WO2012104991A1 - Program test method, program test system, and program - Google Patents

Abstract

The present invention is a program test method in which a computer for causing a plurality of programs to execute processing of test data used for testing each of the programs records a first data portion that is a portion of the test data that is changed by the test from the test data used by the test of the program executed immediately prior, calculates a second data portion that is a portion of the test data to be accessed by the test among the test data that is read for another program that is next to be tested, and selecting the other program in which the difference between the first data portion and the second data portion is at a minimum as the program next to be tested.

Description

Program test method, program test system, and program

The present invention relates to a program test method, a program test system, and a program.

When developing a computer program (hereinafter simply referred to as “program”), various tests are executed for the purpose of verifying the operation of the program. In addition to new development, it was created in the past for the program at the time of function expansion (enhancement) that is executed in a timely manner from the viewpoint of preventing function degradation (degradation) after starting program operation. The test is run as a regression test. Generally, new test items are added to the test to be executed as the program is updated after the operation is started, so that the execution time of the regression test keeps increasing. For this reason, for example, even if the regression test is executed by batch processing at night when the program is stopped, there are cases where the test does not end by the next morning when the operation should be resumed.

∙ From the viewpoint of ensuring the quality of the program, all test items must be implemented in the regression test, so the number of test items to be implemented cannot be reduced. For this reason, an increase in test execution time has led to an increase in development cost, and there is a strong demand for shortening the test execution time.

Furthermore, there is a test-specific requirement in the testing of a program that it is necessary to ensure the independence between the components constituting the target program. For this reason, in testing different components, even if the same test data is used, the test data must be reloaded every time the test of each component is executed, and the load time of the same content test data However, inefficiencies that occur redundantly are particularly problematic.

Patent Document 1 discloses a configuration that reduces the time required for program test and automates it for the purpose of providing a program test system and a program test method that can efficiently test a program. Yes. Specifically, the test data used for the test is once loaded into the database to create a real table. Thereafter, a test is executed, and a virtual table is created when the test target program tries to access the real table. The test target program accesses only the virtual table and does not access the test data of the real table. As a result, when the test data is returned to the initial state, it is only necessary to create and delete the virtual table, and it is not necessary to load the test data from the database, thereby reducing the test time.

JP 2006-227820 A

However, in the configuration of Patent Document 1, instead of loading test data, the virtual table created in the immediately preceding test is deleted and a new virtual table is created each time. As a result, even if there is no change in the contents of the virtual table created immediately before and there is no need to delete it, the virtual table immediately before is deleted. Therefore, there was a problem that it was not an efficient solution.

The present invention has been made to solve the above-mentioned and other problems. When executing a test of a program, the test is performed by reducing the time required for loading test data for each program as much as possible. The purpose is to reduce the execution time itself and execute the test efficiently.

One aspect of the present invention for solving the above-mentioned and other problems is to make test data used for each test available for each program in order to test a plurality of programs, and for each test First data that is a part of the test data changed by the test among the test data used by the test of the program executed immediately before by the computer that causes each program to execute data processing Second data which is a part of the test data to be accessed by the test among the test data to be read for the other program to be subjected to the test to be recorded and to be subsequently executed The other portion in which the difference between the first data portion and the second data portion is minimized. Serial continue is a method of testing a program to be selected as a target to run the test program.

According to the present invention, when executing a test of a program, the time required for loading test data for each program is reduced as much as possible, thereby reducing the test execution time itself and efficiently executing the test. Can do.

FIG. 1 is a schematic diagram showing a general execution procedure of a program test. FIG. 2 is a schematic diagram showing an outline of processing executed by the program test system according to the embodiment of the present invention. FIG. 3 is a schematic configuration diagram of the program test system 1 according to the embodiment of the present invention. FIG. 4 is a diagram illustrating an example of a hardware configuration of the computer 10. FIG. 5 is a diagram showing an example of the software configuration of the client computer 100 included in the program test system 1. FIG. 6 is a diagram showing an example of the software configuration of the server computer 200 included in the program test system 1. FIG. 7 is a diagram showing a configuration example of the test script 141 used in the program test system 1. FIG. 8 is a diagram illustrating a configuration example of the test group information storage table 142. FIG. 9 is a diagram illustrating a configuration example of the test script information storage table 143. FIG. 10 is a diagram showing an example of the test result 144 obtained by the program test system 1. FIG. 11 is a diagram illustrating a configuration example of the business application table 241. FIG. 12 is a diagram showing a configuration example of the test data 242 used in the program test system 1. FIG. 13 is a diagram illustrating a configuration example of the load time information storage table 243. FIG. 14 is a diagram showing a configuration example of the database operation information storage table 244. FIG. 15 is a diagram showing a configuration example of the database update information storage table 245. FIG. 16 is a diagram illustrating a configuration example of the update location table 1500. FIG. 17 is a diagram illustrating a configuration example of the restoration location table 1550. FIG. 18 is a flowchart illustrating an example of the test execution order changing process. FIG. 19 is a flowchart showing an example of the runtime information recording process. FIG. 20 is a flowchart illustrating an example of the test group selection process. FIG. 21 is a flowchart showing an example of the data loading process.

DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments for carrying out the present invention will be described with reference to the accompanying drawings.
First, an overview of a program test performed by the program test system according to the present embodiment will be described with reference to FIGS. FIG. 1 is a schematic diagram showing a general execution procedure of a program test, and FIG. 2 is a schematic diagram showing an outline of a program test executed by the program test system of the present embodiment.

In FIG. 1, it is assumed that a program to be subjected to a program test executed by the present embodiment (hereinafter referred to as “target program”) has three components A, B, and C. When the test data (hereinafter referred to as “test data”) used for testing each of the components A, B, and C is data a, b, and c, the conventional target program is tested. First, test data a is loaded onto a computer on which the target program is running and then component A is operated to perform a test. After the test on component A is completed, test data b is loaded and component B is loaded. A procedure was taken in which the test was executed and then the test of component C was executed by loading test data c. This procedure is the same even if the data a, b, and c are the same test data.

Here, “component” means a functional module that realizes a single function in the target program. A form in which each component itself is an application that constitutes higher-order system software is also included. “Load” of test data means the time required for the test data stored in any storage medium to be used by the target program after being loaded into the computer on which the target program operates. It has a broad concept including the user's operation as a trigger for the above.

