JP2016071774A - Verification support device, verification support method, and computer program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce a time required for an operation for verifying whether an inspection result of a source code of a computer program (program) is correct.SOLUTION: A data creation part 2 of a verification support device 1 extracts groups of source code lines relating to processing determined to have a risk of generating a failure, on the basis of inspection result information added to an inspected source program, as failure candidate code groups from the inspected source program. The data creation part 2 sorts a plurality of the source code lines constituting the failure candidate code groups into expression groups including expression lines, which are the source code lines that generate failures, and factor groups including factor lines, which are the source code lines that is to be the factors of the failures. A calculation part 3 calculates a similarity between the expression groups and a similarity between the factor groups in a plurality of the failure candidate code groups extracted by the data creation part 2. A support part 4 presents information that includes information of the similarities calculated by the calculation part 3.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a technique for inspecting a computer program (software) under development.

  In recent software (computer program) development, with the expansion of development scale, the work of extracting logical errors (problems) in the source code in software (source program) has become very time-consuming and time-consuming. . For this reason, in order to assist the extraction work, a tool that points out a bug in the source code by analyzing the source code without executing the software (hereinafter also referred to as a static analysis tool) has been developed. . For example, when the source code of a source program to be inspected is input, such a tool constructs a unique program flow inside the tool (system), inspects for bugs and security vulnerabilities, and outputs inspection results.

  By using such a static analysis tool, it is possible to shorten the software development time. However, the general-purpose static analysis tool outputs a large number of indications (inspection results) that there is a possibility that a problem may occur in the source program to be inspected even though the problem does not actually occur. For this reason, the program developer may take unnecessary time for verification (confirmation) of the indication (inspection result).

  In order to reduce the burden on such a program developer, for example, Patent Document 1 discloses the following technique. The technology is a technique for grouping multiple program parts that have been pointed out based on the location information and type information of the program part that is pointed out (determined) by the static analysis tool when there is a risk of a problem occurring. is there. Patent Document 2 discloses a technique for calculating a similarity between a plurality of program parts that have been pointed out based on syntax information of source code and presenting a pair of program parts having a high similarity. It is disclosed.

JP 2012-248050 A Japanese Patent No. 512538

  In the technique disclosed in Patent Document 1, the program parts pointed out to be defective are presented in a grouped state. Information that is a criterion for determining the grouping is position information and type information of the program part. Since these pieces of information are not useful information for verifying whether or not a failure will occur, even if program parts that have been pointed out are presented in groups, the time required for verifying those program parts can be shortened. It is difficult to plan.

  Further, in the technique of Patent Document 2, a pair of program parts having a high degree of similarity is presented. Even if it is determined that one of the pairs is a normal program part that does not cause a problem by verification, It cannot be said that there will be no problems. For this reason, it is necessary to verify both of the paired program parts, and it is difficult for the technique disclosed in Patent Document 2 to shorten the verification work time.

  The present invention has been devised to solve the above problems. That is, the main object of the present invention is to provide a technique for shortening the time required for the work of verifying the inspection result by the static analysis tool.

In order to achieve the above object, the verification support apparatus of the present invention provides:
Execution according to the inspected source program based on inspection result information added to the inspected source program in which a plurality of source code lines representing respective processing steps of the plurality of processes executed by the computer are arranged Among the plurality of possible processes, a group of the source code lines related to the process that may cause a defect is extracted as a defect candidate code group from the inspected source program, and the defect candidate code group Are divided into an expression group including an expression line that is the source code line in which a defect occurs and a factor group including a factor line that is the source code line that causes the defect. A data creation unit;
When there are a plurality of defect candidate code groups extracted by the data creation unit, a calculation unit that calculates the similarity between the expression groups and the factor groups in the extracted defect candidate code groups,
A support unit that presents information including information on the similarity calculated by the calculation unit.

Further, the verification support method of the present invention includes:
Execution according to the inspected source program based on inspection result information added to the inspected source program in which a plurality of source code lines representing respective processing steps of the plurality of processes executed by the computer are arranged Among the plurality of possible processes, a group of the source code lines related to the process that may cause a defect is extracted from the inspected source program as a defect candidate code group,
A plurality of source code lines constituting the defect candidate code group, an expression group including an expression line that is the source code line in which the defect occurs, and a factor line that is the source code line that causes the defect Divided into factor groups including
When there are a plurality of the extracted defect candidate code groups, calculate the similarity between the expression groups and the factor groups in the extracted defect candidate code groups,
Information including the calculated similarity information is presented.

