JP2019139268A - Program analysis method, program analysis apparatus and program analysis program - Google Patents

Abstract

To appropriately generate a business rule for a modified program.SOLUTION: A storage unit 11 stores a second program P2 that is a modified first program P1 and has inconsistence with a first business rule table corresponding to the first program P1, and information indicating a definition rule of business terms R1. A processing unit 12 generates an input and output table T2 indicating a branch condition regarding a conditional sentence in the second program P2 and an output according to the branch condition, and defines business terms to a combination of the conditional sentences included in the input and output table T2 based on the information indicating the definition rule R1. The processing unit 12 generates a business rule table T4 corresponding to the second program P2 using the defined business terms and the input and output table T2.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a program analysis method, a program analysis apparatus, and a program analysis program.

  In an information processing system, software that realizes various functions is used. Software programs may be updated as functions are added or modified. Therefore, a technique for supporting program update is considered.

  For example, a part that is likely to be corrected in daily maintenance work from a source program and a part that depends on the OS (Operating System) that is the main correction target at the time of system update are automatically separated / extracted to reconstruct the system. There is a proposal for a business specification analyzer that reduces the load.

  Note that a single conditional expression created by the user and a conditional expression combined with a single conditional expression are handled as one term in the condition list display area on the display screen, and the logical connection relationship between the two terms by the user There is a proposal of a creation device that creates a database search condition only by repeated setting.

JP 7-104994 A JP-A-11-219371

  In an information processing system that has been operated and maintained for a long period of time, the program is often partially modified. Here, there can be a design document created during development for the program. However, when the program is modified, the design document of the program may not be sufficiently modified. In this case, the modification content of the program is not reflected in the design document, and a flaw may occur between the program and the design document. For example, despite the fact that a data item and logic using the data item have been added to the program, the added data item and logic may leak from a document that describes business terms, business rules, etc. . For this reason, when the user performs maintenance or reconstruction of the current information processing system, the current state of the program may not be correctly grasped from the existing design document.

  In one aspect, an object of the present invention is to make it possible to appropriately create business rules for a modified program.

  In one aspect, a program analysis method is provided. In this program analysis method, the computer is the second program after the modification of the first program, and the second program in which a conflict with the first business rule table corresponding to the first program has occurred Is created and an input / output table indicating the branch condition related to the conditional statement in the second program and the output corresponding to the branch condition is created and included in the input / output table based on the information indicating the definition rule of the business term A business term for the combination of conditional statements is defined, and a second business rule table corresponding to the second program is created using the defined business term and the input / output table.

  In one aspect, business rules can be appropriately created for the modified program.

It is a figure which shows the program analysis apparatus of 1st Embodiment. It is a figure which shows the hardware example of the analysis server of 2nd Embodiment. It is a figure which shows the example of a V-shaped model. It is a figure which shows the example of the documented information range and the difficulty of an understanding. It is a figure which shows the example (the 1) of the document in a V-shaped model. It is a figure which shows the example (the 2) of the document in a V-shaped model. It is a figure which shows the function example of an analysis server. It is a figure which shows the example of symbolic execution. It is a figure which shows the example of the program of analysis object. It is a figure which shows the example of a symbolic execution result. It is a figure which shows the example of an input / output table. It is a figure which shows the example of the intermediate data for business condition candidate extraction. It is a figure which shows the example of a business condition candidate table. It is a figure which shows the example of the intermediate data for business condition selection. It is a figure which shows the example of a business condition selection result. It is a figure which shows the example of a naming rule table | surface. It is a figure which shows the example of a business term table | surface. It is a figure which shows the example of a business rule table | surface. It is a flowchart which shows the process example of an analysis server. It is a flowchart which shows the process example of symbolic execution. It is a flowchart which shows the process example of an input / output table production | generation. It is a flowchart which shows the process example of business condition candidate production | generation. It is a flowchart which shows the process example of business condition selection. It is a flowchart which shows the process example of a business term table output. It is a flowchart which shows the example of a process of business rule table output. It is a figure which shows the example of the input / output of an analysis server. It is a figure which shows the other example of the program of analysis object. It is a figure which shows the other example of the input / output of an analysis server.

Hereinafter, the present embodiment will be described with reference to the drawings.
[First Embodiment]
FIG. 1 is a diagram illustrating a program analysis apparatus according to the first embodiment. The program analysis apparatus 10 supports revision of the design document according to the modification of the program. Here, the program analysis apparatus 10 manages the design document as electronic data. The design document includes a requirement definition document and a basic design document in a V-shaped model for system development. For example, the requirement definition document includes a business rule table. The business rule table is information indicating processing rules (business rules) for specific data items in business. In addition, the requirement definition document and the basic design document include a business term table. The business term table is information indicating definitions of terms used in business rules.

  The program analysis device 10 includes a storage unit 11 and a processing unit 12. The storage unit 11 may be a volatile semiconductor memory such as a RAM (Random Access Memory) or a non-volatile storage such as an HDD (Hard Disk Drive) or a flash memory. The processing unit 12 is, for example, a processor such as a CPU (Central Processing Unit) or a DSP (Digital Signal Processor). However, the processing unit 12 may include an electronic circuit for a specific application such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA). The processor executes a program stored in a memory such as a RAM (or the storage unit 11). A set of processors may be referred to as “multiprocessor” or simply “processor”.

  The storage unit 11 stores the second program P2 after the correction of the first program P1. The storage unit 11 stores a first business rule table T1 corresponding to the first program P1. The storage unit 11 also stores information indicating the business term definition rule R1.

  Here, although the first program P1 is modified, the first business rule table T1 is not modified. For example, the second program P2 is obtained by adding data items and logic for the data items to the first program P1. For this reason, the second program P2 has a conflict with the first business rule table T1 (that is, the contents of both do not match). Here, the numbers on the left side of the first program P1 and the second program P2 in the example of FIG. 1 are line numbers assigned for convenience.

  In the example of FIG. 1, in the second program P2, the definition of the constant C2 is added to the fourth line, and the condition determination using the constant C2 regarding the variable “code” is added to the sixth line. Different from the first program P1. In FIG. 1, an additional portion of the second program P2 with respect to the first program P1 is underlined for emphasis. For this reason, for example, the first business rule table T1 is that the content of the constant C2 added to the second program P2 is not reflected in the business rules related to the variable “code” in the first business rule table T1. It can be said that there is a wrinkle in the correspondence between the program and the second program P2.

  The processing unit 12 generates an input / output table T2 indicating a branch condition (a set of input values for the input variable for each branch) related to the conditional statement in the second program P2 stored in the storage unit 11 and an output corresponding to the branch condition. create. For example, the input / output table T2 includes items of item number, output, and condition (branch condition). The processing unit 12 may create the input / output table T2 based on the result of symbolic execution of the second program P2. Symbolic execution is a technique for analyzing the logic of a program by setting a symbol value in an input variable and executing the program in a simulated manner. The result of symbolic execution includes a variable condition for each path by a conditional branch included in the program, and an output. The processing unit 12 obtains the input / output table T2 by organizing the results. The input / output table T2 is an input / output table corresponding to the second program P2. The processing unit 12 stores information indicating the created input / output table T2 in the storage unit 11.

  Based on the information indicating the definition rule R1, the processing unit 12 defines business terms for a combination of conditional statements (compound conditions) included in the input / output table T2. For example, in the definition rule R1, business terms corresponding to the format of the combination of conditional sentences are set in advance. For example, for the branch condition of the item number “1” in the input / output table T2, the conditional statements “code = 1” and “code” corresponding to the IF statements on the 3rd to 4th lines and the 6th to 7th lines of the second program P2 = 2 ”,“ 0 <range ”and“ range <10 ”.

  In the first example, the definition rule R1 includes a rule that “when the condition“ (lower limit value <variable) AND (variable <upper limit value) ”is matched, it is named“ within variable ””. The character strings “variable”, “lower limit value” and “upper limit value” in the rule are replaced with specific character strings included in the conditions of the input / output table T2. In this example, “range” included in the record of item number “1” in the input / output table T2 is applied to “variable”, “0” to “lower limit value”, and “10” to “upper limit value”. Can do. Therefore, the processing unit 12 names the compound term “(0 <range) AND (range <10)” included in the item number “1” of the input / output table T2 as the business term “within range”. Note that the processing unit 12 defines business terms for composite conditions included in the input / output table T2 that include negative conditions in the composite condition in the input / output table T2. This is because a condition that does not include a negative condition does not affect the program path and is not considered to be involved in the business rules. For example, for the condition “(0 <range) AND (range <10)” of the item number “1” in the input / output table T2, the condition “(0 NOT <range) OR (range) of the item number“ 2 ”is used. NOT <10) ”is a negative condition.

  In the second example, the definition rule R1 includes a rule “name it as an“ effective variable ”when it matches the condition“ (variable = constant 1) [OR (variable = constant 2)] + ””. The “[OR (variable = constant 2)] +” part of the rule indicates that the conditions for the second and subsequent constants can be combined with OR. In addition, the character strings “variable”, “constant 1”, and “constant 2” in the rule are replaced with specific character strings included in the conditions of the input / output table T2. In this example, the “code” included in the record of the item numbers “1” and “2” in the input / output table T2 is “variable”, “1” is “constant 1”, and “2” is “constant 2”. Can be applied to each. Therefore, the processing unit 12 names the compound condition “(code = 1) OR (code = 2)” included in the item numbers “1” and “2” of the input / output table T2 as the business term “effective code”. The condition “(code NOT = 1) AND (code NOT = 2)” of the item number “3” in the input / output table T2 becomes a negative condition of the condition “(code = 1) OR (code = 2)”. ing.

  The processing unit 12 stores information indicating the business term table T3 including the defined business terms in the storage unit 11. In the business term table T3, positive conditions are associated with business terms. The positive condition is a condition having a smaller negation operator among two conditions that are in a negative relationship with each other.

