JP2009104489A - Apparatus and method for testing code clone - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an apparatus and method for testing code clones that efficiently tests code-cloned source codes. <P>SOLUTION: A server 100 has a source code file 2, a code clone output file 3, a test result file 4 and a code clone merging file 7, and includes a data setting part 8, a code clone detection/merge processing part 9 and a test result setting part 10 as function blocks implemented by hardware resources in the server 100. The code clone detection/merge processing part 9 detects and merges code clone sets in software source programs stored in the source code file 2. The test result setting part 10 tests one code clone in the detected code clone set, and sets test case numbers to the tested source code lines to indicate that the entire code clone set has been tested. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to testing of software source code programs including embedded software, and more particularly to a code clone testing apparatus and method for improving the efficiency of code cloned source code testing.

  Code clones are often used in software development. The code clone indicates a duplicate code string in the source code, and a software including a large number of code clones is generated by copying and pasting all or part of the source code. However, if one code clone is changed, it is often necessary to make the same change to the other code clones, which causes an increase in the cost of correcting the software.

Patent Document 1 discloses a technique for measuring the similarity of source code from the output result of the code clone detection tool.
Patent Document 2 discloses a technique regarding a code clone detection method.

Japanese Patent Application Laid-Open No. 2004-228561 discloses a technique that enables a check item that has been performed at a diversion source to be diverted to a diversion destination when the program is diverted.
Non-Patent Document 1 defines a dependency relationship between code clone sets as a chained clone, and discloses a chained clone refactoring method.

Non-Patent Document 2 introduces a code clone detection tool.
JP 2003-216425 A Japanese Patent Laid-Open No. 2003-29978 JP-A-5-250213 Norihiro Yoshida and four others "Supporting Referering Based on Dependencies Between Code Clones" IPSJ Journal Vol.48, No.3, Mar. 2007, pp.1431-1442 Kazuo Kobori “Let's find“ copied ”code using the code clone detection tool” Software Test PRESS Vol.4, pp.124, Technical Review (2007)

  Software source code tests generally include a C0 coverage test that covers a single instruction of the source code, a C1 coverage test that covers true and false branches, and a C2 coverage test that covers combinations of conditions.

  As a source code test method, there are a method of creating a test specification covering the above-mentioned coverage and confirming it for each step, a method of creating a test program and automatically executing it. In addition, tools such as XUNIT that generate, execute, and verify test programs may be used.

  In the source code, there is a code clone duplicated by copy and paste in order to reuse all or part of a certain code. In principle, the reuse part is used as a function for reuse, but there are cases where the code is cloned without being functionalized because there are some parts that are different. Conventionally, since it is not possible to distinguish where the code-clone portion is, the source code is tested regardless of the presence or absence of the code clone, thus allowing labor and use of test equipment for unnecessary tests.

  On the other hand, in Patent Document 1, the similarity of source code can be measured, but it cannot be used for testing. Moreover, in the said patent document 2, although it can utilize for the detection of a code clone, it cannot utilize for a test. Moreover, in the said patent document 3, when diverting a program, although the check item performed by the diversion source can be diverted to diversion destination, it is not mentioned whether it can apply to a test. Furthermore, although the said nonpatent literature 1 has defined the chained clone, since it is the chain between the code clone functions by the reference relation of a variable, and the test is closed with a function, it cannot be used for a test. Non-Patent Document 2 is an introduction of a code clone detection tool, and does not touch the test.

  Accordingly, an object of the present invention is to provide a code clone testing apparatus and method for efficiently testing the code cloned source code.

  A code clone test apparatus according to the present invention detects a code clone set from data setting means for storing a software source program including a code clone set in a source code file, and the software source program stored in the source code file. A code clone detection means and a test that tests one code clone of the detected code clone set, sets a test case number in the tested source code line, and displays that the entire code clone set has been tested Result setting processing means.

  The code clone test method of the present invention includes a step of storing a software source program including a code clone set, a step of detecting the code clone set from the stored software source program, and one of the code clone sets. And testing a single code clone, and setting a test case number to a source code line tested for the code clone to display that the entire code clone set has been tested.