Furthermore, the computer program of the present invention is
Execution according to the inspected source program based on inspection result information added to the inspected source program in which a plurality of source code lines representing respective processing steps of the plurality of processes executed by the computer are arranged A process of extracting from the inspected source program a group of source code lines related to the process that may cause a failure among the plurality of possible processes as a defect candidate code group;
A plurality of source code lines constituting the defect candidate code group, an expression group including an expression line that is the source code line in which the defect occurs, and a factor line that is the source code line that causes the defect Processing to be divided into factor groups including
When there are a plurality of the extracted defect candidate code groups, a process of calculating the similarity between the expression groups and the factor groups in the extracted defect candidate code groups;
A processing procedure for causing a computer to execute a process of presenting information including the calculated similarity information is shown.

  The main object of the present invention is also achieved by the verification support method of the present invention corresponding to the verification support apparatus of the present invention. The main object of the present invention is also achieved by a computer program corresponding to the verification support apparatus and verification support method of the present invention.

  ADVANTAGE OF THE INVENTION According to this invention, the time which the operation | work which verifies the result of the test | inspection with respect to the source code of a computer program can be shortened.

It is a block diagram which simplifies and represents the structure of the verification assistance apparatus of 1st Embodiment which concerns on this invention. It is a block diagram which simplifies and represents the structure of the verification assistance apparatus of 2nd Embodiment which concerns on this invention. It is a figure showing the specific example of the source program of test | inspection, and the specific example of the defect candidate code group (pointing information list) in 2nd Embodiment. It is a figure showing the specific example of the state by which defect candidate code group was divided into the factor group and the expression group in 2nd Embodiment. It is a figure showing the specific example of the validity judgment information which the probability calculation part in 2nd Embodiment utilizes. It is a figure showing the specific example of the similarity which is the result which the calculation part in 2nd Embodiment calculated. It is a figure showing the specific example of the reliability of the test result which is the result which the probability calculation part in 2nd Embodiment calculated. It is a flowchart showing the example of 1 operation | movement of the verification assistance apparatus of 2nd Embodiment.

  Embodiments according to the present invention will be described below with reference to the drawings.

<First Embodiment>
FIG. 1 is a block diagram showing a simplified configuration of a verification support apparatus according to the present invention. The verification support apparatus 1 according to the first embodiment is an apparatus that supports an operation of verifying whether or not an inspection result for a source program (computer program) to be inspected is correct. That is, the source program is a sequence in which a plurality of source code lines representing respective processing procedures (processing flows) of a plurality of processes executed by the computer are arranged. The verification support apparatus 1 according to the first embodiment creates information that supports verification work on the inspection result of the source program when an inspected source program to which the inspection result information by the static analysis tool is added is input, for example. , Present the information.

  The verification support apparatus 1 is a computer apparatus provided with, for example, a CPU (Central Processing Unit). By executing a computer program (hereinafter also referred to as a program) 7 read from the storage device 6, the following functions are provided. have. The storage device 7 may be built in the verification support device 1 or may be externally attached.

  That is, in the first embodiment, the verification support apparatus 1 includes a data creation unit 2, a calculation unit 3, and a support unit 4 as functional units.

  The data creation unit 2 has inspected a group of source code lines related to a process (processing flow) determined to have a possibility of occurrence of a defect based on the inspection result information added to the inspected source program. It has a function to extract from the source program. Here, the group of the extracted source code lines is called a defect candidate code group.

  In addition, the data creation unit 2 includes a plurality of source code lines constituting the defect candidate code group, an expression group including an expression line that is a source code line in which a defect occurs, and a factor line that is a source code line that causes the defect. The function is divided into factor groups including

  The calculation unit 3 has a function of calculating the similarity between expression groups and factor groups in the extracted defect candidate code groups when a plurality of defect candidate code groups are extracted by the data creation unit 2. .

