JP6000213B2 - Code generation system, code generation method, and code generation program - Google Patents

Description

  The present invention relates to a technique for describing a rule by a logical expression and converting the written rule into a program code.

  Patent Document 1 describes that a rule is described in a spreadsheet, and the written rule is converted into a program code. Patent Document 1 describes that a converted program code is executed by a server and a rule is evaluated.

Special table 2010-524134

If the rules are simply converted into program codes, the number of program codes executed on the server increases as the number of rules increases, and the time required for processing to evaluate the rules increases.
An object of the present invention is to shorten the processing time for evaluating rules when the number of rules increases.

A code generation system according to the present invention includes:
A rule definition section for defining a rule combining a plurality of logical expressions;
A program code generation unit that converts the rule defined by the rule definition unit into a program code;
A rule evaluation unit that executes the program code generated by the program code generation unit and evaluates a rule;
The program code generator is
A rule table generating unit that extracts a character string from a logical expression constituting the rule, generates a rule table in which the extracted character string is a key, and associates the key with the rule;
A rule conversion unit that converts the rule into a program code;
The rule evaluation unit
An evaluation request receiving unit that receives an evaluation request including an evaluation value for the rule;
A character string is extracted from the evaluation value included in the evaluation request received by the evaluation request receiving unit, the rule table generated by the rule table generating unit is searched using the extracted character string as a key, and the searched rule And an evaluation processing unit that inputs and evaluates an evaluation value included in the evaluation request in the converted program code.

  In the code generation system according to the present invention, a rule table is generated when a rule is converted into a program code. Then, when executing the program code to evaluate the rules, the rule table is drawn based on the evaluation value included in the evaluation request to narrow down the rules that need to be executed. Therefore, even if the number of rules increases, the number of program codes to be executed can be suppressed, and the processing time for evaluating the rules can be shortened.

1 is a configuration diagram of a code generation system 10 according to a first embodiment. 5 is a flowchart showing processing of a rule definition unit 20 according to the first embodiment. 5 is a flowchart showing processing of a program code generation unit 30 according to the first embodiment. 5 is a flowchart showing processing of a rule evaluation unit 40 and an evaluation request input unit 50 according to the first embodiment. The figure which shows the example of the rule described in the spreadsheet 71. FIG. The figure which shows the example of the test value 72 with respect to the rule shown in FIG. The figure which shows the example of the description rule of a rule. The figure which shows the example of input of a basic rule. The figure which shows the example of input of an and rule. The figure which shows the example of input of an or rule. The figure which shows the input example of a negative rule. The figure which shows the example of input of a conditional branch rule. The figure which shows the example of an input of a conversion rule. The figure which shows the example of input of a calculation rule. Explanatory drawing of attribute information. Explanatory drawing of the mapping information 73. FIG. Explanatory drawing of the mapping information 73. FIG. The figure which shows the example of the mapping information 73. Explanatory drawing of the process which produces | generates the rule controller 74. FIG. Explanatory drawing of the process which produces | generates the JV rule 76. FIG. Explanatory drawing of an evaluation request. The figure which shows the example of an evaluation request | requirement. Explanatory drawing of an evaluation result. The figure which shows the example of an evaluation result. The figure which shows the example which does not include the logical expression about the item of the character string used as a key in a program code, and converts only the remaining logical expressions into a program code. The figure which shows the example which made the rule which shows a part of conditions describe in another column. 3 is a diagram illustrating an example of a hardware configuration of each device of the rule definition unit 20, the program code generation unit 30, the rule evaluation unit 40, and the evaluation request input unit 50 illustrated in the first embodiment.

Embodiment 1 FIG.
FIG. 1 is a configuration diagram of a code generation system 10 according to the first embodiment.
The code generation system 10 includes a rule definition unit 20, a program code generation unit 30, a rule evaluation unit 40, and an evaluation request input unit 50.
The rule definition unit 20, the program code generation unit 30, and the rule evaluation unit 40 may be configured as separate devices, or may be configured as one device. When the rule definition unit 20, the program code generation unit 30, and the rule evaluation unit 40 are configured as separate devices, the rule definition unit 20 and the program code generation unit 30 are connected by a network. The program code generation unit 30 and the rule evaluation unit 40 are connected by a network.
The rule evaluation unit 40 and the evaluation request input unit 50 are connected by a network.

