JP2017045354A - Software development support program, software development support apparatus, and software development support method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To present modification candidate portions based on likelihood of changes being required to be made when a change portion due to a version change of customized reference source code is reflected on the reference source code.SOLUTION: A computer: retains calling source nodes which are among paired nodes in the call relation in an extracted difference, in a first set, the difference being extracted from a first call relation in which a calling source and a called destination of a function contained in a first source code are made paired nodes and a second call relation in which a calling source and a called destination of a function contained in a second source code are made paired nodes; retains, in a second set, a calling source node and a called destination node among paired nodes in the call relation in an extracted difference, the difference being extracted from the first call relation and a third call relation in which a calling source and a called destination of a function contained in third source code are made paired node; calculates a similarity of each pair of each node contained in the first set in the first call relation and each node contained in the second set in the third call relation; and outputs the nodes contained in the second set in the descending order of similarities.SELECTED DRAWING: Figure 6

Description

  The present specification relates to a software development support program, a software development support apparatus, and a software development support method.

  In software development, various software development support technologies are used in order to work efficiently.

  For example, as a first technique, a technique for automatically extracting the relationship between an execution program in charge of a function and source program code and presenting the range of influence on other functions due to the change of the source program code to the developer in an easy-to-read form (For example, Patent Document 1).

  As a second technique, there is a technique for specifying a similar part in a program (for example, Patent Document 2).

  As a third technique, there is a technique for creating a new version by upgrading the version of the source code, in particular, by incorporating the changed portion of the other file into one of the two derived files derived from the original file of the source code (for example, Patent Document 3).

JP 2009-122754 A JP 2009-199172 A JP 09-081371 A

  In recent years, software development is increasingly performed using source code managed externally such as open source software (OSS). In this case, the OSS is used by customizing the OSS. When the version of the OSS is upgraded, it is necessary to reflect the contents of the version upgrade in the customized OSS.

  However, if the modification is made to reflect the contents of the version upgrade in the customized OSS without considering the calling relationship of the functions in the source code, there is a possibility that omission of correction will occur.

  At this time, it is conceivable to use the first technique. In the first technique, functions existing on all paths are specified by color change by sequentially searching for functions that call a specified function.

  However, in the first technique, the range of influence of the designated function is only specified uniformly, so it is not possible to grasp a place with high accuracy to be actually corrected. In the first technique, even if the influence range of the source code before and after the version change can be specified, the correction candidate part cannot be detected when the contents of the upgrade are reflected in the customized OSS.

  As one aspect of the present invention, there is provided a technique for presenting correction candidate locations based on the accuracy to be changed when reflecting a changed location of a customized reference source code due to a version change of the reference source code.

  A software development support program according to an aspect of the present invention causes a computer to execute the following processing. The computer obtains a first source code, a second source code obtained by customizing the first source code, and a third source code having a version different from that of the first source code. The computer has a first call relationship in which a caller and a callee of a function included in the first source code are a pair of nodes, and a caller and a callee of a function included in the second source code are a pair of nodes. The node pair of the difference with the second call relationship is extracted. The computer holds the caller node of the extracted node pair in the first set. The computer extracts a difference node pair between the first call relationship and the third call relationship having the call source and the call destination of the function included in the third source code as a pair node. The computer holds the caller and callee nodes of the extracted node pair in the second set. The computer, based on a predetermined rule, each node pair that combines each node included in the first set in the first call relationship and each node included in the second set in the third call relationship The similarity is calculated. The computer outputs the nodes included in the second set in descending order of similarity.

  According to the technology described in this specification, when a changed reference source code is reflected in a customized reference source code, a correction candidate part is presented based on the accuracy to be changed. Can do.

It is a figure for demonstrating the calling relationship of a method. It is a figure for demonstrating updating the software version customized according to the software version update. It is a figure for demonstrating the case where after a source code is customized, the source code is updated, the method call source is changed, and a correction omission occurs. FIG. 3 shows a source code v2c as a result to be originally obtained. It is a figure for demonstrating the process in this embodiment. An example of the software development support apparatus in this embodiment is shown. An example of the change location detection apparatus in this embodiment is shown. An example of source code v1, v1c, v2 acquired by the source code acquisition unit in the present embodiment is shown. An example of a call relationship graph of a method generated by the call relationship generation unit in the present embodiment is shown. It is a figure for demonstrating the process of the replacement origin extraction part in this embodiment. It is a figure for demonstrating the process of the replacement destination extraction part in this embodiment. It is a figure for demonstrating the process of the similarity calculation part in this embodiment. An example of the similarity information in this embodiment is shown. The output example by the output result part in this embodiment is shown. The example of an output by the output result part in the other example of this embodiment is shown. The flowchart of the whole process in this embodiment is shown. It is a detailed flowchart of the process of S3. It is a detailed flowchart of the process of S4. It is a detailed flowchart of the process of S5. It is a detailed flowchart of the process of S5-3. It is an example of a configuration block diagram of a hardware environment of a computer that executes a program in the present embodiment.