  The support unit 4 has a function of presenting information including information on the similarity calculated by the calculation unit 3. For example, the support unit 4 outputs the similarity information calculated by the calculation unit 3 to a display device (display device) together with a display request. Thus, the similarity information is displayed on the display device, so that the similarity information can be presented to a program developer, for example. Further, the support unit 4 outputs the similarity information to the printer together with the print request. Thus, the similarity information is printed by the printer, so that the similarity information can be presented to, for example, a program developer.

  Since the verification support apparatus 1 according to the first embodiment has the above-described configuration, it is possible to improve the efficiency of verification work on the inspection result of the source program. That is, the verification support apparatus 1 according to the first embodiment further divides the defect candidate code group into the expression group and the factor group, and calculates the similarity between the expression groups and between the factor groups.

  Even if a part of the processing procedure (processing flow) by the defect candidate code group is not valid (in other words, the impossible source code execution process (processing flow) due to the performance limit of the static analysis tool is the defect candidate code group. In the case of the error candidate code group, the inspection result for the defect candidate code group is not valid. Therefore, only when the test result of the expression group and the factor group (result that there is a possibility of occurrence of a defect) are both valid, the test result of the defect candidate code group including the expression group and the factor group (defects) As a result, the defect candidate code group can be determined to be an object to be verified more carefully. In other words, when the test result of at least one of the expression group and the factor group is not valid, the test result for the defect candidate code group is not valid. That is, since the inspection result for the defect candidate code group is assumed to be erroneous and the probability of occurrence of the defect is low, the defect candidate code group is less required to be verified by the program developer.

  For example, if the similarity between the expression group of the defect candidate code group and the expression group whose test result is known to be valid is high (similar), the test of the expression group of the defect candidate code group It is assumed that the result is likely to be valid (validity). Similarly, if the similarity between the factor group of the defect candidate code group and the factor group whose test result is known to be valid is high (similar), the factor group of the defect candidate code group It is assumed that there is a high possibility (validity) that the test result is valid. For this reason, the validity of the test results for the expression group and the factor group of the defect candidate code group is determined based on the similarity information between the expression group and the factor group whose test results are known to be valid. Is possible. If it is determined by the validity judgment using such similarity that the inspection results of the defect candidate code group and the cause group are both valid, the source code of the defect candidate code group is verified. It can be determined that it is a target. On the other hand, when the test result of at least one of the expression group and the factor group of the defect candidate code group is not valid, the defect candidate code group can be excluded from the verification work. As a result, the verification work is reduced by the information presented by the verification support apparatus 1 of the first embodiment, so that the verification support apparatus 1 of the first embodiment shortens the time required for the verification work. Can do.

Second Embodiment
The second embodiment according to the present invention will be described below.

  FIG. 2 is a block diagram illustrating a simplified configuration of the verification support apparatus according to the second embodiment. Similar to the verification support apparatus 1 of the first embodiment, the verification support apparatus 10 is an apparatus that presents information that supports verification work on the inspection result of the source program (application program) by the static analysis tool 25. Moreover, the verification support apparatus 10 is comprised by the computer apparatus similarly to the verification support apparatus 1 of 1st Embodiment, By performing the computer program 12 read from the memory | storage device 11, the following functions are carried out. Can have. The storage device 11 may be built in the verification support device 10 or may be externally attached.

  That is, the verification support apparatus 10 includes an acquisition unit 14, an extraction unit 15, a calculation unit 16, a probability calculation unit 17, an output unit 18, and a storage unit 19 as functional units.

  The storage unit 19 includes a storage medium that stores various types of information (data and programs). Information is written in the storage unit 19 by the acquisition unit 14, the extraction unit 15, the calculation unit 16, the probability calculation unit 17, and the output unit 18, and the acquisition unit 14, the extraction unit 15, the calculation unit 16, and the probability calculation. Information is read from the storage unit 19 by the unit 17 and the output unit 18.

