JPH09244881A - Method for generating data item dictionary while specification of data is recognized - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate individual differences of a creator and to improve the productivity of the data item dictionary by generating the definitions of individual processing for data items in a source program and automatically generating information characteristic of a user. SOLUTION: A data item dictionary generating function 103 having recognized specifications extracts the definitions of individual processing for data items of an existent data item dictionary 101 by using a domain name as a key and acquires user specification information 106 as the information characteristic of the user. Then a data item Japanese name table 107, a standard data item dictionary 104, a knowledge base 102, and user specification information 106 are inputted and used associatively with a data item dictionary automatically generating function 108 on which user specification information for automatically generating a data item dictionary 109 is reflected to hand the definitions of the existent data item dictionary 101 as property over to the automatically generated data item dictionary 109.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to improvement of software productivity, and automatically generates a data item dictionary in which software is divided into data item units, and a program specification is automatically generated using this data item dictionary. On how to do.

[0002]

2. Description of the Related Art A data item dictionary automatic generation method, for example, Japanese Patent Laid-Open No. 7-21068, discloses a data item dictionary automatic generation method in which the work of creating a data item dictionary responsible for data item componentization is mechanized. By doing so, it is possible to reduce the man-hours and simplification of the work, and improve the productivity and reliability of system development. In addition, in automatic data item dictionary generation, a standard data item dictionary defined by the domain name of the data item is searched to create a high quality data item dictionary that does not require specialized knowledge of individuals. It is possible. In addition, by registering the information unique to the business system in a file called user specification information, it is possible to reflect the information unique to the user in the data item dictionary, reducing the man-hours for the work of creating the data item dictionary, and reducing the data item dictionary. We are working to improve reliability. In addition, by creating a systematic data item dictionary,
By fixing the contents of the data item dictionary to the business contents of the system, it is possible to reuse across multiple similar business applications. As described above, the data item dictionary automatic generation method has high reliability and enables the productivity of the work of creating the data item dictionary that can meet the user's request to be improved.

[0003]

The conventional methods have the following problems.

The first problem in the prior art is to manually register the user-specific information in the user specification information when the data item dictionary is generated based on the user-specific information for the data items used in the business system. To do
This means that registering user-specific information is a difficult task. In order to provide the data item dictionary with user-specific information created to satisfy the user's request, not only the standard data item dictionary but also specialized knowledge about a method of automatically generating the data item dictionary is required. Further, since the user himself / herself needs to investigate the software component defined in the data item, there is no means for efficiently creating user-specific information. As a result, since the user-specific information is not sufficiently registered, the number of man-hours for the correction work for each data item dictionary that occurs after automatic generation of the data item dictionary or for the source program after the source program is generated is reduced. It became large, and productivity and reliability could not be improved sufficiently.

The second problem in the prior art is that the expression of Japanese names is abundant, so there are multiple Japanese names even for synonymous data items, and it is customary for users of business and data. It is not always possible to search for an appropriate domain because it has a unique name for the business it has used. In this case, the data item dictionary cannot be generated. Therefore, it is necessary to add a data item dictionary or user-specific information, and it has not been possible to sufficiently improve productivity and reliability.

A third problem in the prior art is that it becomes difficult to grasp the correction location of the diverted data item dictionary when diverting the existing data item dictionary, and the number of man-hours for creating the data item dictionary increases. Usually, when the data item dictionary needs to be corrected, only one item is rarely corrected, and a plurality of data items having similar definitions are often the correction targets. Then, modifying a data item having a similar definition is to specify the Japanese name of the data item dictionary to be modified. At this time, if the same domain name is named at the end of these data item Japanese names, the domain name and the correction content can be registered in the user specification information to make corrections to the data item dictionary all at once. Is possible. However, because the method of identifying the domain name to be corrected was a manual survey by the user, it took a lot of work to identify the Japanese name of the data item to be corrected in the data item dictionary and the domain name. As a result, productivity and reliability cannot be improved sufficiently.

The fourth problem in the prior art is that the automatic generation of the data item dictionary is executed by using the data item Japanese name table, the standard data item dictionary, and the user specification information as input, so that the source program generated from the data item dictionary is used. There is no means to reflect the correction contents of the source program in the data item dictionary when the correction is made directly to the. Therefore,
The inconsistency of information occurs between source programs. At this time, the data item dictionary cannot be used for maintenance.

Further, when there is an existing system created without using the data item dictionary, the conventional method cannot create the data item dictionary based on the system which is the existing asset. In addition, the maintainability of the system itself deteriorates because a data item dictionary that can maintain the existing system cannot be created. Even when the existing system is reconstructed and expanded using the data item dictionary, the number of work steps required to create the data item dictionary is large, and productivity and reliability cannot be sufficiently improved.