The program test system according to the present embodiment reduces the time required for test execution as much as possible by reducing the time required for loading test data a, b, and c as much as possible by changing the test execution order for components A, B, and C. It is what. When executing a test of a program, the execution order of tests may be constrained in order to maintain the consistency of test results. In this embodiment, such an execution order is guaranteed for each component. The order of execution can be freely changed.

FIG. 2 schematically shows an outline of a program test executed by the program test system of the present embodiment. As described above, in this program test system, for a plurality of components (components A to C in FIG. 2) included in the target program, it is determined that the load time of the test data is reduced when the component test execution order is changed. In such a case, the program test system manages the test execution order so that the test execution order for the components is changed.

Now, consider the first test when the target program is run for the first time. At this time, as shown in FIG. 2, the user who is the tester specifies which test group the test is to be executed by using the test tool which is a test execution program installed in the program test system of the present embodiment. The test tool is a set of test scripts necessary for executing the test of the target program, and a commercially available test tool or the like can be used. The “test group” is a set of test scripts that are executed when testing the components A to C of the target program.

The execution order changing unit is one of the functional units that realize the main functions of the present embodiment, and the execution order of the test groups A, B, and C so that the load time of the test data related to each test group is shortened as much as possible. Manage.

The execution order changing unit that has received the information on the first test group specified by the user via the test tool first loads the test data A, and after the loading of the test data A, the test script included in the test group A Are executed sequentially. By executing these test scripts, the business application that is the target program is executed. The business application runs on the program execution platform set in the computer on which the business application is installed. The program execution platform is, for example, a Java (registered trademark) virtual machine that operates on a server computer on which an appropriately selected operating system is mounted.

When the business application is executed by the test script included in the test group A, the business application accesses the test data a loaded in the test database. After the test by the test group A is completed, the tests by the test groups B and C are sequentially executed following the loading of the test data b and c. The program execution platform stores the status when the business application accesses the database by the test groups A, B, and C as runtime information.

Next, the second and subsequent test executions will be described. First, as in the initial test, the user designates which test group of the test groups A to C is used for the test via the test tool. At this time, the execution order changing unit reads the execution time information stored in the program execution base during the initial test. The execution order changing unit determines, based on the execution time information, which test group performs the next test efficiently based on the execution time information on the basis of the effect of reducing the load time of the test data.

More specifically, when determining which test group the test is to be executed, the execution order switching unit needs to load a new test data or only a part of the test data that needs to be loaded. Test whether the test data that needs to be loaded or the entire test data that needs to be loaded should be loaded, and which should be executed Load data and run tests. The test data loading method is determined each time a test for one test group is completed.

As with the initial test, the program execution platform records the access status to the database when the business application operates according to each test group as runtime information. The reason for recording the runtime information in this way is that the access status to the business application database may change due to program updates, modifications, and the like. In the example of FIG. 2, the test execution order is changed to the test groups A, C, and B in accordance with the determination result of the execution order changing unit based on the runtime information.

Next, the program test system of the present embodiment whose outline has been described above will be described in detail according to a specific configuration example. FIG. 3 shows a configuration example of the program test system 1 according to the present embodiment. The program test system 1 includes one or more clients 100, a server 200, and a communication network 300.

The client 100 is a computer in which a test program (test tool) is installed and operating in the program test system 1. The server 200 is a computer on which various business applications are installed and operated. This business application is a test target (target program) of the program test system 1. The communication network 300 is a communication line that connects the client 100 and the server 200 so that they can communicate with each other. The communication network 300 is appropriately selected from various wired or wireless networks including a LAN (Local Area Network), the Internet, and a dedicated line. be able to.

FIG. 4 shows a hardware configuration example of the computer 10 that can be used as the client 100 or the server 200. A computer 10 illustrated in FIG. 4 includes a central processing unit 11, a main storage device 12, an auxiliary storage device 13, an input device 14, an output device 15, and a communication interface (hereinafter “communication I / F”) 16.

The central processing unit 11 includes, for example, a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), and the like (hereinafter simply referred to as “CPU”). The main storage device 12 is, for example, a RAM (Random Access Memory) or a ROM (Read Only Memory), and provides various programs executed by the CPU 11 and storage areas for various data processed by these programs.

The auxiliary storage device 13 is a storage device such as a hard disk drive (Hard Disk Drive, HDD), a semiconductor storage drive (Solid State Drive, SSD), and various programs and tests that implement the functions of the program test system 1 of the present embodiment. Stores data, business application programs and processing data thereof.

The input device 14 is an input device used when executing a data input / output operation according to a function provided by a program described later, and includes, for example, a keyboard or a mouse. As the input device 14, other types of input devices such as a touch panel, a pen tablet, and a voice input device may be provided. The output device 15 is an output device for outputting data processed by the program, and includes an output device of an appropriate format such as a liquid crystal display, a printer, and an audio output device. In particular, the input device 14 is used to select the type of test executed by the user, items, a test target program, and the like. The output device 15 is used to output a test result (success or failure) of the executed test so that the user can confirm it.

The communication I / F 16 is an interface circuit for connecting the computer 10 used as the client 100 or the server 200 to the communication network 300 in FIG. 1, and includes, for example, a NIC (Network Interface Card). The CPU 11, the memory 12, the auxiliary storage device 13, the input device 14, the output device 15, and the communication I / F 16 are communicably connected by an internal bus 17 adopting an appropriate communication standard.

Next, a software configuration example of the client 100 will be described. FIG. 5 shows an example of the software configuration of the client 100. As shown in FIG. 5, each program of an operation system (Operation System) (hereinafter referred to as “OS”) 110, a data I / O unit 120, and an execution order changing unit 130 is installed in the client 100 of the present embodiment. The execution order switching unit 130 further includes functional modules of a test group selection unit 131, a load method selection unit 132, and a test execution unit 133.

The client 100 also includes a test script 141, a test group information storage table 142, a test script information storage table 143, and test results as data used by the execution order changing unit 130 or data processed by the execution order changing unit 130. 144 are stored respectively. Each functional module of the execution order changing unit 130 is read by the CPU 11 into the memory 12 and executed. The test script 141, the test group information storage table 142, the test script information storage table 143, and the test result 144 are stored in the auxiliary storage device 13, and are read out by the execution order changing unit 130 in a timely manner and used for processing.

The data I / O unit 120 has a function of managing data input / output processing between various software operating on the client 100 and the input device 14 and the output device 15. Includes compatible device drivers.