  FIG. 1 is a diagram for explaining a method calling relationship. By analyzing the source code 1, the method name (2-1, 2-2, 2-3) included in the source code 1 is a node, and the call relationship (3-1, 3-2) is for calling another method ) Can be obtained as an edge.

  FIG. 2 is a diagram for explaining the updating of the customized software version in accordance with the software version update. In the following description, for the sake of convenience of explanation, an explanation will be given by taking the version upgrade as an example, but the term “version up” may be replaced with “version down”.

  In FIG. 2, v1 indicates the version of the source course before update. v1c indicates the version of the source code after customization of the source code v1. v2 indicates the version of the source code after the update of the source code v1. v2c indicates a version reflecting the upgrade to the source code v2 in the customized source code v1c.

  When the source code v1 managed externally such as OSS is used, a custom version source course v1c obtained by modifying the contents of the source code v1 such as adding a function may be used (S101).

  When the OSS is updated after modifying the contents of the source code v1 (S102), the custom version source code v1c may be updated along with the update of the OSS (S103).

  At this time, the OSS user first checks the location changed in S102 by detecting the difference between the source codes v1 and v2 using a diff tool or the like that detects the difference between the two files. It is conceivable that the OSS user reflects the portion changed in S102 in the custom version source code v1c in S103. For example, you can use git mergetool to do this automatically.

  However, in the existing method such as git mergetool, as described with reference to FIG. 3, there is a case where omission of correction occurs when the method call source is changed.

  FIG. 3 is a diagram for explaining a case in which after the source code is customized, the source code is updated, the method call source is changed, and a correction omission occurs.

  For example, it is assumed that the source code v1 is customized so that the function c is added to the class B, and the customized method B.c is modified to be called from the method A.a of the class A (S101a).

  Thereafter, it is assumed that the source code v1 is updated so that a method of a new class called NewA is called from Main (S102a). At this time, Class A and Class B remain v1.

  By using git mergetool, the difference between the source codes v1 and v2 is detected to detect the portion changed in S102a, and the detected changed portion is reflected in the source code v1c (S103a). Here, the class A and the class B remain the source code v1c.

  However, as shown in FIG. 4, the function of method B.c must be incorporated in method a of class NewA, but the merge tool cannot detect it.

  FIG. 4 shows the source code v2c of the result to be originally obtained in FIG. It is desirable to be able to reflect the process of calling the method Bc from the method Aa that has been modified when creating the source course v1 to v1c to the class NewA.a when creating the source code v2 to v2c. .

  As described above, it is desirable to appropriately extract files (classes and methods) that require modification of the customized source code as the source code is updated.

  Therefore, in the present embodiment, when the contents of the version change (including version upgrade or downgrade) of the reference source code are reflected on the customized reference source code, the correction is made based on the accuracy to be corrected. Provide technology to present candidate locations. Hereinafter, a technique for detecting a portion that needs to be corrected from a difference between software versions and, more specifically, using a method call relationship when extracting the difference will be described.

  FIG. 5 is a diagram for explaining processing in the present embodiment. First, the software development support apparatus according to the present embodiment constructs a method call relationship graph in each version (v1, v1c, v2) of the source code. Next, as shown in FIG. 5A, the software development support apparatus marks the caller node of the edge having a difference from the comparison of the call relationship between the source codes v1 and v1c as a node to be noted.

  Next, as shown in FIG. 5B, the software development support apparatus compares the calling relationship between the source codes v1 and v2. Based on the following determination method, the software development support apparatus checks the degree of similarity between the node (calling source node, calling destination node) included in the newly added edge and the node marked above. The software development support apparatus outputs a node having a high similarity as a change location candidate.

  Here, the software development support apparatus calculates the level of similarity using the following as determination factors. It can be said that the more the following criteria are satisfied, the higher the degree of similarity.

(I) For each node of interest (caller node / callee node included in newly added edge, marked node) (own node), caller nodes and callee nodes respectively Satisfies either of the following: Here, caller nodes or callee nodes are referred to as “compare nodes”.
The nodes to be compared are the same node.
-Nodes to be compared are methods that implement the same interface.
-The method names of the nodes to be compared are the same.

(II) The nodes being focused on (one's own nodes) satisfy either of the following.
-Nodes of interest are methods that implement the same interface.
-The method names of the nodes being focused on are the same.