[0009]

In order to solve the first problem, the present invention performs the following processing. Focusing on the state or format of data that actually exists or is used, a data item dictionary that defines the name, attributes, and related processing of data is built in advance, and this is used as a standard data item. Name it a dictionary. Data item Japanese names are defined for the data items registered in the standard data item dictionary. The data item Japanese name used in the business is to be named by the word structure "entity name-attribute name-domain name", and "domain name" is registered as the Japanese name registered in the standard data item dictionary. When the user inputs the Japanese name of the data item to be created or modified, the data item Japanese name registered in the standard data item dictionary is searched using this as a key. After searching all the data item dictionaries, what kind of data item dictionaries are searched for in each standard data item dictionary is analyzed focusing on the definition contents, and the appearance frequency of the definition is calculated. As a result of the analysis, the definition of each standard data item dictionary is replaced with the definition with the highest appearance frequency in the searched data item dictionary. The standard data item dictionary whose definition has been changed is output to a file called user specification information as user-specific information in which the domain name is defined in Japanese name.

In order to solve the second problem, in the present invention,
The following processing is performed. If the domain name registered in the standard data item dictionary cannot be searched for the Japanese name of the data item, add the Japanese name of the unsearched data item to the domain name of the standard data item dictionary. The frequency of occurrence of the definition is analyzed and replaced with the definition with the highest frequency of occurrence of the definition. For the data item dictionary that is not searched, the Japanese name is defined as the domain name as it is and is output to the file as the user's unique information.

In order to solve the third problem, in the present invention,
The following processing is performed. In the definition contents of the data item dictionary, a document that summarizes the results of the above analysis in tabular format about the Japanese names and definitions of the data items,
Output to a file named analysis result. From the analysis result, the user can compare and examine the classification by the domain name and the classification by the definition content in all the data item dictionaries. As a result, the user should decide whether the definition of the data item dictionary should be modified based on the two classifications of the domain name and the definition content in the data item dictionary or the domain name of the Japanese name of the data item. You can judge.
Further, it is possible to recognize the range in which different definitions are made even with the same domain name.

In order to solve the fourth problem, in the present invention,
The following processing is performed. For the data items described in the source program, the Japanese name of the data item and the word indicating the definition of independent processing are searched, compiled in a table format, and output to a file named data item dictionary information. Based on this data item dictionary information, the definition of the upper data item dictionary managed by the domain name is generated, and is output to the file as user-specific information.

The present invention has the following effects.

In order to achieve the first object, the present invention provides
It has a function to search the domain name of the input data item dictionary, analyze the appearance frequency of the definition of the independent processing of the data item dictionary, and automatically generate information unique to the user. This is because it is considered possible to stochastically create user-specific information in response to a user's request by selecting one that defines a large number of definitions given to the user's data item dictionary. Therefore, the definition to be given to the domain name as the user-specific information is determined by selecting the definition with the highest appearance frequency from the definitions in the data item dictionary. This function of automatically generating user-specific information can reduce the work of registering the user-specific information in the data item dictionary.

Secondly, in order to achieve the object, the present invention comprises:
Even if a Japanese name that cannot be searched for a domain name is named, it has a function to generate user-specific information having a definition by using the definition of the existing data item dictionary used for generation as it is. The user-specific information may be given a special Japanese name. At this time, the Japanese name of the data item itself is used to generate the user-specific information. As a result, even data items with special names can be automatically registered in the information unique to the user, which can contribute to the improvement of the data item dictionary generation rate thereafter.

Thirdly, in order to achieve the object, the present invention comprises:
It has a function to output the analysis result to the document about the processing process when outputting the user-specific information. When the data item dictionary is managed by domain name, it is possible to clarify the inconsistency of the data definition in the data item dictionary and the correction range. That is, it is possible to provide the user with a clear definition of what the data item dictionary having the same domain name has. This allows
It is not necessary for the user to directly investigate, the corrected portion of the data item dictionary can be specified by the domain name, and the work man-hours for creating the data item dictionary can be reduced.

Fourthly, in order to achieve the object, the present invention comprises:
It has the function of automatically obtaining the data item dictionary information by acquiring the definition of the independent processing of the data item in the source program. The data item dictionary information has the same information as the data item dictionary in the definition of the independent processing in the source program, and can be used for the data item dictionary generation function that recognizes the specification. This makes it possible to generate a data item dictionary from the definition of the data item on the source program, which could not be done by the conventional method. In addition, even for an existing system developed without using a data item dictionary, a data item dictionary can be created directly from the information of the source program, and the source program can be used for maintenance, reconstruction, and expansion. Data item dictionary can be easily created.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a functional schematic diagram showing a system configuration for implementing the present invention.

The data item dictionary generation function (103) that has recognized the specifications inputs the existing data item dictionary (101), the standard data item dictionary (104), the knowledge base (102), and the user specification information (106) to obtain the data. Item dictionary (10
9) is generated and stored in a file called user specification information (106).