As described with reference to FIG. 2, the execution order changing unit 130 reduces the test execution time as much as possible for a test executed on a business application operating on the server 200 described later. In order to realize the main functions required for the solution, the function modules of the test group selection unit 131, the load method selection unit 132, and the test execution unit 133 are included as described above.

The test group selection unit 131 selects a test group to be executed next from a test group that is a set of test scripts that execute a test using the same test data according to a predetermined processing flow described later. It is. The test group selection unit 131 executes processing based on runtime information held on the server 200 side to be described later, that is, database operation information and database update information.

The load method selection unit 132 loads a test data used by each test group into a database management system (DataBase Management System, hereinafter referred to as “DBMS”) stored in the server 200 described later. Is a functional module that executes a process of reading test data into the DBMS by the selected loading method. The load method selection unit 132 uses the processing result of the test group selection unit 131 and the later-described load time information held in the server 200 to read the entire test data used for the test group to be executed next. A selection is made as to whether to execute an entire load process, which is a process, or a partial load process, which is a process for reading only necessary portions of test data. As described above, “loading test data” refers to test data (in this embodiment, an auxiliary storage device 13 such as an HDD, etc.) stored in the server 200 by a data load unit 234 that operates on the server 200 described later. However, there is no particular restriction on the storage location.) This means the time required to start reading from the DBMS and become usable by the business application.

The test execution unit 133 is a functional block having a function of sequentially executing test scripts belonging to the test group selected by the processing result of the test group selection unit 131 and verifying the execution result. When the test execution unit 133 executes the test script, the test execution unit 133 notifies the program execution unit 232 of the server 200 described later of information on which test script is being executed. In the present embodiment, it is assumed that a Web application operates as a business application to be described later on the server 200. Therefore, for example, a test script is specified in an HttpRequest statement transmitted from the client 100 to the server 200 using a cookie. The test script ID information is embedded and transmitted to the server 200. In the server 200, the test script ID is read by the program execution unit 232 or the runtime information recording unit 233 described later.

Further, as a result of executing the test script 141 described later, the test execution unit 133 confirms that the response from the business application of the server 200 is the same as the expected result (whether a normal result is obtained as a test result). To verify. The verified result is stored as a test result 144 described later.

The test script 141 is a file for automatically executing the test of this embodiment, and an example thereof is shown in FIG. In the example of FIG. 7, what input parameter is passed to which page of the Web application is held in an XML (eXtensible Markup Language) format. The format of the test script 141 depends on the test tool employed by the program test system 1. The example of FIG. 7 illustrates a test script with a test script ID of 1001. The business application accesses the Web page specified by “http://management.book/stock/” and enters “title”. A process for searching for an item by the value “ΔΔΔ” is executed.

The test group information storage table 142 is a table that holds in correspondence with which test data each test group uses to execute a test. FIG. 8 shows an example of the test group information storage table 142. In the test group information storage table 142 of FIG. 8, items of a test group 1421, test data 1422, and an executed flag 1423 are recorded. In the item of the test group 1421, an identification code for mutually identifying the test groups included in the program test system 1 is recorded. For example, for the test group A, “TestGroup_A” is recorded.

In the item of test data 1422, an identification code for mutually identifying the test data included in the program test system 1 is recorded. In this embodiment, it is assumed that two types of test data are provided, and “TestData_1” and “TestData_2” are recorded in the test data 1422 corresponding to the test group using the test data. When a new test group is added to the program test system 1, a new record is added to the test group information storage table 142.

The executed flag 1423 records an executed flag that is information indicating whether a test by a corresponding test group has been executed in the past, and 1 is recorded when it has been executed. If the test by the corresponding test group 1421 has not been executed, 0 is recorded. Therefore, before execution of the first test and when a record related to the new test group 1421 is added, 0 is recorded in the executed flag 1423.

The test script information storage table 143 is a table for storing information regarding which test group each test script for executing the test of the target program belongs and where the test script is stored. FIG. 9 shows an example of the test script information storage table 143. In the test script information storage table 143, items of a test script ID 1431, a test group 1432, and a file name 1433 are recorded in association with each other. The test script ID 1431 is an identification code assigned to identify the test scripts 141 from each other. The test group 1432 has the same identification code as the test group 1421 recorded in the test group information storage table 142. The file name 1433 records the storage location where the test script 141 specified by the corresponding test script ID 1431 is stored, using the file name including the data path. When test scripts 141 belonging to the same test group 1432 are executed, the test scripts ID 1431 is executed in ascending order.

FIG. 10 shows an example of the test result 144. The test result 144 indicates whether or not the test script 141 is executed for the target program, the execution result matches the normal execution result, and whether or not the correct test result is obtained by the test script execution. In the present embodiment, as an example of the test script 141, a test script 141 that causes a business application as a book management system described later to search for a book whose title (title) is “ΔΔΔ” is illustrated in FIG. The search result of the search processing executed by the business application by the test script 141 in FIG. 7 is automatically confirmed by another test script 141 for determining whether the search result is appropriate, and is output as the test result 144. . The example of FIG. 10 indicates that a normal execution result was obtained as a result of executing the test script 141 of FIG. 7 (success). The information of the test result 144 can be output from the output device 15 of the client 100, for example, and browsed by the user.

Next, the configuration of the server 200 in this embodiment will be described with reference to FIG. FIG. 6 shows a software configuration example of the server 200. The server 200 is loaded with programs for realizing the functions of the OS 210, the data I / O unit 220, the business application 231, the program execution unit 232, the runtime information recording unit 233, and the data load unit 234. The server 200 includes a DBMS 240, business application data 241, test data 242, load time information table 243, database operation information storage table (hereinafter “DB operation information storage table”) 244, and database update information storage table (hereinafter referred to as “database operation information storage table”). , “DB update information storage table”) 245 is stored in the auxiliary storage device 13, for example, and is read out by the business application 231, the program execution unit 232, the runtime information recording unit 233 and the data loading unit 234 in a timely manner. To be used.

The data I / O unit 220 has a function of managing data input / output processing between various software running on the server 200 and the input device 14 and the output device 15. Includes compatible device drivers.

The business application 231 is a target program to be tested by the program test system 1 of the present embodiment. In this embodiment, it is assumed that the business application 231 is a Web application that manages book inventory information, but the business application 231 may be an application program for any application. The business application 231 calls and processes business application data 241 (to be described later) via the DBMS 240 (to be described later). Further, the business application 231 executes processing described in the test script 141 according to an instruction from the test execution unit 133 of the client 100 at the time of test execution.