  A display device 61 and an input device 62 are connected to the rule definition unit 20, and the rule definition unit 20 is operated by a business administrator who manages rules. Further, a display device 63 and an input device 64 are connected to the evaluation request input unit 50, and the evaluation request input unit 50 is operated by a user who requests rule evaluation.

The rule definition unit 20 includes a rule input unit 21, a syntax analysis unit 22, a comprehensive verification unit 23, an individual verification unit 24, and a verification result display unit 25.
The rule input unit 21 inputs one or more spreadsheets 71, a test value 72, and mapping information 73.

The program code generation unit 30 includes a spreadsheet management unit 31, a mapping information management unit 32, a syntax analysis unit 33, a rule controller generation unit 34, and a JV rule generation unit 35.
The spreadsheet management unit 31 manages the spreadsheet 71 input by the rule input unit 21. The mapping information management unit 32 manages the mapping information 73 input by the rule input unit 21. The rule controller generator 34 generates a rule controller 74 (evaluation processor). The rule controller 74 includes a rule table 75. The JV rule generation unit 35 generates a JV rule 76 (Java (registered trademark) rule).

The rule evaluation unit 40 includes an evaluation request reception unit 41, an evaluation result reply unit 42, a JV rule management unit 43, and a decimal number calculation unit 44.
The JV rule management unit 43 manages the rule controller 74 and the JV rule 76.

  The evaluation request input unit 50 includes an evaluation request transmission unit 51, an evaluation result reception unit 52, and an evaluation result display unit 53.

2 to 4 are flowcharts showing processing of the code generation system 10 according to the first embodiment.
FIG. 2 is a flowchart showing processing of the rule definition unit 20 according to the first embodiment. FIG. 3 is a flowchart showing processing of the program code generation unit 30 according to the first embodiment. FIG. 4 is a flowchart showing processing of the rule evaluation unit 40 and the evaluation request input unit 50 according to the first embodiment.

First, the processing of the rule definition unit 20 will be described with reference to FIG.
(S11: Rule input process)
The rule input unit 21 inputs rules to the spreadsheet 71 according to the operation of the business manager via the input device 62. The rule is described by a logical expression according to a predetermined description rule. The rule input unit 21 also inputs attribute information accompanying the rule together with the rule.

(S12: Test value input process)
The rule input unit 21 inputs a test value 72 for simply testing the rule input in S11 in accordance with the operation of the business manager via the input device 62.

(S13: Parsing process)
The syntax analysis unit 22 analyzes the syntax of the rule input in S11.
Analyzing the syntax means verifying whether the rule is syntactically correct, such as whether the rule conforms to the description rule or whether the variable types match. The function of the syntax analysis unit 22 is not limited to this. For example, the function of the syntax analysis unit 22 is realized by a check function for mathematical expressions or the like that is pre-installed in spreadsheet application software.

(S14: Comprehensive verification process)
The comprehensive verification unit 23 inputs the test value 72 input in S12 to the rule input in S11 and verifies the rule.

(S15: Individual verification process)
The individual verification unit 24 inputs the test value 72 input in S12 to each logical expression constituting the rule input in S11, and verifies each logical expression.

(S16: Verification result display process)
The verification result display unit 25 shows the result of analyzing the syntax in S13, the result of verifying the rule in S14, and the result of verifying each logical expression in S15 in the spreadsheet into which the rule is input, and displays the display device 61. To display.

  If the result displayed in S16 is not the desired result, the business administrator returns the process to S11, corrects the rule, the test value 72, etc., and confirms the verified result again. By repeating this, it is possible to create a rule having no error (bug) as generally assumed.