The data item dictionary generation function (108) reflecting the user specification information, which is a conventional technique, includes a data item Japanese name table (107) and a standard data item dictionary (10).
6), knowledge base (102), user specification information (10
6) is input to automatically generate the data item dictionary (109).

The data item dictionary generation function (103) recognizing the specification extracts the definition for the independent processing of the data item of the existing data item dictionary (101) by using the domain name as a key and makes the user unique to the data item dictionary. The user specification information (106) is acquired as the information. The present invention is used in cooperation with a data item dictionary automatic generation function (108) that reflects user specification information, which is a conventional technique. As a result, the definition of the existing data item dictionary (101) as an asset can be easily inherited to the automatically generated data item dictionary (109), which has the effect of promoting the reuse of the asset.

Prior to the description of the specific procedure for generating the data item dictionary according to the present invention, the data item dictionary (109), the standard data item dictionary (104), and the analysis result table (10).
5) The user specification information (106) will be described in detail.

Using FIG. 2, the data item dictionary (109)
Will be described. The data item dictionary (109) defines and stores, for each data item, a software component for independent processing relating to each data item. Here, the data item refers to the smallest unit of data that has a business meaning, such as name and date of birth. The independent processing is an operation that is performed only on the data item, and is a process that is not related to the data items other than itself.

The data item dictionary (109) comprises an arbitrary number of data item dictionary defining sections (201). Items to be defined in the data item dictionary definition unit (201) include a data item Japanese name (202) and a software component name (20).
3), input format (204), output format (205), and storage format (206). Data item Japanese name (202)
Defines the Japanese name given to the data.
The software component name (203) defines the name of the software component assigned to the data item.
The input format (204) defines a format in which the software component handles data in the input form or the input screen. The output format (205) defines the format in which the software component handles data on the output form or output screen. Storage format (206)
Defines the format in which software components handle data in databases and files.

The contents to be defined will be described with reference to FIG. 2B by taking the data item "work start date" as an example. In this example, the data item "work start date" is defined as the data item Japanese name "work start date" as the data item Japanese name (202). Software parts (2
In the item of 03), the name “date” of the software component used in the data item “work start date” is defined. In the input format (204), “YYYYMMDD” indicating the 8-digit half-width data format is defined. The storage format (205) is
"NNZY year ZM month Z" showing the data format of Japanese calendar full-width display
D day "is defined. In the storage format (206), “YYYYMMDD” indicating the 8-digit half-year data format is defined. By defining the above information in the data item dictionary definition unit (201), a single process to be performed on the data item "work start date" is defined. The data item dictionary (109) defines the data items handled by the system.
It is.

The standard data item dictionary (10
4) will be explained.

The "standardization of data" in the standard data item dictionary is to define target data in order to enable the data to be treated as a shared resource in system development and operation. Therefore, the data item definition section sets standards for the Japanese names of data items and establishes a naming method so that the data items can be consistent in meaning interpretation. In addition, in each definition part of software component name, input format, output format, and storage format, it is necessary to specify the data item so that it has stability not affected by application, system configuration, system software, corporate organization, etc. As far as possible, it defines the contents that can be decided from the data side, not from the application point of view.

Here, the name standard of the data item in the data item definition section will be described. Naming that allows data to be consistent is to prevent duplication of data,
It is a nomenclature that facilitates identification from similar data and recognition of the semantic content of data. As a name standard for realizing this, a method in which the word structure of the data item is "entity name-attribute name-domain name" is adopted. An entity is a concept that exists in the real world and is recognizable by humans. An attribute is a characteristic of an entity, and is modified to represent what the entity looks like. A domain is a value defined by a specific regulation and can be shared by multiple attributes. According to this name standard, data items can be named consistently by associating the attributes of entities in the real world with specific attribute value sets and defining the values. For example, in the case of the item name “work start date”, the entity name is “work”, the attribute name is “start”, and the domain name is “date”. This means that when the entity "work" is focused on the attribute "start", it has a specific value depending on the domain "date".
In this way, the value of the data item depends on the domain, and the determination of the domain that specifies the attribute of the entity as the value is the most basic of data definition.

The processing content exists in the real world, with emphasis on the actual existence or the state and format in which it is used, regardless of the processing content of the computer and how data is used in individual applications. It is tailored to the substance.

FIG. 3A shows the standard data item dictionary (10
4) shows an example of definition for the data item "date" in 4). In the data item dictionary definition part (201) of the data item “year / month / day”, the Japanese name of the data item is “year / month / day”, and it has versatility with respect to the software component, the input format, the display format, and the storage format. There is a high definition of single processing. The standard data item dictionary (104) is a data item dictionary (109) constructed by collecting definitions with high versatility in addition to “date”. (FIG. 3B) Using FIG. 4, the analysis result table (105)
Will be described. FIG. 4 is an example of the analysis result table (105). The record of the analysis result table (105) is
Domain name (401), data item Japanese name (40
2), software component name (203), input format (20
4), display format (205) and storage format (206). The analysis result table (105) is output as a result of performing the data item dictionary automatic generation function (103) that recognizes the specification. Existing data item dictionary (101)
The definition of can be recognized in the form of a list. In addition, the domain name (40
By outputting 1), the data item dictionary (10
It is a table for judging the definition performed in 9) from the domain name (401).

