JP2785317B2 - Software standardization method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a software standardization method, and in particular, organizes and integrates existing software files that are not stored in a database and data items in the files, and creates a database system. Consistently support the entire conversion process,
Alternatively, a data dictionary is created by analyzing existing software, and the data dictionary is used to convert the existing software into a format that is easy to maintain, thereby enabling a software standardization method that can efficiently support database generation. .

(Prior art)

In conventional office software, data of the same content is scattered in a plurality of files, and most of the processing of each program is processing of transferring data from file to file.

Software of such a format generally has a case in which a plurality of file data having the same contents are scattered, and different names and attributes are set for those data items. For this reason, it is difficult to understand the contents of the software, and when changing the attribute of a certain data item, it is necessary to find out the data with the same contents from all the programs and correct it. Had a problem.

As a method for dealing with such a problem, there is a method of converting software of a conventional format into a database system and centrally managing data. In this method, in order to convert existing software into a database system, it is necessary to arrange and integrate files in the existing software and data items in the files to perform a logical design of a database.

As another method, the names and attributes of data items in the existing software are standardized and unified, registered in the data dictionary, centrally managed, and the standard data items registered in the data dictionary are added to the standard data items registered in the data dictionary. There is also a method for improving the efficiency of software maintenance work by replacing and standardizing data items.

As a device that supports the work by these methods, a tool that analyzes a program in existing software and outputs, based on the analysis information, a structural diagram of the program, information indicating a data flow, or the like. There is a tool that supports the tracking of the flow of data interactively based on the analysis information of a program.

The procedure for creating new database software is described in, for example, “Data-centric system design technique,
Yamada et al., The Expanding Software Revolution, Nikkei McGraw-Hill, pp. 185-201 (September 1986) ".

[Problems to be solved by the invention]

In the above conventional technology, the data in the existing software is organized and integrated to design a database, or the data in the existing software is organized and integrated to create a data dictionary, and this is used to utilize the data dictionary. It only streamlined some of the tasks of converting to data standard format, and did not consistently support the entire task. Therefore, most of the work is performed manually, and there is a problem in work efficiency.

It is an object of the present invention to remedy such problems and to consistently support the whole process of organizing and integrating data in existing software to design a database, or to reduce the data in existing software. A software standardization method that can improve the efficiency of database generation by organizing and integrating, creating a data dictionary, and using it to help convert existing software into a format that is easier to maintain. Is to provide.

[Means for solving the problem]

In order to achieve the above object, the software standardization method of the present invention extracts each data item having the same and the same name in a file record of existing software, and displays the data items on a display device,
Each name of each data item having the same name is converted to a standard name input from the input device and displayed on the display device,
A standardization method for existing software in a system that supports standardization of each data in software based on interactive operation of an input device and a display device, wherein relation information indicating each of the data items having a dependent relationship is represented by relation information indicating the dependent relationship. A first step of generating an expression using a standard name and displaying the same on a display device, and dividing the relation information based on the division specification input via the input device to use a data item ( Key data item)
And a second step of generating normalized relation information consisting of data items dependent on the key data item and displaying the normalized relation information on a display device, and integrating the normalized relation information having the same key data item among the normalized relation information. A third step of generating integrated relation information by giving a name input via the input device, and normalizing and integrating the file records of the existing software.

[Action]

In the software standardization method of the present invention, a database logical design operation from existing software, that is, (1) extraction of data items to be made into a database in existing software, and (2) normalization of data necessary for making a database (normalization relation) ), And (3) the computer consistently supports the entire operation including the creation of the logical schema of the database corresponding to each normalized relation.

This greatly improves the efficiency of database logic design work from existing software.

In the software standardization method of the present invention,
Work to standardize and unify data in existing software, ie, (1) extraction of data items to be standardized in existing software, (2) standardization of data items and registration in data dictionary, (3) data dictionary A computer is used to consistently support each operation of converting file record definition information and programs into standard formation using registered standard data items.

This makes it possible to efficiently create a data dictionary from existing software and standardize data in the existing software.