  The processing unit 12 creates a second business rule table T4 corresponding to the second program P2 by using the defined business terms and the input / output table T2. For example, based on the business term table T3, the processing unit 12 converts the “(code = 1) OR (code = 2)” part in the condition of the item number “1” of the input / output table T2 to “valid code = True”. Replace with Here, “True” indicates a positive condition for the business term condition defined in the business term table T3. Similarly, based on the business term table T3, the processing unit 12 converts the “(0 <range) AND (range <10)” portion in the condition of the item number “1” of the input / output table T2 to “within range = Replace with “True”. Then, the condition of the item number “1” in the input / output table T2 is expressed as “(valid code = True) AND (within range = True)”.

  Further, based on the business term table T3, the processing unit 12 represents the condition of the item number “2” in the input / output table T2 as “(valid code = True) AND (within range = False)”. The reason for “within range = False” is that “(0 NOT <range) OR (range NOT <10)” in the condition of item number “2” in the input / output table T2 is the business term “within range”. This is because it is a negative condition for the condition.

  Further, the processing unit 12 represents the condition of the item number “3” in the input / output table T2 as “(effective code = False)” based on the business term table T3. The reason for “effective code = false” is that “(code NOT = 1) AND (code NOT = 2)” in the condition of the item number “3” in the input / output table T2 corresponds to the condition of the business term “effective code”. Because it is a negative condition.

  The processing unit 12 stores the created second business rule table T4 in the storage unit 11. Further, the processing unit 12 associates the business term table T3 and the second business rule table T4 with the second program P2.

  The business term table T3 and the second business rule table T4 newly created in this way have contents reflecting corrections to the second program P2. For example, the description of the condition relating to “valid code” in the second business rule table T4 appropriately reflects the condition related to the newly added constant C2 according to the business term table T3. In this way, according to the program analysis apparatus 10, it is possible to appropriately create a business rule for the corrected program.

Hereinafter, a server computer (analysis server) used as the program analysis apparatus 10 will be exemplified to describe the function of program analysis in more detail.
[Second Embodiment]
FIG. 2 is a diagram illustrating a hardware example of the analysis server according to the second embodiment. The analysis server 100 includes a CPU 101, a RAM 102, an HDD 103, an image signal processing unit 104, an input signal processing unit 105, a medium reader 106, and a NIC (Network Interface Card) 107. Each piece of hardware is connected to the analysis server 100 bus.

  Here, the analysis server 100 is an example of the program analysis apparatus 10 according to the first embodiment. The CPU 101 is an example of the processing unit 12 according to the first embodiment. The RAM 102 or the HDD 103 is an example of the storage unit 11 according to the first embodiment.

  The CPU 101 is hardware that controls information processing of the analysis server 100. The CPU 101 may be a multiprocessor. The CPU 101 is, for example, a CPU, DSP, ASIC, or FPGA. The CPU 101 may be a combination of two or more elements among a CPU, a DSP, an ASIC, an FPGA, and the like.

  The RAM 102 is a main storage device of the analysis server 100. The RAM 102 temporarily stores at least part of an OS program and application programs to be executed by the CPU 101. The RAM 102 stores various data used for processing by the CPU 101.

  The HDD 103 is an auxiliary storage device of the analysis server 100. The HDD 103 magnetically writes and reads data to and from the built-in magnetic disk. The HDD 103 stores an OS program, application programs, and various data. The analysis server 100 may include other types of auxiliary storage devices such as a flash memory and an SSD (Solid State Drive), or may include a plurality of auxiliary storage devices.

  The image signal processing unit 104 outputs an image to the display 21 connected to the analysis server 100 in accordance with a command from the CPU 101. As the display 21, a CRT (Cathode Ray Tube) display, a liquid crystal display, or the like can be used.

  The input signal processing unit 105 acquires an input signal from the input device 22 connected to the analysis server 100 and outputs it to the CPU 101. As the input device 22, for example, a pointing device such as a mouse or a touch panel, a keyboard, or the like can be used.

  The medium reader 106 is a device that reads programs and data recorded on the recording medium 23. As the recording medium 23, for example, a magnetic disk such as a flexible disk (FD) or HDD, an optical disk such as a CD (Compact Disc) or a DVD (Digital Versatile Disc), or a magneto-optical disk (MO) is used. Can be used. Further, as the recording medium 23, for example, a non-volatile semiconductor memory such as a flash memory card can be used. For example, the medium reader 106 stores a program or data read from the recording medium 23 in the RAM 102 or the HDD 103 in accordance with an instruction from the CPU 101.

The NIC 107 communicates with other devices via the network 24. The NIC 107 may be a wired communication interface or a wireless communication interface.
FIG. 3 is a diagram illustrating an example of a V-shaped model. In the field of system development, development process planning and development work are performed according to the V-shaped model. In the V-shaped model, the requirement definition is first performed. In the requirement definition, the requirements required by users are clarified. Next, the design for realizing the requirements clarified in the requirement definition is refined step by step through the basic design and detailed design processes, and then the program manufacturing process is performed. This is the flow of development work from the upper left to the lower of the V-shape. When the program is manufactured, the quality is confirmed by associating the design of each process with the test (unit test, integration test, and system test) from the bottom to the top right of the V shape.

Here, a product is created at each stage of the V-shaped model. The deliverables include the following documents:
The product of the requirement definition is a requirement definition document D1. The requirement definition document D1 is created for the purpose of clarifying requirements and functions to be realized in the development target system between the system orderer and the system developer. The requirement definition document D1 may be described using business terms rather than system development terms so that even an orderer who is not familiar with system development can easily understand.

  The product of the basic design is the basic design document D2. The basic design document D2 describes basic design items for realizing the requirements of the requirement definition document. The basic design document D2 is a basic and general design document, and is described by using business terms and basic design terms without describing a level that can be executed by a computer.

  The product of the detailed design is a detailed design document D3. The detailed design document D3 describes a design in which the design contents of the basic design document D2 are refined to a level that can be implemented in a program. It is described using terms at the level corresponding to the program variable.

  The product of manufacture is program D4. The program D4 is a source program for executing the design contents of the detailed design document D3 by a computer, and is described using a program language.

Each product is detailed while maintaining the consistency of the description contents in the order from the requirement definition document D1 to the program D4 as the development process progresses.
FIG. 4 is a diagram showing examples of documented information ranges and difficulty levels of understanding. As documents are refined from requirement definition D1 to program D4, the range of documented information increases and undocumented information decreases. The undocumented information includes business knowledge that is commonly or implicitly shared between the orderer and the developer. In the requirement definition document D1 created at the initial stage of development, related parties such as the orderer and the developer commonly share business knowledge. For this reason, even if the requirement definition document D1 is described using business knowledge such as business terms that are not strictly defined, the orderer or developer can understand the contents. On the other hand, a description using business terms becomes a simple description, and there is one aspect that it is easy to form a common understanding between the orderer and the developer.

  In order to build a system that can be executed using a computer, it is necessary to strictly define the meaning of business knowledge. For this reason, in order to develop a program executable by a computer, it is necessary to describe information related to the system in the program D4. As a result, as in the detailed design document D3 and the program D4, the description of the text in the latter half of the development becomes detailed, and the amount of description content increases. As a result, the detailed design document D3 and the program D4 are generally more difficult to understand.

  FIG. 5 is a diagram showing an example (part 1) of a document in the V-shaped model. The business rule D11 is described in the requirement definition document D1. The business rule D11 indicates a specific business criterion related to the function of the system. Since the purpose of the requirement definition document D1 is to clarify development requirements between the orderer and the developer, for example, the requirement definition document D1 is described using business terms such as “valid code” and “within period”. In the example of the business rule D11 in FIG. 5, the case where it is described in a decision table (table format) in order to improve readability, listability, etc. is shown. May be described in a natural language such as “1”.

  The business term D12 defines the meaning of “effective code” and “within period” of the business rule D11 in the requirement definition document D1 and the basic design document D2. For example, “within period” means “between August 1, 2017 and August 31, 2017”. In this way, it may be described in a natural language, or may be described as “(20110801 <period) AND (period <20110831)” using a conditional expression like the business term D12. It can be said that the business rule D11 and the business term D12 are integrated to form knowledge about the business rule. In the original design document, the contents of the business term D12 may be incorporated in the business rule D11.

  The detailed business rule D13 is described in the basic design document D2. The business rule D13 is obtained by developing the business term D12 into a conditional expression so that the business rule D11 can be described in a program language.

  FIG. 6 is a diagram illustrating an example (part 2) of the document in the V-shaped model. The program D14 is a part of the program D4, and is described in a program language so that the detailed business rule D13 can be executed by a computer. In the program D14, description using COBOL is shown as an example of a program language.

  Here, in a system that is operated and maintained for a long period of time, partial modifications such as change or addition of functions are intermittently performed in order to cope with repair of defects or changes in business contents. When making this partial modification, it is desirable that the developer maintain consistency of all documents and programs. However, it is not easy for a developer to properly update all documents and programs while maintaining consistency, due to limitations such as renovation costs and development periods.

  In such a situation, even if a program directly related to the function of the system is modified, a document that is not directly related to the function is not sufficiently modified, and the content of the document is deficient. That is, the contents of the program modification are not reflected in the design document, and a flaw occurs between the program and the design document. As described above, if there is a flaw between the program and the design document, the information described in the design document cannot be used when the program is subsequently modified.

Therefore, the analysis server 100 provides a function for appropriately recreating a design document such as a business rule in accordance with the correction of the program.
FIG. 7 is a diagram illustrating an example of functions of the analysis server. The analysis server 100 includes a program storage unit 110, an analysis information storage unit 120, a naming rule storage unit 130, an analysis result storage unit 140, and an analysis unit 150.

  The program storage unit 110, the analysis information storage unit 120, the naming rule storage unit 130, and the analysis result storage unit 140 are realized using storage areas of the RAM 102 and the HDD 103. The analysis unit 150 is realized by the CPU 101 executing a program stored in the RAM 102.