  According to the present invention, it is possible to omit a test that has conventionally been performed on a portion copied by copy and paste, and it is also possible to efficiently test a source code that has been copied and partially changed. Further, according to the present invention, a source code line tested for a code clone duplicated by copy and paste in order to reuse all or a part of a certain code is clarified, and a test of a duplicate source code can be prevented. In addition, for the source code copied and pasted by copy and paste, the untested part of the changed part becomes clear, and the test specifications and test cases that need to be changed can be identified. Can be created and executed semi-automatically.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a diagram showing a schematic configuration of a system that embodies a code clone test apparatus according to an embodiment of the present invention. In the system configuration of FIG. 1, the code clone test apparatus 1 is configured in the server 100, and the code clone test apparatus 1 is coupled to the client 300 via the network 200. The code clone test apparatus 1 stores the source code file 2, the code clone output file 3, the test result file 4, the test specification file 5, the change test specification file 6, the code in the storage device which is a hardware resource in the server 100. Data setting processing unit 8, code clone detection as a functional block realized by a central processing unit (CPU), an input / output device, a storage device, etc., which are provided with a clone merge file 7 and are hardware resources in the server 100・ Merge processing unit 9, test result setting processing unit 10,
A change test case extraction processing unit 11 and a change test specification creation processing unit 12 are provided. Here, the server 100 is realized by computer hardware resources known to those skilled in the art. For example, a CPU, a storage device (including ROM, RAM, HDD), an input / output device (communication I / F) as computer hardware resources. , Display, printer, mouse, keyboard, etc.).

  The network 200 links the client 300 and the server 100 (code clone test apparatus 1), and a well-known network technology such as a LAN or a public network (including the Internet) can be used.

  The client 300 is connected to the server 100 (code clone test apparatus 1) via the network 200, and starts processing and displays data for the server 100 (code clone test apparatus 1) using computer hardware resources of a known client. Set the data.

The source code file 2 is a file for storing source code to which the present invention is applied, and FIG. 2 shows a configuration example of the source code file 2. This will be described later.
The code clone output file 3 is a file for storing data detected using a code clone detection tool such as CCFinderX introduced in Non-Patent Document 2, and a configuration example of the code clone output file 3 is shown in FIG. Show. This will be described later.

  The test result file 4 is a file for storing the test result executed in correspondence with the source code as a test case number. FIG. 4 shows a configuration example of the test result file 4. This will be described later.

The test specification file 5 is a file for storing test specifications related to the code clone portion. FIG. 5 shows a configuration example of the test specification file 5. This will be described later.
The change test specification file 6 is a file for storing the test specification of the part changed from the code clone, and FIG. 6 shows a configuration example of the change test specification file 6. This will be described later.

  The code clone merge file 7 is a file for storing data of a set of code clone sets determined to be mergeable between code clone sets. FIG. 11 shows a configuration example of the code clone merge file 7. This will be described later.

  The data setting processing unit 8 executes processing for setting source code data corresponding to the source code file 2. FIG. 12 is a flowchart for explaining the operation of the data setting processing unit 8. This will be described later.

  The code clone detection / merge processing unit 9 detects a code clone set by a code clone detection tool such as the CCFinderX described above, and executes a process of determining whether or not to merge between code clone sets. FIG. 13 shows a flowchart for explaining the operation of the code clone detection / merge processing unit 9. This will be described later.

  The test result setting processing unit 10 registers the test specification or test program of the source code including the code clone part in the test specification file 5, executes the test, and sets the test case number corresponding to the tested source code to the test result file 4 Execute the process stored in. FIG. 14 is a flowchart for explaining the operation of the test result setting processing unit 10. This will be described later.

  The changed test case extraction processing unit 11 executes a process of extracting a test case corresponding to the source code of the changed portion of the merged code clone set and storing it in the changed test specification file 6. FIG. 15 is a flowchart for explaining the operation of the changed test case extraction processing unit 11. This will be described later.

  The change test specification creation processing unit 12 executes a process of semi-automatically creating a test specification corresponding to the source code of the changed part of the merged code clone set and storing it in the change test specification file 6. FIG. 16 is a flowchart for explaining the operation of the change test specification creation processing unit 12. This will be described later.