〔Example〕

 Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

 First, a first embodiment of the present invention will be described.

FIG. 1 is an explanatory diagram showing functions and processing procedures of a database generation support system in a first embodiment of a software standardization method of the present invention, and FIG. 2 is a database generation support system to which the software standardization method of the present invention is applied. FIG. 3 is a diagram showing an example of a data definition section of a source program stored in the existing program storage section of FIG. 1, and FIG.
FIG. 5 shows an example of a procedure section of a source program stored in the existing program storage section of FIG. 1, FIG. 5 shows an example of existing file record information created from the file record definition of the order file of FIG. 3, and FIG. FIGS. 3 and 4 show examples of data item related information generated from the source program, FIG. 7 shows an example of a display screen at the time of data standardization in the first embodiment of the present invention, and FIG. FIG. 9 is a diagram showing an example of standardized file record information created by converting existing file record information, FIG. 9 is an explanatory diagram showing functions and processing procedures of a normalized relation creating section in FIG. 1, and FIG. 10 is a relation extracting section in FIG. FIG. 11 is a diagram showing an example of data relation information extracted by FIG. 11, FIG. 11 is an explanatory diagram showing a method of creating first normal type relation information in the first embodiment of the present invention, and FIG. FIG. 13 is an explanatory diagram showing a method of creating the third normal relation information in FIG. 13, FIG. 13 is a diagram showing an example of normal relation information created from the data relation information shown in FIG. 10, and FIG. 14 is a normal form shown in FIG. FIG. 15 is an example diagram of logical schema information created based on the integrated relationship information shown in FIG. 14, and FIG. 15 is an example diagram of logical schema information created based on the integrated relationship information shown in FIG.

As shown in FIG. 2, the database design support system of this embodiment includes a display device 1, an input device 2, a memory 3, a CPU 4,
An external storage device 5 is provided to consistently support the entire operation of designing and integrating a database in existing software.

The functional configuration and processing procedure are shown in FIG.

In FIG. 1, 101 is an existing program storage unit, 102 is an existing program analysis unit, 103 is an existing file record information storage unit, 104 is a data item related information storage unit, 105 is a data item standardization unit, and 106 is standardized file record information. A storage unit, 107 is a normalized relation creation unit, 108 is a logical schema information creation unit, and 109 is a database logical schema information storage unit.

 Next, a processing procedure in this embodiment will be described.

First, the existing program storage unit 101 stores at least one source program in existing software to be made into a database.

This source program is shown, for example, in FIGS. 3 and 4, and shows a data definition part and a procedure part of the same program described in COBOL.

In FIG. 3, 301 to 303 are file record definitions,
304 is the file name, 305 is the level number, 306,309,310,31
4,315 is the data item name, 307,308,312,313 is the attribute, 311
Is a program name. In the attribute 307, “S” indicates a sign (+, −), “9” indicates a number,
“(6)” indicates the number of digits. “10” of the attribute 308 indicates the number of repetitions.

In FIG. 4, reference numeral 401 denotes a condition part of the procedure part, and 402 denotes a condition part.
409 409 are data item names.

Next, the existing program analysis unit 102 in FIG. 1 analyzes the existing source program in the existing program storage unit 101, extracts the file record definition information therein, and stores it in the existing file record information storage unit 103. ,
Analyzes the transfer and comparison relationships of each data item in the file record in the existing source program, associates the names and attributes of file data items presumed to have the same contents, and stores them in the data item related information storage unit 104 I do.

The existing file record information stored in the existing file record information storage unit 103 is represented as shown in FIG. 5 in a tabular form. In this embodiment, the information is based on the “order file” of the file record definition 301 in FIG. Creating. In this case, the file name column, level column, data item name column,
The contents of the repeat column and the attribute column are generated from the definition of each data item of the file record.

In FIG. 5, reference numeral 501 denotes the contents of a file name column generated from the “order file” of the file name 304 in FIG. 3,
502 is the content of the level column generated from “02” of the level number 305, 503 is the content of the data item name column generated from “I-JUCHU-CODE” of the data item name 306, and 504 is “S9” of the attribute 307.
(6) "is the content of the attribute column generated from"", and 505 is the content of the repeat column generated from" 10 "of the attribute 308.