The program execution unit 232 provides a program execution base necessary for executing the business application 231. In the present embodiment, a Java virtual machine operating on an application server is assumed, but the present invention is not particularly limited to this configuration. When the program execution unit 232 receives from the business application 231 a SQL statement instructing processing to the DBMS 240 described later, the test script ID 1431 corresponding to the content of the SQL statement is notified to the runtime information recording unit 233 described later. Let me record. If a predetermined process to be executed before executing the SQL statement is defined, the program execution unit 232 performs the process. After the predetermined processing is completed, an SQL statement is issued to the DBMS 240. Thereafter, the SQL statement and the corresponding test script ID 1431 are notified again to the runtime information recording unit 233, and if a predetermined process after execution of the SQL statement is defined, the process is performed.

When the SQL statement is issued from the business application 231, the runtime information recording unit 233 records the content of the issued SQL statement and the analysis result thereof in the DB operation information storage table 244 described later as database operation information. The runtime information recording unit 233 records predetermined information before and after the SQL statement issued from the business application 231 is executed by the DBMS 240 as DB update information in the DB update information storage table 245 described later. .

The data load unit 234 reads later-described test data 242 used by each test group 1421 for testing the business application 231 that is the target program from a storage location in the auxiliary storage device 13, and is accessed later by the business application 231. Is stored in the business application data table 241. Further, the data load unit 234 measures the time required for loading the test data 242 and records it as load time information in a load time information storage table 243 described later.

The DBMS 240 receives a SQL statement including a database operation command from the business application 231 and performs data input / output processing on a business application table 241 described later. The business application table 241 is a table storing data that is accessed and used by the business application 231 running on the server 200. FIG. 11 shows a configuration example of the business application table 241. The business application 231 of the present embodiment assumes an application program for performing book inventory management as described above. Therefore, the business application table 241 in FIG. 11 stores the items of id 2411, title 2412, publisher 2413, price 2414, and stock 2415 necessary for managing the inventory of the books in association with each other, and the “Books” table Name is given. “Id” 2411 is an identification code for specifying each record recorded in the business application table 241. “Title” 2412 operates the book application, “publisher” 2413 operates the publisher of the corresponding book, “price” 2414 operates the price of the corresponding book, and “stock” 2415 operates the business application 231 for the corresponding book. The number of stocks held by each organization is recorded. Note that the type of data stored in the business application table 241 differs depending on the use and purpose of the business application 231, and a configuration in which two or more business application tables 241 are provided is also possible. It is.

The test data 242 is data used when the program test system 1 of the present embodiment executes a test of the business application 231, and is read into the business application data table 241 by the data load unit 234 during the test execution. FIG. 12 shows an example of the test data 242. Test data 242 illustrated in FIG. 12 is held in the auxiliary storage device 13 as a csv format file. The test data 242 may be prepared and held in a dedicated storage table.

The load time information storage table 243 is a table that records the time required for the data load unit 234 to read the test data 242 into the business application table 241 together with related information. An example of the load time information storage table 243 is shown in FIG. In the load time information storage table 243, items of test data 2431, load type 2432, row information 2433, column information 2434, and load time 2435 are recorded in association with each other. In the test data 2431, an identification code of the test data 242 loaded by the data loading unit 234 is recorded. In the load type 2432, the mode of the load process executed for the corresponding test data 242 is an entire load that is a process for loading the entire test data 242, or a process for loading only a part of the test data 242. The type of partial load is recorded. In the line information 2433, the line information of the loaded part of the test data 242 in the partial load process is recorded using the ID of the test data 242. In the column information 2434, column information of a portion loaded in the test data 242 in the partial load processing is recorded by a column label (title 2412, publisher 2413, price 2414, or stock 2415) of the business application table 241. In the load time 2435, the time required for the load process is recorded. In the case of the entire load, the time required to load the csv file as the test data 242 into the DBMS 240 is recorded. In the case of partial load, a total value of a time for generating an SQL statement for changing (restoring) only a necessary portion of the business application table 241 and a time for executing the SQL statement is recorded.

The DB operation information storage table 244 is a table that records information regarding what SQL statements the business application 231 issued to the DBMS 240 and what operations were performed on the business application table 241. The data is recorded by the runtime information recording unit 233. FIG. 14 shows an example of the DB operation information storage table 244. In the DB operation information storage table 244 of FIG. 14, items of a test script ID 2441, an SQL statement 2442, an access pattern 2443, an access table 2444, an access condition 2445, row information 2446, and column information 2447 are recorded in association with each other.

The test script ID 2441 is an identification code assigned to identify the test scripts 141 from each other. In the SQL statement 2442, an SQL statement issued from the business application 231 to the DBMS 240 during the test execution is recorded. The SQL statement 2442 is detected by the program execution unit 232 and notified to the runtime information recording unit 233. The access pattern 2443 records the contents of the operation executed by the corresponding SQL statement 2442. In the example of FIG. 14, any of select (selection), update (update), delete (deletion), and insert (insertion) may be recorded, but other operations defined in SQL may be recorded. The access table 2444 records the table name of the table that is the target of the operation of the corresponding SQL sentence 2442. In the present embodiment, “Books” which is the name of the business application table 241 is recorded.

In the access condition 2445, a condition for specifying a target of an operation executed by the corresponding SQL sentence 2442 is recorded as contents described by a where clause of the corresponding SQL sentence 2442. In the row information 2446 and the column information 2447, positions in the table “Books” where data acquired by executing the corresponding SQL statement 2442 exists are recorded. In the row information 2446, primary key information is recorded, and in the column information 2447, the corresponding column name is recorded. Referring to the record of test script ID 2441 = 101 in FIG. 14, an operation for acquiring a title of a book whose “stock (stock) is less than 5” from the table “Books” is recorded. In this case, three IDs of 15, 165, and 210 are recorded in the row information 2446 as ids of the corresponding records, and “title” is recorded in the column information 2447 as the column name.

The access pattern 2443, the access table 2444, and the access condition 2445 can be acquired by analyzing the corresponding SQL sentence 2442. Further, the row information 2446 and the column information 2447 can be acquired by executing the corresponding SQL statement 2442 using the access table 2444 and the access condition 2445.

