JP5525459B2 - Program modification support method and program modification support device - Google Patents

Description

  The present invention relates to a program modification support method and a program modification support apparatus.

In general software development, when there is a request for modification, such as modification or correction, for a source code of a program, a programmer in charge first investigates and examines the modification to be performed at the design stage. The remodeling method is summarized in the specifications.
In the next implementation stage, the person in charge of programming decides the implementation policy based on this specification, and modifies the source code based on the implementation policy.
In the final review stage, reviewers validate the modifications made to the source code.

At this time, even if the person in charge of programming decides the mounting policy correctly according to the contents of the specification, there is a possibility that the modification differs from the mounting policy such as omission of modification or modification to the wrong place. In other words, there is a possibility of remodeling different from the content of the specification. In addition, even in the review stage, if there are a wide range of remodeling locations, the reviewer may miss the erroneous remodeling.
Therefore, a mechanism for checking the validity of the modifications made to the source code is required.

  For example, Patent Document 1 discloses a program specification matching inspection apparatus that can check the consistency between a source code and a program specification in detail and mechanically.

JP-A-8-30448

In the apparatus described in Patent Document 1, it is necessary to create a specification on the apparatus in order to associate the source code with the specification that is the basis of the inspection. For this reason, a specification created on the device is newly required for specifying the modification location and verifying the modification contents.
In addition, the apparatus described in Patent Document 1 requires a specification with detailed description at the same level as the actual coding. For this reason, there is a problem that it cannot be used if the details of coding are not fixed at the time of specification creation.

  Therefore, an object of the present invention is to provide a program remodeling support method and a program remodeling support apparatus that do not require a new specification to be created on the apparatus.

In order to solve the above-mentioned problems, the present invention provides a program modification support method for a program modification support device that supports program modification, wherein the program modification support device is based on modification schedule data that is information on a planned modification site. Generating prediction data having information on a place where the planned remodeling is performed, generating post-remodeling data based on the source code after remodeling that is the source code of the program after remodeling, and generating the prediction data and The modified data are compared, and when a mismatched portion is detected, a message indicating that a mismatched portion has been detected is displayed on the display unit.
Other solutions will be described as appropriate in the embodiments.

  According to the present invention, it is possible to provide a program remodeling support method and a program remodeling support apparatus that do not require creation of a new specification on the apparatus.

It is a figure which shows the structural example of the program modification assistance apparatus which concerns on this embodiment. It is a flowchart which shows the procedure of the whole process in the program modification assistance process which concerns on this embodiment. It is a figure which shows the example of a source code which concerns on this embodiment, and an example of structure data. It is a flowchart which shows the procedure of the structure analysis process in a structure analysis part. It is a figure which shows an example of the conversion to an intermediate language command (example of C language). It is a figure which shows the example of the remodeling place input screen which concerns on this embodiment. It is a figure which shows the example of the modification information input screen which concerns on this embodiment. It is a figure which shows the example of the modification plan data which concern on this embodiment. It is a figure which shows the example of the prediction structure data which concern on this embodiment. It is a flowchart which shows the procedure of the remodeling place specific process which concerns on this embodiment. It is a figure for demonstrating the example of the determination method of the content of remodeling. It is a figure which shows the example of a remodeling place specific process result. It is a flowchart which shows the procedure of the modification content verification process which concerns on this embodiment. It is an example which shows the example of the determination pattern in a modification content verification process. It is a figure which shows the example of the verification result display screen displayed on a display part by the modification content verification part.

  Next, modes for carrying out the present invention (referred to as “embodiments”) will be described in detail with reference to the drawings as appropriate.

[Device configuration]
FIG. 1 is a diagram illustrating a configuration example of a program modification support device according to the present embodiment.
The program modification support device 1 includes a modification schedule registration processing unit 10, a structure analysis unit 30, a modification content verification processing unit 20, an input unit 40, and a display unit 50.
The input unit 40 is an input device such as a keyboard or a mouse.
The display unit 50 is a display device such as a display.
The structure analysis unit 30 is realized by a part of a general compiler function, and analyzes the source code and outputs structure data to be described later.

The modification schedule registration processing unit 10 includes a modification schedule registration unit 11 for registering a part to be modified in the source code as modification schedule data 800, and a structure when a modification scheduled from the modification schedule data 800 is performed. And a remodeling scheduled reproduction unit 12 that predicts data and generates predicted structure data (predicted data) 900.
The remodeling schedule registration unit 11 includes the source code of the program to be remodeled input by the user via the input unit 40 (source code 101 before remodeling), and the planned remodeling input by the user via the input unit 40. The remodeling schedule data 800 is stored as data based on the remodeling input data 103 which is information relating to the contents.
More precisely, (1) based on the pre-modification structure data 104, the modification schedule registration unit 11 displays the flowchart display area 602 (or source code display area 601) (FIG. 6) of the modification location input screen 600 on the display unit 50. . (2) The user views the flowchart display area 602 (or the source code display area 601) of the modification location input screen 600 and inputs the modification input data 103 via the input unit 40. (3) The modification schedule registration unit 11 stores modification schedule data 800 based on the modification input data 103. This procedure will be described in detail later.
In this embodiment, the user who has seen the source code display area 601 and the flowchart display area 602 of the remodeling location input screen 600 described later in FIG. 6 inputs the remodeling input data 103 via the input unit 40. If there is another method for obtaining the remodeling input data 103, it is not always necessary to display the source code display area 601 and the flowchart display area 602 based on the pre-modification structure data 104. Alternatively, only the source code display area 601 may be displayed.
The pre-modification structure data 104 is structure data generated by performing a later-described structural analysis process on the pre-modification source code 101. The modification schedule data 800 and the structure data will be described later.

  The remodeling schedule reproduction unit 12 is prediction information that predicts structural data when a planned remodeling is performed based on the remodeling plan data 800 generated by the remodeling plan registration unit 11 and the structure data 104 before remodeling. Generate structure data 900.