FIG. 6 shows an example of a software development support apparatus in the present embodiment. The software development support apparatus 11 includes an acquisition unit 12, a first extraction unit 13, a second extraction unit 14, a calculation unit 15, and an output control unit 16.
The acquisition unit 12 acquires a first source code, a second source code obtained by customizing the first source code, and a third source code having a version different from that of the first source code. An example of the acquisition unit 12 is a source code acquisition unit 22.

  The first extraction unit 13 extracts a node pair that is a difference between the first call relationship and the second call relationship. The first call relationship is a call relationship in which a call source and a call destination of a function included in the first source code are paired nodes. The second call relationship is a call relationship in which the call source and the call destination of the function included in the second source code are paired nodes. The first extraction unit 13 holds the caller node of the extracted node pair in the first set. An example of the first extraction unit 13 is a replacement source extraction unit 24.

  The second extraction unit 14 extracts a node pair that is a difference between the first call relationship and the third call relationship. The third call relationship is a call relationship in which the call source and the call destination of the function included in the third source code are paired nodes. The second extraction unit 14 holds the caller and callee nodes of the extracted node pair in the second set. An example of the second extraction unit 14 is a replacement destination extraction unit 25.

  Based on a predetermined rule, the calculation unit 15 combines a node pair obtained by combining each node included in the first set in the first call relationship with each node included in the second set in the third call relationship. The similarity of each is calculated. An example of the calculation unit 15 is a similarity calculation unit 26.

  The output control unit 16 outputs the nodes included in the second set in descending order of similarity. An example of the output control unit 16 is a result output unit 27.

  By configuring in this way, when the changed reference source code that has been changed due to the version change of the reference source code is reflected, correction candidate locations can be presented based on the accuracy to be changed.

  In addition, for each node pair that has been combined, the calculation unit 15 includes, for each node included in the node pair that has been combined, the nodes that have called each node of the node pair that has been combined, and the call destinations of each node in the node pair that has been combined. Compare nodes. The calculation unit 15 assigns the similarity to each of the compared node pairs based on a predetermined rule, and the similarity of the node pair obtained by combining the total values of the similarities assigned to the compared node pairs. Degree. For a given rule, the degree of similarity depends on whether each of the compared node pairs is the same node, whether the methods implement the same interface, or whether the method names are the same. The value of is specified.

  With this configuration, it is possible to determine the similarity between the second source code and the third source code with reference to the first source code.

  The output unit 16 outputs the nodes included in the second set as correction candidate locations. At the same time, the output unit 16 is a caller node included in the combined node pair in the first set, and is called out of the difference node pairs between the first call relation and the second call relation. The callee node corresponding to the original node is output as a correction reflection target for the correction candidate location.

  With this configuration, it is possible to output a correction reflection target for a correction candidate location.

  FIG. 7 shows an example of a change location detection apparatus in the present embodiment. The change location detection device 21 is connected to an input interface (I / F) 28 and an output I / F 29. The input I / F 28 is an input device such as a keyboard. The output I / F 29 is an output device such as a display device.

  The change location detection device 21 includes a storage unit (not shown). The storage unit (not shown) includes v1 call relation information 31, v1c call relation information 32, v2 call relation information 33, v1-v1c change contents 34, v1-v2 change contents 35, replacement source information 36, replacement destination information 37, Similarity information 38 and threshold information 39 are stored.

  The change location detection device 21 includes a source code acquisition unit 22, a call relationship generation unit 23, a replacement source extraction unit 24, a replacement destination extraction unit 25, a similarity calculation unit 26, and a result output unit 27.

  The source code acquisition unit 22 acquires the source codes v1, v1c, and v2 input by the input I / F 28.

  The call relationship generation unit 23 constructs a method call relationship from the source code for each of the source codes v1, v1c, and v2, and generates v1 call relationship information 31, v1c call relationship information 32, and v2 call relationship information 33.

  The replacement source extraction unit 24 changes v1-v1c by changing the difference between the call relationship (v1 call relationship information 31) of the original version of the source code v1 and the call relationship (v1c call relationship information 32) of the customized version of the source code v1c. Acquired as content 34. Based on the v1-v1c change content 34, the replacement source extraction unit 24 records the call source of the changed edge as a caution location (replacement source information 36).

  The replacement destination extraction unit 25 acquires the difference between the calling relationship of the original version of the source code v1 (v1 calling relationship information 31) and the calling relationship of the updated version of the source code v2 as the v1-v2 change content 35. The replacement destination extraction unit 25 records the method added in the updated version based on the v1-v2 change content 35 as a detection candidate location (replacement destination information 37).

  The similarity calculation unit 26 calculates the method similarity obtained by the replacement source extraction unit 24 and the replacement destination extraction unit 25.

  The result output unit 27 notifies that the replacement destination information 37 that holds the similarity exceeding the preset threshold is the replacement target. Further, the result output unit 27 lists replacement candidates in descending order of similarity.