  The program storage unit 110 stores a program 111 to be analyzed. The program 111 is a program after a certain program is corrected. For this reason, the definition content of business terms and the content of business rules in the existing design document created for the program before correction are not consistent with the program 111.

  The analysis information storage unit 120 stores various types of information used for the processing of the analysis unit 150. For example, the analysis information storage unit 120 stores a symbolic execution result 121, an input / output table 122, a business condition candidate table 123, and a business condition selection result 124. These pieces of information may be tabular data.

  The symbolic execution result 121 is information indicating the result of the symbolic execution of the program 111 by the analysis unit 150. The symbolic execution result 121 includes processing path information of the program extracted by symbolic execution.

The input / output table 122 is information indicating the relationship between input and output in the program 111 extracted from the symbolic execution result 121 by the analysis unit 150.
The business condition candidate table 123 is information indicating business condition candidates extracted from the input / output table 122 by the analysis unit 150.

The business condition selection result 124 is information indicating the business condition selected from the business condition candidate table 123 by the analysis unit 150.
The naming rule storage unit 130 stores a naming rule table 131. The naming rule table 131 is information indicating a naming rule for condition names for business conditions.

  The analysis result storage unit 140 stores the business term table 141 and the business rule table 142 created by the analysis unit 150. The business term table 141 is information indicating the definition of business terms. The business rule table 142 is information indicating the content of business rules.

  When the analysis target program 111 is input, the analysis unit 150 analyzes the program 111 and outputs a business term table 141 and a business rule table 142. The analysis unit 150 includes a symbolic execution unit 151, an input / output table generation unit 152, a business condition candidate generation unit 153, a business condition selection unit 154, a business condition naming unit 155, and an input / output table rewriting unit 156.

  The symbolic execution unit 151 extracts a processing path of the program 111 by generating a symbolic execution result 121 by executing the program 111 to be analyzed symbolically. The symbolic execution unit 151 stores the symbolic execution result 121 in the analysis information storage unit 120.

  The input / output table generation unit 152 aggregates and organizes the symbolic execution results 121 to generate an input / output table 122 indicating the relationship between the input and output of the program 111, and the input / output table 122 is stored in the analysis information storage unit 120. Is stored.

  The business condition candidate generation unit 153 generates business condition candidates from the conditions included in the input / output table 122 and registers them in the business condition candidate table 123. The business condition candidate generation unit 153 stores the business condition candidate table 123 in the analysis information storage unit 120. Here, the business conditions are a group of conditions estimated to be relatively strong in relation to the user's business (for example, frequently updated) in the program 111. The business condition candidate is a composite condition that is a combination of a plurality of conditions. Therefore, business conditions are also complex conditions. By making the composite condition a candidate for business condition, a change in the condition updated by the user (for example, addition of a condition to a variable) can be appropriately detected. For example, there are many cases in which code is changed to a complex condition by a programmer or the like due to the addition of a new function, etc. Therefore, there is a possibility that the update contents of the code can be detected appropriately by detecting the complex condition Rise.

  The business condition selection unit 154 uses, as a business condition, a candidate for which a negative condition of the candidate exists among the plurality of candidates among the plurality of business condition candidates registered in the business condition candidate table 123. The business condition selection result 124 is generated. The business condition selection unit 154 stores the business condition selection result 124 in the analysis information storage unit 120.

  The business condition naming unit 155 names a condition name based on the naming rules in the naming rule table 131 for the business conditions included in the business condition selection result 124, and generates a business term table 141 indicating the naming result. The business condition naming unit 155 stores the business term table 141 in the analysis result storage unit 140.

  The input / output table rewriting unit 156 generates the business rule table 142 by replacing the description of the conditions included in the input / output table 122 with the business conditions named by the business condition naming unit 155. The input / output table rewriting unit 156 stores the business rule table 142 in the analysis result storage unit 140.

  FIG. 8 is a diagram illustrating an example of symbolic execution. One of the program analysis methods is symbolic execution (symbol execution). In symbolic execution, a variable that stores an input value to a program is set with a symbol (symbol value) indicating a constant instead of a specific numerical value or a specific character, and the executable path of the program is extracted. This is a technique for analyzing programs. Regarding symbolic execution, for example, the following documents can be referred to.

(Reference) Tetsuo Tamai, Koichi Fukunaga, “Symbolic Execution System”, Information Processing, pp. 18-28, January 15, 1982 Next, a program analysis example by symbolic execution will be described.

  The program E1 is an example of a program to be analyzed. However, in FIG. 8, the program E1 is illustrated in the form of a flowchart instead of a description in a program language so that the logic can be easily understood.

  The program E1 receives a two-character string as the value of the variable “gender” and outputs one character as the value of the variable “result”. The program E1 includes two conditional branches. In the first conditional branch, it is determined whether the value of the variable “gender” is “male”. If the value of the variable “gender” is “male”, the value of the variable “result” is updated to “M”; otherwise, the value of the variable “result” is updated to “” (blank). . In the second conditional branch, it is determined whether the value of the variable “gender” is “female”. If the value of the variable “gender” is “female”, the value of the variable “result” is updated to “F”; otherwise, the value of the variable “result” does not change. Then, the program E1 ends.

  The program E1 has three paths A, B, and C. Any one of these three passes is executed according to the value of the variable “gender”. In path A, the value of the variable “gender” is “male”. In this case, the first branch condition is determined as YES, and the second branch condition is determined as NO. In the path A, the value of the variable “result” is “M”.

  Path B is a case where the value of the variable “gender” is “female”. In path B, the first branch condition is determined as NO, and the second branch condition is determined as YES. In the path B, the value of the variable “result” is “F”.

  Path C is a case where the value of the variable “gender” is neither “male” nor “female”. In path C, it is determined that both the first branch condition and the second branch condition are NO. In this execution path, the value of the variable “result” is “” (blank).

  Since the program E1 includes two serial conditional branch statements, there are four execution path candidates if the contents of the branch condition are ignored. However, since the value of the variable “gender” is “male” and the value of the variable “sex” is “female”, both of the two branch conditions are not determined to be YES. Therefore, of the four execution path candidates, there are three effective execution paths that may be executed.

  In a conditional branch sentence, when there is at least one variable value that satisfies a condition for proceeding in a certain branch direction (YES direction or NO direction), the branch direction may be said to be “satisfiable”. On the other hand, when there is no value of a variable that satisfies the condition for proceeding in a certain branch direction, the branch direction may be said to be “not satisfiable”.

  In the program E1, since the value of the variable “gender” is not restricted in the first conditional branch, both the two branch directions can be satisfied. On the other hand, the satisfiability in the second conditional branch depends on the branch direction selected in the first conditional branch. When the YES direction is selected in the first conditional branch, the YES direction of the second conditional branch is no longer satisfactory and only the NO direction can be satisfied. When the NO direction is selected in the first conditional branch, both the branch directions of the second conditional branch can be satisfied. Therefore, it can be determined that there are three effective execution paths.

  Here, an example in which the program E1 is symbolically executed will be described. The symbolic execution unit 151 assigns an abstract symbol (symbol value) indicating an arbitrary constant instead of a specific character to the variable “gender” indicating the input to the program E1. Here, the same “sex” as the variable name is used as a symbol. The symbolic execution unit 151 executes the instructions of the program E1 in a pseudo manner using symbols.

  There is no path branch from the start point of the program E1 to before the first conditional branch. For the first conditional branch, there is no condition (path condition) for selecting the path at this point, and both the two branch directions can be satisfied. Therefore, the symbolic execution unit 151 divides the path in the first conditional branch and selects one path. Here, it is assumed that the symbolic execution unit 151 selects a path in the YES direction.

  Next, for the second conditional branch, since the path condition at this point is that the symbol “sex” is “male”, the YES direction cannot be satisfied, and the NO direction can be satisfied. Therefore, the symbolic execution unit 151 selects the NO direction. Thereafter, the first path (path A) ends. The symbolic execution unit 151 determines that the symbol “sex” is “male” and the symbol “sex” is not “female” for the path condition of the first pass. Also, the symbolic execution unit 151 determines that the value of the variable “result” is “M” for the processing result (path output) of the first pass. The symbolic execution unit 151 registers these determination results in the symbolic execution result E2 (corresponding to the record of the item number “1”).

  When the first path ends, the symbolic execution unit 151 returns to the nearest division point of the path and selects an unselected path. Here, the symbolic execution unit 151 returns to the first conditional branch and selects a path in the NO direction.

  Regarding the second conditional branch, the path condition at this point is that the symbol “sex” is not “male”, so that both the YES direction and the NO direction can be satisfied. Therefore, the symbolic execution unit 151 further divides the path in the second conditional branch and selects one path. Here, it is assumed that the symbolic execution unit 151 selects an execution path in the YES direction. Thereafter, the second path (path B) ends. The symbolic execution unit 151 determines that the path condition of the second pass is that the symbol “sex” is not “male” and the symbol “sex” is “female”. Further, the symbolic execution unit 151 determines that the value of the variable “result” is “F” for the path output of the second path. The symbolic execution unit 151 registers these determination results in the symbolic execution result E2 (corresponding to the record of item number “2”).

  When the second path ends, the symbolic execution unit 151 returns to the second conditional branch and selects the execution path in the NO direction. Thereafter, the third path (path C) ends. The symbolic execution unit 151 determines that the symbol condition of the third pass is not “male” and the symbol “sex” is not “female”. Further, the symbolic execution unit 151 determines that the value of the variable “result” is “” (blank) for the path output of the third path. The symbolic execution unit 151 registers these determination results in the symbolic execution result E2 (corresponding to the record of the item number “3”). Then, since there is no unselected execution path, the symbolic execution ends.

  The symbolic execution unit 151 stores the path condition and the value of each variable at that point at the path division point. The symbolic execution unit 151 takes over the saved path condition and the value of each variable when one path is finished and backtracks to the division point, and the other path corresponding to the unselected branch direction is taken over. Continue searching.