(S17: Mapping information input process)
The rule input unit 21 inputs the mapping information 73 in accordance with the operation of the business manager via the input device 62.
The mapping information 73 is information indicating the correspondence between the name of the rule used in the spreadsheet and the class or function used in the program code.

FIG. 5 is a diagram illustrating an example of rules described in the spreadsheet 71. FIG. 6 is a diagram showing an example of the test value 72 for the rule shown in FIG.
In FIG. 5, one rule is described in each row of the spreadsheet. Each rule is described together with attribute information (here, ID, name, error code, rule type).
Then, in each row of the spreadsheet, a test value 72 is input to the rule described in the row, and an evaluation result indicating a result of evaluating the rule and each of the rules described in the row A test value 72 is input to the logical expression, and an individual evaluation result indicating a result of evaluating each logical expression is displayed.
That is, the rule and attribute information fields are input in S11, and the test value 72 is input in S12. Then, the result of evaluating the rule in S14 is displayed in the evaluation result column in S16, and the result of evaluating each logical expression in S15 is displayed in the individual evaluation result column in S16.
Although not shown in FIG. 5, if it is confirmed that there is an error in the syntax in S13, there may be an error in the syntax in the rule column, the evaluation result column, and the individual evaluation result column in S16. Is displayed.

FIG. 7 is a diagram illustrating an example of a rule description rule.
In FIG. 7, basic rules, and rules, or rules, negative rules, conditional branch rules, conversion rules, and calculation rules are defined as rule description rules.
The basic rule defines the basic description rule, and is the same as the description rule of the mathematical formula used in general spreadsheet application software such as Excel (registered trademark). The and rule defines the description of the and condition, and is defined to be described as “basic rule and basic rule and...”. The or rule defines the description of the or condition, and is described as “basic rule or basic rule or...”. The negative rule specifies a negative description, and it is specified to be described as “NOT (basic rule)”. The conditional branch rule defines the description of the conditional branch, and is defined as “IF (conditional expression, value, value)”. The conversion rule defines a description of conversion of a character string or the like, and is defined to be described as “scheduled date / year / scheduled date / month / scheduled date / day”, for example. The calculation rule defines the description of the calculation formula, and for example, it is defined to be described as “fee * (1-discount rate)”.

FIG. 8 to FIG. 14 are diagrams showing input examples of the basic rule, the and rule, the or rule, the negative rule, the conditional branch rule, the conversion rule, and the calculation rule. 8 to 14 also show evaluation results and individual evaluation results.
As shown in FIG. 8C, if there is a syntax error in the input rule, it is displayed that there is an error in the syntax in the rule column, evaluation result column, and individual evaluation result column. . Here, a mark indicating that there is an error is displayed in the rule column, and “#NAME?” Indicating that there is an error in the name is displayed in the evaluation result column and the individual evaluation result column. Yes.
Since the function of the syntax analysis unit 22 is realized by the function pre-installed in the spreadsheet application software, the information displayed when there is an error in the syntax is displayed by the business administrator using the spreadsheet application software. It is very easy to understand if you are used to it. Therefore, the rule can be easily corrected as compared with the case where the rule is corrected based on an error related to a syntax error output by the compiler of the program.

FIG. 15 is an explanatory diagram of attribute information.
In FIG. 15, ID, name, error code, and rule type are defined as attribute information.
The ID is an identifier for uniquely identifying the rule. The name is the name of the rule used when referring to the evaluation result of the rule from another rule. The error code is a code for notifying the user of the application which rule is established when the rule is established. The rule type is information specifying at which timing the rule is executed.
The attribute information is used by the business administrator to manage the rules. Some attribute information such as the rule type may be used as a rule application condition.
Note that the attribute information is not limited to that shown in FIG. 15, and other attributes that are conditions for applying the rule, such as an application period and an applicable job type, may be defined.