  As a result, the replacement source extraction unit 24 captures and marks the change in the difference in the call relationship, so that it can be recognized as a point to be noted even if there is no change when the original source code v1 is updated. In addition, the replacement destination extraction unit 25 and the similarity calculation unit 26 can determine whether or not the location obtained by the replacement source extraction unit 24 is reflected.

  FIG. 8 shows an example of the source codes v1, v1c, v2 acquired by the source code acquisition unit in the present embodiment. The source code acquisition unit 22 acquires three versions of source code v1, v1c, and v2 that are input through the input I / F 28.

  FIG. 9 shows an example of a method call relationship graph generated by the call relationship generation unit in the present embodiment. In FIG. 9, arrows are shown from the method call source to the call destination.

  The call relationship generation unit 23 constructs a method call relationship graph (v1 call relationship information 31, v1c call relationship information 32, v2 call relationship information 33) for each acquired source code (v1, v1c, v2).

  The v1 call relation information 31 holds relation information where the method “A.a ()” is called from the method “Main.main ()” and the method “B.b ()” is called from the method “A.a ()”.

  In the v1c calling relation information 32, the method “Aa ()” is called from the method “Main.main ()”, and the methods “Bb ()” and “Bc ()” are called from the method “Aa ()”. Holds relationship information.

  In the v2 call relation information 33, the method “NewA.a ()” is called from the method “Main.main ()”, and the method “Bb ( ) "Is held and related information is retained.

In the description of the present embodiment, the call relation graph is represented as a diagram for easy understanding. However, in order to enable processing by the computer, for example, as a list of caller / callee pairs as follows: A call relationship graph may be maintained. In the following example, the v2 call relation information 33 in FIG. 9 is used as a list of caller / callee pairs.
[<Main.main (), NewA.a ()>,
<NewA.a (), Bb ()>,
<Aa (), Bb ()>]

  FIG. 10 is a diagram for explaining processing of the replacement source extraction unit in the present embodiment. Based on the generated call relationship (v1 call relationship information 31, v1c call relationship information 32), the replacement source extraction unit 24 compares the call relationships of the source codes v1 and v1c, and calls the edges having different contents. The original node is held as replacement source information 36.

  In the example of FIG. 10, the replacement source extraction unit 24 compares the v1 call relation information 31 with the v1c call relation information 32 and adds the call relation: Aa-> Bc, deletion as the v1-v1c change content 34. Called call relationship: None is acquired. The replacement source extraction unit 24 holds, as replacement source information 36, the call source node A.a of the edge (A.a-> B.c) having a difference in content from the v1-v1c change content 34.

  FIG. 11 is a diagram for explaining processing of the replacement destination extraction unit in the present embodiment. The replacement destination extraction unit 25 compares the call relationship between the source codes v1 and v2 based on the v1 call relationship information 31 and the v2 call relationship information 33, and replaces the node included in the newly added edge with the replacement destination information 37. Hold as.

  The replacement destination extraction unit 25 obtains a change in the call relationship between the source codes v1 and v2 in the same manner as the comparison of the call relationship between the source codes v1 and v1c by the replacement source extraction unit 24, and newly adds an edge. Is stored as replacement destination information.

  In the example of FIG. 11, the replacement destination extraction unit 25 compares the v1 call relation information 31 with the v2 call relation information 33, and adds the call relation: Main.main-> NewA as the v1-v2 change content 35. .a, NewA.a-> Bb, deleted call relation: Main.main-> Aa is acquired.

  The replacement destination extraction unit 25 holds, as replacement destination information 37, nodes Main.main, NewA.a, and B.b included in the newly added edge from the v1-v2 change content 35.

  FIG. 12 is a diagram for explaining the processing of the similarity calculation unit in the present embodiment. The similarity calculation unit 26 calculates the similarity between each method of the replacement source information 36 in the v1 call relation information 31 and each method of the replacement destination information 37 in the v2 call relation information 33, and the similarity as shown in FIG. 13. The degree information 38 is retained.

  The similarity is calculated from the three viewpoints of (1) comparison of caller nodes, (2) comparison of callee nodes, and (3) comparison of own nodes, and the sum is obtained.

The contents of node comparison are as follows.
(I) The same node (only (3) above does not occur substantially)
(Ii) A method that implements the same interface.
(Iii) The method name is the same.

  It is assumed that the lower the comparison content, the higher the degree of similarity. For example, 3 points, 2 points, and 1 point are set.

  In FIG. 12, the replacement source information 36 includes a method “A.a”. The replacement destination information 37 includes methods “Main.main”, “NewA.a”, and “B.b”. Here, each node (A.a) in the v1 call relation information 31 and each node (Main.main, NewA.a, B.b) in the v2 call relation information 33 are compared.