  The acquisition unit 14 has a function of acquiring, from, for example, the static analysis tool 25, an inspected source program to which information (inspection result information) of the inspection execution result by the static analysis tool 25 is added. Furthermore, the acquisition unit 14 has a function of creating the following indication information list and writing it in the storage unit 19. The indication information list is information in which a defect candidate code group (indication flow) is associated with an identification number (indication number) assigned to the defect candidate code group in order to identify the defect candidate code group. . The defect candidate code group represents a process (processing flow) that is pointed out (determined) by the static analysis tool 25 that there is a possibility of a defect, as in the defect candidate code group described in the first embodiment. A group of source code lines.

  FIG. 3 is a diagram illustrating a specific example of a source program to be inspected and a specific example of an indication information list created by the acquisition unit 14 based on a result of the source program being inspected by the static analysis tool 25.

  In the example shown in FIG. 3, information on “expression” and “factor” is included as test result information (additional information) by the static analysis tool 25 in the source program that has been tested by the static analysis tool 25. It has been added. In the “expression” information, the added source code line (in the example of FIG. 3, the source code lines of line numbers 7 and 10) is a source code line (expression line) that may cause a problem. Is information. The “factor” information indicates that the added source code line (the source code line of line number 2 in the example of FIG. 3) is the source code line (cause line) that causes the malfunction. Information.

  The extraction unit 15 has a function of dividing a plurality of source code lines constituting the defect candidate code group into a factor group and an expression group. An expression group is a group of source code lines representing a processing procedure (processing flow) that causes a defect, and includes an expression line. The factor group is a group of source code lines representing a processing procedure (processing flow) that causes the defect, and includes a factor line.

  FIG. 4 is a diagram illustrating a specific example of a defect candidate code group and a specific example in which the extraction unit 15 groups the defect candidate code group. As shown in this specific example, for example, the extraction unit 15 causes a cause line (a source code line to which “factor” information is added) from a source code line that starts a processing procedure represented by a defect candidate code group. A group of source code lines representing the processing procedure up to is extracted as a factor group. In addition, the extraction unit 15 sets, as an expression group, a group of source code lines from a source code line representing a processing procedure next to a factor line to an expression line (a source code line to which “expression” information is added). Extract. In other words, the extraction unit 15 divides the defect candidate code group into two groups based on the factor rows.

  The extraction unit 15 further has a function of writing information representing a defect candidate code group including information grouped in this way into the storage unit 19 as an update indication information list 22.

  Further, the extraction unit 15 has a function of creating a validity judgment information list 23 as shown in FIG. 5A based on the grouping result information. The validity judgment information list 23 is, for example, table data in which the program developer verifies the validity of the test result for each of the cause group and the expression group of the defect candidate code group, and the result of the verification is written. The validity judgment information list 23 created by the extraction unit 15 is created before the program developer inputs the validity judgment result. Therefore, in the validity judgment information list 23 created by the extraction unit 15, as shown in FIG. 5A, the validity judgment result information associated with all the groups is information indicating “undecided”. (Horizontal bar (-)).

  In the second embodiment, the acquisition unit 14 and the extraction unit 15 extract a defect candidate code group from a source program that has been inspected by the static analysis tool 25, and further divide the defect candidate code group into a factor group and an expression group. Configure the data creation unit.

  The calculation unit 16 has a function of calculating the similarity between factor groups and expression groups in all defect candidate code groups. FIG. 6A is table data representing a specific example of the similarity between factor groups calculated by the calculation unit 16. FIG. 6B is table data representing a specific example of the similarity between expression groups calculated by the calculation unit 16.

  There are various methods by which the calculation unit 16 calculates the similarity, and any method may be adopted here. For example, the longest common subsequence is used in units of rows. There is a calculation method. This calculation method is described in, for example, Reference: T.A. Kolmen and 8 other authors, “Algorithm Design and Analysis Method (Algorithm Introduction)”, Modern Science Co., Ltd., shown in March 2007. FIGS. 6A and 6B are calculation examples of similarity calculated using the longest common subsequence.

  Alternatively, the calculation unit 16 may calculate the similarity between groups by a method of calculating the similarity based on the edit distance (Levenstein distance) and the n-gram frequency. As described above, there are various methods for the calculation unit 16 to calculate the degree of similarity, and among these methods, a method capable of analyzing according to the characteristics of the defect candidate code group is appropriately selected and the calculation unit 16 selects the method. You may employ | adopt as a method of calculating a similarity. In addition, as a reference when selecting a method for calculating similarity, Masayuki Takeda and Tomoko Fukuda, "Knowledge literary wearing mobile suits from classical Waka", IPSJ Journal, Vol. 43, No. 9, September 2002, p. 941-949.