FIGS. 16 and 17 are explanatory diagrams of the mapping information 73, and FIG. 18 is a diagram illustrating an example of the mapping information 73.
The mapping information 73 defines the correspondence between the name of the rule described in Japanese or the like used in the spreadsheet and the Java (registered trademark) class or function used in the program code. 16 to 18, the mapping information 73 is defined in the XML format. 16 to 18 show examples of the definition of the mapping information 73, and the mapping information 73 may be defined in other formats.

Next, the processing of the program code generation unit 30 will be described with reference to FIG.
(S21: Spreadsheet acquisition process)
The spreadsheet management unit 31 acquires the spreadsheet 71 input by the rule input unit 21 and stores it in the storage device.

(S22: Mapping information acquisition process)
The mapping information management unit 32 acquires the mapping information 73 input by the rule input unit 21 and stores it in the storage device.

(S23: Rule controller generation process)
The rule controller generation unit 34 generates a rule controller 74 having a rule table 75 based on the spreadsheet 71 acquired in S21. The rule controller 74 is a program that narrows down the JV rules 76 to be executed by drawing the rule table 75.

(S24: JV rule generation process)
The JV rule generation unit 35 converts the rules described in the spreadsheet 71 acquired in S21 into Java (registered trademark) program code based on the mapping information 73 acquired in S22, and compiles the JV rules. 76 is generated.

At the time of compilation in S24, the syntax analysis unit 33 analyzes the syntax of the rule acquired in S21.
The syntax analysis executed here is basically the same as the syntax analysis executed by the syntax analysis unit 22 in S13.
However, while the function of the syntax analysis unit 22 is realized by a function that is preinstalled in the application software of the spreadsheet, the function of the syntax analysis unit 33 is not limited to this. This is realized by the parsing function. Therefore, although there is a possibility that some syntax errors cannot be found in S13, in S23, syntax errors that cannot be found in S13 can also be found.
If the syntax is incorrect, an error is output to the file. If an error is output, the process is returned to S11 and the rule is corrected.

  In general, a unit test is performed on the JV rule 76 generated in S24 using a tool such as JUnit. If an error is found in the unit test, the process is returned to S11 and the rule is corrected.

FIG. 19 is an explanatory diagram of processing for generating the rule controller 74.
The rule controller generation unit 34 extracts a character string designated as a condition from the rules described in the spreadsheet 71 in the logical expression connected by the and condition. The rule controller generation unit 34 generates a rule table 75 in which the extracted character strings are concatenated and used as keys to store the keys and rules in association with each other. The rule controller generation unit 34 may calculate a hash value of the concatenated character string and use the calculated hash value as a key.
In the case of FIG. 19, a logical expression of “insurance type =“ personal insurance ””, “travel destination =“ USA ””, “purpose =“ business ””, and “insurance amount> 1000” under the and condition. Is specified. Therefore, “personal insurance”, “America”, and “business”, which are character strings designated as conditions in each logical expression, are extracted. 1000 specified as the condition in “insurance amount> 1000” is not a character string but a numerical value, and is not extracted. The extracted “personal insurance”, “USA”, and “business” are connected to form a key, and the key and Rule001 are associated with each other.

  Then, the rule controller generation unit 34 generates a rule controller 74 that specifies the corresponding rule by drawing the generated rule table 75. Note that the source program itself that realizes the function of the rule controller 74 is created in advance, and the rule controller generation unit 34 designates the generated rule table 75 as an input for the previously created source program. The rule controller 74 is generated by compiling with a normal Java (registered trademark) compiler.

Note that when an or condition is included as a rule, a plurality of keys may be generated for one rule.
For example, the rules are “insurance type =“ personal insurance ”” and “travel destination =“ USA ”” and (“purpose =“ business ”” or “purpose =“ tourism ””) and “insurance amount> 1000”. If "Insurance Type =" Individual Insurance "", "Destination =" America "" and "Purpose =" Business "", the strings "Individual Insurance", "America" and "Business" are extracted. And a key is generated. In addition, the character strings “personal insurance”, “USA” and “tourism” in “insurance type =“ personal insurance ””, “travel destination =“ USA ””, and “purpose =“ tourism ”” are extracted. A key is generated. This rule is associated with both of the two generated keys in the rule table.