The remodeling content verification processing unit 20 verifies whether or not the remodeled source code 102 has been remodeled according to the schedule. The remodeling location identification unit (post-remodeling data generation unit) 21 and the remodeling content verification Part 22.
The remodeling location specifying unit 21 specifies the actual remodeling from the pre-remodeling structure data 104 and the post-remodeling structure data 105, which are the results of analysis by the structural analysis unit 30 from the remodeled source code 102. Information is output as remodeling place specifying data (data after remodeling) 1100.
The remodeling content verification unit 22 verifies whether remodeling has been performed according to the schedule based on the remodeling location specifying data 1100 and the predicted structure data 900, and the verification result data 106 as a result is displayed on the display unit 50. Display.
In FIG. 1, the display unit 50 appears to display only the verification result data 106, but the modified input data 103, the predicted structure data 900, and the like are also displayed as appropriate.

  The modification schedule registration processing unit 10, the modification content verification processing unit 20, the structure analysis unit 30, and the units 11, 12, 21, and 22 are stored in a ROM (Read Only Memory) or an HDD (Hard Disk Drive). The program is embodied by being expanded in a random access memory (RAM) and executed by a central processing unit (CPU).

[Overall processing]
FIG. 2 is a flowchart showing a procedure of overall processing in the program modification support processing according to the present embodiment (see FIG. 1 as appropriate).
First, a user such as a programming person obtains the pre-modification source code 101 using a terminal or the like (S1). In addition, since step S1 and step S3a are the processes by a user, the process is shown with the broken-line square.
Next, the user inputs the created pre-modification source code 101 to the program modification support device 1, and the structure analysis unit 30 of the program modification support device 1 performs the structure analysis processing of the source code 101 before modification (S2). The structure data 104 is output. The structural analysis process will be described later with reference to FIGS.
Note that source code created by someone else in the past may be used as the pre-modification source code 101, or source code created in the past may be used as the pre-modification source code 101. Moreover, the user who performs a process may differ in step S1 and step S2.
Then, the user creates the modified source code 102 using a terminal or the like (S3a).
Next, the user inputs the created modified source code 102 to the program modification support device 1, and the structure analysis unit 30 of the program modification support device 1 performs the structure analysis processing of the modified source code 102 (S4a). The structure data 105 is output. The structural analysis process will be described later with reference to FIGS.

On the other hand, a remodeling schedule registration process is performed by inputting a part to be remodeled, a remodeling content, and the like into the program remodeling support apparatus 1 (S3b). As a result of the modification schedule registration process, a modification schedule reproduction process is performed to generate predicted structure data 900 that is the structure data that is predicted to be generated as a result of the planned modification based on the modification schedule data 800 that is generated ( S4b).
The modification schedule registration process will be described later with reference to FIGS. 6 to 8, and the modification schedule reproduction process will be described later with reference to FIG.
Note that either step S3a to S4a or step S3b to S4b may be performed first, or work may be performed in parallel.

Then, the remodeling location specifying unit 21 of the program remodeling support apparatus 1 compares the pre-remodeling structure data 104 with the post-remodeling structure data 105 to perform a remodeling location specifying process for specifying a location where remodeling has actually been performed. (S5). The modification location specifying process will be described later with reference to FIGS.
Subsequently, the modification content verification unit 22 of the program modification support device 1 performs modification content verification processing for verifying whether or not the modification actually performed is a modification according to the schedule (S6).
The modification content verification processing will be described later with reference to FIGS.

[Structural analysis processing]
Next, with reference to FIGS. 3 to 5, the structural analysis process (steps S <b> 2 and 4 a in FIG. 2) relating to the present embodiment will be described with reference to FIG. 1 as appropriate. As described above, it is assumed that the structural analysis process is performed using a general compiler, but a dedicated program may be prepared. Hereinafter, instructions on the source code will be referred to as “instructions” as appropriate.

(Example of source code and structural data)
FIG. 3 is a diagram showing an example of source code and structural data according to the present embodiment. FIG. 3A corresponds to the source code example, and FIG. 3B corresponds to the source code example of FIG. FIG. 3 (c) shows an example of an intermediate language instruction. In addition, although C language is assumed as a source code, other programming languages are applicable. Further, the source code shown here indicates both the source code 101 before modification and the source code 102 after modification.
This structure data is generated as a result of the structure analysis process described later in FIG.

As shown in FIG. 3A, the source code 301 (described in C language) includes one or more functions 303. The function 303 is a series of instructions combined into one, and is called a procedure or a procedure in other programming languages.
In the source code 301, the function 303 is one function. The function 303 includes instructions such as an assignment operation 304, a comparison operation 305, a branch instruction 306, and four arithmetic operations 307.

  On the other hand, the structure data 302 shown in FIG. 3B includes a source code file name 308 and one or more function information 309 (one in the example of FIG. 3B). The function information 309 is a collection of information related to the function 303 included in the source code, and one function information 309 is generated for the function 303.

  The function information 309 is composed of one or more basic blocks 310, 311, and 313. The basic blocks 310, 311, and 313 are a set of a series of instructions that do not branch the processing flow in the program processing flow in the source code 301. The assignment operation 304 to the branch instruction 306 in the source code 301 correspond to the basic block 310 of the structure data 302. Further, the four arithmetic operations 307 and the substitution operation 322 in the source code 301 correspond to the basic block 311. Further, an assignment operation 312 in the source code 301 corresponds to the basic block 313.

Further, the basic block includes one or more intermediate language instructions 314, a basic block identifier 315, a basic block identifier 316 to be processed immediately before and after the basic block, and the like.
The intermediate language instruction is an instruction for expressing the instruction included in the source code 301 on the structure data 302, and is generally output by compilation.

As shown in FIG. 3C, the intermediate language instruction includes a name 317 indicating the type of instruction on the source code, a variable or value 318 used by the instruction on the source code, and an instruction position 319 on the source code. Configure. Incidentally, FIG. 3C is an intermediate language instruction 314 corresponding to the instruction 304 “a = 0;” of FIG. 3A, and is also an intermediate language instruction 314 of the basic block 310. The intermediate language instruction 314 indicates that “0” is copied to a variable “a” in “10th row, second column”.
Further, in the C language, there may be a case where a plurality of processes of adding the result and substituting the result are simultaneously described like the four arithmetic operations 307 of the source code 301. As described above, when a plurality of processes are described at the same time, an intermediate language instruction is changed to an intermediate language instruction for each process, such as an intermediate language instruction 320 representing an addition process and an intermediate language instruction 321 representing an assignment process. Express separately.