User specification information (106) using FIG.
Will be described. The user specification information (106) is the standard data item dictionary (104) to which user-specific information is added or modified. In the example, the standard data item dictionary “year / month / day” (501) contains software parts “year / month / day”, input format “YYYYMMDD”, and display format “NNZY year Z”.
“M month ZD day” and storage format “YYYYMMDD” (502) are defined. On the other hand, user specification information "date"
(503) is a user-specific display format "NYY.MM.DD"
This is an example changed to (504).

FIG. 6 is a flowchart showing the processing procedure of the data item dictionary generation function (103) which has recognized the specifications. This function is a process of analyzing the definition of the existing data item dictionary (101) and outputting the user specification information (106). The details of the processing method for realizing the present invention will be sequentially described with reference to FIG.

In the data item dictionary input step (601) of the present invention, the data item Japanese name is read from the existing data item dictionary (101) into the data item Japanese name table (701) as input information.

The operation data item name editing step (602) in the present invention will be described with reference to FIG. In the business data item name editing step (602), the business data item Japanese name described in the read data item Japanese table (701) is converted into the standard data item dictionary (1
Edit so as to conform to the data item name defined in 04). FIG. 7 shows a step (6) for editing the business data item name.
It is an example of the search data item Japanese name table (707) edited in 02). The reason why the business data item name editing process (602) is performed is to increase the search rate of the standard data item dictionary (104). This example is shown in FIG.
Is a search data item Japanese name table (707) for the data item dictionary Japanese name table (701).
"Employee appointment date" (702) "Employee appointment date"
(704) "Citizen's move-in date" (705) "Resident's move-in date (for this month)" (706) has the same search data item Japanese name and data item Japanese name . This is "employee appointment date" (703) "employee appointment date" (704) "citizen transfer date" (705)
This is because there are no editing processes to be performed for the four items of “Date of moving in / out of residents (current month)” (706). On the other hand, "work start date-Japanese calendar" (712) is converted to "work start date (Japanese calendar)" (709). This is because the process of “editing“ -Japanese calendar ”into“ (Japanese calendar) ”” is performed. Note that this is only one example, and in addition to this, in the process of creating the search data item Japanese name table (707), a process for facilitating the search of the standard data item dictionary (104) is performed.

In the search step (303) of the standard data item dictionary (104), the search data item Japanese name table (707) and the standard data item dictionary (104) are used as input information, and the data item dictionary Japanese name table ( The data item in the standard data item dictionary (104) is searched using the Japanese name of each data item described in 701) as a key.

The domain name search (603) in the standard data item dictionary according to the present invention will be described with reference to FIGS. 8 and 9. FIG. 8 is a standard data item dictionary (10) corresponding to the data item “employee appointment date” (708) described in the search data item Japanese name table (707).
It is an example of searching the data item of 4). In addition, FIG.
It is a definition example of a standard data item dictionary (104) to be searched. Search data item Japanese name table (707)
Japanese name table for search data items defined in (10
7) and the Japanese names of all the data items stored in the standard data item dictionary (104) are compared one by one from the end of the word, and a completely matched data item is searched. If there are multiple standard data items searched, the standard data item having the maximum length of the matched character string is used as the search result. In the example of FIG. 8, the input Japanese name for search “Employee appointment date” (7
08), the standard data item dictionary (104) is searched, and as a result, “year / month / day” (90
1), “Month and Day” (902), and “Sun” (903) 3
Search for one standard data item name (104). Therefore,
In this example, from the searched Japanese name (801), "year / month / day" (901) having the longest matching character string is the final search result.

In the present invention, the process of temporarily defining the user specification information with the retrieved domain name (605) and the process of temporarily defining the user specification information with the data item Japanese name (606)
Expands the user specification information (106) into a class and the existing data item dictionary (101) into an instance under the data item dictionary class (1001) on the knowledge base (102), and executes the process.

First, the process (605) of temporarily defining the user specification information with the retrieved domain name will be described with reference to FIG. The figure shows the object structure of the data item dictionary class (1001) on the knowledge base. For the search data item Japanese name “work date” described in the search data item Japanese name table (707), the domain name “date” (901) is acquired and the search is performed. The user specification information (106) that defines the domain name "year, month, day" as a Japanese name is added to the class "year, month, day" (1
002). Next, the data item dictionary “work start date” used in the search is defined as an instance “work start date” (1003) subordinate to the “year date” class. At this time, the user specification information “date” (10
02) The independent processing definition (1004) is undefined, and the independent processing definition (1005) of the instance "work start date" (1003) is that of the existing data item dictionary "work start date". . After that, "Date" is added to the domain name.
Similarly, the search data item Japanese name “employee appointment date” and “citizen transfer date” are also defined in the instances subordinate to the user specification information “date”. Of all the existing data item dictionaries (101), for the data item dictionaries (109) for which the domain name can be searched, the user specification information (106) similarly defining the domain name is used as a class, and the existing data item dictionaries (101) are set. Repeat the process of provisionally defining the instance.