FIG. 20 is an explanatory diagram of a process for generating the JV rule 76.
The JV rule generation unit 35 changes each name included in the rules described in the spreadsheet 71 to a class name or a function name based on the mapping information 73. Then, the JV rule generation unit 35 converts the rule into a Java (registered trademark) program code in accordance with a predetermined logic. Then, the JV rule generation unit 35 generates a JV rule 76 from the converted program code using a normal Java (registered trademark) compiler.
When converting into a program code, the JV rule generation unit 35 includes each attribute information as a comment in a predetermined format in the program code. This is so that the spreadsheet 71 can be reversely generated later from the program code.

Next, processing of the rule evaluation unit 40 and the evaluation request input unit 50 will be described with reference to FIG.
(S31: Code acquisition process)
The JV rule management unit 43 acquires the rule controller 74 generated by the rule controller generation unit 34 and the JV rule 76 generated by the JV rule generation unit 35 and stores them in the storage device.

(S32: Evaluation request transmission process)
The evaluation request transmission unit 51 transmits an evaluation request to the rule evaluation unit 40 via the network in accordance with a user operation via the input device 64. The evaluation request includes an evaluation value that is an input to the rule.

(S33: Evaluation request reception process)
The evaluation request receiving unit 41 receives the evaluation request transmitted in S32.

(S34: Rule specifying process)
The JV rule management unit 43 executes the rule controller 74. Then, the rule controller 74 extracts a character string from the evaluation value included in the evaluation request received in S33, and generates a key by concatenating the extracted character strings. Then, the rule controller 74 draws the rule table 75 with the generated key to identify the corresponding rule.

(S35: Code execution process)
The JV rule management unit 43 receives the evaluation value included in the evaluation request received in S33, and executes the JV rule 76 generated from the rule specified in S34.
At this time, the JV rule management unit 43 causes the decimal number calculation unit 44 to perform numerical calculation when the JV rule 76 includes numerical calculation. The decimal arithmetic unit 44 performs numerical arithmetic using a decimal number without using the floating point arithmetic function of the CPU. When a floating point operation is performed, an error may occur in the operation, and as a result, there is a possibility that the determination by the rule is wrong. Therefore, by using the decimal number arithmetic unit 44 and performing the numerical operation with the decimal number without performing the floating point calculation, the determination by the rule is correctly performed.

(S36: Evaluation result transmission process)
The evaluation result reply unit 42 transmits the result of the execution of the JV rule 76 in S35 to the evaluation request input unit 50 via the network.

(S37: Evaluation result reception process)
The evaluation result receiving unit 52 receives the evaluation result transmitted in S36.

(S38: Evaluation result display process)
The evaluation result display unit 53 displays the evaluation result received in S37 on the display device 63.

FIG. 21 is an explanatory diagram of an evaluation request, and FIG. 22 is a diagram illustrating an example of an evaluation request.
21 and 22, the evaluation request includes a due date and input data 1 to n as evaluation values. For the input data 1 to n, it is determined in advance what is the column to be set. For example, in FIG. 22, the country is set for the input data 1, the gender is set for the input data 2, and the age is set for the input data n.

  In S34, the rule controller 74 extracts a character string from the evaluation value and concatenates the extracted character strings to generate a key. Here, it may be determined in advance which input data of the input data 1 to n is connected to the character string set to generate the key. Further, all combinations of character strings included in the input data 1 to n may be generated as keys, and the corresponding rule may be specified based on each key.

FIG. 23 is an explanatory diagram of the evaluation result, and FIG. 24 is a diagram illustrating an example of the evaluation result.
23 and 24, the evaluation results include output data 1 to n. For the output data 1 to n, it is determined in advance which rule results are set for. For example, in FIG. 24, the result of rule 1 is set for output data 1, the result of rule 2 for output data 2, the result of rule 3 for output data 3, and the result of rule n for output data n. .
In the evaluation result, only the result regarding the rule for which the JV rule 76 is executed in S35 is set.