(Structure analysis process flowchart)
FIG. 4 is a flowchart showing the procedure of the structure analysis process in the structure analysis unit. The structure analysis unit 30 forms part of the functions of the existing compiler, and the processing in FIG. 4 is also a processing provided in the existing compiler. To do. The structural analysis process is a process performed on each of the source code 101 before modification and the source code 102 after modification. In the process of FIG. 4, the source code indicates that the source code 101 before modification or the source code 102 after modification.

First, the structure analysis unit 30 performs lexical analysis on the character string described in the source code (S101). Here, the structure analysis unit 30 divides the character string in the source code into units called tokens, which are the minimum components of the program language, such as keywords that define the program language, identifiers, operators, variables, and function names. To do. If the source code is C language, the token is divided in units of semicolons, for example.
Next, the structure analysis unit 30 performs syntax analysis on the token string generated by the lexical analysis in step S101 (S102). Here, in accordance with the grammar of the programming language of the source code, the structure analysis unit 30 determines what kind of calculation formula the token sequence constitutes, and what is the range of one function, etc. Analyze the structure. At this time, it is verified whether or not the sequence of tokens conforms to the grammar of the programming language. Specifically, for example, in the case of C language, the analysis from the first brace to the brace corresponding to the brace is recognized as one function.

Then, the structure analysis unit 30 performs semantic analysis based on the result of the syntax analysis in step S102 (S103). Here, the structure analysis unit 30 analyzes what names are used for the functions and variables, what type the variables have, and what point the branch instruction branches to. Also at this time, the structure analysis unit 30 verifies whether or not the function or variable conforms to the grammar of the programming language.
Subsequently, the structure analysis unit 30 acquires the function start point and end point in the source code based on the result of the semantic analysis in step S103, and generates function information (S104). Here, the structure analysis unit 30 registers only the function name as the function information in the function information, and adds a basic block and an intermediate language instruction in subsequent processing in this function.

Next, the structure analysis unit 30 verifies the inside of the function in the source code, and performs a post-intermediate instruction conversion that converts an instruction or process included in the source code into a corresponding intermediate language instruction (S105).
Then, the structure analysis unit 30 verifies the intermediate language instruction extracted in step S105 and determines whether it is necessary to generate a basic block (S106). The case where the basic block needs to be generated is, for example, a case where the intermediate language instruction converted in step S105 is located at the head of the function or a case where it follows the branch instruction.
If it is necessary to generate a basic block as a result of step S106 (S106 → Yes), the structure analyzing unit 30 generates a basic block according to the definition of the basic block described with reference to FIG. 3 (S107), and proceeds to step S109. .

As a result of step S106, when the generation of the basic block is unnecessary (S106 → No), the structure analysis unit 30 inserts the intermediate language instruction generated by the intermediate language instruction conversion at the end of the basic block to be processed. (S108).
Then, the structure analysis unit 30 determines whether all instructions in the function to be processed have been analyzed (S109).
If there is an unanalyzed instruction in the function to be processed as a result of step S109 (S109 → No), the structure analysis unit 30 returns the process to step S103 and analyzes the next instruction.
As a result of step S109, when all the instructions in the function to be processed are analyzed (S109 → Yes), the structure analysis unit 30 determines whether the analysis has been completed for all the functions in the source code. Determine (S110).
As a result of step S110, when there is an unanalyzed function in the source code (S110 → No), the structure analysis unit 30 returns the process to step S103 and analyzes the next function.
As a result of step S110, when the analysis is completed for all functions (S110 → Yes), the structure analysis unit 30 ends the structure analysis process.

  The structure analysis unit 30 performs the process of FIG. 4 for each of the source code 101 before modification and the source code 102 after modification, so that the program modification support device 1 generates the structure data 104 before modification and the structure data 105 after modification. To do.

(Example of conversion to an intermediate language instruction)
FIG. 5 is a diagram showing an example of conversion to an intermediate language instruction (example of C language). In FIG. 5, the left column shows instructions in the source code, and the right column shows the corresponding intermediate language instructions.
For example, in the case of an assignment instruction as indicated by reference numeral 501, the corresponding intermediate language instruction is as indicated by reference numeral 502.
In addition, in the case of an addition instruction that assigns a calculation result as indicated by reference numeral 503 to a variable, the corresponding intermediate instruction is an intermediate instruction that indicates addition as indicated by reference numeral 504 and an assignment to a variable such as reference numeral 505. Is divided into two intermediate language instructions.
In the case of a branch instruction as indicated by reference numeral 506, the corresponding intermediate language instruction includes an intermediate word instruction indicating processing for determining a condition as indicated by reference numeral 507 and an intermediate language indicating processing for branching according to the condition as indicated by reference numeral 508. Divided into two of the instructions.
In the case of a loop instruction that repeatedly executes the same processing according to a condition such as reference numeral 509, the corresponding intermediate word instruction includes an intermediate word instruction for condition determination processing such as reference numeral 510 and a process that branches according to a condition such as reference numeral 511. This is divided into an intermediate word instruction and an intermediate word instruction indicating a branch instruction for returning to the start of the loop from the end of the processing in the loop as indicated by reference numeral 512.

As described above with reference to FIGS. 3 to 5, the structure analysis unit 30 generates structural data obtained by converting the source code into the form of intermediate language instructions, thereby absorbing differences in coding methods that are different for each user. Can be Note that coding means writing a program (according to the grammar of the programming language).
In addition, by doing this, it is possible to remove parts such as comments that are not directly related to the processing of the program. Therefore, by converting the source code into the form of the intermediate language instruction, the source code 101 before modification and the source code 102 after modification can be verified more accurately in the subsequent processing.
In the intermediate language instruction conversion process (step S3b in FIG. 2) in step S105 in FIG. 4, the structure analysis unit 30 converts the instruction in the source code into an intermediate language instruction according to the conversion rule as shown in FIG.

[Modification schedule registration process]
Next, the modification schedule registration processing according to the present embodiment will be described with reference to FIGS.
(Modification location input screen)
FIG. 6 is a diagram showing an example of a modified location input screen according to the present embodiment.
The remodeling location input screen 600 has a source code display area 601 and a flowchart display area 602.
The source code display area 601 displays the source code 101 before modification, and the flowchart display area 602 displays a flowchart corresponding to the source code displayed in the source code display area 601. Note that the flowchart displayed in the flowchart display area 602 is automatically generated by the program modification support device 1, but is a well-known technique, so description of the flowchart generation processing is omitted. Further, the flowchart display area 602 can be omitted.