  The overview of the system configuration in FIG. 1 has been described above, and the details will be described below. The flow of processing between the components in the code clone test apparatus 1 of FIG. 1 is indicated by a bold arrow, and the flow of data is indicated by a thin arrow.

  FIG. 2 is a diagram showing a configuration example of a source code file according to the embodiment of the present invention. As shown in FIG. 2, the source code file 2 includes a source code file name 21 and a source code 22. Although not shown in the drawing, a series of source codes corresponding to the source code file name 21 is stored in the source code 22 portion.

  FIG. 3 is a diagram showing a configuration example of the code clone output file according to the embodiment of the present invention. As shown in FIG. 3, the code clone output file 3 includes a code clone set name 31, a source code file name 32, a start line number (illustrated number 33), and an end line number (illustrated number 34). The code clone set name 31 is assigned the same name (for example, cloneset1) for the same code set. The source code file name 32 stores the source code file name of the same code. The line number of the code where the same code starts is stored in the start line number (illustrated number 33), and the line number of the code where the same code ends is stored in the end line number (illustrated number 34).

  FIG. 4 is a diagram showing a configuration example of a test result file according to the embodiment of the present invention. As shown in FIG. 4, the test result file 4 includes a source code file name 41, a line number (illustrated number 42), and a test target function / test case number (illustrated number 43). The source code file name 41 stores the source code file name where the code clone exists. In the line number (number 42 in the figure), a series of line numbers from the start line number to the end line number of the code clone output file 3 or the code clone merge file 7 if the code clone set name exists in the code clone merge file 7 Stores a series of line numbers from the start line number to the end line number. The test target function / test case number (illustrated number 43) stores the test target function name and test case number corresponding to the unit of the test specification corresponding to the row number (illustrated number 42). The test case number is plural when a plurality of test cases correspond to the source code line.

  FIG. 5 is a diagram showing a configuration example of a test specification file according to the embodiment of the present invention. As shown in FIG. 5, the test specification file 5 includes a test target function / source code file name 51, a test case number or test program number (illustrated number 52), an input value (illustrated number 53), and a verification value (illustrated number 54). Consists of The test target function / source code file name 51 stores the test target function name and the source code file name to identify the test specification. In the test case number or the test program number (number 52 in the figure), a number for identifying the test case or a number for identifying the test program when there is a test program for executing the test case is stored. The input value (number 53 in the figure) stores the input value necessary for passing the route of the test case. A verification value (verification number 54) stores a verification value for verifying whether the source code is valid.

  FIG. 6 is a diagram showing a configuration example of a change test specification file according to the embodiment of the present invention. As shown in FIG. 6, the changed test specification file 6 stores the changed portion of the test specification file 5 shown in FIG. 5, and includes a test target function / source code file name 61, a test case number or a test program number (shown in FIG. 6). Number 62), an input value (illustrated number 63), and a verification value (illustrated number 64). The test target function / source code file name 61 stores the test target function name and the source code file name to identify the test specification. In the test case number or the test program number (number 62 in the figure), a number for identifying the test case or a number for identifying the test program when there is a test program for executing the test case is stored. The input value (number 63 in the figure) stores the input value necessary for passing the route of the test case. The verification value (in the figure, number 64) stores a verification value for verifying whether the source code is valid.

  FIG. 7 is a diagram showing a code clone detection display example obtained by the code clone detection process of the code clone detection / merge processing unit 9 according to the embodiment of the present invention. As shown in FIG. 7, the code clone source code (see lines 111 to 122 shown in FIG. 7) detected using a code clone detection tool such as the CCFinderX described above is detected by, for example, shading. Display the code clone source code clearly.

  FIG. 8 is a diagram showing a code clone detection / merge display example obtained by the code clone detection / merge processing of the code clone detection / merge processing unit 9 according to the embodiment of the present invention. In the display example shown in FIG. 8, the left and right sides of the illustrated shaded rectangular portion, that is, the shaded rectangular portions 81 and 82, the shaded rectangular portions 83 and 84, and the shaded rectangular portions 85 and 86 are the same code clone set. Is shown. In addition, three vertical code clone sets (shaded rectangular portions 82, 84, 86) indicate merge code clone sets obtained by merging the left three clone sets.