The DB update information storage table 245 indicates that when the status of the business application table 241 to be operated is changed by the SQL statement issued from the business application 231, in other words, the operation content included in the SQL statement is updated / updated. This is a table for recording changes in data related to the business application table 241 in the case of delete / insert. The data recorded in the DB update information storage table 245 is used when the data load unit 234 performs a partial load process on the test data 242.

FIG. 15 shows an example of the DB update information storage table 245. In the DB update information storage table 245, test script ID 2451, access timing 2452, id 2453, title 2454, publisher 2455, price 2456, and stock 2457 are recorded in association with each other. The test script ID 2451 corresponds to the test script ID 2441 recorded in the DB operation information storage table 244. The access timing 2452 records whether the update information is recorded in the storage table 245 before or after the execution of the test script 141 specified by the corresponding test script ID 2451.

The items id 2453, title 2454, publisher 2455, price 2456, and stock 2457 correspond to id 2411, title 2412, publisher 2413, price 2414, and stock 2415 recorded in the business application data table 241, respectively. When the SQL statement issued from the business application 231 is update, data before and after the execution of update is recorded in units of rows. When the SQL statement issued from the business application 231 is “delete”, data before execution of “delete” is recorded in units of rows. When the SQL statement issued from the business application 231 is insert, data after execution of insert is recorded in units of rows. For example, for the record with the test script ID 2451 = 2, since it can be seen from the SQL statement 2442 in FIG. 14 that the update operation is executed, data before and after the update execution is recorded. As a result, as shown in FIG. 15, the numerical value 0 changes to 30 for stock 2457.

Next, the update location table 1500 and the restoration location table 1550 will be described. The update location table 1500 and the restoration location table 1550 are temporary variables stored in a part of the memory area allocated to the test execution order changing unit 130 of the client 100. However, the update location table 1500 and the restoration location table 1550 are tables as in the example of this embodiment. It can also be held in the form. The update location table 1500 and the restoration location table 1550 are the operation results of the business application table 241 by the other test group 1421 executed immediately before the test data 1422 to be loaded for the test group 1421 to be executed next. Based on the above, it is determined whether to perform the entire load that is the process of loading the entire test data for the test group 1421 to be executed next or the partial load that is the process of partially loading the test data. Used to determine.

FIGS. 16 and 17 show examples of the update location table 1500 and the restoration location table 1550, respectively. The update location table 1500 and the restoration location table 1550 both record the same data items with the same configuration as the business application table 241. Specifically, items of id 2411, title 2412, publisher 2413, price 2414, and stock 2415 in FIG. 11 are assigned to the rows of the update location table 1500 and the restoration location table 1550, respectively. In the updated location table 1500 and the restored location table 1550, all recorded data are 0 in the initial state.

When 1 is recorded as data in the update location table 1500, the SQL statement issued by the test script 141 included in the test group 1421 for the test data 242 loaded for the test group 1421 executed immediately before It is shown that the data recorded with 1 is changed by the operation instruction, update (update), delete (delete), and insert (insertion) included in. The update location table 1500 is recorded by the test group selection unit 131 in the execution order changing unit 130 of the client 100.

In addition, the test group selection unit 131 refers to the DB update information storage table 245 for the test group candidate to be executed next for the target program, and the SQL statement issued according to the test script 141 included in each test group candidate Examine the data part that will be accessed by. If it is determined that the data in which the update location table 1500 is recorded and the access location (data location determined by the row information / column information) by the test script 141 included in the test group candidate are duplicated, the duplicate location is It is necessary to restore the test script 141 before executing the next test group 1421. Therefore, 1 is recorded in the restoration location table 1550 as the data of the portion determined as the overlapping location. Note that the restoration location is defined as a location that may be used by the test script 141 included in the test group 1421 to be executed next, so the SQL statement for determining the duplicate location is substantially included in the SQL statement. A select operation that does not rewrite data is also included.

Next, the contents of data processing by the program test system 1 of the present embodiment having the above configuration will be described. First, with reference to FIG. 18, the test execution order changing process, which is the main function of the system 1, will be described. FIG. 18 shows an example of a data processing flow for realizing the test execution order changing process of the present embodiment. The test execution order changing process is executed by the test execution order changing unit 130 executed in the client 100.

First, information on each test group 1421 to be tested this time is acquired from the test group information storage table 142 in FIG. Next, it classifies for each test data 1422 which uses those test groups 1421 (S1610). In the case of the present embodiment, as shown in FIG. 8, TestGroup_A, TestGroup_C, TestGroup_D, and TestGroup_F that are Test Groups 1421 that use TestData_1, and TestGroup_B and TestGroup_E that are Test Groups 1421 that use TestData_2, based on the test data 1422 used. Are classified into two sets.

Next, the test by the test group 1421 included in each classified set is repeated until all the test groups 1421 are completed (S1620 loop).

The execution order changing unit 130 determines whether there is a test group 1421 that has not been executed in the selected set (S1621). Whether or not a test has been executed for a specific test group 1421 can be determined by referring to the test group information storage table 142 and determining whether or not the executed flag 1423 recorded therein is 1. Now, considering a set including TestGroup_A, TestGroup_C, TestGroup_D, and TestGroup_F, it is determined that TestGroup_F has not been executed.

When it is determined in S1621 that there is an unexecuted test group 1421 (S1621, Yes), the test is repeatedly executed until the test is executed for all the test groups 1421 included in the same set (S1630). loop). First, an unexecuted test group 1421 is extracted from the test group information storage table 142. If there are a plurality of unexecuted test groups 1421, one of them is selected. In the example of FIG. 8, since the unexecuted test group 1421 related to TestData_1 is only TestGroup_F, the test is executed for this (S1630).

Next, before executing a test on the selected unexecuted test group 1421, the data load unit 234 executes a data load process for reading the test data 1422 (TestData_1 in this embodiment) corresponding to the test group 1421 into the DBMS 240. (S1631). The details of the data loading process will be described later.

Next, the test execution unit 133 in the client 100 sequentially executes the test scripts 141 included in the selected test group 1421 and records execution time information in the DB operation information storage table 244 and the DB update information storage table 245. A time information recording process is executed (S1632). The details of the runtime information recording process will be described later. The test script 141 included in the test group 1421 and its storage location are acquired from the test script information storage table 143 illustrated in FIG. The test execution unit 133 records, for example, in the auxiliary storage device 13 as a test result 144 whether an appropriate result is obtained as a result of executing the test script.

Next, the test execution unit 133 records 1 in the executed flag 1423 corresponding to the test group 1421 that executed the test in the test group information storage table 142 (S1633). The above process is repeated until there is no unexecuted test group 1421.