  As described above, in the code generation system 10 according to the first embodiment, the rule controller 74 including the rule table 75 is generated when a rule is converted into a program code. When executing the program code and evaluating the rules, the rule controller 74 draws the rule table 75 based on the evaluation value included in the evaluation request to narrow down the rules that need to be executed. Therefore, even if the number of rules increases, the number of program codes to be executed can be suppressed, and the processing time for evaluating the rules can be shortened.

Further, in the code generation system 10 according to the first exemplary embodiment, it is possible to describe a complex rule in a spreadsheet by further connecting logical expressions connected with and or or and the like with and and or or the like. Also, complex rules including conditional branches can be described in the spreadsheet. Therefore, it is very convenient.
On the other hand, when such a complicated rule is described, there is a high possibility that the described rule includes an error. However, in the code generation system 10 according to the first exemplary embodiment, the result of evaluating the entire rule and the result of evaluating each logical expression constituting the rule with the test value 72 as an input before converting the rule into the program code Can be easily obtained. Therefore, it is possible to notice an error in a rule early, and it is possible to reduce the occurrence of rework and efficiently describe the rule.

That is, generally, a unit test is performed on the JV rule 76 generated by the JV rule generation unit 35, and the rule error is first noticed at this point. In this case, the spreadsheet rule is corrected again, the JV rule generation unit 35 generates the JV rule 76 again, and the unit test is performed.
On the other hand, in the code generation system 10 according to the first embodiment, when a rule is described in a spreadsheet and the test value 72 is input, it can be recognized that there is an error in the rule. Therefore, it is possible to describe rules efficiently by reducing the occurrence of rework.
Of course, verification by the test value 72 is simple, and if a unit test is performed, a new error may be found. However, since the verification by the test value 72 is performed, the number of errors found in the unit test is reduced, and the occurrence of rework is reduced.

  Moreover, since not only the result of evaluating the rules but also the result of evaluating each logical expression is displayed on the spreadsheet 71, it is possible to easily identify an erroneous part. Therefore, rules can be described efficiently.

  In the above description, in S24, the JV rule generation unit 35 simply converts the rules into program codes. However, when the rule table 75 is drawn and the corresponding rule is specified, it can be considered that the conditions used when the rule table 75 is drawn are already satisfied. Therefore, when converting the rule into the program code, the JV rule generation unit 35 may convert only the remaining logical expression into the program code without including the logical expression for the character string used as the key.

FIG. 25 is a diagram illustrating an example in which a logical expression for a character string item used as a key is not included in a program code, and only the remaining logical expressions are converted into a program code. FIG. 25 shows an example of converting Rule001 shown in FIG. 19 into a program code.
In FIG. 19, “Individual Insurance”, “America”, and “Business” are extracted from Rule 001, and “Individual Insurance”, “America”, and “Business” are linked to form a key. Therefore, as shown in FIG. 25, the JV rule generation unit 35 has “insurance type =“ personal insurance ””, “travel destination =“ USA ””, and “ The purpose = “business” is not converted into a program code, but only the remaining “insurance amount> 1000” is converted into a program code.
In this case, by subtracting the rule table 75 in S34 and specifying the corresponding rule, “insurance type =“ personal insurance ”” and “travel destination =“ USA ”” and “purpose =“ business ”” are applicable. Then, it is considered that the determination has been made, and the determination regarding the remaining “insurance amount> 1000” is made in S35.

  Further, in the above description, as shown in FIG. 5, all rules are described by logical formulas in the rule column of the spreadsheet so that the rules can be generally written. However, for example, when there are conditions used in many rules, the conditions may be provided with a separate column. This facilitates the description of the rule and reduces the possibility that the rule description contains an error.