The user inputs a place where the modification is performed on the pre-modification source code 101 by operating the input unit 40 by a mouse operation or the like. For example, when the instruction 603 is rewritten, the user selects the line of the instruction 603. In addition, when the instruction 603 and the instruction 604 are rewritten, the user selects the lines of the instruction 603 and the instruction 604. When a new command is added between the commands 603 and 604, the user performs an operation such as clicking between the commands 603 and 604.
At this time, the screen for accepting the input by the user can be either the source code display area 601 or the flowchart display area 602. When performing in the flowchart display area 602, the same process as the process in the source code display area 601 may be performed for each process in the flowchart.
Further, the display of the window is not limited to the format in which the source code display area 601 and the flowchart display area 602 are simultaneously displayed as shown in FIG. 6, and the source code display area 601 and the flowchart display area 602 are each or only one of them. May be displayed independently.
Furthermore, if necessary for implementation, a window for displaying other information may be displayed.
When the input on the modification location input screen 600 is accepted, the modification schedule registration processing unit 10 displays the modification information input screen shown in FIG.

(Modification information input screen)
FIG. 7 is a diagram illustrating an example of a modification information input screen according to the present embodiment.
The modification information input screen 700 is a screen for accepting input of detailed information related to the modification schedule of the source code.
The remodeling information input screen 700 includes a remodeling place display area 701 for displaying a place where remodeling is performed, a remodeling content input area 702 for accepting input of remodeling contents, a summary input area 703 for receiving a summary of remodeling contents, and a basis for remodeling A specification information input area 704 for receiving input of a description portion of the specification.
Note that the remodeling information input screen 700 shown in FIG. 7 is an example of display, and if necessary for implementation, an area for accepting input of information other than the above may be displayed. Further, the location of each area is not necessarily the same as that of the modification information input screen 700, and it is not necessary to display all areas simultaneously.

In the remodeling location display area 701, information regarding the location of the instruction selected on the remodeling location input screen 600 in FIG. 6 is displayed. In the modified location display area 701, the file name and the line are displayed, and among these, the line number in the source code of the instruction selected on the modified location input screen 600 of FIG. 6 is displayed. The file name and the line can be changed via the input unit 40.
As shown in FIG. 7, in the remodeling content input area 702, “delete”, “overwrite”, “addition”, and the like can be selected, and selection of these remodeling contents can be accepted. When inputting more detailed information, the user inputs detailed information in the summary input area 703.
The summary input area 703 is information input by the user via the input unit 40, and detailed information on the planned modification is input. The outline input area 703 can be omitted.

The specification information input area 704 is information input by the user via the input unit 40, and is information such as a reference page of a specification related to the modification. Here, the “specification” may be a specification at the time of creating the source coat 101 before remodeling, or a specification describing the contents of remodeling (not essential in the present embodiment). In the present embodiment, the specifications at the time of creating the pre-modification source code may be referred to, but it is not necessary to create new specifications.
In the example of FIG. 7, “addition” is performed between lines 100 to 101.

  When there is one remodeling place, the user inputs the remodeling place in FIG. 6 and the remodeling information in FIG. 7 once. When there are a plurality of remodeling locations, the user inputs the remodeling location in FIG. 6 and the remodeling information in FIG. 7 by the number of remodeling locations.

(Remodeling schedule data)
FIG. 8 is a diagram illustrating an example of remodeling scheduled data according to the present embodiment.
The modification schedule data 800 is data generated by the modification schedule registration unit 11 based on the information (modification input data 103) received on the modification information input screen 700 of FIG. Information.
The remodeling schedule data 800 includes an identifier 801 for associating the remodeling location with the remodeling input data 103, remodeling location information 802 as remodeling location information, remodeling content information 803 indicating addition or deletion of processing, and remodeling content summary information 804. , Specification information 805 which is information on a description portion of the specification that is the basis for the modification.
Among these, the identifier 801 is given by the remodeling schedule registration unit 11. The modification location information 802 is information of the modification location display area 701 in the modification information input screen 700 of FIG. 7, the modification content information 803 is information of the modification content input area 702 of FIG. 7, and the overview information 804 is the overview of FIG. Information on the input area 703, and specification information 805 is information on the specification information input area 704.

In this way, by including the specification information 805 and the summary information 804 in the remodeling scheduled data 800, even if the details of the coding are not fixed at the time of creating the specification in the pre-modification source code 101, It becomes possible to verify the location and contents of the modification.
Further, by including the identifier 801 and the specification information 805 in the remodeling scheduled data 800, it becomes possible to associate the remodeling location with the content of the remodeling and the description of the related specification. It becomes possible to confirm the description contents of the book. Here, as described above, the “specification” is a specification at the time of creating source code before modification. In the present embodiment, the specifications at the time of creating the pre-modification source code may be referred to, but it is not necessary to create new specifications.

[Modification schedule reproduction process]
Next, the modification schedule reproduction process (step S4b in FIG. 2) will be described with reference to FIG. In the modification schedule reproduction process, when the modification schedule reproduction unit 12 modifies the pre-modification source code 101 according to the schedule input and registered in the modification schedule registration process described with reference to FIGS. The predicted structure data 900 is generated.

FIG. 9 is a diagram illustrating an example of predicted structure data according to the present embodiment.
The remodeling schedule reproduction unit 12 predicts and generates predicted structure data 900 based on the remodeling schedule data 800 (FIG. 8). At this time, the remodeling scheduled reproduction unit 12 introduces an instruction called a pseudo intermediate language instruction 901 as an intermediate language instruction for embedding the information of the modification content in the same format as the intermediate language instruction. That is, since the modification to be performed in accordance with the modification schedule registration process is not confirmed, the intermediate language instruction cannot be confirmed. Therefore, the modification schedule reproduction unit 12 generates the predicted structure data 900 using the pseudo intermediate language instruction 901. To do.