  The calculation unit 16 has a function of writing the similarity information calculated as described above into, for example, table data as shown in FIGS. 6A and 6B held in the storage unit 19. The form of the similarity information held by the storage unit 19 is not limited to the form of table data.

  The output unit 18 outputs information stored in the storage unit 19 and information prompting the program developer to input information necessary for the control operation of the verification support apparatus 10 to, for example, a display device (not shown). It has a function. The display device has a function of displaying information received from the output unit 18.

  For example, the output unit 18 has a function of outputting information prompting a program developer to input validity determination information using an input device such as a keyboard to a display device (display device). . Based on the information displayed on the display device by the function of the output unit 18, the validity determination information input by the program developer using the input device is stored in the storage unit 19 in FIG. ) Is written in the validity judgment information list 23 as shown in FIG.

  FIG. 5B is a diagram illustrating an example of a state in which the validity determination information is input (written) in the validity determination information list 23. In FIG. 5B, a static analysis is performed for a group (for example, refer to the factor group with the identification number “1”) that the program developer determines to be a source code execution process that may actually cause a problem. Information (circle mark (◯)) indicating that the inspection result of the tool 25 is correct, that is, “with validity” is associated as validity determination information. In addition, a group (for example, identification number “1”) determined by the program developer that it is a source code execution process in which an inspection result of the static analysis tool 25 cannot exist, that is, the inspection result is incorrect. The information indicating the “no validity” of the test result (cross mark (×)) is associated with the validity determination information. Further, as described above, information (horizontal bar (−)) indicating “undecided” is associated as validity determination information with a group not verified by the program developer.

The probability calculation unit 17 calculates the test result by the static analysis tool 25 based on the similarity information calculated by the calculation unit 16 and the validity determination information by the program developer for the factor group and the expression group. It has a function to calculate the degree of reliability (reliability). Here, the probability calculation unit 17 calculates the reliability of the test result for a factor group or expression group that is not verified by the program developer (that is, a group that is associated with “undecided” as validity judgment information). To do. For example, in the example shown in FIG. 5B, the probability group for the defect candidate code group having the identification number “2”, the expression group, and the factor group of the defect candidate code group having the identification number “3” The calculation unit 17 performs a calculation operation. For example, the probability calculation unit 17 calculates the reliability P (i) in the defect candidate code group having the identification number “i” in each of the factor group and the expression group according to the following equation (1).

Note that N in Equation (1) represents the number of all defect candidate code groups. F (i, j) represents the similarity calculated by the calculation unit 16 between the factor groups or the expression groups of the defect candidate code group with the identification number “i” and the defect candidate code group with the identification number “j”. . B (j) represents a numerical value according to the following conditions.

  Further, the probability calculation unit 17 has a function of writing the calculated result (the reliability of the inspection result) in the storage unit 19. FIG. 7 is a diagram illustrating an example of the reliability list 24 representing the calculation result held by the storage unit 19. For example, when the output unit 18 outputs the reliability list 24 to the display device, the reliability list 24 is displayed on the display device and presented to the program developer. In the second embodiment, the probability calculation unit 17 and the output unit 18 constitute a support unit that presents information that supports verification work on the inspection result by the static analysis tool 25.

  The verification support apparatus 10 of the second embodiment is configured as described above. Next, an example of the operation in the verification support apparatus 10 will be described with reference to FIG. FIG. 8 is a flowchart illustrating an example of an operation in the verification support apparatus 10, and this flowchart represents a processing procedure of a computer program executed by the verification support apparatus (computer apparatus) 10.

  For example, when the acquisition unit 14 of the verification support apparatus 10 acquires an inspected source program including inspection result information inspected by the static analysis tool 25 (step S101), the defect candidate code group is extracted from the source program. (Step S102). Then, the acquisition unit 14 creates an indication information list 21 as illustrated in FIG. 3, for example, and writes the information of the indication information list 21 in the storage unit 19.