  FIG. 9 is a diagram showing a test result setting example by the test result setting processing unit 10 according to the embodiment of the present invention. As shown in FIG. 9, the test case number of the test performed corresponding to the line number of the source code is set in the right margin of the source code line. In the illustrated example, it is shown that test case number 1 and test case number 2 are set corresponding to the performed test, and the source code where the test was not performed is blank.

  FIG. 10 is a diagram showing a test result setting / display example obtained by the processing of the test result setting processing unit 10 according to the embodiment of the present invention. In the example shown in FIG. 10, test case numbers (test case numbers 1, 2, and 3 are represented by (a)) of test results corresponding to the code clone detection / merge display example shown in FIG. 8 are newly set. The test case number “1a” is displayed at the right end of FIG. In the example shown, the first screen (left part) and the second screen (right part) are distinguished by the numbers at the top of the figure and the two screens are displayed on the left and right. The test case number (illustrated number 91) of line number 105 and the test case number (illustrated number 92) of line number 117 are extracted as the test case numbers to be changed and executed on the second screen (right part). The test case number (illustrated number 93) of the line number 255 and the test case number (illustrated number 94) of the line number 267 indicate that the test case number “1a” is newly set.

FIG. 11 is a diagram showing a configuration example of a code clone merge file according to the embodiment of the present invention. As shown in FIG. 11, the code clone merge file 7 includes a merge name 71, a code clone set name 72, a source code file name 73, a start line number (illustrated number 74), and an end line number (illustrated number 75). . The merge name 71 has the same name for the same code clone set order. The code clone set name 72 stores code clone set names having the same code clone set order. The source code file name 73 stores a source code file name having a code clone having the same code clone set order. In the starting line number (illustrated number 74), the starting line number of the first code clone set in the order of the clone set name is stored. Further, the ending line number (number 75 in the figure) stores the ending line number of the last code clone set in the order of the code clone set name. The illustrated example corresponds to the code clone detection / merge display example shown in FIG.

  Based on the detailed description of each component described above, the operation of the code clone test apparatus according to the embodiment of the present invention will be described below with reference to the flow of operations of the processing units and the flowcharts describing the operations shown below. . Although not shown, an initial screen is displayed on the user terminal of the client 300 by logging in to the server 100, and the code clone test apparatus 1 is activated on this initial screen. That is, the data setting processing unit 8 is first started at the user terminal of the client 300. Each processing unit shown below also performs processing based on the processing flow and data flow shown in FIG.

  FIG. 12 is a flowchart for explaining the operation of the data setting processing unit 8 according to the embodiment of the present invention. In the flowchart shown below, step is abbreviated as “S”. 12, in step S1, the target source code is written in the source code file 2, and the source code file name 21 (see FIG. 2) is registered in the source code file 2. Then, the process ends.

  FIG. 13 is a flowchart for explaining the operation of the code clone detection / merge processing unit 9 according to the embodiment of the present invention. In step S11 in FIG. 13, the target source code is read from the source code file 2. In step S12, a code clone is detected by a code clone detection tool (not shown). Note that a code clone detection tool (not shown) such as CCFinderX is prepared in the server 100 and made accessible as in the code clone test apparatus 1, or made accessible through the network 200. . In step S13, data obtained by a code clone detection tool (not shown) is written in the code clone output file 3.

  In step S14, the code clone output file 3 is read. In step S15, it is determined whether or not there are a plurality of code clone set names in the same source code file name. If there are not a plurality of code clone set names, the process jumps to step S20. If there are a plurality of code clone set names, in step S16, it is determined whether or not the number of lines in the interval of the code clone set names is smaller than the threshold value. If larger, the process jumps to step S20. In this case, the threshold can be arbitrarily set by the user. If it is smaller than the threshold, in step S17, the name and order of the code clone set names are stored (temporarily) in, for example, a work memory (not shown).

  In step S18, it is determined whether other source code file names have the same order as the names of the code clone set names stored above (temporarily). If there is no same, the process proceeds to step S20. jump. If there is the same one, in step S19, a merge name is added to the sequence of the code clone set name, merge name 71 of code clone merge file 7, code clone set name 72, source code file name 73, start line number (illustrated Number 74) and the end line number (illustrated number 75) are written in each item (see FIG. 11). In step S20, all of the source code file names 32 (see FIG. 3) of the code clone output file 3 are confirmed. If the code clone output file 3 is confirmed, the process ends. If there is an unconfirmed item, the process returns to step S15.