The pseudo intermediate language instruction 901 includes a prefix 902 indicating that it is a pseudo intermediate language instruction, a pseudo intermediate language instruction name 903, and pseudo processing content 904.
A prefix 902 indicates a pseudo intermediate language instruction, and is given by the remodeling scheduled reproduction unit 12.
The pseudo intermediate language instruction name 903 corresponds to the modification content information 803 in FIG. That is, if “add” is selected in the remodeling content input area 702 of FIG. 7, “insert” is displayed as in the example of FIG. 9, and “delete” is selected if “delete” is selected.
The pseudo intermediate language instruction name 903 can express what kind of modification is scheduled to be performed at the place where the pseudo intermediate language instruction 901 is inserted.
The pseudo-processing content 904 is inserted with the identifier 801 in FIG. As described above, by using the identifier 801 in the pseudo processing content 904, the location where the pseudo intermediate language instruction 901 is inserted can be associated with the specification information (specification information 805) in the remodeling scheduled data 800. .
As described above, as shown in FIG. 9, by inserting a pseudo intermediate language instruction in the remodeling scheduled reproduction process, the contents to be remodeled with respect to the modified source code 102 can be verified without the need for a new specification in the process described later. become able to.

  In the process for generating the predicted structure data 900, the remodeling schedule reproduction unit 12 refers to the remodeling schedule data 800 in FIG. Since only the processing corresponding to 803 is performed, the flowchart is omitted. For example, if the remodeling content information 803 is “insert”, the remodeling schedule reproduction unit 12 leaves a line at the corresponding location, and inserts an identifier 801 as the prefix 902, pseudo-intermediate instruction name 903, and pseudo-processing content 904. . If the modification content information 803 is “delete”, the modification schedule reproduction unit 12 deletes the corresponding line, and inserts the identifier 801 as the prefix 902, pseudo-intermediate instruction name 903, and pseudo-processing content 904 at the deleted location.

[Modification location identification processing]
Next, referring to FIG. 10 and FIG. 12, the remodeling place specifying process (step S5 in FIG. 2) will be described with reference to FIG. 1 as appropriate. In the remodeling place specifying process, the remodeling place specifying unit 21 specifies the remodeling place in the actual remodeling from the pre-remodeling structure data 104 and the post-remodeling structure data 105.

(Flow chart of remodeling location identification process)
FIG. 10 is a flowchart showing the procedure of the remodeling place specifying process according to the present embodiment. In this process, the remodeling location specifying unit 21 compares the pre-remodeling structure data 104 and the remodeling structure data 105 generated in the structural analysis process of FIG. Performs processing to insert pseudo intermediate language instructions.

First, the remodeling location specifying unit 21 acquires one intermediate language command (modified intermediate language command) from the reconstructed structure data 105 obtained by analyzing the structure of the remodeled source code 102 by the structural analysis processing of FIG. 4 (S201).
Next, the remodeling location specifying unit 21 acquires one intermediate language command (pre-remodeling intermediate language command) from the pre-remodeling structure data 104 obtained by analyzing the structure of the source code 101 before remodeling by the structural analysis processing of FIG. 4 (S202). ).

  Subsequently, the remodeling location identifying unit 21 compares the post-remodeling intermediate language instruction acquired in step S201 with the pre-remodeling intermediate language instruction acquired in step S202. It is determined whether or not they are the same intermediate language instruction (S203).

  As a result of step S203, when the modified intermediate language command and the pre-modified intermediate language command are not the same intermediate language command (S203 → No), the modified location specifying unit 21 assumes that the modification has been performed at this location. It is deemed that the modification contents are determined (S204). In step S204, it is determined what kind of remodeling has been performed at the location to be processed. Examples of the contents of remodeling include “addition”, “deletion”, and “change”.

The determination method at this time will be described with reference to FIG.
As shown in FIG. 11A, when the pre-modification intermediate language instruction that is currently processed is “A2” and the modified intermediate language instruction that is the processing target is “B1”, the processing target If the modified intermediate language instruction after the modified intermediate language instruction “B1” has the same “A2” as the processed intermediate language instruction before modification, the modified location specifying unit 21 “adds” Is determined.

  As shown in FIG. 11B, when the pre-remodeling intermediate language instruction to be processed is “A2” and the post-remodeling intermediate language instruction to be processed is “A3”, it is a processing target. If the pre-remodeling intermediate language command after the pre-remodeling intermediate language command “A2” is “A3”, which is the same as the post-remodeling intermediate language command to be processed, the remodeling location identifying unit 21 reads “delete”. judge.

  As shown in FIG. 11C, when the pre-remodeling intermediate language instruction to be processed is “A2” and the post-remodeling intermediate language instruction to be processed is “C1”, it is a processing target. If the intermediate language commands before and after the modified intermediate language command and the modified intermediate language command match after the remodeling and before the remodeling, the remodeling location specifying unit 21 determines “change”.

  Next, the remodeling location specifying unit 21 inserts (marks) a pseudo intermediate language instruction into the remodeling location specifying data 1100 (S205), and proceeds to step S207. Step S205 will be described later with reference to FIG.

As a result of step S203, when the modified intermediate language instruction and the intermediate language instruction before modification are the same intermediate language instruction (S203 → Yes), the modified location identifying unit 21 regards that the modification has been performed, The modified intermediate language instruction is inserted at the end of the modified location specifying data 1100 (S206), and it is determined whether or not all the modified intermediate language instructions and the unmodified intermediate language instructions (intermediate instruction) have been analyzed (S207). ).
If there is an intermediate language instruction that has not been analyzed as a result of step S207 (S207 → No), the remodeling location identifying unit 21 returns the processing to step S201, and the next reconstructed intermediate language instruction and pre-remodeled intermediate language instruction Perform analysis.
As a result of step S207, when all the intermediate language instructions are analyzed (S207 → Yes), the remodeling location specifying unit 21 outputs the remodeling location specifying data 1100 (S208), and the process is terminated.

(Remodeling location specific processing results)
FIG. 12 is a diagram illustrating an example of the result of the remodeling place specifying process.
FIG. 12A shows the structural data in the upper row and the source code in the lower row. The left row shows the result of the remodeling location specifying process before remodeling, the middle row after remodeling, and the right column.
The result of the remodeling place specifying process in FIG. 12A becomes the remodeling place specifying data 1100 in FIG.
From FIG. 12A, the structure data generated from the pre-modification source code 101 (structure data before modification) and the structure data generated from the modified source code 102 (structure data after modification) are compared in step S203 in FIG. Then, it can be seen that the intermediate language instruction 1101 is inserted. This indicates that the instruction 1102 is inserted in the modified source code 102.
The remodeling location specifying unit 21 replaces the intermediate language instruction 1101, which is an instruction inserted by remodeling in this way, with a pseudo intermediate language instruction 1103.