On the other hand, if it is determined in S1621 that there is no test group 1421 that has not been executed in the set for the same test data 1422 (S1621, No), the execution order changing unit 130 performs all executed tests included in the set. For the group 1421 (the test group in which 1 is recorded as the executed flag 1423 in the test group information storage table 142 of FIG. 8), the following processing is repeatedly executed (loop of S1640). In the example of FIG. 8, since 1 is recorded in the executed flag 1423 for TestGroup_A, TestGroup_C, and TestGroup_D among TestGroup_A, TestGroup_C, TestGroup_D, and TestGroup_F corresponding to the test data TestData_1, these three test groups 1421 are loops of S1640. It becomes the object of processing.

First, the execution order changing unit 130 executes a test group selection process, which is a process for determining which test group 1421 the test is executed next for the selected executed test group 1421 ( S1641). The test group selection process is a process of selecting the test group 1421 so that the load time for reading the test data 1422 for the test group 1421 to be executed next is minimized. A restoration location table 1550 used for executing a test by the group 1421 is generated and notified to the data load unit 234 of the server 200. Specific processing contents of the test group selection processing will be described later.

Next, the data load unit 234 of the server 200 executes the data load process similar to S1631 according to the content notified from the execution order changing unit 130 (S1642). After the loading of the test data is completed, the test execution unit 133 of the client 100 sequentially executes the test scripts 141 included in the selected test group 1421, and the execution time information is stored in the DB operation information storage table 244 and the DB update information storage table 245. A runtime information recording process for recording is executed (S1643). The test execution unit 133 records whether or not an appropriate result is obtained as a result of executing the test script as the test result 144 in, for example, the auxiliary storage device 13 and ends the process.

According to the above test execution order changing process, the test group 1421 having the shortest load time of the test data 1422 can be selected as the subsequent test group 1421 based on the test execution result by each test group 1421. The test execution time for H.231 can be shortened as much as possible.

Next, the runtime information recording process, the test group selection process, and the data load process executed in the test execution order changing process of the present embodiment will be described sequentially.

First, the runtime information recording process will be described with reference to FIG. FIG. 19 shows an example of a data processing flow for realizing the runtime information recording process of the present embodiment. In the execution time information recording process, database operation information and update information via the DBMS 240 by the business application 231 are selected in order to select the load time of the corresponding test data 1422 in the test group determination in the test group selection process described later. It is a process to acquire. The runtime information recording process is executed by the program execution unit 232 of the server 200 in which the business application 231 is running. The program execution unit 232 corresponds to the program execution base in FIG.

The program execution unit 232 acquires the SQL statement issued by the business application 231 to the DBMS 240 from the business application 231, and sets the test script ID 1431 attached to the test script 141 that generated the SQL statement to the test script information storage table 143. (S1705).

Next, the program execution unit 232 analyzes the acquired SQL sentence and records the analysis result in the DB operation information storage table 244 (S1710). The analysis of the SQL sentence can be performed using the technique of an SQL parser implemented in a general DBMS. By analyzing the SQL statement, the DB operation information storage table 244 includes an access pattern 2443, an access table 2444, an access condition 2445, row information 2446 as an access location, and column information 2447 as described with reference to FIG. When the operation content included in the SQL statement to be analyzed is “insert” or “delete”, the corresponding column information 2447 is always all the columns set in the business application table 241. Further, since the line information 2446 is a primary key of the business application table 241, the item of the access condition 2445 does not always include the line information 2446. Therefore, if it is determined that the line information 2446 is not included in the item of the access condition 2445, a select statement using the access condition 2445 is issued once to check the line information 2446.

Next, the program execution unit 232 determines the operation content recorded in the access pattern 2443, and performs the following processing according to the determination result (S1720). When it is determined that the access pattern 2443 is select (S1720 = select), the program execution unit 232 issues the SQL statement issued from the business application 231 as it is to the DBMS 240 and ends the processing (S1730). Since no change is caused in the business application table 241 by executing the select statement, recording in the DB update information storage table 245 as described later is not performed.

When it is determined that the access pattern 2443 is update (S1720 = update), the program execution unit 232 issues an SQL statement using the access table name 2444 and the access condition 2445, and the business application 231 that is actually the target program. Examines which part of the business application table 241 is to be updated. The program execution unit 232 records the row information 2446 and the column information 2447 regarding the obtained access location in the DB update information storage table 245 (S1740). Next, the program execution unit 232 issues the SQL statement issued from the business application 231 to the DBMS 240 for execution (S1741). Next, the program execution unit 232 issues an SQL statement using the access table 2444 and the access condition 2445, and records information on the line including the updated portion in the DB update information storage table 245 as a state after execution ( S1742).

If it is determined that the access pattern 2443 is delete (S1720 = delete), the program execution unit 232 issues an SQL statement using the access table 2444 and the access condition 2445, and the business application 231 that is actually the target program is issued. It is checked which part of the business application table 241 is to be deleted, and information related to the deleted line is recorded in the DB update information storage table 245 as a state before execution (S1750). Then, the program execution unit 232 issues the SQL statement issued from the business application 231 as it is to the DBMS 240 and ends the processing (S1751).

When it is determined that the access pattern 2443 is insert (S1720 = insert), the SQL statement issued from the business application 231 is issued to the DBMS 240 and executed (S1760). Next, the program execution unit 232 issues an SQL statement using the access table 2444 and the access condition 2445, examines what rows are actually inserted into the business application table 241, and obtains information on the inserted rows. The post-execution state is recorded in the DB update information storage table 245, and the process ends (S1761).

In S1720, when it is determined that the access pattern 2443 is neither select, update, delete, or insert (S1720 = others), the program execution unit 232 issues the SQL statement issued from the business application 231 as it is to the DBMS 240. The process ends (S1770). If the same information as the information to be recorded in the DB operation information storage table 244 or the DB update information storage table 245 has already been recorded, the information to be recorded later is not recorded in order to avoid duplication. It is good.

According to the above-described runtime information recording process, the location where the business application 231 actually accesses the business application table 241 can be specified based on the test script 141, and the test data 1421 can be specified in the test group selection process described later. A determination material for determining whether to load the whole can be obtained.

Next, the test group selection process will be described. FIG. 20 shows an example of a data processing flow for realizing the test group selection process in the present embodiment. The test group selection process is a data process executed by the test group selection unit 131 installed in the execution order changing unit 130 of the client 100, where the test group 1421 to be executed next by the client 100 is selected. Is done.