  Here, the pass condition of the first pass is “gender is male” and “gender is not female”. This can be simplified to the condition “gender is male”. The pass condition of the second execution pass is “gender is not male” and “gender is female”. This can be simplified to the condition that “sex is female”. The pass condition of the third pass is “gender is not male” and “gender is not female”. In this manner, the input / output table generation unit 152 generates the input / output table E3 related to the program E1 by performing simplification or the like on the symbolic execution result E2 and organizing it. The input / output table E3 is information indicating a correspondence relationship between a condition and an output corresponding to an input related to the program E1.

  It should be noted that existing mathematical expression processing techniques can be used to determine satisfiability in conditional branch sentences and to simplify path conditions. For example, the symbolic execution unit 151 can use an SMT (Satisfiability Modulo Theories) solver, an SAT (Satisfiability) solver, or the like to determine satisfiability. Further, the input / output table generation unit 152 can use a Quine-McCluskey (QM) method or the like for simplification of the path condition. According to the QM method, for example, with respect to the logical expressions A, B, and C, the logical expression is simplified as (A∧B∧C) ∨ (A∧B∧¬C) = A∧B. Can do. Alternatively, the input / output table generation unit 152 can simplify the path condition by the following procedure.

  (Procedure 1) The conditional expression of the path condition is expanded to a disjunctive normal form (DNF: Disjunctive Normal Form). The expanded conditional expression is a disjunction of a conjunctive clause (referred to as a DNF term). That is, the format is “conditional expression = DNF term (1) （DNF term (2) ∨. As a specific example, the conditional expression is “((X> 1) ∧ (Y <0)) ∨ ((X> 1) ∧ (Y> 10)) ∨ ((X> 1) ∧ (Y NOT <0 ) ∧ (Y NOT> 10)) ”. In this case, the DNF terms are ((X> 1) ∧ (Y <0)), ((X> 1) ∧ (Y> 10)), ((X> 1) ∧ (Y NOT <0) ∧ ( Y NOT> 10)).

  (Procedure 2) From the conditional expression, a comparison conditional expression having only a comparison operator having no NOT (negative operator) is extracted as an atomic comparison conditional expression. The atomic comparison conditional expressions extracted for the above conditional expressions are (X> 1), (Y <0), and (Y> 10).

  (Procedure 3) With the combination of disjunctions in the DNF term, the atomic comparison conditional expression and its negative conjunction are factored as factor condition candidates. When the above conditional expression is factored with (X> 1) as a factor condition, the conditional expression “(X> 1) ∧ ((Y <0) ∨ (Y> 10) ∨ ((Y NOT <0) ∧ ( Y NOT> 10))) ”.

  (Procedure 4) It is determined whether another factor condition that is said to be the above factor condition is true, and if it is true, the fact that the other factor condition is true is substituted to select the DNF term. Simplify the word. For the example of (Procedure 3), the other factor condition that is referred to as the factor condition (X> 1) is “(Y <0) ∨ (Y> 10) ∨ ((Y NOT <0) ∧ (Y NOT> 10)) = TRUE ”, and the other factor conditions are true (always true regardless of the value of Y). Therefore, by substituting the result of (Procedure 4) into the conditional expression obtained in (Procedure 3), the conditional expression shown in (Procedure 1) is simplified as (X> 1).

  FIG. 9 is a diagram illustrating an example of a program to be analyzed. The program 111 is an example of a program after correction with respect to the above-described program D14. Here, each line included in the program 111 is indicated by a line number given to the left side of the program 111 for convenience. Further, in FIG. 9, correction points for the program D14 in FIG. 6 are enclosed by a rectangle for emphasis.

  For example, the definition of the constant C2 is added to the 12th line of the program 111. Further, on the 26th to 34th lines of the program 111, a condition determination regarding the variable “code” using the constant C2 and a process according to the condition determination are added.

  FIG. 10 is a diagram illustrating an example of a symbolic execution result. The symbolic execution result 121 is a result of symbolic execution of the program 111. The symbolic execution result 121 includes items of item number, path output, and path condition.

  In the item number item, a record identification number is registered. In the path output item, output at the corresponding path is registered. In the path condition item, a condition for executing the corresponding path is registered.

  For example, in the symbolic execution result 121, the item number is “1”, the path output is “result = 1”, and the path condition is “((code = 11) AND (20110801 <period) AND (period <20110831))”. A record is registered. This record indicates that the output of the program 111 is “result = 1” when the conditions of “code = 11”, “20110801 <period”, and “period <20110831” are satisfied.

In the symbolic execution result 121, a path output and a path condition are registered for each path corresponding to a combination of branch destinations of conditional branches included in the program 111.
FIG. 11 is a diagram illustrating an example of an input / output table. The input / output table 122 includes items of item number, output, and condition. In the item number item, a record identification number is registered. In the output item, a path output value in the symbolic execution result 121 is registered. However, in the input / output table 122, when there are a plurality of records having the same path output value in the symbolic execution result 121, a plurality of conditions included in the plurality of records are integrated into one. In the condition item, a result obtained by transforming the path condition for each path output in the symbolic execution result 121 into a multiplicative normal form (CNF: Conjunctive Normal Form) is registered. In CNF, conditional expressions are expressed in the form of disjunctive conjunctions. The input / output table generation unit 152 simplifies the path condition of the symbolic execution result 121 for each path output having the same value, and converts the reduced result into CNF, so that the condition contents in the input / output table 122 are changed. Ask. Existing methods (for example, Domorgan's law, distribution law, and Tseitin transformation) can be used for transformation to CNF.

  For example, in the input / output table 122, the item number is “1”, the output is “result = 1”, and the condition is “((code = 11) OR (code = 22)) AND ((2011081 <period) AND (period <20110831)) "is registered. In this record, when the “code” is “11” or “22”, the condition of “20110801 <period” and “period <20110831” is satisfied, the output of the program 111 is “result = 1”. Indicates that

  In the input / output table 122, records corresponding to the outputs “result = 2” and “result = 3” in the symbolic execution result 121 are also registered. “Code = 11”, “Code = 22”, “2011081 <period”, “period <20110831”, and the like included in the conditions of the input / output table 122 can be referred to as conditional statements.

  FIG. 12 is a diagram illustrating an example of intermediate data for extracting business condition candidates. The intermediate data 122a is intermediate data for extracting business condition candidates. The intermediate data 122 a is generated by the business condition candidate generation unit 153 based on the input / output table 122 and stored in the analysis information storage unit 120. The intermediate data 122a includes items of condition number, candidate condition, negative relationship, and positive condition.

  In the condition number item, an identification number of a condition (referred to as a candidate condition) from which a business condition candidate is extracted is registered. The content of the candidate condition is registered in the item of candidate condition. The contents of the candidate conditions are generated by the business condition candidate generation unit 153 using a predetermined calculation based on the input / output table 122. In the negative relationship item, the condition numbers of other candidate conditions having a negative relationship with the candidate condition of the corresponding record are registered. If there is no other candidate condition in a negative relationship, the item in the negative relationship is not set. Here, no setting is indicated by a hyphen symbol “-”. In the item of the positive condition, when there is another candidate condition having a negative relationship, identification information indicating whether the positive condition is satisfied is registered. If applicable, “Applicable” is registered in the item of the correct condition. When not applicable (when there is no negative relationship or when there is no positive condition), the item of the positive condition is not set ("-"). Here, the positive condition is a candidate condition having a smaller number of negation operators among two candidate conditions having a negative relationship with each other. If there is no other candidate condition in a negative relationship, the positive condition item is always set.

  For example, in the intermediate data 122a, the condition number is “condition 1”, the candidate condition is “((code = 11) OR (code = 22)) AND (20110801 <period) AND (period <20110831)”, and the negative relation is A record with “−” and a positive condition “−” is registered. This record indicates the contents of the candidate condition with the condition number “condition 1”, and indicates that the candidate condition has no other candidate condition having a negative relationship.

In the example of the intermediate data 122a, the candidate conditions with the condition numbers “condition 2”, “condition 3”, and “condition 6” do not have other candidate conditions that are similarly negative.
Further, in the intermediate data 122a, the condition number is “condition 4”, the candidate condition is “((20110801 <period) AND (period <20110831))”, the negative relation is “condition 7”, and the positive condition is “corresponding”. A record is registered. This record indicates the contents of the candidate condition with the condition number “condition 4”. Further, in this record, the candidate condition with the condition number “condition 4” includes other candidate conditions with the condition number “condition 7” having a negative relationship, and candidates with the condition numbers “condition 4” and “condition 7”. Among the conditions, it indicates that the candidate condition with the condition number “condition 4” corresponds to the positive condition.

In the example of the intermediate data 122a, the two candidate conditions with the condition numbers “condition 5” and “condition 8” are also in a negative relationship with each other, and the candidate condition with the condition number “condition 5” is the positive condition.
FIG. 13 is a diagram illustrating an example of the business condition candidate table. The business condition candidate table 123 is generated by the business condition candidate generation unit 153 based on the intermediate data 122a. The business condition candidate generation unit 153 sorts the records in the order in which the conditions in the candidate conditions of the intermediate data 122a are long (in the order of many nodes) and in the order of few negative operators. The result of this sorting is the business condition candidate table 123 (candidate conditions in the business condition candidate table 123 are referred to as business condition candidates). The business condition candidate table 123 includes items of item number, condition number, business condition candidate, negative relationship, and positive condition.

  In the item number item, a record identification number is registered. In the condition number item, an identification number of a candidate condition is registered. In the business condition candidate item, the contents of the business condition candidate are registered. In the item of negative relation, the condition numbers of other business condition candidates having a negative relation with the business condition candidate of the corresponding record are registered. The setting rule for the negative relationship item is the same as that of the intermediate data 122a. In the item of the positive condition, when there is another business condition candidate in a negative relationship, identification information indicating whether or not the positive condition is satisfied is registered. The setting rule for the item of the correct condition is the same as that of the intermediate data 122a.