The same file record information is generated from the “customer master” of the file record definition 302 and the “product master” of the file record definition 303, and is stored in the existing file record information storage unit 103.

The data item related information stored in the data item related information storage unit 104 is represented in a table format, for example, as shown in FIG. 6, and is generated from the source program shown in FIGS.

In FIG. 6, 601 and 602 are the contents of the data item name (name of the related data item) column of the data item related information;
3,605 is the contents of the file name (the name of the file to which the data item belongs) column, and 604,606 is the contents of the belonging program name (the program name in which the open connection is detected) column. The left column and the right column show data items whose contents are estimated to be equal.

For example, “I-KOKYAKU-CODE” of the data item name 601 and “K-CODE” of the data item name 602 correspond to the condition part 40 in FIG.
By comparing the contents at one place, it is presumed that the values are equal, and the values are set as shown in FIG. The “order file” of the file name 603 is set by including the “order file” of the file name 304 and the “I-KOKYAKU-CODE” of the data item name 309 in the file record definition 301 in FIG. . The same applies to the “customer file” with the file name 605. “MATSUPD” of the belonging program names 604 and 606 is obtained from the program name 311 in FIG.

When the contents are transferred and compared via the intermediate data, the contents are similarly stored in the data item related information storage unit 104.

Next, the data item standardization unit 105 inputs a standard data name and a standard data attribute to be used to make a database for the data items in each existing file record information stored in the existing file record information storage unit 103.
, And converts the existing file record information into the standardized file record information storage unit 106. In this case, data items in different file record information but having the same contents have the same standard name and standard attribute.

In this embodiment, the original name, attribute, desired file name, and program name of each data item are displayed on the display device 1, and standardized data names and standard attributes are given to each data from the input device 2. Related data estimated to have is obtained from the data item related information storage unit 104 and displayed simultaneously,
Supports giving the same standard name and attributes.

 The screen display method in this case is shown in FIG.

The left column of the screen displays the data item name (existing data item name), data attribute (existing attribute), belonging file name (file name), and belonging program name of the existing data. Enter the standard name (standardized data name) and the standard data attribute (standard attribute).

That is, "I-KOKYAKU-COD-COD-COD" of the contents 601 of the data item name column registered as the data item related information in FIG.
"E-" and "K-CODE" of the contents 602 of the data item name column related thereto, and the existing attribute, file name, and associated program name are displayed.
The file name and the affiliation program name are obtained from the corresponding existing file record information (FIG. 5).

In FIG. 7, the standardized data name "customer code" and the standard attribute "9"
(6) ”is specified. The same standard data name and standard attribute are given to two existing data items, but if different data are considered, different names and attributes are given. Is also good.

Also, when the standardized file record information stored in the standardized file record information storage unit 106 is tabulated,
For example, it is represented as shown in FIG. 8, and is created by converting the existing file record information shown in FIG.

In FIG. 8, reference numerals 801, 803, and 805 denote contents of a standard data item name column obtained by converting the contents of the data item name column of FIG.
Reference numeral 802 denotes the contents of the standard attribute column obtained by converting the contents of the attribute column of FIG. 5, and 804 denotes the contents of the file name column generated from FIG.

In the present embodiment, “I-KOKYAKU-CODE” 506 in FIG.
And 504, "S9 (6)" are converted into "801 customer code" and 802 "9 (6)" according to the designation in FIG. The same applies to the names and attributes of other data items.

Next, the normalized relation creating unit 107 in FIG. 1 divides, arranges, and integrates the standardized file record information in the standardized file record information storage unit 106, and creates normalized relation information to be made into a database.

The functional configuration is shown in FIG. 9 and includes a relation extracting unit 901, a relation information storing unit 902, a relation normalizing unit 903, a relation integrating unit 904, and an integrated relation storing unit 905.

Further, a processing procedure of the normalized relation creation unit 107 will be described.