  Thereafter, the extraction unit 15 groups each defect candidate code group into a factor group and an expression group based on test result information (additional information) from the static analysis tool 25 (step S103). Then, the extraction unit 15 creates an update indication information list 22 as shown in FIG. 4 and writes this information in the storage unit 19.

  Thereafter, the calculation unit 16 calculates the similarity between the factor groups and the expression groups based on the grouping information based on the updated indication information list 22 (step S104). Then, the calculation unit 16 writes information on the calculation results as shown in FIGS. 6A and 6B in the storage unit 19, for example.

  On the other hand, the extraction unit 15 creates a validity determination information list 23 as shown in FIG. 5A and writes the information of the validity determination information list 23 in the storage unit 19.

  Thereafter, the output unit 18 outputs information on the defect candidate code group and information that prompts the user to input the validity determination information using an input device such as a keyboard to the display device (display device). As a result, the display device displays information on the defect candidate code group and information prompting input of the validity determination information. Thereby, when the program developer acquires the validity determination information by inputting the validity determination information (step S105), the verification support apparatus 10 uses the acquired validity determination information as the validity determination information in the storage unit 19. The list 23 is written as shown in FIG.

  After that, the probability calculation unit 17 determines whether or not there is a group associated with “undecided” information in the validity determination information list 23 at a predetermined timing (step S106). The predetermined timing is, for example, when information is written in the validity judgment information list 23 and a calculation execution command is input to the verification support apparatus 10 by using, for example, an input device by a program developer. It is.

  If the probability calculation unit 17 determines that there is an “undetermined” group, the probability determination unit 17 determines that the “undetermined” is based on the similarity information by the calculation unit 16 and the validity determination result information by the program developer. The reliability of the inspection result for the group "" is calculated (step S107). The probability calculation unit 17 writes the calculation result in the storage unit 19.

  Thereafter, the output unit 18 outputs the calculation result and a display request for the calculation result to the display device (display device) (step S108). Thereby, a display apparatus displays a calculation result. At this time, for example, information prompting the program developer to input the validity determination information for the “undecided” group is also displayed on the display device. Thereby, for example, when it is detected that the validity determination information is newly written in the storage unit 19, the probability calculation unit 17 repeats the operations after step S106.

  If the probability calculation unit 17 determines that there is no “undecided” group by the determination operation in step S106, the operation is terminated. As a result, the verification support apparatus 10 ends a series of operations related to calculation and output of reliability from acquisition of the verified sort program.

  Similar to the first embodiment, the verification support apparatus 10 of the second embodiment can contribute to the efficiency of the inspection result verification work by the static analysis tool 25. In particular, the verification support apparatus 10 according to the second embodiment includes a probability calculation unit 17 and a configuration that realizes quantifying and displaying the reliability of the inspection result by the static analysis tool 25. This configuration can provide the program developer with information effective for determining whether or not the verification process of the defect candidate code group is actually necessary, and thus contributes to further efficiency of the verification process.

<Third Embodiment>
Embodiments according to the present invention will be described below. In the description of the third embodiment, the same reference numerals are given to the same names as the components of the verification support apparatus of the second embodiment, and the duplicate description of the common parts is omitted.

  In the second embodiment described above, the extraction unit 15 groups the source code lines of each defect candidate code group, and then, as shown in FIG. 4, the source code lines are included as part of the grouped information. It is the aspect included as it is. Instead, in the third embodiment, the extraction unit 15 parses the source code of each group in the defect candidate code group, and abstracts a part of the source code included in each group based on the parse analysis. It has a configuration to make it. In other words, if the source program does not contain a branch structure syntax (for example, if statement or for statement in C language), a similar defect (special variant) occurs due to a difference in the branch condition in the source program. There is nothing to do. That is, even if the source code is abstracted in a range not including the branch structure, the similarity between groups as described in the second embodiment is not adversely affected. From this, in this 3rd Embodiment, the extraction part 15 abstracts a part of source program part which does not contain a branch structure in each group, for example by the encoding using a hash function.

  The other configuration of the third embodiment is the same as that of the second embodiment.