  The business condition candidate generation unit 153 is the sorting result of the intermediate data 122a as described above. For this reason, the specific setting content of each item is the same as that of the intermediate data 122a. However, in the item number item, numbers are set in ascending order from the top to the bottom of each record after sorting.

  FIG. 14 is a diagram illustrating an example of intermediate data for selecting business conditions. The intermediate data 123a is intermediate data for selecting business conditions. The intermediate data 123 a is generated by the business condition selection unit 154 based on the business condition candidate table 123 and stored in the analysis information storage unit 120. The intermediate data 123a includes items of item number, positive condition input / output table item number, negative input / output table item number, and business condition determination.

  In the item number item, a record identification number is registered. The item of the input / output table item number of the normal condition includes the location of the job condition candidate corresponding to the corresponding item number of the job condition candidate table 123 in the input / output table 122 (the record item number in the input / output table 122, or , Information indicating the combination of the item number and the position in the condition) is registered. The positive condition indicates the same condition as the business condition candidate. When the corresponding business condition candidate matches a part of the condition of a record in the input / output table 122, the item number in the input / output table 122 and the position corresponding to the corresponding business condition candidate among the conditions corresponding to the item number Is registered in the item of the input / output table item number of the positive condition.

  The item of the negative input / output table item number exists in the input / output table 122 of the negative of the business condition candidate corresponding to the corresponding item number of the business condition candidate table 123 (that is, the negative condition indicating the negative of the business condition candidate). Information indicating the location is registered. If the denial of the relevant business condition candidate matches a part of the condition of a record in the input / output table 122, the relevant business condition candidate is negated from the item number of the input / output table 122 and the condition corresponding to the item number. And the information indicating the position that matches are registered in the negative I / O table item number item.

  In the business condition determination item, identification information indicating whether or not the business condition is selected is registered. When the business condition is selected, “true” is set in the business condition determination item. If it is not selected as the business condition, “false” is set in the business condition determination item. Here, of the two business condition candidates that are in a negative relationship with each other, the positive condition is selected as the business condition. For this reason, business condition candidates corresponding to the item numbers “6” and “8” of the intermediate data 123a are excluded from the selection of business conditions. Therefore, information other than the item number is not set in the business condition candidate records corresponding to the item numbers “6” and “8” (other than the item number, “−” is not set).

  Note that the input / output table item number of the positive condition and the negative input / output item number are set when the business condition determination is “true” and the set value is registered, and when the business condition determination is “false”. None.

  For example, a record in which the item number is “1”, the positive input / output table item number and the negative input / output table item number are not set “−”, and the business condition determination is “false” is registered in the intermediate data 123a. The This record indicates that the business condition candidate of item number “1” in the business condition candidate table 123 has not been selected as the business condition (the same applies to records of item numbers “2” to “4”).

  In the intermediate data 123a, the item number is “5”, the positive input / output table item number is “1 right end”, the negative input / output table item number is “2 right end”, and the business condition determination is “true”. "Is registered. This record indicates that the same condition as the job condition candidate of item number “5” in the job condition candidate table 123 is in the rightmost node of the condition of item number “1” in the input / output table 122. Further, the condition indicating that the business condition candidate of the item number “5” is negative is the rightmost node of the condition of the item number “2” in the input / output table 122. Further, it indicates that the business condition candidate of the item number “5” is selected as the business condition.

  Further, in the intermediate data 123a, the item number is “7”, the input / output table item number of the positive condition is “left end of 1, 2 left end”, the input / output table item number of negative is “3”, and the business condition determination is performed. A record “true” is registered. This record has the same condition as the job condition candidate of item number “7” in the job condition candidate table 123, the leftmost clause of the item number “1” in the input / output table 122, and the item number “2” in the input / output table 122. Indicates that it is in the leftmost node. In addition, the condition indicating that the business condition candidate of the item number “7” is negative is the condition of the item number “3” in the input / output table 122. Further, it indicates that the business condition candidate of the item number “7” is selected as the business condition.

  FIG. 15 is a diagram illustrating an example of a business condition selection result. The business condition selection result 124 is generated based on the intermediate data 123 a by the business condition selection unit 154. The business condition selection result 124 includes items of item number, business condition, appearance place of the positive condition, and appearance place of the negative condition.

  In the item number item, a record identification number is registered. The contents of the business condition are registered in the business condition item. In the item of the appearance location of the normal condition, the position (information indicating the item number of the record and the position within the condition of the record) of the input / output table 122 where the business condition appears is registered. In the item of the appearance condition of the negative condition, the position (information indicating the item number of the record and the position within the condition of the record) of the input / output table 122 where the negative of the business condition appears is registered.

  For example, in the business condition selection result 124, the item number is “1”, the business condition is “((20110801 <period) AND (period <20110831))”, and the place where the positive condition appears is “1) I / O table condition 1 "Right end of", and the appearance of the negative condition is registered as "1) Right end of input / output table condition 2". This record indicates the business condition of item number “1”, and indicates that the business condition corresponds to the rightmost clause of the condition of item number “1” in the input / output table 122. Further, it is indicated that the condition indicating the denial of the business condition corresponds to the rightmost node of the condition of item number “2” in the input / output table 122.

  In the business condition selection result 124, the item number is “2”, the business condition is “((code = 11) OR (code = 22)”, and the place where the positive condition appears is “1”. Records of the left end, 2) the left end of input / output table condition 2, and the appearance location of the negative condition are “1) input / output table condition 3” are registered. This record indicates the business condition of item number “2” and indicates that the business condition appears in the leftmost section of the condition of item number “1” in the input / output table 122. It also indicates that the business condition appears in the leftmost section of the condition of item number “2” in the input / output table 122. In addition, the condition indicating that the business condition is negative corresponds to the condition of the item number “3” in the input / output table 122.

  FIG. 16 is a diagram illustrating an example of a naming rule table. The naming rule table 131 includes items of item numbers and naming rules. In the item number item, a record identification number is registered. In the naming rule item, a business term naming rule for business conditions is registered.

  For example, in the naming rule table 131, when the item number is “1” and the naming rule is “business condition matches“ (lower limit <variable) AND (variable <upper limit) ”, it is named“ inside variable ”. This record defines the business condition as “in variable” when the format of the business condition related to “variable” matches the format of the naming rule of item number “1”. Indicates. Here, the actual variable name is inserted into the “variable” portion of the character string “inside variable”.

  In the naming rule table 131, if the item number matches “2” and the naming rule “business condition is“ (variable = constant 1) [OR (variable = constant 2)] + ””, “effective variable” is indicated. Naming. This record defines the business condition as “valid variable” when the format of the business condition related to “variable” matches the format of the naming rule of item number “2”. Indicates. The “[OR (variable = constant 2)] +” part of the naming rule indicates that the conditions for the second and subsequent constants can be combined with OR. Here, the actual variable name is inserted in the “variable” portion of the character string “valid variable”.

  FIG. 17 is a diagram illustrating an example of a business term table. The business term table 141 is generated by the business condition naming unit 155 based on the business condition selection result 124 and the naming rule table 131. The business term table 141 includes items of item number, business term, and condition.

  In the item number item, a record identification number is registered. Named business terms are registered in the business term field. In the condition item, a condition (business condition) corresponding to the business term is registered.

  For example, in the business term table 141, a record is registered in which the item number is “1”, the business term is “within period”, and the condition is “(20110801 <period) AND (period <20110831)”. This record indicates that the business term “within period” is named for the condition of item number “1”. The business term “within period” is a result of applying the naming rule of item number “1” in the naming rule table 131 to the condition of item number “1”.

  In the business term table 141, a record with an item number “2”, a business term “valid code”, and a condition “(code = 11) OR (code = 22)” is registered. This record indicates that the business term “valid code” is named for the condition of item number “2”. The business term “valid code” is a result of applying the naming rule of item number “2” in the naming rule table 131 to the condition of item number “2”.

FIG. 18 is a diagram illustrating an example of a business rule table. FIG. 18A shows the business rule table 142 in the first format. FIG. 18B shows a business rule table 143 in the second format.
The business rule table 142 is obtained by replacing the contents of the conditions in the input / output table 122 with business terms defined in the business term table 141. Specifically, each section included in the condition of the input / output table 122 is replaced with the positive form and negative form of the business condition. In the normal format, the format is “business term = True”, and in the negative format, the format is “business term = False”.

  The business rule table 143 is obtained by converting the contents of the conditions of the business rule table 142 into a format called a decision table. In the business rule table 143, the condition column in the business rule table 142 is divided into columns for each business term, where “business term = True” in the business rule table 142 is “T” and “business term = False”. Each part is replaced with “F”.

The input / output table rewriting unit 156 generates a business rule table 142 and a business rule table 143 based on the input / output table 122 and stores them in the analysis result storage unit 140.
Next, a processing procedure by the analysis server 100 will be described.

FIG. 19 is a flowchart illustrating a processing example of the analysis server. When the analysis unit 150 receives an input to start analysis of the program 111 by the user, the analysis unit 150 executes the following procedure.
(S1) The symbolic execution unit 151 extracts the processing path of the program 111 and generates a symbolic execution result 121 by executing the analysis target program 111 stored in the program storage unit 110 symbolically. The symbolic execution unit 151 stores the symbolic execution result 121 in the analysis information storage unit 120.

  (S2) The input / output table generation unit 152 generates the input / output table 122 based on the symbolic execution result 121 stored in the analysis information storage unit 120. The input / output table generation unit 152 stores the input / output table 122 in the analysis information storage unit 120.

  (S3) The business condition candidate generation unit 153 generates intermediate data 122a based on the input / output table 122 stored in the analysis information storage unit 120. The business condition candidate generation unit 153 stores the intermediate data 122a in the analysis information storage unit 120. The business condition candidate generation unit 153 generates the business condition candidate table 123 based on the intermediate data 122 a stored in the analysis information storage unit 120. The business condition candidate generation unit 153 stores the business condition candidate table 123 in the analysis information storage unit 120.