First, the relation extraction unit 901 extracts, from each standardized file record information in the standardized file record information storage unit 106, data item information (relation information) configured for each file record, and stores the data item information in the relation information storage unit 902. I do.

The relation information stored in the relation information storage unit 902 is represented as a table, for example, as shown in FIG.

10, 1001, 1006 to 1008 denote the item numbers of the data relation information created from the standardized file record information of FIG. 8, 1002, 1005 denote the contents of the relation name column, 100
Numeral 3 denotes the content of the relation content column, and 1004 denotes the content of the original file name column, and indicates the relation information extracted from one piece of standardized file record information corresponding to each item number.

The relation name is a name for identifying the relation information, and is obtained from the 01-level data name in each standardized file record information in this embodiment. For example, 10
The “order data” 02 is obtained from the “order data” 803 in FIG.

The relation contents indicate data items constituting each relation information and their relations. “｛｝” In Fig. 10
Indicates that this part is repeated, and "A + B + C" indicates
Indicates that the relation contents are composed of A, B, and C data items.

Note that the relation content of 1003 is obtained from FIG. The “Order Details” of 805 in FIG. 8 is repeated (1
0) is specified, the repetition is specified for the three standard data items below (product code, order unit price, order quantity). The original file name column indicates the original file from which the relation information was extracted. For example, 10
The “order file” of 04 is obtained from the “order file” of 804 in FIG.

Next, the relation normalizing unit 903 divides each data relation information stored in the relation information storage unit 902 by using the display device 1 and the input device 2 so that the data relation information becomes a normal format data relation information. .

In this case, the normal form relation has three levels of a first normal form, a second normal form, and a third normal form.
The creation formats are shown in the following (1) to (3).

(1) Creation of First Normal Form Relation The first normal form relation is a relation configured so that the relation does not include another relation.

To put a general denormal relation into this form,
The part containing the repetition in the relation needs to be split into another relation. However, at the time of division, information should not be lost. That is, each data relation includes a key data item necessary to identify the relation.

For example, since “order confirmation data” 1005 in FIG. 10 includes a portion indicating repetition, it can be divided into two first normal-type relations of order confirmation data and order confirmation details as shown in FIG. The underlined data items are key data items. Further, since the relation of the order confirmation details cannot be identified only by the product code, it is necessary to include the order code as key data.

(2) Creation of a second normal form relation A data relation satisfying the first normal form relation and in which all data items in the relation are dependent on all key items for identifying the relation is defined as a second normal relation. This is called a shape relation.

That is, if there is a data item that depends only on a specific key item in the first relation having a plurality of key items, the second normal form cannot be obtained unless the relation is divided.

For example, the "order confirmation data" 1101 in FIG. 11 is a second normal type relation because the key item is unique. The “order confirmation details” 1102 has two key items, an order code, a product code, and a product code. However, the data items of the order quantity, order unit, and designated delivery date are all the two key items described above. , So it is in the second normal form.

(3) Creation of Third Normal Form A data relation that satisfies the second normal form and whose data items in the relation do not depend on data items other than key items is called a third normal form data relation.

A data relation having a data item subordinate to a data item other than the key item will not be in the third normal form unless divided.

For example, since the data item of the customer name in 1101 in FIG. 11 is dependent on the data item of the customer code in the data, the data in FIG. As shown, "Order confirmation data" and "Order confirmation data 2"
Needs to be divided into two relations.

Generally, in order to facilitate software creation and maintenance, it is better to make the data relation in the database the second normal form and the third normal form as much as possible.
Since the data relation is subdivided, the required number of file accesses increases and the performance may decrease.

In the present embodiment, since it is assumed that the divided data relations will be integrated later, it is assumed that the third normal data relation is once created as much as possible.

In the processing of the relation normalizing unit 903, the computer supports the processing of detecting a repetitive part and dividing it into a first normal form. However, the processing is divided into a second normal form and a third normal form. The key items and the new relation name required for the operation are manually specified by dialog.

Thus, the normal form relation information created by dividing the non-normal form relation information shown in FIG. 10 through dialogue is represented as shown in FIG. 13 in the form of a table.