  Actually, the remodeling location specifying unit 21 compares the intermediate language commands of the structure data 104 before remodeling and the structure data 105 after remodeling in order from the top. If the intermediate language instruction of the data 105 is added and a mismatch is detected, the modification content is determined according to the determination method described in FIG. 11, and the pseudo intermediate language instruction corresponding to the modification content is added to the modification location specifying data 1100. Add.

  As shown in FIG. 12 (b), the pseudo intermediate language instruction 1103 in the remodeling place specifying process result is the same as the pseudo intermediate language instruction 901 (FIG. 9) used in the prediction structure data 900 and the like. It has an instruction name 1112 and a pseudo process content 1113. However, the pseudo-processing content 1113 of the pseudo-intermediate instruction 1103 in the modified location specifying process result is that the modified intermediate language instruction 1101 itself in the post-modification structure data 105 is inserted into the predicted structure data 900 and the like. This is different from the pseudo intermediate language instruction 901 used in FIG.

  As described above, the remodeling location in the post-remodeling structure data 105 is clarified by replacing the remodeling location after the remodeling with the pseudo intermediate language command by the remodeling location specifying process described in FIGS.

[Modification content verification process]
Next, with reference to FIGS. 13 to 15, the modification content verification process (step S <b> 5 in FIG. 2) will be described with reference to FIG. 1 as appropriate. In the modification content verification process, the modification content verification unit 22 compares the modification location identification data 1100 generated in the process of FIG. 10 with the predicted structure data 900 generated in the modification schedule reproduction process described in FIG. Thus, it is determined whether remodeling has been performed according to the schedule, and the determination result is displayed on the display unit 50.

(Modification content verification process flowchart)
FIG. 13 is a flowchart showing the procedure of the modification content verification process according to the present embodiment.
First, the remodeling content verification unit 22 acquires a basic block (basic block after remodeling) from the remodeling location specifying data 1100 (S301).
Next, the modification content verification unit 22 retrieves a similar basic block that is a basic block similar to the modified basic block acquired in step S301 from the predicted structure data 900 in which the modification content is reproduced by the modification schedule reproduction unit 12 ( S302), it is determined whether or not a similar basic block has been detected (S303). As a reference for detecting a similar basic block, the number of matching intermediate language instructions in the basic block is a certain number or more, and the basic target to be processed in the remodeling location specifying data 1100 and the predicted structure data 900 For example, the number of matching intermediate language instructions is equal to or greater than a certain number in the basic blocks before or after the blocks.

  If no similar basic block is detected as a result of step S303 (S303 → No), the remodeling content verification unit 22 regards the basic block acquired in step S301 as an unscheduled remodeling and verifies “NG” as the verification result. The data 106 is registered (S304), and the process proceeds to step S309. At this time, in the verification result data 106, the identifier 315 (FIG. 3) of the corresponding basic block is registered together with the information “NG”.

When a similar basic block is detected as a result of step S303 (S303 → Yes), the remodeling content verification unit 22 determines that the modified basic block acquired in step S301 and the intermediate language included in the detected similar basic block The instructions are compared (S305).
Then, the remodeling content verification unit 22 determines whether or not the format of the intermediate language instruction matches between the remodeled basic block acquired in step S201 and the similar basic block (S306). Here, the intermediate language instruction includes a pseudo intermediate language instruction, and the format relates to whether or not it is a pseudo intermediate language instruction.
If the result of step S306 is a mismatch (S306 → No), the remodeling content verification unit 22 registers “NG” in the verification result data 106 (S304), and the process proceeds to step S309. At this time, if the similar basic block side is a pseudo intermediate language instruction, the remodeling content verification unit 22 does not match the identifier of the pseudo processing content 904 (FIG. 9) of the predicted structure data 900 and the intermediate language instruction in which a mismatch is detected. The line number is registered in the verification result data 106 together with the information “NG”. If the similar basic block side is not a pseudo intermediate language instruction, the line number of the intermediate language instruction in which the mismatch is detected is registered in the verification result data 106 together with the information “NG”. Here, the registered line number is the line number of the modified structural data 105.

If the result of step S306 is a match (S306 → Yes), the remodeling content verification unit 22 determines whether or not the content matches (S307). Here, the contents are “insert”, “delete”, and the like. For example, if both are “insert”, the modification content verification unit 22 determines “match”, and if it is “insert” and “delete”, the modification content verification unit 22 determines “mismatch”.
If the result of step S307 is a mismatch (S307 → No), the remodeling content verification unit 22 advances the process to step S304, and registers “NG” in the verification result data 106 (S304). Here, if the similar basic block side is a pseudo intermediate language instruction, the remodeling content verification unit 22 identifies the identifier of the pseudo processing content 904 (FIG. 9) of the predicted structure data 900 and the line number of the intermediate language instruction in which a mismatch is detected. Is registered in the verification result data 106 together with the information “NG”. If the similar basic block side is not a pseudo intermediate language instruction, the line number of the intermediate language instruction in which the mismatch is detected is registered in the verification result data 106 together with the information “NG”. Here, the registered line number is the line number of the modified structural data 105.

  If the result of step S307 is a match (S307 → Yes), the remodeling content verification unit 22 registers “OK” in the verification result data 106 (S308). At this time, the modification content verification unit 22 registers the line number of the corresponding intermediate language instruction in the verification result data 106 together with the information “OK”. Here, the registered line number is the line number of the modified structural data 105.

Then, the remodeling content verification unit 22 determines whether all intermediate language instructions in the basic block to be processed (basic block after remodeling, similar basic block) have been analyzed (S309).
If there is an intermediate language instruction that has not been analyzed as a result of step S309 (S309 → No), the modified content verification unit 22 returns to step S305 to analyze the next intermediate language instruction.
As a result of step S309, when all the intermediate language instructions are analyzed (S309 → Yes), the remodeling content verification unit 22 determines whether all the basic blocks in the remodeling location specifying data 1100 have been analyzed (S310). .
If there is a basic block that has not been analyzed in the modified location specifying data 1100 as a result of step S310 (S310 → No), the modified content verification unit 22 returns the process to step S301 and analyzes the next basic block.
As a result of step S310, when all the basic blocks in the remodeling location specifying data 1100 are analyzed (S310 → Yes), the remodeling content verification unit 22 outputs the verification result data 106 (S311), and performs remodeling content verification processing. finish.