  As in the second embodiment, the verification support apparatus 10 of the third embodiment is effective for narrowing down the defect candidate code group to be actually verified from the defect candidate code group based on the inspection of the static analysis tool 25. Such information can be provided to program developers, for example. As a result, the verification support apparatus 10 according to the third embodiment can also shorten the time required for the verification work.

  Also, the verification support apparatus 10 of the third embodiment abstracts a part of the source program in each group so that the similarity calculation is not adversely affected. For this reason, the verification support apparatus 10 of the third embodiment can improve the efficiency of calculating the similarity.

<Other embodiments>
In addition, this invention is not limited to the 1st-3rd embodiment, Various embodiment can be taken. For example, in the second and third embodiments, the output unit 18 of the verification support apparatus 10 outputs the reliability of the inspection result obtained by the probability calculation unit 17. On the other hand, the output unit 18 may output the similarity between the groups calculated by the calculation unit 16 instead of the reliability of the inspection result. In this case, based on the output information, the display device displays the similarity between the groups. In such a case, the probability calculation unit 17 in the verification support apparatus 10 may be omitted. Further, the output unit 18, for example, according to the instruction information input by the device operator (program developer) using the input device, the reliability of the inspection result by the probability calculation unit 17 and the similarity by the calculation unit 16 One of these may be output.

DESCRIPTION OF SYMBOLS 1,10 Verification support apparatus 2 Data preparation part 3,16 Calculation part 4 Support part 15 Extraction part 17 Probability calculation part

Claims (6)

Execution according to the inspected source program based on inspection result information added to the inspected source program in which a plurality of source code lines representing respective processing steps of the plurality of processes executed by the computer are arranged Among the plurality of possible processes, a group of the source code lines related to the process that may cause a defect is extracted as a defect candidate code group from the inspected source program, and the defect candidate code group Are divided into an expression group including an expression line that is the source code line in which a defect occurs and a factor group including a factor line that is the source code line that causes the defect. A data creation unit;
When there are a plurality of defect candidate code groups extracted by the data creation unit, a calculation unit that calculates the similarity between the expression groups and the factor groups in the extracted defect candidate code groups,
A verification support apparatus comprising: a support unit that presents information including information on the similarity calculated by the calculation unit.   Instead of presenting information including the similarity information calculated by the calculation unit, the support unit performs the similarity information calculated by the calculation unit and the defect candidate code group. The verification support apparatus according to claim 1, wherein a reliability of the inspection result information for the defect candidate code group is calculated based on the given validity determination information, and the calculation result is presented.   The said support part calculates the reliability of the said test result information selectively about the said defect candidate code group which detected that the presence or absence of the validity was not judged based on the said validity judgment information. The verification support apparatus described.   The data creation unit parses the source code lines in the expression group and the factor group in the defect candidate code group, and is a source program selected based on the syntax analysis and according to a predetermined condition The verification support apparatus according to claim 1, further comprising a function of abstracting a part. Execution according to the inspected source program based on inspection result information added to the inspected source program in which a plurality of source code lines representing respective processing steps of the plurality of processes executed by the computer are arranged Among the plurality of possible processes, a group of the source code lines related to the process that may cause a defect is extracted from the inspected source program as a defect candidate code group,
A plurality of source code lines constituting the defect candidate code group, an expression group including an expression line that is the source code line in which the defect occurs, and a factor line that is the source code line that causes the defect Divided into factor groups including
When there are a plurality of the extracted defect candidate code groups, calculate the similarity between the expression groups and the factor groups in the extracted defect candidate code groups,
A verification support method for presenting information including the calculated similarity information. Execution according to the inspected source program based on inspection result information added to the inspected source program in which a plurality of source code lines representing respective processing steps of the plurality of processes executed by the computer are arranged A process of extracting from the inspected source program a group of source code lines related to the process that may cause a failure among the plurality of possible processes as a defect candidate code group;
A plurality of source code lines constituting the defect candidate code group, an expression group including an expression line that is the source code line in which the defect occurs, and a factor line that is the source code line that causes the defect Processing to be divided into factor groups including
When there are a plurality of the extracted defect candidate code groups, a process of calculating the similarity between the expression groups and the factor groups in the extracted defect candidate code groups;
A computer program in which a processing procedure for causing a computer to execute processing for presenting information including information on the calculated similarity is represented.

Images