First, the test group selection unit 131 initializes a test group candidate to be used for a test to be executed next, a restoration capacity, and an update location table 1500 (S1810). In this embodiment, the restoration capacity is a place changed by the test group 1421 executed previously in the business application table 241 and restored as a place that should be restored at the minimum when the next test data 1422 is read. The number of locations recorded in the location table 1550 (cells where 1 is recorded in the restored location table 1550).

Next, the test group selection unit 131 records the location of the business application table 241 accessed by the executed test group 1421 in the update location table 1500 with reference to the DB update information storage table 245 (S1820). Thus, basic data for comparison with a location on the business application table 241 that may be accessed by a subsequent test group candidate is prepared.

Next, the test group selection unit 131 repeatedly executes the following processing for all the unexecuted test groups 1421 (S1830 loop). First, the test group selection unit 131 selects an arbitrary one from the unexecuted test groups 1421 (S1831), and initializes the restoration location table 1550 by clearing it to zero (S1832). Next, the test group selection unit 131 may be referred to or updated by the update location (location where 1 is recorded as data) recorded in the update location table 1500 and the selected unexecuted test group 1421. Compared with a certain place, an overlapping place is recorded in the restoration place table 1550 (S1833). For the update location recorded in the update location table 1500 and referenced or updated by the next unexecuted test group 1421, it is necessary to restore the value before the update in order to maintain the reliability of the test result. is there. In the restoration location table 1550, locations that require such data restoration, in other words, locations that must be newly loaded before executing a test by the next test group are recorded. The restoration capacity represents the number of restoration points that need to be restored on the restoration point table 1550. A portion that may be referred to or updated by the unexecuted test group 1421 is acquired by referring to the DB update information storage table 245.

Next, the test group selection unit 131 determines whether or not the access condition 2445 defined by the SQL statement issued by the selected unexecuted test group 1421 includes “range specification” or “subquery” (S1834). “Range specification” means that a certain numerical range is specified for the data to be operated in the where clause included in the SQL statement. In the example of the DB operation information storage table 244 in FIG. “Stock <5 (the number of stock is less than 5)” corresponding to the test script ID 2441 = 101 corresponds. Further, “sub-query” means that another SQL command is included for the data to be operated in the where clause included in the SQL statement.

When it is determined that the access condition 2445 includes “range specification” or “subquery” (S1834, Yes), the test group selection unit 131 displays the entire column on the business application table 241 specified by the access condition 2445. Is recorded in the restoration location table 1550 (S1835). On the other hand, when it is determined that the access condition 2445 does not include both “range specification” and “sub-query” (S1834, No), the test group selection unit 131 restores only a part that matches the access condition 2445 to the restoration part table. It is recorded in 1550 (S1836). When it is determined that “range specification” or “sub-query” is included, the entire column specified in the access condition 2445 is recorded in order to specify the overlapping portion included in the corresponding column. This is because an SQL statement including “or a sub-query” is executed, which may cause a problem that the time required for the test group selection process is prolonged.

Next, the test group selection unit 131 counts the locations where 1 is recorded as data in the restoration location table 1550 as duplicate locations, and records the count as the restoration capacity, for example, in the main storage device 12 (S1837).

Next, the test group selection unit 131 compares the restoration capacity for the test group 1421 for which the restoration capacity is acquired this time with the restoration capacity recorded for other test group candidates for which the restoration capacity has been obtained before ( S1838). When it is determined that the restoration capacity for the test group 1421 that has acquired the restoration capacity this time is smaller than the restoration capacity recorded for another test group candidate that has obtained the restoration capacity before (S1838, Yes) The group selection unit 131 records the current test group 1421 as a new test group candidate (S1839). When it is determined that the restoration capacity for the test group 1421 that has acquired the restoration capacity this time is greater than or equal to the restoration capacity recorded for another test group candidate that has obtained the restoration capacity before (S1838, No). The test group selection unit 131 does nothing. The above process is repeated until there is no unexecuted test group 1421.

When the processing included in the loop of S1830 is completed for all the unexecuted test groups 1421, the test group selection unit 131 selects the test group candidate that is finally recorded as the test group 1421 to be executed next, and performs processing Is finished (S1840).

According to the test group selection process described above, it is possible to select the test group 1421 that is expected to have the shortest load time of the corresponding test data 1422 as the test group 1421 to be executed next. It is possible to reduce the time required for the test execution of the program as much as possible.

Next, the data loading process will be described. FIG. 20 shows an example of a data processing flow for realizing the data load processing of this embodiment. The data load process is a process executed by the load method selection unit 132 installed in the client 100, and the load method selection unit 132 loads test data 1422 used when executing a test by a test group 1421. In this case, it is determined whether the entire test data 1422 is loaded or only a part of the test data 1422 is loaded, and the data loading process according to the determination result is executed by the data loading unit 234 of the server 200.

First, the loading method selection unit 132 determines whether the test group 1421 to be processed is not executed (S1900), and if it is determined that the test group 1421 is not executed (S1900, Yes), refer to the test group information storage table 142. Thus, the corresponding test data 1422 is specified. Then, the loading method selection unit 132 loads the entire test data 1422 on the data loading unit 234, records the actual measurement value of the time required for the loading in the loading time information storage table 243, and ends the processing (S1910). When the target is an unexecuted test group 1421, the first test is executed for the test group 1421, so it is necessary to load the corresponding test data 1422 as a whole.

On the other hand, when it is determined that the test group 1421 to be processed is not unexecuted (S1900, No), the load method selection unit 132 refers to the load time information storage table 243, and the test data 1422 part for the test group 1421 It is determined whether or not an actual measurement value of the load time 2435 when a specific load (hereinafter referred to as “partial load”) is executed is recorded (S1920). Whether or not the actual measurement value is recorded at the load time 2435 for the partial load is determined if the load information 2433 and the column information 2434 match the data to be loaded by the partial load, and the load time 2435 has a value. Suppose that actual measurement values are recorded. If it is determined that the actual load value 2435 for the partial load is recorded (S1920, Yes), the load method selection unit 132 acquires the actual value and sets it as the partial load time for the test group 1421 ( S1921).