In FIG. 13, reference numerals 1301 and 1302 denote normal-form relation information created by dividing the non-normal-form relation information 1001 in FIG. 10, and reference numerals 1305-1307 denote 1008 non-normal-form relation information. Also, 1303 is from 1006, 1304 is 1007
Is the data relation information created as it is.

The meaning of each column is almost the same as in FIG.
The relation content column in the figure has been replaced with a normalized relation content column. In the contents of the normalized relation, underlined data items indicate key data items.

Next, the relation integration unit 904 in FIG. 9 integrates the normalized relation information having the same key data item among the normalized relation information in the relation information storage unit 902 to create a data relation to be created in the database. Then, it is stored in the integrated relation storage unit 905. The name of the integrated data relation information is input from the input device 2.

The data relation information obtained by integrating the normal form relation information shown in FIG. 13 is represented as shown in FIG. 14 in a tabular form.

In FIG. 14, reference numerals 1401 to 1404 denote integrated relation information created from the normal form relation information in FIG.
1405 indicates the content of the relation name column, 1406 indicates the content of the integrated relation content column, and 1407 indicates the content of the original file name column.

Reference numeral 1401 denotes the normal relation information of 1303 and 1306 in FIG. 13, 1402 denotes the normal relation information of 1301 and 1305, and 1403 denotes the data relation obtained by integrating the normal relation information of 1302 and 1307. Information. In this case, duplication of the same data item is eliminated.
The integrated relation information 1404 is created by transcribing the normal form relation information 1304 as it is.

When the integrated data relation information is stored in the integrated relation storage unit 905, the logical schema information creation unit 108 in FIG. The logical schema information of the database corresponding to the relation is created based on the configuration and name of the file and the attribute information of each data item in the file record information in the standardized file record information storage unit 106. Accumulate in 109.

For example, when the logical schema information created based on the integrated relation information 1401 in FIG. 14 is tabulated, it is represented as shown in FIG.

In FIG. 15, reference numeral 1501 denotes the contents of the database name field obtained by specifying from the input device 2, and the 01-level data “customer data” 1502: a logical schema obtained from “customer data” of the integrated relation information 1405 in FIG. Information, 1503 is 14th
The attribute is obtained by detecting the standard attribute information in the standardized file record information in FIG. 8 from the original file name 1407 in FIG. The other 02-level data item names are 1406 in FIG.
Obtained from the contents of the integrated relation. Each attribute 1503 is obtained by searching for the standard attribute information of the data item information in the standardized file record information from the content of the original file name column in FIG.

As described above, in the present embodiment, it is possible to reduce the number of work steps by consistently supporting the entire database logical design work when converting existing software into a database system.

Further, in the first embodiment, when the same data in one file is defined by another name by the program, it cannot be detected that the data is the same, and the same data name, Since it is difficult to standardize and unify attributes, a method to solve this will be described.

In this solution, the function of the existing program analysis unit 102 in FIG. 1 is changed as follows.

That is, when extracting existing file record information,
Calculates the relative position and length from the beginning of each data item, detects data items with the same relative position and length from the file record information of the same file extracted from another program, and correlates them Then, it is stored in the data item related information storage unit 104. Further, similarly to the first embodiment, the file data items to be compared and transferred are also stored in the data item information storage unit 104 in association with each other.

Accordingly, the data item standardizing unit 105 can simultaneously present the file data items having the same substance to the user, and thus can give standardized and unified names and attributes to the same data.

Next, a description will be given of a method for coping with a software environment that allows a program to refer to the same file with different file names according to the second embodiment of the present invention.

FIG. 16 is an explanatory diagram showing functions and processing procedures of the database generation support system in the second embodiment of the standardization method of the present invention.

In FIG. 16, 101, 103 to 109 are the same as those in FIG. A job control parameter storage unit 1601 stores job control parameters for associating physical names assigned to each file entity. 1602 uses job control parameters to associate files having different names in different programs. Software analysis unit.