  (S4) The business condition selection unit 154 generates intermediate data 123a based on the business condition candidate table 123 stored in the analysis information storage unit 120. The business condition selection unit 154 stores the intermediate data 123a in the analysis information storage unit 120. The business condition selection unit 154 generates a business condition selection result 124 based on the input / output table 122 and the intermediate data 123a stored in the analysis information storage unit 120. The business condition selection unit 154 stores the business condition selection result 124 in the analysis information storage unit 120.

  (S5) The business condition naming unit 155 generates a business term table 141 based on the business condition selection result 124 stored in the analysis information storage unit 120 and the naming rule table 131 stored in the naming rule storage unit 130. To do. The business condition naming unit 155 stores the business term table 141 in the analysis result storage unit 140.

  (S6) The input / output table rewriting unit 156 stores the business rule tables 142 and 143 based on the input / output table 122 stored in the analysis information storage unit 120 and the business term table 141 stored in the analysis result storage unit 140. Generate. The input / output table rewriting unit 156 stores the business rule tables 142 and 143 in the analysis result storage unit 140. And the process of the analysis server 100 is complete | finished.

Next, the procedures of steps S1 to S6 will be described.
FIG. 20 is a flowchart illustrating a processing example of symbolic execution. The symbolic execution process is executed in step S1 of FIG.

(S10) The symbolic execution unit 151 acquires the analysis target program 111 stored in the program storage unit 110.
(S11) The symbolic execution unit 151 performs symbolic execution of the program 111. As illustrated in FIG. 8, the symbolic execution unit 151 sets a symbol value to an input variable of the program 111 and extracts a process path that can be executed by the program 111. Get the relationship.

  (S12) The symbolic execution unit 151 records the path condition and path output of the corresponding processing path in the symbolic execution result 121 at the end point of each processing path. The symbolic execution unit 151 stores the symbolic execution result 121 in the analysis information storage unit 120. Then, the symbolic execution ends.

FIG. 21 is a flowchart illustrating a processing example of input / output table generation. The input / output table generation process is executed in step S2 of FIG.
(S20) The input / output table generating unit 152 sets a pair of path outputs and path conditions (Out _n , Cnd _n ) (from N to N (where N is the total number of records in the symbolic execution result 121)). n = 1,..., N).

(S21) The input / output table generation unit 152 substitutes 1 for n (n = 1).
(S22) The input / output table generator 152 determines whether n is N or less (whether n ≦ N). If n is N or less, the process proceeds to step S23. If n is greater than N, the process proceeds to step S25.

(S23) The input / output table generation unit 152 extracts (Out _n , Cnd _n ), and combines Cnd _n with OR (logical sum) for each value of Out _n .
(S24) The input / output table generator 152 adds 1 to n (n = n + 1). Then, the process proceeds to step S22.

(S25) The input / output table generator 152 simplifies the conditions for each output value. As illustrated in FIG. 8, the input / output table generation unit 152 can simplify the conditions using an existing technique.
(S26) The input / output table generator 152 generates the input / output table 122 by changing the condition for each output value after the reduction in step S25 to CNF. An existing technique can be used for transformation to CNF. For example, a method using Domorgan's law, distribution law, Tseitin transformation, or the like is known for transformation to CNF.

  (S27) The input / output table generator 152 outputs the generated input / output table 122. The input / output table generation unit 152 stores the input / output table 122 in the analysis information storage unit 120. Then, the input / output table generation ends.

  Here, the result of step S23 with respect to the symbolic execution result 121 illustrated in FIG. 10 will be specifically described. By repeatedly executing step S23 on the symbolic execution result 121, two path conditions corresponding to the path output “result = 1” are combined by OR. More specifically, the input / output table generating unit 152 performs “((20170801 <period)) AND (code = 11) AND (period <20170831)) OR ((20110801 <period) AND” with respect to “result = 1”. (Code = 22) AND (period <20110831)) ”.

  Similarly, four path conditions corresponding to the path output “result = 2” are combined by OR. More specifically, the input / output table generation unit 152 performs “((20170801 NOT <period) AND (code = 11)) OR ((20110801 NOT <period) AND (code = 22) for“ result = 2 ”. )) OR ((code = 11) AND (period NOT <20110831)) OR ((code = 22) AND (period NOT <20110831)) ".

  Similarly, one path condition corresponding to the path output “result = 3” is combined by OR. More specifically, the input / output table generation unit 152 obtains “(code NOT = 11) AND (code NOT = 22)” for “result = 3”.

In the case of this example, the OR combination results of the path conditions for each output value are already simple, so even if simplification is performed in step S25, the original conditions are not changed.
As described above, the input / output table generation unit 152 creates the input / output table 122 based on the result of the symbolic execution of the program 111. The result of symbolic execution includes a plurality of conditions (path conditions) relating to input variables for each of a plurality of outputs (path outputs) by the program 111. In the example of the symbolic execution result 121, there are two path conditions for the path output “result = 1”, and there are four path conditions for the path output “result = 2”. The input / output table generating unit 152 creates the input / output table 122 by executing, for each output, processing for converting a plurality of conditions for one output into one branch condition (condition in the input / output table 122). In the example of the input / output table 122, two path conditions for “result = 1” are converted into one condition (branch condition). Similarly, four path conditions for “result = 2” are converted into one condition (branch condition). Since “result = 3” has only one path condition, it is registered in the input / output table 122 as it is.

FIG. 22 is a flowchart illustrating a processing example for generating business condition candidates. The business condition candidate generation process is executed in step S3 of FIG.
(S30) The business condition candidate generation unit 153 obtains a set of outputs and conditions (Out _n , Cnd _n ) (n = 1,..., N) from the input / output table 122 (N is the input / output table 122). Total number of records).

(S31) The business condition candidate generation unit 153 substitutes 1 for n (n = 1).
(S32) The business condition candidate generation unit 153 determines whether n is N or less (whether n ≦ N). If n is N or less, the process proceeds to step S33. If n is greater than N, the process proceeds to step S41.

  (S33) The business condition candidate generation unit 153 generates a condition (candidate condition) in which a combination of disjunction clauses of the condition represented by CNF in the input / output table 122 is linked. A specific example for the input / output table 122 will be described later.

(S34) The business condition candidate generation unit 153 selects one unprocessed candidate condition generated for n.
(S35) The business condition candidate generation unit 153 determines whether the candidate condition selected in step S34 does not exist in the business condition candidate list (that is, the intermediate data 122a). If not, the process proceeds to step S36. If it exists, the process proceeds to step S39.

(S36) The business condition candidate generation unit 153 adds the candidate condition selected in step S34 to the business condition candidate list (intermediate data 122a).
(S37) The business condition candidate generation unit 153 determines whether the candidate condition added in step S36 is a negative of the existing candidate condition. When the added candidate condition is negative of the existing candidate condition, the process proceeds to step S38. If the added candidate condition is not a negative of the existing candidate condition, the process proceeds to step S39.

  (S38) The business condition candidate generation unit 153 records a negative relationship regarding the candidate condition added in step S36 and other candidate conditions that are negatively related to the candidate condition. In other words, the business condition candidate generation unit 153 sets identification information of other candidate conditions for the candidate condition added this time in the business condition candidate list (intermediate data 122a), and also sets the current condition for other candidate conditions. Set the identification information of the added candidate condition.

  (S39) The business condition candidate generation unit 153 determines whether all the candidate conditions generated in step S33 have been processed for n. If all have been processed, the process proceeds to step S40. If all have not been processed, the process proceeds to step S34.

(S40) The business condition candidate generation unit 153 adds 1 to n (n = n + 1). Then, the process proceeds to step S32.
(S41) The business condition candidate generation unit 153 refers to the business condition candidate list (intermediate data 122a), and sets an identifier indicating the positive condition to the lesser of the negative operators, out of two candidate conditions having a negative relationship. To do.

  (S42) The business condition candidate generation unit 153 sorts the candidate conditions in the business condition candidate list (intermediate data 122a) into the business condition candidate table 123, and outputs the business condition candidate table 123. The business condition candidate generation unit 153 stores the business condition candidate table 123 in the analysis information storage unit 120. The business condition candidate generation unit 153 sorts the candidate conditions according to a predetermined rule. For example, the business condition candidate generation unit 153 sorts in order from the longest condition included in the candidate condition to the smallest negative condition. Then, the business condition candidate generation ends.

  Here, the results of steps S33 to S39 for the input / output table 122 illustrated in FIG. 11 will be specifically described. By executing step S33 for the input / output table 122, the next candidate condition is generated for the output “result = 1”. The set of CNF disjunctive clauses for the output “result = 1” is {((code = 11) OR (code = 22)), (20110801 <period), (period <20110831)}. There are the following five candidate conditions generated by combinations of these CNF disjunctive clauses (combinations by conjunction).

(Condition 1) ((Code = 11) OR (Code = 22)) AND (20110801 <period) AND (period <20110831)
(Condition 2) ((Code = 11) OR (Code = 22)) AND (20110801 <Period)
(Condition 3) ((Code = 11) OR (Code = 22)) AND (Period <20110831)
(Condition 4) ((20110801 <period) AND (period <20110831))
(Condition 5) ((Code = 11) OR (Code = 22))
In step S36, since the above conditions 1 to 5 are not existing conditions, they are added to the intermediate data 122a. In addition, since the conditions 1 to 5 are not existing negations in step S38, the processing of the condition of the output “result = 1” ends.

  Also, according to the input / output table 122, the set of CNF disjunctive clauses for the output “result = 2” is {((code = 11) OR (code = 22)), ((20110801 NOT <period) OR ( Period NOT <20110831))}. There are the following three candidate conditions generated by a combination of these CNF disjunctive clauses.

(Condition 6) ((Code = 11) OR (Code = 22)) AND (20110801 NOT <Period) OR (Period NOT <20110831)
(Condition 7) ((20110801 NOT <period) OR (period NOT <20110831))
(Condition X) ((Code = 11) OR (Code = 22))
In Step S36, the conditions 6 and 7 are not existing conditions and are added to the intermediate data 122a. However, since the condition X is existing (matches the condition 5), it is not added to the intermediate data 122a. In step S38, since condition 7 is negative of condition 4, identification information (identification information of the other party's condition) indicating that there is a negative relationship with each other is set in intermediate data 122a. Then, the processing for the condition of output “result = 2” ends.