  First, a comparison between the node (A.a) 42 in the v1 call relationship information 31 and the node (Main.main) 44 in the v2 call relationship information 33 will be described. If attention is paid to the node (Aa) 42 of the v1 call relation information 31, the call source node of the node (Aa) 42 is the node (Main.main) 41, the call destination node is the node (Bb) 43, and the node itself is a node. (Aa) 42.

  On the other hand, paying attention to the node (Main.main) 44 of the v2 call relation information 33, there is no caller node of the node (Main.main) 44, and the callee node is the node (NewA.a) 45, Is a node (Main.main) 44.

  The call source nodes of the node (A.a) 42 in the v1 call relation information 31 and the node (Main.main) 44 in the v2 call relation information 33 are compared. Since the caller node of the node (Aa) 42 is the node (Main.main) 41, but there is no caller node of the node (Main.main) 44, the above (i) (ii) (iii) Does not fall under, and no points will be awarded.

  Comparing the call destination nodes, they are the node (B.b) 43 and the node (NewA.a) 45, which do not correspond to the above (i), (ii) and (iii), and are not added.

  When comparing their own nodes, they are the node (A.a) 42 and the node (Main.main) 44, which do not correspond to the above (i), (ii) and (iii), and are not added.

  Next, a comparison between the node (A.a) in the v1 call relationship information 31 and the node (NewA.a) in the v2 call relationship information 33 will be described. As described above, regarding the node (Aa) 42 of the v1 call relation information 31, the call source node is the node (Main.main) 41, the call destination node is the node (Bb) 43, and its own node is the node (Aa). 42.

  On the other hand, paying attention to the node (NewA.a) 45 of the v2 call relation information 33, the caller node of the node (NewA.a) 45 is the node (Main.main) 44, and the callee node is the node (B, b). ) 47 and itself is a node (NewA.a) 45.

  Comparing the call source nodes of the node (Aa) in the v1 call relationship information 31 and the node (Main.main) in the v2 call relationship information 33, the node (Main.main) 41 and the node (Main.main) 44 are obtained. They are the same node (corresponding to (i) above). In this case, three points are added.

  Comparing the call destination nodes, the node (B.b) 43 and the node (B.b) 47 are the same node (corresponding to the above (i)). In this case, three points are added.

  When comparing their own nodes, they are the node (A.a) 42 and the node (NewA.a) 45, which have the same method name (corresponding to the above (iii)). In this case, one point is added.

  The comparison between the node (A.a) 42 in the v1 call relationship information 31 and the node (B.b) 47 in the v2 call relationship information 33 is performed in the same manner.

  Note that the validity of the score according to the comparison contents may change depending on the area of the application program, and may be changed. In addition, when there are a plurality of call sources and call destinations, the highest one is calculated as the similarity.

  FIG. 13 shows an example of similarity information in this embodiment. For each combination of each node of the replacement source information 36 in the v1 call relation information 31 and each node of the replacement destination information 37 in the v2 call relation information 33, the similarity information 38 is obtained by calling each other and between the call destination nodes. The degree of similarity calculated according to the comparison result between its own nodes is held.

  As described with reference to FIG. 12, the node (Aa) 42 in the v1 call relation information 31 and the node (Main.main) 44 in the v2 call relation information 33 are as follows. As a result of comparison, none of them correspond to the above (i), (ii) and (iii) and are 0 points. Accordingly, in this case, the total value is also 0 points.

  The node (A.a) 42 in the v1 call relation information 31 and the node (NewA.a) 45 in the v2 call relation information 33 are as described in FIG. That is, the caller nodes are the same node (addition of 3 points), the callee nodes are the same node (addition of 3 points), and their own nodes have the same method name (addition of 1 point). As a result, the total value is 7 points.

  As for the node (Aa) 42 in the v1 call relation information 31 and the node (Bb) 47 in the v2 call relation information 33, as shown in FIG. 12, the comparison results between the caller nodes, the callee nodes, and their own nodes. , None of the above (i), (ii) and (iii), and 0 points. Accordingly, in this case, the total value is also 0 points.

  The result output unit 27 sets a node on the v2 call relation information 33 side of a pair whose total value of similarity information is equal to or greater than a preset threshold value as a correction candidate location. The threshold is registered in advance as threshold information 39.

  For example, 6 is set as the threshold value. In this case, the pair whose similarity information total value is equal to or greater than the threshold is the node (A.a) and the node (NewA.a), and the node on the v2 call relation information 33 side is the node (NewA.a). Therefore, the node (NewA.a) is a candidate for correction.