In the first embodiment, a software environment is assumed in which each program uses the same file name for the same file. However, in the present embodiment, a software environment that allows a program to refer to the same file by different file names is used. Can be dealt with.

In other words, the existing software analysis unit 1601 analyzes the logical name of the file used in the program and the job control parameter that associates the physical name given to each file entity, or the information of the command procedure, and analyzes the information in separate programs. To associate files with different names.

Further, the same data items in the file record information representing the same entity are associated with each other using the analysis method, and are stored in the data item related information storage unit 104.

As described above, in the first and second embodiments, in order to detect file data items having the same contents scattered in a plurality of programs, a method of obtaining data to be mutually transferred and compared by data flow analysis of the programs is employed. doing.

Next, a method of inferring that data items having similar or matching names have the same content, presenting the same to the user, and supporting giving the same standard name and attribute will be described.

As this method, for example, in the existing program analysis unit 102 in the first embodiment, among the data items in each file record extracted from the existing program, those having the same or similar names are associated, and the data item related information is stored. It is conceivable that the information is accumulated in the unit 104. In addition, among the data items in each file record extracted from the existing program, not only those whose names are identical or similar but also those whose attributes are identical or similar are detected and stored in the data item related information storage unit 104. An accumulation method is also conceivable.

This increases the possibility that data items having the same contents are actually stored in the data item related information storage unit 104,
In the data item standardizing unit 105, the user is less likely to be guided to standardize and unify the names and attributes of data having no identical contents.

Further, when the related data is detected in the existing program analysis unit 102, the certainty that the related data has the same content is obtained, and the certainty information is stored in the data item related information storage unit 104, and the data item standardization unit As a method 105, a method of presenting the certainty information to the user together with the name of the related data and the like can be considered.

Thereby, the user can easily determine whether or not the related data has the same content, and can give a correct standard data name and attribute. As a method of determining the certainty, for example, when two data items are simply compared, a lower value is set than when the data is transferred, or when the names are merely similar, furthermore, It is conceivable to set the certainty low.

Next, a method in which the normalized relation creation unit 107 supports the process of identifying a keyword item for identifying data relation information from the data items of the data relation information will be described.

For example, in the existing program analysis unit 102 of the first embodiment, a data item that may be a key is estimated by program analysis, and this information is stored in the existing file record information storage unit 103. Further, the data item standardization unit 105 transfers this information to the standardized file record information storage unit 106, and the normalized relation creation unit 107
When extracting the data relation, this information is added to the target data item, and the user is informed that there is a possibility that this is a key data item.

According to this method, the key data item estimation in the existing program analysis unit 102 is performed by, for example, VSAM (virtual st
orage access method) file can be easily estimated because it is specified in the file record definition. Also, data items in a file record that are compared with data items in another file record and that affect control logic in the program are likely to be key data items. Also, “… CODE”,
A data item having a name such as “… KEY” is also likely to be a key data item. A key data item is estimated using these pieces of information.

In addition, since the certainty of the key item estimation differs depending on the case, when presenting to the user, the certainty may be added with the certainty. This can be easily realized by obtaining certainty information by the existing program analysis unit 102.

In addition to the above, as a method of supporting data normalization, the relation normalizing unit 903 of the normalized relation creating unit 105 may support division into a second normal form and a third normal form. .

For example, it is detected that there are a plurality of identical data items in a plurality of data relation information, and these data items are separated from the respective data relation information to divide the data items into a second normal form and a third normal form. To help you.

From the above embodiments, operability can be further improved by accumulating / selecting each execution means according to the performance of the target existing software.

With these methods, logical schema information is generated from existing software for the purpose of converting all existing software into a database system.

Next, according to the third embodiment of the present invention, not only the case where existing software is converted into a database system, but also unification and standardization of data in a record and conversion of a file record structure into a normal form into a general file. The method used is described.

FIG. 17 is an explanatory diagram showing functions and processing procedures of a file design support system in a third embodiment of the software standardization method of the present invention. FIG. 18 is an example of new file record information in the third embodiment of the present invention. FIG.

