashbacks of the test platform during the execution of the test point and the duration of the execution operation. If the number of flashbacks of the test point is greater than a preset threshold or the duration of the execution operation of the test point is greater than a preset threshold, mark the test point as a performance abnormality test point and send the performance abnormality test point to the problem display model. Step 5: Display the functional abnormality test points, interface abnormality test points and performance abnormality test points in the problem display model, and scan them in real time to locate the abnormal test points where software testing problems occur, and prioritize the abnormal test points.

2. The cross-platform software compatibility testing and optimization method according to claim 1, characterized in that: The specific process of compatibility testing in step 1 is as follows: The test platform includes: different operating systems, hardware environments and browsers; when conducting compatibility tests on different test platforms, develop matching test cases. Test cases refer to a collection of specific test points in terms of software function implementation, interface presentation and performance in cross-platform usage scenarios; each test case contains several test points, and each test point corresponds to an execution operation behavior.

3. The cross-platform software compatibility testing and optimization method according to claim 2, characterized in that: The specific process of functional testing in step 2 is as follows: Start the software on different test platforms and simulate the user's operation behaviors according to the execution operations contained in the test cases. The operation behaviors include: inputting data, clicking buttons, and selecting menu options. For each test platform, multiple repetitive tests are performed when the corresponding test cases are tested; For each test point, obtain the execution operation time of the test point and the time for the test platform to respond to the corresponding operation, and subtract them to obtain the response delay time difference of the test point execution operation, sum and average the response delay time differences of the test point execution operations for multiple repetitive tests, and obtain the average response delay time difference of the test point execution operations; at the same time, obtain the number of successful execution operations of the test point during multiple repetitive tests, and obtain the success rate of the test point execution operations by dividing the number of successful execution operations of the test point by the total number of repetitive tests; preset each test point execution operation has a corresponding response delay time difference threshold and execution operation success rate threshold, compare the average response delay time difference of the test point execution operation with the corresponding preset threshold, and compare the success rate of the test point execution operation with the corresponding preset operation success rate threshold, when the average response delay time difference of the test point execution operation is greater than the corresponding preset threshold, or the success rate of the test point execution operation is less than the corresponding preset threshold, mark the test point as a functionally abnormal test point, and display the functionally abnormal test point by sending it to the problem display model.

4. The cross-platform software compatibility testing and optimization method according to claim 3, characterized in that: The process of judging whether the operation is successful is as follows: The response delay time difference of the test point execution operation is less than the corresponding preset test point execution operation response delay time difference threshold, and after the test point executes the operation, the result of the operation is consistent with the expected result of the execution operation; The result of the operation is consistent with the expected result of the operation, including: consistency in functional integrity, consistency in data accuracy, and consistency in the effectiveness of the error handling mechanism.

5. The cross-platform software compatibility testing and optimization method according to claim 4, characterized in that: The specific process of performance testing in step 4 is as follows: For each test point, a threshold for the number of operations to be performed is preset, and each threshold for the number of operations to be performed for each test point has a threshold for the number of flashbacks and a threshold for the length of time it takes to complete the operation. For each test point, monitor the number of times the test platform crashes and the time it takes to complete the operation during each test of the corresponding operation quantity threshold. Sum the number of crashes and the time it takes to complete the operation for all the operations during the performance test, and take the average as the number of crashes and the time it takes to complete the operation for the test point during the performance test. The number of flashbacks and the time to complete the operation of the test point are compared with the corresponding preset flashback number threshold; at the same time, the time to complete the execution of the operation is compared with the corresponding preset execution of the operation completion time threshold; If the number of test point flashbacks is greater than the corresponding preset flashback threshold or the test point execution operation completion time is greater than the corresponding preset execution operation completion time threshold, the test point will be marked as a performance abnormality test point, and the performance abnormality test point will be sent to the problem display model.