  Also, as shown in FIG. 10, in the v1-v1c change contents 34, the call-destination side node (B.c) including the v1 side node of the pair that is equal to or greater than the threshold value is subject to correction reflection.

  FIG. 14 shows an output example by the output result unit in the present embodiment. As shown in FIG. 14A, the result output unit 27 outputs the correction candidate portion 51a obtained above. As shown in FIG. 14B, the result output unit 27 relates to the correction reflection target for the correction candidate portion as shown in FIG. 14 based on the correction candidate portion and the correction reflection target obtained above. Information 52a indicating that correction is necessary is output. In the above description, the result output unit 27 sets the node on the v2 call relation information 33 side of the pair whose total value of similarity information is equal to or greater than a preset threshold as a correction candidate location, but the total value of similarity information The node on the v2 call relation information 33 side of the pair with the highest value may be used as a candidate for correction.

  FIG. 15 shows an output example by the output result unit in another example of the present embodiment. First, the output result unit 27 sets a plurality of nodes on the side of the v2 call relationship information 33 of the pair whose total value of similarity information is equal to or greater than a preset threshold as correction candidate locations.

  Then, as shown in FIG. 15A, the result output unit 27 outputs a replacement candidate list 51b in which the plurality of correction candidate portions are arranged and stored in descending order of the total value of the similarities. In the replacement candidate list 51b, correction candidate locations having a higher similarity score are registered so that the priority is higher for the correction candidate locations having a similarity equal to or higher than the threshold. Depending on the design specifications, the total value of the similarity information may not be cut off by the threshold value.

  In addition, the result output unit 27 acquires a correction reflection target corresponding to each correction candidate location registered in the replacement candidate list 51b. Then, the result output unit 27 gives priority to the information 52b indicating that the correction candidate portion needs to be corrected with respect to the correction candidate portion as shown in FIG. 15B based on the correction candidate portion and the correction reflection target. You may output in order of order.

  FIG. 16 shows a flowchart of overall processing in the present embodiment. As described with reference to FIG. 8, the source code acquisition unit 22 acquires the source codes v1, v1c, and v2 (S1).

  As described with reference to FIG. 9, the call relationship generation unit 23 generates call relationship information (v1 call relationship information 31, v1c call relationship information 32, v2 call relationship information 33) of the source codes v1, v1c, and v2 ( S2).

  As described with reference to FIG. 10, the replacement source extraction unit 24 extracts the replacement source information 36 from the comparison between the v1 call relationship information 31 and the v1c call relationship information 32 (S3). Details of S3 will be described with reference to FIG.

  As described with reference to FIG. 11, the replacement destination extraction unit 25 extracts replacement destination information from the comparison between the v1 call relationship information 31 and the v2 call relationship information 33 (S4). Details of S4 will be described with reference to FIG.

  As described with reference to FIG. 12, the similarity calculation unit 26 calculates the similarity between each method of the replacement source information in the v1 call relationship information 31 and each method of the replacement destination information in the v2 call relationship information 33 (S5). ). Details of S5 will be described with reference to FIGS.

  As described with reference to FIG. 15A, the result output unit 27 outputs a replacement candidate list 51b in which all methods of replacement destination information having a similarity equal to or greater than a threshold are arranged in descending order of similarity (S6). Further, the result output unit 27 acquires the correction reflection target corresponding to each correction candidate location registered in the replacement candidate list 51b. Then, the result output unit 27 gives priority to the information 52b indicating that the correction candidate portion needs to be corrected with respect to the correction candidate portion as shown in FIG. 15B based on the correction candidate portion and the correction reflection target. You may output in order of order.

  In S6, the case of outputting the case of FIG. 15 has been described as an example, but the case of FIG. 14 may be output.

  FIG. 17 is a detailed flowchart of the process of S3. The replacement source extraction unit 24 acquires the v1 call relationship information 31 and the v1c call relationship information 32 (S3-1, S3-2). The replacement source extraction unit 24 compares the v1 call relationship information 31 and the v1c call relationship information 32 (S3-3).

  The replacement destination extraction unit 25 determines whether there is a difference edge between the v1 call relationship information 31 and the v1c call relationship information 32 (S3-4). If there is no difference edge between the v1 call relationship information 31 and the v1c call relationship information 32 (“NO” in S3-4), this flow ends.

  If there is a difference edge between the v1 call relationship information 31 and the v1c call relationship information 32 (“YES” in S3-4), the replacement source extraction unit 24 extracts the difference edge as the v1-v1c change content 34 ( S3-5).

  The replacement source extraction unit 24 stores the extracted node on the origin side of the difference edge as replacement source information 36 in the storage unit (S3-6).

  FIG. 18 is a detailed flowchart of the process of S4. The replacement destination extraction unit 25 acquires the v1 call relationship information 31 and the v2 call relationship information 33 (S4-1, S4-2).