  Furthermore, according to the input / output table 122, the condition relating to the output “result = 3” is only one of {((code NOT = 11) AND (code NOT = 22))}. Accordingly, the candidate condition generated in step S33 is one of the following.

(Condition 8) ((Code NOT = 11) AND (Code NOT = 22))
In step S36, condition 8 is not an existing condition and is added to the intermediate data 122a. In step S38, since condition 8 is negative of condition 5, identification information (identification information of the other party's condition) indicating that they are in a negative relationship with each other is set in intermediate data 122a. Then, the processing for the condition of output “result = 3” ends.

  Since all the output conditions in the input / output table 122 have been processed as described above, the business condition candidate generation unit 153 executes step S41. Then, the business condition candidate generation unit 153 compares the condition 4 with the condition 7, sets an identifier indicating the correct condition in the condition 4, compares the condition 5 with the condition 8, and indicates an identifier indicating the correct condition in the condition 5. Set. Then, the business condition candidate generating unit 153 obtains the business condition candidate table 123 by sorting the intermediate data 122a in step S42.

FIG. 23 is a flowchart illustrating an example of processing for selecting business conditions. The business condition selection process is executed in step S4 of FIG.
(S50) The business condition selection unit 154 acquires a set (Out _m , Cnd _m ) (m = 1,..., M) of the output and condition of the input / output table 122 (M is the input / output table 122). Total number of records). m corresponds to the item number in the input / output table 122.

(S51) The business condition selection unit 154 acquires conditions BzCnd _n , n = 1,..., N (N is the total number of records in the business condition candidate table 123) in the business condition candidate table 123. n corresponds to an item number in the business condition candidate table 123. The business condition selection unit 154 generates intermediate data 123a based on the business condition candidate table 123 (at this stage, only the item number values corresponding to the record of each item number in the business condition candidate table 123 are set. A record is registered in the intermediate data 123a).

(S52) The business condition selection unit 154 substitutes 1 for n (n = 1).
(S53) The business condition selection unit 154 determines whether n is N or less (whether n ≦ N). If n is N or less, the process proceeds to step S54. If n is larger than N, the process proceeds to step S65.

(S54) The business condition selection unit 154 determines whether BzCnd _n has been processed. If not, the process proceeds to step S55. If it has been processed, the process proceeds to step S58.

(S55) The business condition selection unit 154 determines whether BzCnd _n has a negative relationship based on the business condition candidate table 123. If there is a negative relationship, the process proceeds to step S56. If there is no negative relationship, the process proceeds to step S58. If there is no negative relationship, the business condition selection unit 154 sets “false” in the business condition determination item of the record of item number n in the intermediate data 123a (the record is processed).

(S56) Based on the business condition candidate table 123, the business condition selection unit 154 determines whether BzCnd _n is a positive condition. If so, the process proceeds to step S59. If not, the process proceeds to step S57.

(S57) The business condition selection unit 154 sets BzCnd _n as processed. Then, the process proceeds to step S58.
(S58) The business condition selection unit 154 adds 1 to n (n = n + 1). Then, the process proceeds to step S53.

(S59) The business condition selection unit 154 selects BzCnd _n as the business condition. The business condition selection unit 154 sets “true” in the business condition determination item of the record of item number n of the intermediate data 123a (the record is already processed).

(S60) The business condition selection unit 154 substitutes 1 for m (m = 1).
(S61) The business condition selection unit 154 determines whether m is M or less (whether m ≦ M). If m is M or less, the process proceeds to step S62. If m is greater than M, the process proceeds to step S58.

(S62) The business condition selection unit 154 determines whether the negation of BzCnd _n or BzCnd _n exists in Cnd _m . When the negation of BzCnd _n or BzCnd _n exists in Cnd _m , the process proceeds to step S63. If the negation of BzCnd _n or BzCnd _n does not exist in Cnd _m , the process proceeds to step S64.

(S63) The business condition selection unit 154 records the appearance location of BzCnd _n or BzCnd _n in negative Cnd _m . Specifically, the business condition selection unit 154 uses BzCnd _n as the intermediate data 123a for the appearance location of Cz _m in Cnd _m (information indicating the item number m or Cnd _m of the record in the input / output table 122 that matches BzCnd _n ). Is registered in the record of item number n. Alternatively, the business condition selection unit 154 may change the appearance location of BzCnd _{n in} the negative Cnd _m (information indicating the position of the record item number m or Cnd _m in the input / output table 122 that matches BzCnd _n ) of the intermediate data 123a. Register in the record of item number n.

(S64) The business condition selection unit 154 adds 1 to m (m = m + 1). Then, the process proceeds to step S61.
(S65) The business condition selection unit 154 generates and outputs a business condition selection result 124 based on the intermediate data 123a. The business condition selection unit 154 stores the business condition selection result 124 in the analysis information storage unit 120. And business condition selection is complete | finished.

Here, in the determination in step S62 and the “record of appearance location” in step S63, the business condition selection unit 154 expresses each condition in a tree structure, and the tree structure of the business condition BzCnd _{n is} a tree structure of Cnd _m . Investigate whether it is part of it. In addition, in the confirmation of the existence of the negative condition, the business condition selection unit 154 generates a negative condition of BzCnd _n based on Domorgan's law and expresses the negative condition in a tree structure, so Determine.

FIG. 24 is a flowchart illustrating an example of a business term table output process. The business term table output process is executed in step S5 of FIG.
(S70) The business condition naming unit 155 acquires the business condition BzCnd _n , n = 1,..., N from the business condition selection result 124. Here, N is the total number of records in the business condition selection result 124.

(S71) The business condition naming unit 155 acquires the naming rule table 131 from the naming rule storage unit 130.
(S72) The business condition naming unit 155 substitutes 1 for n (n = 1).

  (S73) The business condition naming unit 155 determines whether n is N or less (whether n ≦ N). If n is N or less, the process proceeds to step S74. If n is larger than N, the process proceeds to step S77.

(S74) The business condition naming unit 155 determines whether the format of BzCnd _n matches the format indicated by any naming rule in the naming rule table 131. _If the format of BzCnd _n matches any format indicated by the naming rule, the process proceeds to step S75. _If the format of BzCnd _n does not match any format indicated by the naming rule, the process proceeds to step S76.

(S75) The business condition naming unit 155 names business conditions according to a naming rule. The business conditions and business terms named for the business conditions are registered in the business term table 141.
(S76) The business condition naming unit 155 adds 1 to n (n = n + 1). Then, the process proceeds to step S73.

(S77) The business condition naming unit 155 displays the contents of the business term table 141 on the display 21.
(S78) The business condition naming unit 155 determines whether to edit a business term. When editing business terms, the process proceeds to step S79. If the business term is not edited, the process proceeds to step S80. For example, the user can determine whether or not to edit the business terms named by the business condition naming unit 155 with reference to the business terms and business conditions in the business term table 141 displayed on the display 21. . When the business condition naming unit 155 receives an input of a business term editing instruction from the user, the business condition naming unit 155 determines to edit the business term. On the other hand, when the business condition naming unit 155 receives an input indicating that the business term is not edited by the user, the business condition naming unit 155 determines that the business term is not edited.

(S79) The business condition naming unit 155 accepts editing of business terms by the user, and updates the business terms registered in the business term table 141 according to the editing content.
(S80) The business condition naming unit 155 outputs the business term table 141. The business condition naming unit 155 stores the business term table 141 in the analysis result storage unit 140. Then, the business term table output ends.

  Here, as shown in FIG. 21, the condition (branch condition) in the input / output table 122 is represented by CNF (multiplicative standard form). Each condition in the input / output table 122 includes a plurality of disjunct clauses corresponding to the conditional sentence of the program 111 or the negation of the conditional sentence, as described in the explanation of FIG. In the definition of business terms, first, the business condition candidate generation unit 153 generates a business condition candidate table 123. That is, the business condition candidate generation unit 153 generates a plurality of candidates for the combination of conditional sentences (business condition candidates) by combining each of the plurality of disjunctive clauses included in the conditions (branch conditions) of the input / output table 122. . Then, the business condition selection unit 154 selects a business term definition target (that is, a business condition) from the plurality of business condition candidates according to the procedure of FIG.

  Specifically, the business condition selection unit 154 determines whether the negative condition for the first candidate among the plurality of business condition candidates is included in the plurality of business condition candidates. It is determined whether a combination of conditional statements corresponding to candidates is to be defined as a business term.

  Furthermore, when the second candidate indicating negative to the first candidate is included in the plurality of candidates, the business condition selecting unit 154 determines whether the first candidate is a positive condition or not. It is determined whether or not a combination of conditional sentences corresponding to one candidate is to be defined as a business term.

The reason for determining whether or not a negative condition for the first candidate exists in the business condition candidate table 123 is to appropriately extract a composite condition that affects the path of the program 111.
The business condition naming unit 155 names business terms according to the procedure of FIG. 24 for the business term definition target (that is, business conditions) determined in this way. Thereby, business terms can be appropriately selected and business terms can be named for the business conditions.

FIG. 25 is a flowchart illustrating a processing example of a business rule table output. The business rule table output process is executed in step S6 of FIG.
(S90) The input / output table rewriting unit 156 acquires the business condition BzCnd _n , n = 1,..., N from the business condition selection result 124. Here, N is the total number of records in the business condition selection result 124.

(S91) The input / output table rewriting unit 156 substitutes 1 for n (n = 1).
(S92) The input / output table rewriting unit 156 determines whether n is N or less (whether n ≦ N). If n is N or less, the process proceeds to step S93. If n is greater than N, the process proceeds to step S96.