Next, the process (606) for temporarily defining the user specification information with the Japanese name of the data item will be described with reference to FIG. The figure shows the object structure of the data item dictionary class (1001) on the knowledge base. The corresponding domain name is not searched for the search data item Japanese name “Citizen's transfer date month and month” described in the search data item Japanese name table (707).
In that case, the data item Japanese name "Citizen's move-in date" is regarded as the domain name, and the user-specific information "Citizen's move-in date this month" is defined as "Japanese" 1101) is provisionally defined as a class. Next, the existing data item dictionary used for the search "Citizen's transfer date, this month"
Define (1102) as an instance subordinate to that class. At this time, the definition of the independent processing of the user specification information “Citizen's transfer date, current month” (1101) (1103)
Is undefined, instance "Citizen transfer date this month" (1
The definition (1104) of the independent processing of 102) defines the existing data item dictionary “Citizen's transfer date, current month”. Thereafter, for the existing data item dictionary (109) for which the domain name could not be searched, processing for temporarily defining the user specification information (106) by the data item Japanese name as a class and the existing data item dictionary (101) as an instance repeat.

The user specification information correction process (608) according to the present invention will be described with reference to FIG. The figure shows
The object structure of the data item dictionary class (1001) on the knowledge base is shown. For each user specification information (106) provisionally defined in the class, the existing data item dictionary (10
The appearance frequencies of the definitions used in 9) are totaled. As a result, the definition with the highest appearance frequency is defined for each class. FIG. 12 shows the user specification information “date”
(1001) and "Citizens move in this month" (1101)
The data item dictionary definition correction process (608) will be described with reference to FIG. Subordinate to the class "Date" are three data item dictionaries "Work start date", "Employee appointment date" and "Citizen transfer date" (1209), which are used there. Of the existing software components, input formats, display formats, and storage formats, the one with the highest frequency of occurrence is defined as the class “date”. In this case, the software component “date” (1) that has the same definition in all existing data item dictionaries
201), input format “CYYMMDD” (1202), storage format “YYYYMMDD” (1204), and display format “NYY.MM.D” used in two of the three existing data item dictionaries.
“D” (1203) is the user specification information “date” (100
It becomes the definition of 2). Similarly, since the instance of the user specification information "Citizen's move-in date this month" is "Citizen's move-in date this month", the software component name "This month check date" Day "(1
205), input format "CYYMMDD" (1206), display format "NYY.MM.DD" (1207), storage format "YYYYMMDD"
(1208) is the definition of the user specification information "Citizen transference date, this month".

The user specification information output process (609) in the present invention will be described with reference to FIG. After the user specification information correction processing (608), the definition of the user specification information is output to a file. In the case of the example, the information held by "year / month / day" (1002) and "citizen transfer date / month / date this month" (1101) defined in the class as user specification information is edited into a file. As a result, as shown in the figure, "date" (1
301) and "Citizen's transfer date this month" (1302)
It is output as user specification information (105).

The analysis result table output process (610) in the present invention will be described with reference to FIG. For the definitions made in the existing data item dictionary (101), edit the retrieved domain name, data item Japanese name, software component, input format, display format, storage format into records and sort the records by domain name. What you did
Output to a tabular file like the example. In the case of item number 1 in the example, for the existing data item dictionary "work start date" subordinate to the domain name "date", the software component "date", input format "CYYMMDD", display format "NYY. MM.
"DD" and the storage format "YYYYMMDD" are defined. The blank column of the domain name of item No. 4 indicates that the domain name cannot be searched.

The above is the embodiment in which the user specification information is output by the data item dictionary generation function (103) which recognizes the specification.

The data item automatic generation function (108) reflecting the user specification information handled in the present invention is a conventional technique. Here, the processing result of the data item dictionary automatic generation function (108) reflecting the user specification information (106) automatically generated according to the present invention will be described. Roughly speaking, when the data item dictionary (109) is automatically generated for the data items registered in the data item Japanese name table (107), it is a processing step of outputting the data item dictionary based on the user specification information. .

Details of the data item dictionary generation function (108) reflecting the user specification information in the present invention will be described with reference to FIGS. 15, 16, 17, and 18.

FIG. 15 shows an example of the data item Japanese name table (107). The date of accommodation start (1
501), end date of accommodation (1502), reservation date (1503), reservation date and current month (1504),
Five items of accommodation amount (1505) are registered as Japanese names of data items that generate a data item dictionary.