(Modification content verification judgment pattern)
FIG. 14 is an example showing an example of the determination pattern used in step S305 of FIG. Note that the step numbers in this description are the step numbers in FIG.
In FIG. 14, when the intermediate language instruction of the remodeling location specifying data 1100 is a pseudo intermediate language instruction, this indicates that the remodeled source code 102 has been remodeled. This indicates that the source code 102 after the modification has not been modified. In addition, when the intermediate language instruction of the predicted structure data 900 is a pseudo intermediate language instruction, it indicates that the modification is scheduled to be performed, and when it is not a pseudo intermediate language instruction, the modification is not scheduled. Show.
In the “pattern A”, the intermediate language instruction of the remodeling location specifying data 1100 and the intermediate language instruction of the prediction structure data 900 are not pseudo intermediate language instructions (intermediate language instructions), and the contents are also matched (step). S306 → Yes; Step S307 → Yes). This “pattern A” indicates that the place is not a place to be remodeled and has not been remodeled, and the remodeling content verification unit 22 registers “OK” in the verification result data 106 (step S308).
“Pattern B” is a pattern in which the intermediate language instruction of the remodeling location specifying data 1100 is not a pseudo intermediate language instruction (intermediate language instruction), and the intermediate language instruction of the predicted structure data 900 is a pseudo intermediate language instruction. In this case, the contents are naturally inconsistent. This “pattern B” indicates that the necessary modification is not performed on the place where the modification is required, and the modification content verification unit 22 registers “NG” in the verification result data 106 (step S304). ).

“Pattern C” is a pattern in which the intermediate language instruction of the remodeling location specifying data 1100 is a pseudo intermediate language instruction, and the intermediate language instruction of the predicted structure data 900 is not a pseudo intermediate language instruction (intermediate language instruction). It is a pattern that does not match. This “pattern C” indicates that unnecessary modification has been performed in a place where modification is not necessary, and the modification content verification unit 22 registers “NG” in the verification result data 106 (step S304). .
“Pattern D” is a pattern in which the intermediate language instructions of the remodeling location specifying data 1100 and the predicted structure data 900 are both pseudo intermediate language instructions and the contents match (step S306 → Yes; step S307 → Yes). . This “pattern D” indicates that appropriate remodeling has been performed at a place where remodeling is necessary, and the remodeling content verification unit 22 registers “OK” in the verification result data 106 (step S308).

“Pattern E” is a pattern in which both of the intermediate language instructions of the remodeling location specifying data 1100 and the predicted structure data 900 are pseudo intermediate language instructions, but their contents are inconsistent (step S306 → Yes; step S307 → No). . This “Pattern E” indicates that the modification has been made at the planned location, but the contents are different, and the modification content verification unit 22 registers “NG” in the verification result data 106. (Step S304).
“Pattern F” is a pattern in which the intermediate language instructions in the remodeling location specifying data 1100 and the predicted structure data 900 are not pseudo intermediate language instructions (intermediate language instructions), but their contents do not match (step S306 → Yes; step S307 → No). This “pattern F” indicates that there is a suspicion that a source code different from the pre-modification source code 101 to be modified is modified and registered as the post-modification source code 102. Registers “NG” in the verification result data 106 (step S304).

(Verification result display screen)
FIG. 15 is a diagram illustrating an example of a verification result display screen displayed on the display unit by the modification content verification unit.
The verification result display screen 1400 has a post-modification source code display area 1401 and a verification result display area 1402.
The verification result display area 1402 includes an OK area 1403 that is displayed when the modification is correctly performed according to the modification schedule, and an NG area 1404 that is an area displayed when a modification different from the modification schedule is performed. have.
The OK area 1403 includes an ID display area 1405 for identifying a modification schedule, a determination result display area 1406, a specification information display area 1407 for displaying a description location of a specification that is the basis for modification, and a message display area 1408 for prompting a review. Have.
As in the OK area 1403, the NG area 1404 includes an ID display area 1408 for identifying a modification schedule, a determination result display area 1409, a specification information display area 1410 for displaying a description location of a specification that is the basis for modification, and an abnormality determination. A determination content display area 1411 indicating the content of the item is displayed.

Among these, the contents displayed in the ID display areas 1405 and 1408 are identifiers registered in the verification result data 106. If a mismatch is detected at a location where the predicted structure data 900 is not a pseudo-middle language instruction, the ID display areas 1405 and 1408 are blank. The contents of the specification information display areas 1407 and 1410 are the specification information 805 of the remodeling scheduled data 800 shown in FIG. These data are information acquired by the modification content verification unit 22 searching the modification schedule data 800 using the identifier in the verification result data 106 as a key. Note that the specification displayed in the specification information 805 is a specification created when creating the pre-modification source code.
The determination result display areas 1406 and 1409 are information displayed by holding the result in step S306 of FIG.
The message display area 1408 and the determination content display area 1411 are information displayed by preparing a message for each pattern shown in FIG. 14, for example.

In the post-modification source code display area 1401, the case where the modification location is normal (“OK”) and the abnormal (“NG”) are distinguished and highlighted, and further associated with each item in the verification result display area 1402. I do. Such association is performed by the remodeling content verification unit 22 referring to the line number registered in the verification result data 106.
In the example of FIG. 15, the highlight display location 1421 and the OK area 1403 are connected by a straight line 1412, and the highlight display location 1422 and the NG area 1404 are connected by a straight line 1413, so Is associated with the report contents.

  Note that FIG. 15 is an example of display, and an area for displaying other information may be added if necessary for mounting. Further, it is not necessary to display all the areas at the same time. For example, the area to be displayed may be selectable by a user click input or the like. Also, the method of indicating the remodeling location is not limited to the highlight display.

  In this way, by clearly showing the contents of the planned remodeling and location, the content of the actual remodeling, and the location where consistency is not achieved, It is possible to prompt the user to modify the source code 102 after remodeling or re-input the remodeled input data 103.

  The modification content verification unit 22 described with reference to FIGS. 10 to 15 makes it possible to verify the modification contents for the modified source code 102 without requiring a new specification. In addition, even if the details of coding are not confirmed in the creation of the specifications when creating the source code before modification, the modification location and modification contents can be verified.

(Summary)
As described above, according to the present embodiment, it becomes possible to verify the modified contents of the modified source code 102 without requiring a new specification, and the detailed coding contents in the creation of the specifications at the time of creating the modified source code Even if is not confirmed, it will be possible to verify the location and contents of the modification.
As a result, an erroneous modification in the programming stage can be detected in the review stage, and a review omission in the review stage can be prevented.