(S93) The input / output table rewriting unit 156 replaces the occurrence location of the positive condition of BzCnd _n in the input / output table 122 with a description using business terms based on the business term table 141.
(S94) The input / output table rewriting unit 156 replaces the appearance location of the negative condition of BzCnd _n in the input / output table 122 with a description using business terms based on the business term table 141.

(S95) The input / output table rewriting unit 156 adds 1 to n (n = n + 1). Then, the process proceeds to step S92.
(S96) The input / output table rewriting unit 156 determines whether to change the format of the input / output table after rewriting (after replacement). When the format is changed, the process proceeds to step S97. If the format is not changed, the process proceeds to step S98.

(S97) The input / output table rewriting unit 156 changes the format of the input / output table after rewriting to the format of the decision table.
(S98) The input / output table rewriting unit 156 outputs the rewritten input / output table as the business rule table 142 (the business rule table 143 when changed to the decision table in step S97). The input / output table rewriting unit 156 stores the business rule table 142 (or the business rule table 143) in the analysis result storage unit 140. Then, the business rule table output ends.

  In this way, the input / output table rewriting unit 156 searches for the appearance location of a combination of conditional statements (corresponding to business conditions) in the conditions (branch conditions) in the input / output table 122. Then, the input / output table rewriting unit 156 generates the business rule tables 142 and 143 by replacing the description of the appearance location of the input / output table 122 with business terms corresponding to the combination of the conditional statements (business conditions). .

  FIG. 26 is a diagram illustrating an example of input / output of the analysis server. As described in the second embodiment, the analysis server 100 analyzes the program 111 with respect to the input of the program 111 and obtains the business term table 141 and the business rule table 143 (or the business rule table 142). Output.

  Here, as illustrated in FIG. 9, the program 111 includes a definition of the constant C2 and a description of processing using the constant C2 with respect to the program D14 of FIG. When such partial modification of the program D14 is performed, the contents of the business rules D11 and business terms D12 corresponding to the program D14 and the contents of the modified program 111 become inconsistent. For example, the business term D12 has no definition of a valid code related to the constant C2 (= 22), and therefore the constant C2 (= 22) is not considered in the condition of the valid code of the business rule D11. That is, there is a flaw between the existing business rules D11 and business terms D12 and the modified program 111.

  Therefore, the analysis server 100 receives the program 111 and outputs a business term table 141 and a business rule table 143 (or business rule table 142) corresponding to the program 111. In the example of the business term table 141, the valid code is defined in association with the constant C2 (= 22) added by the partial modification. Therefore, the condition for the valid code in the business rule table 143 is also the constant C2 (= 22). Is considered. Therefore, the business term table 141 and the business rule table 143 are consistent with the contents of the program 111, and there is no flaw. In this way, the analysis server 100 can appropriately create the business term table 141 and the business rule table 143 (or the business rule table 142) for the modified program 111.

  FIG. 27 is a diagram illustrating another example of a program to be analyzed. The program 112 is a program after partial modification obtained by partial modification of the program D14 of FIG. FIG. 27 exemplifies a case where the following two function additions have been made.

(Addition 1) “22” is added to the valid code, and the valid code is set to “11” or “22”.
(Addition 2) “33” is added to the valid code, and the valid code is set to “11”, “22”, or “33”.

Assume that the following two partial modifications are mainly made to the program D14 in order to cope with these function additions.
(Modification 1) In the IF statement condition on the 18th line of the program 112 after the modification, “code = C2” is added to the condition where “code = C1” originally, and “(code = C1) OR ( Code = C2) ”. Here, C2 is a constant, and a definition is added to the 12th line of the program 112.

  (Renovation 2) The ELSE IF block is added to the 24th to 29th lines of the program 112 after the renovation. Here, C3 is a constant, and a definition is added to the 13th line of the program 112.

Further, in the example of the program 112, the description in the 17th and 18th lines of the program D14 in FIG.
As illustrated in FIG. 27, the program D14 may be modified a plurality of times, and the modification method is not unique. For example, the modification 2 can be dealt with by a method of modifying the condition of the IF sentence like the modification 1 instead of adding the ELSE IF block. However, the programmer in charge of the program D14 may be changed during many years of maintenance, and the modification can be performed by various description methods.

  FIG. 28 is a diagram illustrating another example of input / output of the analysis server. The analysis server 100 analyzes the program 112 with respect to the input of the program 112, and outputs a business term table 141a and a business rule table 143a. However, the analysis server 100 can also output a business rule table in which the business rule table 143a is changed to a decision table format.

  According to the business term table 141a, the valid code is defined in association with the constant C2 (= 22) and the constant C3 (= 33) added by the partial modification. Therefore, the condition for the valid code in the business rule table 143a also takes into account the constant C2 (= 22) and the constant C3 (= 33). Therefore, the business term table 141a and the business rule table 143a are consistent with the contents of the program 112, and there is no flaw. As described above, the analysis server 100 appropriately creates the business term table 141a and the business rule table 143a (or the business rule table obtained by changing the business rule table 143a into the decision table format) even for the modified program 112. it can.

  By the way, in the activities of information processing system maintenance and restructuring, in order to understand and understand the current state of the information processing system, knowledge of the business handled by the information processing system and information on how to implement the information processing system (business terminology) Table and business rule table). Such information is originally described in the document, but if there is a deficiency that the originally created document has not been properly updated, the system developer must trust the document. I can't refer to it. In addition, it is often difficult for an expert such as an initial developer in charge of the information processing system to make an inquiry to an expert because it is dissipated during a long operation period.

  Analysis of the current program that is the actual state of information processing system implementation to compensate for the lack of information for maintenance and reconstruction work due to such lack of documents and the absence of experts. May attempt to restore the information.

  However, as described above, in a program that has undergone partial modification over many years, business information is not neatly described, and is described in a distributed form throughout the program (for example, program 111 or program 112). Sometimes. In this case, it is difficult to acquire business information using the program description as it is. In addition, there is a gap between the design document and the current program, and information described in the design document may not be used.

  As described above, even when it is difficult to obtain business information from the current program, the analysis server 100 can appropriately create business knowledge (business term table or business rule table) from the current program. For example, the analysis server 100 displays the created business term table and business rule table using the display 21 and presents them to the system developer. The system developer can appropriately understand and understand the current state of the information processing system from the business term table and the business rule table displayed by the analysis server 100, and can efficiently perform maintenance and reconstruction activities of the information processing system. Be able to proceed. In this way, the analysis server 100 can support appropriate grasp of existing software assets by the user.

  The information processing according to the first embodiment can be realized by causing the processing unit 12 to execute a program. The information processing according to the second embodiment can be realized by causing the CPU 101 to execute a program. The program can be recorded on a computer-readable recording medium 23.

  For example, the program can be distributed by distributing the recording medium 23 on which the program is recorded. Alternatively, the program may be stored in another computer and distributed via a network. For example, the computer stores (installs) a program recorded in the recording medium 23 or a program received from another computer in a storage device such as the RAM 102 or the HDD 103, and reads and executes the program from the storage device. Good.

DESCRIPTION OF SYMBOLS 10 Program analyzer 11 Memory | storage part 12 Processing part P1 1st program P2 2nd program R1 Definition rule T1 1st business rule table T2 Input / output table T3 Business term table T4 2nd business rule table

Claims (9)

Computer
A second program after correction of the first program, wherein the second program has a conflict with the first business rule table corresponding to the first program;
An input / output table indicating a branch condition related to a conditional statement in the second program and an output corresponding to the branch condition is created, and the information included in the input / output table is based on information indicating a definition rule for a business term. Define the business terms for the combination of conditional statements,
Creating a second business rule table corresponding to the second program using the defined business terms and the input / output table;
Program analysis method.   The program analysis method according to claim 1, wherein the input / output table is created based on a result of symbolic execution of the second program. The result of the symbolic execution includes a plurality of conditions related to input variables for each of a plurality of outputs by the second program,
In the generation of the input / output table, a process of converting the plurality of conditions for one output into one branch condition is executed for each output.
The program analysis method according to claim 2. The branch condition is a multiplicative standard form,
In the definition of the business term, a plurality of candidates for the combination of the conditional sentences are generated by combining each of a plurality of disjunct clauses included in the branch condition, and the business term is defined from the plurality of candidates Select
The program analysis method according to any one of claims 1 to 3.   In the definition of the business term, the conditional sentence corresponding to the first candidate is determined according to whether or not a negative condition for the first candidate among the plurality of candidates is included in the plurality of candidates. The program analysis method according to claim 4, wherein it is determined whether or not the combination of the business terms is to be defined.   In the definition of the business term, when a second candidate indicating negative to the first candidate is included in the plurality of candidates, according to the determination as to whether or not the first candidate is a positive condition, The program analysis method according to claim 5, wherein it is determined whether or not the combination of the conditional sentences corresponding to the first candidate is to be defined as the business term.   In the creation of the second business rule table, the occurrence location of the combination of the conditional statements in the branch condition in the input / output table is searched, and the description of the occurrence location in the input / output table is described in the conditional statement. The program analysis method according to claim 1, wherein the business term corresponding to a combination is substituted. The second program after the modification of the first program, the second program in which there is a conflict with the first business rule table corresponding to the first program, and business term definition rules A storage unit for storing information indicating
An input / output table indicating a branch condition related to a conditional statement in the second program and an output corresponding to the branch condition is created, and the conditional statement included in the input / output table based on information indicating the definition rule A processing unit that defines the business terms for a combination of the first and second business rule tables corresponding to the second program using the business terms and the input / output table defined;
A program analysis apparatus. A second program after correction of the first program, wherein the second program has a conflict with the first business rule table corresponding to the first program;
An input / output table indicating a branch condition related to a conditional statement in the second program and an output corresponding to the branch condition is created, and the information included in the input / output table is based on information indicating a definition rule for a business term. Define the business terms for the combination of conditional statements,
Creating a second business rule table corresponding to the second program using the defined business terms and the input / output table;
A program analysis program that causes a computer to execute processing.

Images