FIG. 16) shows the user specification information (106), "date" (1601), "date and middle of month" (160).
It is an example of 2). In this example, the user specification information (106) output by the data item dictionary generation function (103) recognizing the specification is “year / month / day” (1301) and “citizen transfer date / month / current month” (1302). However, the Japanese name "Citizen's transfer date, current month" (1302) uses the data item Japanese name as it is and the user specification information (106).
Since it is generated in, it is not appropriate as a domain name.
Therefore, at this time, it is necessary to name a domain name considered appropriate at the discretion of the user. Here, it is assumed that "the date, the month, and the minute" (1602), which is considered to be appropriate, has been renamed to the domain name, and the following examples will be described.

FIG. 17 shows the object structure on the knowledge base (102) when the data item dictionary generation function (108) using the user specification information is performed.
Data item Among the data items in the Japanese name table, "accommodation start date" (1703) "accommodation end date" (17
04) The user specification information "date" (1701) is searched for in the "accommodation reservation date" (1705), and the user specification information "date" is included in the "accommodation reservation date this month" (1706). This month's portion "(1702) is searched. At this time, the data item dictionary of each item in the data item Japanese name table (107) has the contents in which the definition of the user specification information (106) is copied.

FIG. 18 shows the generated data item dictionary (1
09) is shown. Among the generated data item dictionaries, "accommodation start date" (1801), "accommodation end date" (1802), "accommodation reservation date" (1803),
"Accommodation reservation date this month" (1804), 4 items,
Since the user specification information “year / month / day” (1701) and “year / month / day and current month” (1702) are searched, the data item dictionary is output with the content based on the user specification information. Since the data item dictionary “accommodation amount” (1805) is not searched for in the user specification information, it is generated based on the standard data item dictionary.

The above is the description of the processing steps of the data item dictionary generation function using the user specification information.

FIG. 19 is a flow chart showing the second embodiment of the present invention. The source program analysis function (1902) of the present invention means a business source program (190
This is a function of inputting 1) and outputting to a file called data item dictionary information (1903). The data item dictionary information (1903) is used by the data item dictionary generation function (103) that recognizes the following specifications, and the business source program (1
The user specification information (106) reflecting the specifications of the data item 901) can be generated.

FIG. 20 shows a definition description example of a single process for a data item on the business source program. From this description, the description to be stored in the definition of the data item dictionary is read using the source program analysis function. For example,
From the description (2001) of the independent processing performed for the data item "work start date", the software component name "date", input format "CYYMMDD", display format "NYY.MM.DD",
Acquires the storage format "YYYYMMDD". The process is performed for all data items (2002, 2003, 2004)
Do the same.

FIG. 21 shows the source program analysis function (1
902) is data item dictionary information (1903) output from the business source program example of FIG. This data item dictionary information (1903) is stored in the same file format as the data item dictionary. From the contents of the definition description of the independent processing acquired by the source program analysis function (1902), for example, the description of "work start date" (2001), the Japanese name of the data "work start date" (2101) and software Part name “Date”, input format “CYYMMD
D ”, display format“ NYY.MM.DD ”, storage format“ YYYYMMDD ”
(2102) and store each in the same file format as the data item dictionary. All data items (210
3, 2105, 2107) are processed in the same manner and output as data item dictionary information (1903).

The data item dictionary information (1903) is a data item dictionary generation function (103) that recognizes the subsequent specifications.
Used as the input of. At this time, the same processing as the existing data item dictionary (101) used in the first embodiment is performed. Data item dictionary information (190
From 1), the processing contents and results performed are exactly the same as in the first embodiment.

The above is the description of the source program analysis function using the second embodiment.

[0057]

Since the present invention has the processing procedure described above, it has the following effects.

First, the function of automatically generating user-specific information makes it possible to improve the productivity of the data item dictionary. In addition, the user can register information unique to the user without having detailed knowledge of the data item dictionary automatic generation method, and there is no individual difference between creators.

Secondly, even if the naming of data items is special, it is possible to generate them by the function of acquiring the definition made in the existing data item dictionary and outputting it to the information unique to the user. It is possible to automatically generate a data item dictionary that has not existed, and it is possible to improve the generation rate of the data item dictionary.

Thirdly, the definition of the data item dictionary or the definition error, which was difficult to grasp as a whole, can be provided as clear information to the user by outputting it in a table format as the analysis result. It became possible.

Fourthly, the following effect can be obtained by the function of automatically generating user-specific information by utilizing the definition of the software component in the source program. First, it is possible to reduce the man-hours for maintaining the consistency between the definition of the data item in the source program and the definition registered in the data item dictionary. Furthermore, even in a system constructed without using a data item dictionary, a data item dictionary that can be used for maintenance, management, or system reconstruction can be created. In addition, when integrating multiple systems or subsystems that have been constructed separately, the man-hours required to integrate the definitions of the data items in each system into a single data item dictionary can be reduced to improve the productivity of subsequent programs. Maintenance and management can be facilitated. In addition, it becomes possible to easily extract the definition of the data from the system constructed as the prototype and use it in the undeveloped part of the system to improve the productivity of the program thereafter.