FIG. 26 is a diagram illustrating an example in which rules indicating some conditions are described in another column.
In FIG. 26, separate fields are provided for the insurance type, travel destination, and purpose. For the insurance type, travel destination, and purpose, whether or not the character string is described in each column is determined by the and condition. For example, the rule described in the first line is “insurance type =“ individual insurance ”” and “travel destination =“ USA ”” and “purpose =“ business ”” and “insurance” as in Rule 001 shown in FIG. The rule “amount> 1000” is shown.

  When a rule indicating some conditions is described in a separate column, a rule character string described in the separate column may be concatenated as a key of the rule table 75.

  As can be seen from the rule description rule shown in FIG. 7, the rule description can include a rule that returns a true / false value, a rule that returns a character string, and a rule that returns a numerical value. Further, not only a rule that returns a constant value, but also a rule that includes a conditional branch that changes a value to be returned depending on a condition can be described. Moreover, the evaluation result of a certain rule can be used as an input for another rule, and an association rule describing the relationship between a plurality of rules can be described.

  Also, for example, when two days are specified, if a function used in a rule is required, such as a function that counts the number of days between two days, the program code for that function must be developed separately. You may make it available by calling from the rule.

FIG. 27 is a diagram illustrating an example of a hardware configuration of each device of the rule definition unit 20, the program code generation unit 30, the rule evaluation unit 40, and the evaluation request input unit 50 illustrated in the first embodiment.
Each device is a computer, and each element of each device can be realized by a program.
As a hardware configuration of each device, an arithmetic device 901, an external storage device 902, a main storage device 903, a communication device 904, and an input / output device 905 are connected to the bus.

  The arithmetic device 901 is a CPU (Central Processing Unit) that executes a program. The external storage device 902 is, for example, a ROM (Read Only Memory), a flash memory, a hard disk device, or the like. The main storage device 903 is, for example, a RAM (Random Access Memory). The communication device 904 is, for example, a communication board. The input / output device 905 is, for example, a mouse, a keyboard, a display device, or the like.

The program is normally stored in the external storage device 902, and is loaded into the main storage device 903 and sequentially read into the arithmetic device 901 and executed.
The program includes a rule input unit 21, a syntax analysis unit 22, a comprehensive verification unit 23, an individual verification unit 24, a verification result display unit 25, a spreadsheet management unit 31, a mapping information management unit 32, a syntax analysis unit 33, and a rule controller generation unit. 34, JV rule generation unit 35, evaluation request reception unit 41, evaluation result reply unit 42, JV rule management unit 43, decimal number calculation unit 44, evaluation request transmission unit 51, evaluation result reception unit 52, and evaluation result display unit 53 It is a program that realizes the functions described.
Furthermore, an operating system (OS) is also stored in the external storage device 902. At least a part of the OS is loaded into the main storage device 903, and the arithmetic device 901 executes the above program while executing the OS.
In the description of the first to fifth embodiments, “to input”, “analysis of”, “verification of”, “display of”, “acquisition of”, “generation of”, “to” Information, data, signal values, and variable values indicating the results of the processing described as “reception of”, “transmission of”, “execution of”, and the like are stored in the main storage device 903 as files.

  Note that the configuration in FIG. 27 is merely an example of the hardware configuration of each device, and the hardware configuration of each device is not limited to the configuration described in FIG. 27 but may be another configuration.

  10 code generation system, 20 rule definition unit, 21 rule input unit, 22 syntax analysis unit, 23 comprehensive verification unit, 24 individual verification unit, 25 verification result display unit, 30 program code generation unit, 31 spreadsheet management unit, 32 mapping Information management unit, 33 Parsing unit, 34 Rule controller generation unit, 35 JV rule generation unit, 40 Rule evaluation unit, 41 Evaluation request reception unit, 42 Evaluation result reply unit, 43 JV rule management unit, 44 Decimal number calculation unit, 50 evaluation request input unit, 51 evaluation request transmission unit, 52 evaluation result reception unit, 53 evaluation result display unit, 61 display device, 62 input device, 63 display device, 64 input device, 71 spreadsheet, 72 test value, 73 mapping Information, 74 rule controller, 75 rule table, 76 JV rule.