  The replacement destination extraction unit 25 compares the v1 call relationship information 31 with the v2 call relationship information 33 (S4-3). When there is a difference edge between the v1 call relation information 31 and the v2 call relation information 33, the replacement destination extraction unit 25 extracts the difference edge as the v1-v2 change content 35 (S4-4).

  The replacement destination extraction unit 25 determines whether there is an edge that exists only in the v2 call relation information 33 in the v1-v2 change content 35 (S4-5). In the v1-v2 change content 35, when there is no edge that exists only in the v2 call relation information 33 ("NO" in S4-5), this flow ends.

  In the v1-v2 change content 35, when there is an edge that exists only in the v2 call relation information 33 ("YES" in S4-5), the replacement destination extraction unit 25 selects an edge that exists only in the v2 call relation information 33. Both the start and end nodes are stored as replacement destination information 37 (S4-6).

  FIG. 19 is a detailed flowchart of the process of S5. The similarity calculation unit 26 acquires replacement source information 36 and replacement destination information 37 (S5-1). The similarity calculation unit 26 obtains a Cartesian product set of the node included in the replacement source information 36 and the node included in the replacement destination information 37 (S5-2).

  The similarity calculation unit 26 performs similarity calculation processing for all the sets of Cartesian product sets (S5-3), and stores the similarity in the storage unit as similarity information 38 (S5-4).

  FIG. 20 is a detailed flowchart of the process of S5-3. The similarity calculation unit 26 initializes the similarity with 0 (S5-3-1).

  The similarity calculation unit 26 acquires the caller node (A) of each of the two elements included in one set (S5-3-2). The similarity calculation unit 26 acquires a call destination node (B) of each of the two elements (S5-3-3).

  The similarity calculation unit 26 determines whether the node is the same, the interface is the same, or the method is the same between the two elements, the caller node (A), and the callee node (B). (S5-3-4, S5-3-6, S5-3-8).

  When the two nodes to be compared are the same, the similarity calculation unit 26 adds 3 to the similarity (S5-3-5). When the interfaces of the two nodes to be compared are the same, the similarity calculation unit 26 adds 2 to the similarity (S5-3-7). When the method names of the two nodes to be compared are the same, the similarity calculation unit 26 adds 1 to the similarity (S5-3-5).

  FIG. 21 is an example of a configuration block diagram of a hardware environment of a computer that executes a program according to the present embodiment. The computer 60 functions as the software development support device 11 and the change location detection device 21. The computer 60 includes a CPU 62, a ROM 63, a RAM 66, a communication I / F 64, a storage device 67, an output I / F 61, an input I / F 65, a reading device 68, a bus 69, an output device 71, and an input device 72.

  Here, CPU indicates a central processing unit. ROM indicates a read-only memory. RAM indicates random access memory. I / F indicates an interface. A CPU 62, ROM 63, RAM 66, communication I / F 64, storage device 67, output I / F 61, input I / F 65, and reading device 68 are connected to the bus 69. The reading device 68 is a device that reads a portable recording medium. The output device 71 is connected to the output I / F 61. The input device 72 is connected to the input I / F 65.

  As the storage device 67, various types of storage devices such as a hard disk, a flash memory, and a magnetic disk can be used. The storage device 67 or the ROM 63 stores a program according to the present embodiment that causes the CPU 62 to function as the acquisition unit 12, the first extraction unit 13, the second extraction unit 14, the calculation unit 15, and the output control unit 16. . More specifically, the storage device 67 or the ROM 63 functions as the source code acquisition unit 22, the call relationship generation unit 23, the replacement source extraction unit 24, the replacement destination extraction unit 25, the similarity calculation unit 26, and the result output unit 27. A program according to the present embodiment is stored. In addition, the storage device 67 may store the correction candidate location 51a, the replacement candidate list 51b, and information 52a and 52b indicating that the correction candidate location needs to be corrected regarding the correction reflection target.

  Information is temporarily stored in the RAM 66. The RAM 66 stores v1 call relation information 31, v1c call relation information 32, v2 call relation information 33, v1-v1c change contents 34, v1-v2 change contents 35, replacement source information 36, replacement destination information 37, similarity information 38, Threshold information 39 is stored.

  The CPU 62 reads out the program according to the present embodiment from the storage device 67 or the ROM 63 and executes the program.

  The communication I / F 64 is an interface such as a port for connecting to a network and communicating with other devices.