FIG. 14 is a flowchart for explaining the operation of the test result setting processing unit 10 according to the embodiment of the present invention. In FIG. 14, in step S21, the code clone output file 3, the code clone merge file 7, and the source code file 2 are read. In step S22, a test specification for a function corresponding to a code clone having one source code file name among a plurality of source code file names having the same code clone set name is registered in the test specification file 5. If the code clone set name has a merge name, register the test specification for the function to which the code clone of one source code file name among the multiple source code file names with the same merge name corresponds in the test specification file 5 To do.

  In step S23, a test is executed for the test specification registered in the process of step S22. In this case, there are a case where the test is performed manually, a case where the test is automatically performed by a test program, and a case where the test is automatically performed by a test tool (not shown) such as XUNIT. Note that the test tool (not shown) such as the XUNIT described above is prepared in the server 100 and can be accessed as in the code clone test apparatus 1, or can be accessed via the network 200. . In step S24, the executed test case number is written in the test result file 4 in association with the line of the source code file name that executed the test. In this case, the line number to be written is written from the start line number to the end line number of the code clone output file 3 for the code clone set name without the merge name, and the code clone for the code clone set name with the merge name. Writing from the start line number to the end line number of the merge file 7 is performed. In step S25, the test case number is similarly written in the test result file 4 corresponding to the other source code file name line of the same code clone set name. Then, the process ends.

  FIG. 15 is a flowchart for explaining the operation of the changed test case extraction processing unit 11 according to the embodiment of the present invention. In step S31 in FIG. 15, the test result file 4 and the code clone merge file 7 are read. In step S32, the same merge name 71 (see FIG. 11) of the code clone merge file 7 describes the test case number of the test result file 4 from the start line number (illustrated number 74) to the end line number (illustrated number 75). Extract a range of line numbers that have not been processed. The line number corresponds to a line between code clone sets. In step S33, although the test case number is described in the line between the code clone sets extracted in step S32, the test target function and the test case number are extracted. In step S34, the test specification corresponding to the test target function and the test case number extracted in the process of step S33 is read from the test specification file 5 and written to the changed test specification file 6. Then, the process ends.

  FIG. 16 is a flowchart for explaining the operation of the modified test specification creation processing unit 12 according to the embodiment of the present invention. In step S41 in FIG. 16, the modified test specification file 6 is read. In step S42, the test case of the change test specification file 6 is corrected. If the test case cannot be dealt with, such as when a branch is added, a new test case is added to the changed test specification file 6. Then, the process ends.

  In combination with what has been described above, the code clone test apparatus according to the embodiment of the present invention uses a code clone detection tool such as the above-described CCFinderX for the code clone set of the source code set in the source code file. To detect. When one code clone of the detected code clone set is tested, since the other code clones are the same source code, the test can be omitted assuming that the code clones have already been tested. In some cases, a part of the code clone is changed and used, and in this case, detection can be made by arranging a plurality of code clone sets in the same order. When detecting, by setting a threshold value for the number of source code lines between code clone sets, detection other than intended can be eliminated. In the case of a merged code clone set, it is necessary to test a part that has been partially changed, and a test specification is generated semi-automatically.

The test result is set by executing a test on the test specification corresponding to the code clone set or the merged code clone set and specifying the test case number corresponding to the source code. It is also possible to automatically grasp the test execution and coverage by using a test tool such as XUNIT.

  As a method for testing a part of the merged code clone set that has been partially changed, the changed part can be detected as source code between code clone sets, so that it is possible to grasp which test case of which test specification the source code is. The changed part includes code that does not participate in branching and code that includes branching. Code changes that do not participate in branching can be handled by adding, deleting, or changing test case input values or verification values. On the other hand, the change part of the code including the branch includes addition / deletion / change of the test case.