On the other hand, when it is determined that the load time 2435 actual measurement value for the partial load is not recorded (S1920, No), the load method selection unit 132 calculates the restoration capacity for the test group 1421 to be processed (S1922), A predicted value of the time required for partial loading of the corresponding test data 1422 is calculated using the calculated restoration capacity (S1923). The predicted value of the partial load time can be calculated by, for example, Equation 1 below.
(Partial load time predicted value) = (Partial load capacity) × α (seconds) Equation 1
Here, “partial load capacity” is the capacity of data to be partially loaded in the test data 1422, and α is a conversion coefficient empirically determined for the partial load data capacity of a unit in order to obtain the partial load required time. The actual load time value for loading the entire test data 1422 and the total test data capacity indicating the capacity of the entire test data 1422 may be used to calculate the following formula 2.
(Partial load time predicted value) = (Total load time) × [(Partial load capacity) / (Test data total capacity)] × α (second)

Next, the load method selection unit 132 uses the partial load time acquired in S1921 or S1923 to refer to the load time information storage table 243 and measure the actual load time 2435 recorded for the corresponding test group 1421. (S1930). If it is determined that the total load time is shorter than the partial load time (S1930, Yes), the load method selection unit 132 performs partial load processing according to the recorded contents of the restoration location table 1550 for the corresponding test data 1421. The load unit 234 executes the measurement, records the actual measurement value in the load time information storage table 243, and ends the processing (S1931). On the other hand, when it is determined that the partial load time is equal to or greater than the total load time (S1930, No), the load method selection unit 132 causes the data load unit 234 to execute the corresponding load processing of the entire test data 1421 and the actual measurement value thereof. Is stored in the load time information storage table 243, and the process is terminated (S1910).

According to the data load process described above, the entire load time, which is the time required to load the entire test data 1422 used for all the test groups 1421, is measured and recorded. Since it is possible to determine whether or not the test data 1422 can be shortened compared to the entire load time when the test data 1422 is partially loaded, it is possible to more reliably reduce the time required for data loading and increase the effect of reducing the test execution time. it can.

As described above, according to the present embodiment, when executing a test of a program, the test execution time itself is reduced by reducing the time required for loading test data for each component constituting the program as much as possible. And test can be executed efficiently.

Note that the embodiment of the invention described above is for facilitating understanding of the present invention, and does not limit the present invention. The present invention can be changed and improved without departing from the gist thereof, and equivalents thereof are also included in the present invention.

Claims (15)

1. A computer for causing each program to execute processing of test data used for testing each of the plurality of programs,
   Recording a first data portion which is a portion of the test data changed by the test among the test data used by the test of the program executed immediately before,
   Obtaining a second data portion which is a portion of the test data to be accessed by the test among the test data read for the other program to be subsequently executed;
   Selecting the other program for which the difference between the first data portion and the second data portion is minimized to be subjected to the subsequent test,
   How to test the program.
2. The program reads the test data into a database,
   The computer
   Obtaining the first or second data portion based on an operation instruction executed on the database when the program is tested using the test data stored in the database;
   The program test method according to claim 1.
3. The computer is
   When it is determined that the test data is only referred to by the test when the first data portion is recorded, the referenced test data portion is not recorded as the first data portion. ,
   The program test method according to claim 1.
4. The computer is
   When it is determined that the operation instruction executed on the database when the program is tested includes a secondary operation instruction related to data range specification or operation target data. Including the data having the same attribute as the target data as the first or second data portion,
   The method for testing a program according to claim 2.
5. The computer is
   The amount of data to be read for the difference between the first data portion and the second data portion obtained for any of the programs and the test read for another program to be subsequently executed When comparing the data amount of data and determining that the amount of data read for the difference is smaller than the amount of data for reading the test data, read the data corresponding to the difference,
   The program test method according to claim 1.
6. The computer is
   The amount of data to be read for the difference between the first data portion and the second data portion obtained for any of the programs is converted into the reading time, and the other programs to be subsequently executed are converted. When it is determined that the reading time for the difference is shorter than the actual reading time of the test data, compared to the actual reading time value of the test data that is read for the purpose, the data corresponding to the difference is read. ,
   The program test method according to claim 1.
7. The computer is
   Measure the time required for reading the test data for each program and record the measured value.
   The method for testing a program according to claim 6.
8. A program test system for causing each program to execute processing of test data used for testing each program for a plurality of programs,
   Recording a first data portion which is a portion of the test data changed by the test among the test data used by the test of the program executed immediately before,
   Obtaining a second data portion which is a portion of the test data to be accessed by the test among the test data read for the other program to be subsequently executed;
   A test execution order changing unit that selects another program that is the smallest difference between the first data portion and the second data portion as a target to be continuously tested;
   Program test system.
9. The program reads the test data into a database,
   The test execution order changing unit
   Obtaining the first or second data portion based on an operation instruction executed on the database when the program is tested using the test data stored in the database;
   The program test system according to claim 8.
10. The test execution order changing unit
    When it is determined that the test data is only referred to by the test when the first data portion is recorded, the referenced test data portion is not recorded as the first data portion. ,
    The program test system according to claim 8.
11. The test execution order changing unit
    When it is determined that the operation instruction executed on the database when the program is tested includes a secondary operation instruction related to data range specification or operation target data. Including the data having the same attribute as the target data as the first or second data portion,
    The program test system according to claim 9.
12. The test execution order changing unit
    The amount of data to be read for the difference between the first data portion and the second data portion obtained for any of the programs and the test read for another program to be subsequently executed When comparing the data amount of data and determining that the amount of data read for the difference is smaller than the amount of data for reading the test data, read the data corresponding to the difference,
    The program test system according to claim 8.
13. The test execution order changing unit
    The amount of data to be read for the difference between the first data portion and the second data portion obtained for any of the programs is converted into the reading time, and the other programs to be subsequently executed are converted. When it is determined that the reading time for the difference is shorter than the actual reading time of the test data, compared to the actual reading time value of the test data that is read for the purpose, the data corresponding to the difference is read. ,
    The program test system according to claim 8.
14. The test execution order changing unit
    Measure the time required for reading the test data for each program and record the measured value.
    The program test system according to claim 13.
15. A computer that causes each program to execute processing of test data used for testing each of the plurality of programs,
    Recording a first data portion which is a portion of the test data changed by the test among the test data used by the test of the program executed immediately before,
    Obtaining a second data portion which is a portion of the test data to be accessed by the test among the test data read for the other program to be subsequently executed;
    Causing the other data program having the smallest difference between the first data portion and the second data portion to be selected as a target to be continuously tested.
    program.

Images