  The program for realizing the processing described in the above embodiment may be stored in, for example, the storage device 67 from the program provider side via the communication network 70 and the communication I / F 64. Moreover, the program which implement | achieves the process demonstrated by the said embodiment may be stored in the portable storage medium marketed and distribute | circulated. In this case, the portable storage medium may be set in the reading device 68 and the program read by the CPU 62 and executed. As a portable storage medium, various types of storage media such as a CD-ROM, a flexible disk, an optical disk, a magneto-optical disk, an IC card, a USB memory device, and a semiconductor memory card can be used. The program stored in such a storage medium is read by the reading device 68.

  As the input device 72, a keyboard, a mouse, an electronic camera, a web camera, a microphone, a scanner, a sensor, a tablet, a touch panel, or the like can be used. The output device 71 can be a display, a printer, a speaker, or the like.

  The network 70 may be a communication network such as the Internet, a LAN, a WAN, a dedicated line, a wired line, and a wireless line.

  According to the present embodiment, files (classes and methods) that need to be corrected at the time of update tracking can be extracted over a wide range.

  The present invention is not limited to the above-described embodiment, and various configurations or embodiments can be taken without departing from the gist of the present invention.

DESCRIPTION OF SYMBOLS 11 Software development support apparatus 12 Acquisition part 13 1st extraction part 14 2nd extraction part 15 Calculation part 16 Output control part 21 Change location detection apparatus 22 Source code acquisition part 23 Call relation generation part 24 Replacement source extraction 25 Replacement destination extraction Part 26 similarity calculation part 27 result output part 28 input I / F
29 Output I / F
31 v1 call relation information 32 v1c call relation information 33 v2 call relation information 34 v1-v1c change contents 35 v1-v2 change contents 36 Replacement source information 37 Replacement destination information 38 Similarity information 39 Threshold information

Claims (5)

On the computer,
Obtaining a first source code, a second source code obtained by customizing the first source code, and a third source code having a version different from that of the first source code;
A first call relationship in which a function call source and a call destination included in the first source code are paired nodes and a function call source and a call destination included in the second source code are set as a pair node. A node pair that is different from the second call relationship is extracted, and the caller node of the extracted node pairs is held in the first set;
A node pair of a difference between the first call relationship and a third call relationship having a call source and a call destination of a function included in the third source code as a pair node is extracted, and the extracted node pair Keep the caller and callee nodes in the second set;
Based on a predetermined rule, a node pair that combines each node included in the first set in the first call relationship and each node included in the second set in the third call relationship Calculate each similarity,
A software development support program for executing a process of outputting nodes included in the second set in descending order of similarity. In the calculation of the similarity, for each combined node pair, the nodes included in the combined node pair, the caller nodes of each node of the combined node pair, and each node of the combined node pair Call destination nodes of each other, and whether each of the compared node pairs is the same node, whether the methods implement the same interface, or whether the method names are the same The similarity is assigned to each of the compared node pairs based on the predetermined rule in which the value of the similarity is defined, and the total value of the similarities assigned to each of the compared node pairs is calculated. The software development support program according to claim 1, wherein the similarity of the combined node pair is used. In the output, the node included in the second set is output as a correction candidate portion, and the caller node included in the combined node pair in the first set, 2. The call destination node corresponding to the call source node in the difference node pair between the call relationship and the second call relationship is output as a correction reflection target for the correction candidate portion. Or the software development support program described in 2. An obtaining unit for obtaining a first source code, a second source code obtained by customizing the first source code, and a third source code having a version different from that of the first source code;
A first call relationship in which a function call source and a call destination included in the first source code are paired nodes and a function call source and a call destination included in the second source code are set as a pair node. A first extraction unit that extracts a node pair that is different from the second call relationship and holds the caller node of the extracted node pair in a first set;
A node pair of a difference between the first call relationship and a third call relationship having a call source and a call destination of a function included in the third source code as a pair node is extracted, and the extracted node pair A second extractor that holds the caller and callee nodes in a second set;
Based on a predetermined rule, a node pair that combines each node included in the first set in the first call relationship and each node included in the second set in the third call relationship A calculation unit for calculating each similarity,
An output control unit for outputting the nodes included in the second set in descending order of the similarity;
A software development support apparatus comprising: Computer
Obtaining a first source code, a second source code obtained by customizing the first source code, and a third source code having a version different from that of the first source code;
A first call relationship in which a function call source and a call destination included in the first source code are paired nodes and a function call source and a call destination included in the second source code are set as a pair node. A node pair that is different from the second call relationship is extracted, and the caller node of the extracted node pairs is held in the first set;
A node pair of a difference between the first call relationship and a third call relationship having a call source and a call destination of a function included in the third source code as a pair node is extracted, and the extracted node pair Keep the caller and callee nodes in the second set;
Based on a predetermined rule, a node pair that combines each node included in the first set in the first call relationship and each node included in the second set in the third call relationship Calculate each similarity,
A software development support method for executing a process of outputting nodes included in the second set in descending order of the similarity.