  In addition, this invention is not limited to the example described in embodiment, Various modifications are included. For example, in the present embodiment, the present invention has been described in detail for easy understanding, and is not necessarily limited to the one having all the configurations described.

Further, the names and formats of the intermediate language instructions described in this embodiment are examples, and the names and formats may be changed or information may be added depending on the implementation.
In addition, each configuration, function, processing unit, processing unit, and the like of the present embodiment may be realized by hardware by designing a part or all of them with, for example, an integrated circuit. Each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor. Information such as programs, tables, and files for realizing each function is stored in a memory, a hard disk, a recording device such as an SSD (Solid State Drive), an IC (Integrated Circuit) card, an SD (Secure Digital) card, a DVD (Digital Versatile Disc) can be stored.

1 Program modification support device 10 Modification schedule registration processing section 11 Modification schedule registration section 12 Modification modification reproduction section (data generation section after modification)
20 Modification Content Verification Processing Unit 21 Modification Location Identification Unit 22 Modification Content Verification Unit 30 Structural Analysis Unit 40 Input Unit 50 Display Unit 101 Source Code Before Modification 102 Source Code After Modification 103 Modification Input Data 104 Structure Data Before Modification 105 Structure Data After Modification 106 Verification result data 800 Planned modification data 900 Predicted structure data (predicted data)
1100 Modification location specific data (data after modification)

Claims (6)

A program remodeling support method for a program remodeling support device for supporting program remodeling,
The program modification support device is
Convert the instructions in the source code before modification, which is the source code of the program before modification, into intermediate language instructions output by the function of the compiler to generate structure data before modification,
Prediction data having information on the location where the planned remodeling will be performed based on the remodeling planned data that is information related to the planned remodeling location and the structure data before remodeling, which are input via the input unit. Is generated using the intermediate language instruction,
Using the intermediate language instruction to generate modification location identification data based on the modified source code that is the source code of the program after modification,
The information on the location where the modification is performed in the prediction data is compared with the location of the modification specified by the modification location specifying data, and if a mismatched portion is detected, the fact that the mismatched location has been detected A program modification support method characterized by displaying on a display unit. A program remodeling support method for a program remodeling support device for supporting program remodeling,
The program modification support device is
Via the input section, information on the specifications related to the planned modification is entered,
The information about the entered specifications is reflected in the remodeling planned data, which is information on the remodeling planned location,
Information on the modification schedule data is linked to the prediction data having information on the location where the planned modification is performed based on the modification schedule data,
Generate modification location identification data based on the modified source code that is the source code of the program after modification,
The information regarding the location where the remodeling is performed in the prediction data is compared with the location of the remodeling specified by the remodeling location specifying data, and when a noncoincidence location is detected, based on the prediction data, the disagreement A program modification support method, characterized in that information on a specification of the modification planned data including information on a location is displayed on a display unit. A program remodeling support method for a program remodeling support device for supporting program remodeling,
The program modification support device is
Convert the instructions in the source code before modification, which is the source code of the program before modification, into intermediate language instructions output by the function of the compiler to generate structure data before modification,
An instruction in the modified source code that is the source code of the program after modification is converted into the intermediate language instruction to generate modified structure data,
Based on the remodeling schedule data, which is information on the planned remodeling location, generate prediction data having information on the location where the planned remodeling will be performed,
Compare the structure data before remodeling and the structure data after remodeling, and mark the location where inconsistency is detected,
The information of the marking, by storing the modified location specific data based on said modified after the source code, prior to generating the Kiaratame granulation place specifying data,
The information regarding the location where the remodeling is performed in the prediction data is compared with the marking information in the remodeling location specifying data, and when a mismatched portion is detected, the display unit displays that the mismatched location has been detected. A program modification support method characterized in that A program modification support device for supporting program modification,
A structure analysis unit that converts the instructions in the source code before modification, which is the source code of the program before modification, into intermediate language instructions output by the function of the compiler to generate structure data before modification;
Prediction data having information on the location where the planned remodeling will be performed based on the remodeling planned data that is information related to the planned remodeling location and the structure data before remodeling, which are input via the input unit. A remodeling scheduled reproduction unit that generates using the intermediate language instruction,
A modified location identifying unit that generates modified location identification data based on the modified source code, which is the source code of the program after modification, using the intermediate language instruction;
The information on the location where the modification is performed in the prediction data is compared with the location of the modification specified by the modification location specifying data, and if a mismatched portion is detected, the fact that the mismatched location has been detected Modification content verification section to be displayed on the display section,
A program modification support device characterized by comprising: A program modification support device for supporting program modification,
A modification schedule registration unit that receives information on the specification sheet regarding the planned modification through the input unit, and reflects the information regarding the input specification sheet in the modification schedule data, which is information on the planned modification area,
A modification schedule reproduction unit that links information related to the modification schedule data to prediction data having information on a location where a planned modification is performed based on the modification schedule data;
A modified location identifying unit that generates modified location identifying data based on the modified source code that is the source code of the program after modification;
The information regarding the location where the remodeling is performed in the prediction data is compared with the location of the remodeling specified by the remodeling location specifying data, and when a noncoincidence location is detected, based on the prediction data, the disagreement A modification content verification unit that displays information on the specification of the modification planned data including information on the location on the display unit;
A program modification support device characterized by comprising: A program modification support device for supporting program modification,
This is the source code of the program after modification by converting the instructions in the source code before modification, which is the source code of the program before modification, into intermediate language instructions output by the function of the compiler to generate structure data before modification. A structure analysis unit that converts instructions in the modified source code into the intermediate language instructions to generate modified structure data;
Based on the remodeling schedule data that is information on the remodeling planned location, the remodeling planned reproduction unit that generates prediction data having information on the location where the planned remodeling is performed,
Said modified front structure data, said comparing and remodeling after structure data, and mark the mismatch is detected location, the information of the marking, by storing the modified location specific data based on the source code after the retrofit , a modified location detection part for pre-generate Kiaratame granulation place specifying data,
The information regarding the location where the remodeling is performed in the prediction data is compared with the marking information in the remodeling location specifying data, and when a mismatched portion is detected, the display unit displays that the mismatched location has been detected. Remodeling content verification section
A program modification support device characterized by comprising:

Images