6. The cross-platform software compatibility testing and optimization method according to claim 5, characterized in that: In step 5, the process of locating and processing the function abnormality test points, interface abnormality test points, and performance abnormality test points displayed in the problem display model is as follows: The software icon for cross-platform testing is placed at the center of the problem display model. Three coordinate axes extend outward from the software icon, namely the function test direction axis, the interface test direction axis, and the performance test direction axis, and the angle between each direction axis is 120°; On the function test direction axis, the interface test direction axis and the performance test direction axis, different sizes of marking circles and colors are set for marking according to the types and severity of the function abnormality test points, the interface abnormality test points and the performance abnormality test points, and at the same time, the functional problems are prioritized on the function test direction axis according to the size of the marking circle. The larger the marking circle, the closer the position on the function test direction axis is to the software icon, and the higher the priority of the functional problem handling; When processing the marked circle in the problem display model, the problem display model is scanned in real time. When a circle is marked in the problem display model, the direction axis, color and size corresponding to the marked circle are retrieved to obtain the test points corresponding to the marked circle, the abnormal types of the test points and the abnormal evaluation value data corresponding to the abnormal types of the test points. The test points corresponding to the marked circle, the abnormal types of the test points and the abnormal evaluation value data corresponding to the abnormal types of the test points are compressed into a data packet and sent to the maintenance personnel. After receiving the data packet, the maintenance personnel optimize the cross-platform test software according to the data in the data packet. When a single marked circle appears in the problem display model, the data corresponding to the marked circle is directly sent to the maintenance personnel; When multiple marked circles appear on the same direction axis in the problem display model, the test points corresponding to the marked circle with the largest radius, the abnormal types of the test points, and the abnormal evaluation value data corresponding to the abnormal types of the test points are compressed into a data packet and sent to the maintenance personnel; the maintenance personnel optimize the cross-platform test software according to the data in the data packet according to the order in which the data packets are sent; When marked circles appear on the function test direction axis, the interface test direction axis, and the performance test direction axis in the problem display model, the priority processing method is as follows: A proportional normalization adjustment coefficient is preset for each of the preset function test direction axis, interface test direction axis and mark circle on the performance test. The radius of the mark circle on each direction axis is multiplied by the corresponding normalization adjustment coefficient to obtain the radius of the mark circle on each direction axis after normalization adjustment. The priority is determined by comparing the radii of the mark circles on each direction axis after normalization adjustment, that is, the larger the radius, the higher the priority of problem handling. The problem handling priorities are sorted according to the size of the priority, and the test points corresponding to the mark circle with the highest priority, the abnormal types of the test points and the abnormal evaluation value data corresponding to the abnormal types of the test points are compressed into a data packet and sent to the maintenance personnel. The maintenance personnel optimize the cross-platform test software according to the data in the data packets according to the order in which the data packets are sent.

7. The cross-platform software compatibility testing and optimization method according to claim 6, characterized in that: The specific process of analyzing each abnormal function test point, its corresponding color and the size of the marking circle is as follows: A color set of function abnormality test key points is preset, wherein each function abnormality test key point in the color set corresponds to a color, and for each function abnormality test key point, the corresponding color is obtained by substituting the function abnormality test key point into the preset function abnormality test key point color set; For each function abnormality test point, obtain the difference between the average response delay time difference of its execution operation and the corresponding preset threshold value, and mark it as the function abnormality test point response deviation value XP. At the same time, obtain the success rate CG of the execution operation of the abnormal test point, and use the formula CXP= (XP×b1-CG×b2)×r to obtain the function abnormality evaluation value CXP, where b1 and b2 are preset weight coefficients, and r is the function adjustment coefficient. Different function abnormality test points have different r values. A number of functional abnormality evaluation value intervals are preset, each functional abnormality evaluation value interval corresponds to a marking circle radius, and for each functional abnormality test point, its corresponding functional abnormality evaluation value is substituted into the preset number of functional abnormality evaluation value intervals for matching, and its corresponding marking circle radius is output.

8. The cross-platform software compatibility testing and optimization method according to claim 7, characterized in that: The specific process of analyzing the corresponding color and the size of the marking circle for each interface abnormality test point is as follows: A preset interface abnormality test key point color set, wherein each interface abnormality test key point in the interface abnormality test key point color set corresponds to a color, and for each interface abnormality test key point, the corresponding color is obtained by substituting the interface abnormality test key point into the preset interface abnormality test key point color set; For each interface abnormality test point, obtain the difference between the detection point deviation evaluation value and the preset deviation evaluation threshold of the corresponding detection point, and mark it as the interface test deviation value ZP. Use the formula: ZMP=ZP×z to obtain the interface abnormality evaluation value ZMP, where z is the interface adjustment coefficient, and the z value corresponding to different interface abnormality test points is different; Several interface anomaly evaluation value intervals are preset, each interface anomaly evaluation value interval corresponds to a marking circle radius, and for each interface anomaly test point, its corresponding interface anomaly evaluation value is substituted into several preset interface anomaly evaluation value intervals for matching, and its corresponding marking circle radius is output.

9. The cross-platform software compatibility testing and optimization method according to claim 7, characterized in that: The specific process of analyzing each performance abnormality test point, its corresponding color and the size of the marked circle is as follows: A preset performance abnormality test key point color set, wherein each performance abnormality test key point in the performance abnormality test key point color set corresponds to a color, and for each performance abnormality test key point, the corresponding color is obtained by substituting the performance abnormality test key point into the preset performance abnormality test key point color set; For each performance abnormality test point, obtain the difference between the number of flashbacks greater than the corresponding preset flashback threshold, and mark it as the abnormal performance test point flashback deviation value TP. At the same time, obtain the difference between the execution operation completion time greater than the corresponding preset execution operation completion time threshold, and mark it as the abnormal performance test point operation time deviation value WP. Use the formula WTP=(WP×c1+TP×c2)×w to obtain the performance abnormality evaluation value WTP, where c1 and c2 are preset weight coefficients, and w is the performance adjustment coefficient. Different performance abnormality test points have different w values. Several performance anomaly evaluation value intervals are preset, each performance anomaly evaluation value interval corresponds to a marking circle radius, and for each performance anomaly test point, its corresponding performance anomaly evaluation value is substituted into several preset interface anomaly evaluation value intervals for matching, and its corresponding marking circle radius is output.

Images