  For the test cases of the test specification in the changed part of the code not involved in the branching, the corresponding test case can be automatically extracted by the test case of the merged code clone set on which the test is performed. This test case is semi-automatic because the input value or verification value is added, deleted, or changed manually. Test specifications can be automatically extracted for test cases of test specifications in the changed part of the code including the branch, but the test cases are added / deleted / changed manually and become semi-automatic.

  In addition, when a bug is found in the code cloned part and changed, the changed part is the source code between the end line of the previous code clone set and the start line of the subsequent code clone set between the code clone sets to be merged It can be easily detected as a line, bugs caused by forgetting to change can be prevented, and the quality of software development can be improved.

It is a figure which shows the structure outline | summary of the system which implements the test apparatus of the code clone which concerns on embodiment of this invention. It is a figure which shows the structural example of the source code file which concerns on embodiment of this invention. It is a figure which shows the structural example of the code clone output file which concerns on embodiment of this invention. It is a figure which shows the structural example of the test result file which concerns on embodiment of this invention. It is a figure which shows the structural example of the test specification file which concerns on embodiment of this invention. It is a figure which shows the structural example of the change test specification file which concerns on embodiment of this invention. It is a figure which shows the code clone detection display example obtained by the code clone detection process of the code clone detection and merge process part which concerns on embodiment of this invention. It is a figure which shows the example of a code clone detection and merge display obtained by the code clone detection and merge process of the code clone detection and merge process part which concerns on embodiment of this invention. It is a figure which shows the example of a test result setting by the test result setting process part which concerns on embodiment of this invention. It is a figure which shows the example of a test result setting and display obtained by the setting process of the test result setting process part which concerns on embodiment of this invention. It is a figure which shows the structural example of the code clone merge file which concerns on embodiment of this invention. It is a flowchart explaining operation | movement of the data setting process part which concerns on embodiment of this invention. It is a flowchart explaining operation | movement of the code clone detection and merge process part which concerns on embodiment of this invention. It is a flowchart explaining operation | movement of the test result setting process part which concerns on embodiment of this invention. It is a flowchart explaining operation | movement of the change test case extraction process part which concerns on embodiment of this invention. It is a flowchart explaining operation | movement of the change test specification creation process part which concerns on embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Code clone test apparatus 2 Source code file 3 Code clone output file 4 Test result file 5 Test specification file 6 Change test specification file 7 Code clone merge file 8 Data setting process part 9 Code clone detection and merge process part
10 Test result setting processing section
11 Change test case extraction processing part
12 Test result setting processing section
100 servers
200 networks
300 clients

Claims (9)

Data setting means for storing a software source program including a code clone set in a source code file;
Code clone detection means for detecting the code clone set from the software source program stored in the source code file;
Test result setting processing means for testing one code clone of the detected code clone set, setting a test case number in the tested source code line, and displaying that the entire code clone set has been tested,
A test apparatus for a code clone, comprising:   2. The code clone set according to claim 1, further comprising: a code clone detection / merging processing means for detecting and merging a code clone set having the same sequence of a plurality of code clone sets stored in the source code file. Testing equipment.   The code clone detection / merge processing means has a threshold used for detecting the same sequence of code clone sets, and the threshold does not detect the number of source code lines between code clone sets other than intended. The code clone set test apparatus according to claim 2.   A code clone merge file storing the plurality of merged code clone sets and a series of source codes across the plurality of merged code clone sets stored in the code clone merge file are read and tested. 4. The code clone test apparatus according to claim 2, further comprising a modified test case extraction processing means for automatically detecting a portion that is not present.   5. The code clone test apparatus according to claim 4, further comprising a modified test specification creation processing means for semi-automatically generating a test specification of source code for which the test is not performed. Storing a software source program including a code clone set;
Detecting the code clone set from the stored software source program;
Testing one code clone of the code clone set;
Setting a test case number in the source code line tested for the code clone and displaying that the entire code clone set has been tested;
A method for testing a code clone, comprising:   7. The method for testing a code clone set according to claim 6, further comprising a step of detecting and merging code clone sets having the same sequence in a plurality of code clone sets.   8. The method for testing code clones according to claim 7, further comprising a step of automatically detecting a portion of the source code that is merged in the plurality of code clone sets that has not been tested.   9. The code clone test method according to claim 8, further comprising a step of semi-automatically generating a test specification of a source code for which the test is not performed.