[Brief description of drawings]

FIG. 1 is a functional configuration of a system that implements the present invention.

FIG. 2 is a definition example of a data item dictionary.

FIG. 3 is a definition example of a standard data item dictionary.

FIG. 4 is a table format of an analysis result table.

FIG. 5 is a definition example of user specification information.

FIG. 6 is a flowchart showing a procedure for generating a data item dictionary.

FIG. 7 is a definition example of a data item Japanese name and search data Japanese name table.

FIG. 8 is an explanatory diagram of a method for searching a domain name from a data item Japanese name.

FIG. 9 is a definition example of a standard data item dictionary that is a search destination of a domain name.

FIG. 10 is a diagram showing an object structure of a knowledge base when a main name is searched.

FIG. 11 is an object configuration diagram of a knowledge base when a domain name is not searched.

FIG. 12 is a diagram showing an object structure of a knowledge base after a user specification information correction process.

FIG. 13 is an example of generation of user specification information.

FIG. 14 is an output example of an analysis result table.

FIG. 15 is a definition example of a data item Japanese name table.

FIG. 16 is a definition example of user specification information used for generating a data item dictionary.

FIG. 17 is a knowledge base object configuration diagram when a data item dictionary is generated by reflecting user specification information.

FIG. 18 is a generation example when a data item dictionary is generated by reflecting user specification information.

FIG. 19 is a functional configuration diagram when a source program analysis function is used. .

FIG. 20 is a description example of a source program to be analyzed.

FIG. 21 is an output example of data item dictionary information.

[Explanation of symbols]

101 ... Existing data item dictionary, 102 ... Knowledge base, 1
03 ... A data item dictionary generation function that recognizes the specifications,
104 ... Standard data item dictionary, 105 ... Analysis result table, 106 ... User specification information, 107 ... Data item Japanese name table, 108 ... Data item dictionary generation function reflecting user specifications, 109 ... Data item dictionary, 2
01 ... Definition format of data item dictionary, 202 ... Data item Japanese name definition area, 203 ... Software component name definition area, 204 ... Input format definition area, 205 ... Display format definition area, 206 ... Storage format definition area, 401 ...
Analysis result table domain name output column, 402 ... Analysis result table data item Japanese name output column, 501 ... Standard data item dictionary "year / month / day", 502 ... Standard data item dictionary "year / month / day" definition example, 503 ... User Specification information “date”, user specification information “date” definition example, 601
... existing data item dictionary input processing step, 602 ... business data item Japanese name editing processing step, 603 ... standard data item dictionary domain name search step, 604 ...
Step 605 of determining success or failure of matching domain name
… Temporarily defining user specification information with the searched domain name, 606 ... Temporarily defining user specification information with data item Japanese name, 607 ... Search processing end confirmation step, 608 ... User specification information correction processing step,
609 ... User specification information output processing step, 610
... analysis result table output processing step, 701 ... data item dictionary Japanese name table, 702 ... data item Japanese name example "employee appointment date", 703 ... data item Japanese name example "work start date- Japanese calendar, 704 ... Data item Japanese name example "Employee appointment date", 705 ... Data item Japanese name example "Citizen transfer date", 706 ... Data item Japanese name example "Citizen move out date" (For this month) ", 707
… Data item dictionary Japanese name table, 708… Search data item Japanese name example “Employee appointment date”, 709
... Search data item Japanese name example "Work start date (Japanese calendar)", 710 ... Search data item Japanese name example "Employee appointment date", 711 ... Search data item Japanese name example "Citizen" Date of move-in ", 712 ... Example of Japanese name of data item for search"Citizen's move-in date (for this month) ", 801 ... Example of Japanese name of standard data item searched, 901 ... Standard data searched Item Japanese name "Date", 902
… Standard data item searched for Japanese name “Monday”, 903
… Standard data item retrieved Japanese name “day”, 10
01 ... Knowledge object "data item dictionary", 1
002 ... Knowledge object "date", 1003 ... Knowledge object "work start date", 1004 ... Knowledge object "work date" definition example, 1005 ... Knowledge object "work start date" definition example, 1101 ... Knowledge object "Citizen's transfer date this month" (class), 1102
… Knowledge object “Citizen transfer date this month” (instance) 1103… Knowledge object “Citizen move date this month” definition example (class), 1104… Knowledge object “Citizen move date this month” definition example (instance),
1201 ... Knowledge object "year / month / day" Software component name modification example 1202 ... Knowledge object "year / month / day" input format modification example 1203 ... Knowledge object "year / month / day"
Display format modification example 1204 ... Knowledge object “year / month / day” storage format modification example 1205… Knowledge object “citizen transfer date / month / current month” software component name modification example, 12
06 ... Knowledge object "Citizen's move-in date this month" input format modification example, 1207 ... Knowledge object "Citizen's move-in date this month's" display format modification example 1208 ... Knowledge object "Citizen move-in date this month's" storage Format modification example, 13
01 ... User specification information generation example "year / month / day" 1302 ... User specification information generation example "citizen transfer date / month / month", 140
1 ... Analysis result table output record example "work start date", 1402 ... Blank output example when domain name is not searched, 1501 ... Data item Japanese name "Accommodation start date", 1502 ... Data item Japanese Name “Accommodation end date”, 1503… Data item Japanese name “Accommodation reservation date”, 1504… Data item Japanese name “Accommodation reservation date for this month”, 1505… Data item Japanese name “Accommodation” "Amount", 1601 ... User specification information definition example "Date", 1
601 ... Example of user specification information definition “year, month, day, current month”, 17
01 ... Knowledge base "year / month / day" definition example (user specification information) 1702 ... Knowledge base "year / month / day / current month" definition example (user specification information) 1703 ... Knowledge base "accommodation start date / month" definition example 1704 ... Knowledge base "accommodation end date" definition example, 1705 ... Knowledge base "accommodation reservation date" definition example, 1706 ... Knowledge base "accommodation reservation date, current month" definition example, 1801 ... Data item dictionary " Generation example of "accommodation start date", 1802 ... Generation example of data item dictionary "Accommodation end date", 1803 ... Generation example of data item dictionary "Accommodation reservation date", 1804 ... Data item dictionary "Accommodation reservation date""Currentmonth" generation example, 1805 ... Data item dictionary "accommodation amount" generation example, 1901 ... Business source program, 1902 ... Source program analysis function, 1903
... data item dictionary information, 2001 ... Example of description of independent processing definition of data item on source "work start date", 2002
… Independent processing definition description example of data item on source “Employee appointment date”, 2003… Independent processing definition description example of data item on source “Citizen transfer date”, 2004… of data item on source Individual processing definition description example "Citizen's transfer date this month" 2101 ... Data item dictionary information "Work start date" 2102 ... Data item dictionary information "Work start date" Definition example 2103 ... Data item dictionary Information "Employee appointment date", 2104 ... Data item dictionary information "Employee appointment date" Definition example, 2105 ... Data item dictionary information "Citizen transfer date", 2106 ... Data item dictionary information "Citizen transfer" “Date” definition example, 2107 ... Data item dictionary information “Citizen's transfer date, this month”, 2108 ... Data item dictionary information “Citizen transfer date, this month's” definition example.