Claims (7)

A rule definition section for defining a rule combining a plurality of logical expressions;
A program code generation unit that converts the rule defined by the rule definition unit into a program code;
A rule evaluation unit that executes the program code generated by the program code generation unit and evaluates a rule;
The program code generator is
A rule table generating unit that extracts a character string from a logical expression constituting the rule, generates a rule table in which the extracted character string is a key, and associates the key with the rule;
A rule conversion unit that converts the rule into a program code;
The rule evaluation unit
An evaluation request receiving unit that receives an evaluation request including an evaluation value for the rule;
A character string is extracted from the evaluation value included in the evaluation request received by the evaluation request receiving unit, the rule table generated by the rule table generating unit is searched using the extracted character string as a key, and the searched rule A code generation system comprising: an evaluation processing unit that inputs an evaluation value included in the evaluation request and evaluates the program code into which the program is converted. The rule table generation unit extracts a character string in a logical expression connected by an AND condition from a logical expression constituting the rule,
The evaluation processing unit extracts an evaluation value for the logical expression connected by the and condition from the evaluation value included in the evaluation request, and the rule table generated by the rule table generation unit using the extracted evaluation value as a key. The code generation system according to claim 1, wherein the code generation system is searched.   The rule conversion unit converts only the remaining logical expression excluding the logical expression obtained by extracting the character string from the rule into a program code, and sets the converted program code from the rule. Or the code generation system of 2. The rule definition unit
A rule input unit for inputting the rules into a spreadsheet;
A test value input unit for inputting a test value for the rule input by the rule input unit;
An overall verification unit that inputs and verifies the test value input by the test value input unit into the rule; and an individual verification unit that inputs and verifies the test value into each logical expression constituting the rule;
4. The code generation system according to claim 1, further comprising: a verification result display unit that displays a result verified by the comprehensive verification unit and the individual verification unit on the spreadsheet. 5. The rule definition unit defines attribute information of the rule together with the rule,
5. The code generation system according to claim 1, wherein, when the rule is converted into a program code, the attribute information of the rule is included in the program code as a comment. 6. . A rule definition process in which the processing device defines a rule combining a plurality of logical expressions;
A processing code, a program code generation step of converting the rule defined in the rule definition step into a program code;
A processing device includes a rule evaluation step for executing a program code generated in the program code generation step and evaluating a rule,
The program code generation step includes
A rule table generation step of generating a rule table in which a character string is extracted from a logical expression constituting the rule, and the key and the rule are associated with each other using the extracted character string as a key;
A rule conversion step of converting the rule into a program code,
The rule evaluation process includes
An evaluation request receiving step of receiving an evaluation request including an evaluation value for the rule;
A character string is extracted from the evaluation value included in the evaluation request received by the evaluation request receiving step, the rule table generated in the rule table generation step is searched using the extracted character string as a key, and the searched rule A code generation method comprising: an evaluation processing step of inputting and evaluating an evaluation value included in the evaluation request to a program code converted from. Rule definition processing that defines a rule that combines multiple logical expressions,
A program code generation process for converting the rule defined in the rule definition process into a program code;
Executing the program code generated by the program code generation process, causing the computer to execute a rule evaluation process for evaluating a rule,
The program code generation process includes:
A rule table generation process for generating a rule table in which a character string is extracted from a logical expression constituting the rule, and the extracted character string is used as a key and the key and the rule are associated;
A rule conversion process for converting the rule into a program code,
The rule evaluation process is:
An evaluation request receiving process for receiving an evaluation request including an evaluation value for the rule;
A character string is extracted from the evaluation value included in the evaluation request received by the evaluation request reception process, and the rule table generated by the rule table generation process is searched using the extracted character string as a key, and the searched rule A code generation program comprising: an evaluation process in which an evaluation value included in the evaluation request is input and evaluated to the program code converted from.

Images