Claims (4)

[Claims] 1. A means for storing software in a file called a data item dictionary, in which software is defined as a software component by dividing a single process for each data item into several types for each data item, and a data item dictionary is provided. The processing method in the system is designed in advance, and it has a standard data item dictionary that defines software parts for individual processing of each data item, and a definition that gives user-specific information for each business data is named a user specification information file. By storing the means and collating the Japanese name of the business data item defined in the existing data item dictionary that has already been created with the Japanese name of the data item in the standard data item dictionary, one character at a time is added to the end of the word structure. It has a means to search for a domain name and a business data item Japanese name that searches for the same domain name. Means to assume a class named user specification information class above the existing data item dictionary, means to define the domain name searched for in the Japanese name of the data item, and existing data item dictionary below the user specification information class Is defined as an instance, and the definition of the single process with the highest occurrence frequency in the existing data item dictionary defined in the lower instance of the user specification information class is determined as the definition of the single process of the user specification information class. A means to output the class definition to a file as user specification information, a means to search the domain name from the end of the Japanese name of the data item of the newly created system and the Japanese name of the data item defined in the user specification information, New by reflecting the user-specific specifications acquired from the user specification information having the searched domain name. A data item that recognizes the specifications of the data, characterized by outputting a data item dictionary to be applied to system development to a file and enabling the definition of the independent processing of the data item in the source program. Dictionary automatic generation method. 2. In claim 1, when an appropriate domain name is not searched from the data item Japanese name in the existing data item dictionary, the Japanese name itself of the data item is recognized as a new domain name, and the matching domain is recognized. A data item dictionary automatic generation method that recognizes data specifications, characterized in that a data item dictionary having names is classified into a set, and a data item dictionary in which a domain name is named as a Japanese name is tentatively defined above this set. . 3. A single processing definition having the highest appearance frequency is identified from a set of data item dictionaries classified from domain names, and a definition having a low appearance frequency different from that definition and a data item having the definition are identified. A method for automatically generating a data item dictionary, characterized by having a means for identifying and classifying a definition of a data item definition that can be inferred from the result of the analysis into a document called an analysis result. 4. The means for analyzing the definition contents of individual data items required to generate a higher-order data item from the description in the source program, and the analysis result can be defined in the data item dictionary. A data item dictionary automatic generation method that recognizes data specifications, which is characterized by generating a data item dictionary based on the Japanese names and definitions of the data items described in the source by converting the data items into various formats.