In FIG. 17, reference numerals 101 to 107 denote the first embodiment (FIG. 1).
1701 is a logical schema information creation unit 10
Reference numeral 1702 denotes a new file record information storage unit which replaces the database logical schema information storage unit 109.

In this embodiment, the new file record information creation unit 1701
Creates new file record information in a general file record format and stores it in the new file record information creation unit 1702.

As in the case of the logical schema information in FIG. 15, the new file record information created from the normal form relation information and the standardized file record information obtained in the same manner as in the first embodiment and described in the COBOL language formation is It is represented as in Figure 18.

As a result, the data in the records of the existing software are unified and standardized, the file record information is converted into a regular format, and a new file is created.

 Next, a fourth embodiment of the present invention will be described.

FIG. 19 is an explanatory diagram showing functions and processing procedures of a database generation support system in a fourth embodiment of the software standardization method of the present invention.

In FIG. 19, 101 to 109 have the same configuration as in FIG. 1, and reference numeral 1901 denotes data dictionary information for extracting data item information in the logical schema information in the database logical schema information storage unit 109 by eliminating duplication. A creation unit 1902 is a data dictionary storage unit that stores the data item information extracted by the data dictionary information creation unit 1901.

In the present embodiment, the data dictionary information creation unit 1901
And a data dictionary storage unit 1902, and register the unified and standardized data item information in the data dictionary by the method shown in the first embodiment. As a result, the data information of the converted software can be centrally managed, and the maintenance work of the software can be made more efficient by using the data dictionary information.

In addition, it is conceivable to create data dictionary information by extracting data item information from the standardized file record information in the standardized file record information storage unit 106 instead of from the database logical schema information storage unit 109. .

 Next, a fifth embodiment of the present invention will be described.

FIG. 20 is an explanatory diagram showing functions and processing procedures of a database generation support system in a fifth embodiment of the software standardization method of the present invention.

In FIG. 20, 101 to 109 are the first embodiment (FIG. 1).
Reference numeral 2001 denotes a database physical information creation unit, and 2002 denotes a database physical information storage unit.

The database physical information creation unit 2001 stores, in addition to the information in the database logical schema information storage unit 109, the storage order of records in the database, a logical storage method such as an additional index, or information on a storage device in which the database is to be stored. The physical information to be physically allocated to the database is created by the input device 2, and stored in the database physical information storage unit 2002.

That is, in the first embodiment, the database logical schema information is created from the existing software information in order to support the logical design work of the database, but in the present embodiment, the functions of the first embodiment are added to the functions of the first embodiment. By adding an information creation unit 2001 and a database physical information storage unit 2002, physical characteristics of the database are given together with the logical schema information to create physical information of the database.

Thereby, the number of man-hours for designing the database can be further reduced.

As described above, in the first to fifth embodiments, the existing software is analyzed, and not only the names and attributes of the data items to be used are converted into the standard format, but also the file records used by the existing software are divided. It creates file record definitions that have been integrated and normalized, and logical schema information of the database.

Next, standardize and unify the names and attributes of data items in existing software, but do not perform normalization by dividing and integrating file records, maintain existing software,
To improve the efficiency of reuse, that is, to standardize and unify data names and attributes in existing software,
A method of extracting data names, attributes, and the like from existing software, converting them to a standard format, registering them in the data dictionary, and using this to convert data items in the existing program to the standard format will be described.

FIG. 21 is an explanatory diagram showing functions and processing procedures of a data dictionary creation / use support system in a sixth embodiment of the software standardization method of the present invention, and FIG.
FIG. 23 is an example of standard data item information created by the data item standardization unit shown in FIG. 23. FIG. 23 is an example diagram of new and old data related information created by the data item standardization unit of FIG. 21. FIG. 24 is data replacement in the program of FIG. FIG. 25 is an example diagram of the procedure definition of the standardized source program created by the data replacement unit in the program of FIG.

In FIG. 21, 101 to 104 have the same configuration as in FIG. 1, 2101 is a data item standardization unit, 2102 is a data dictionary, 2103 is a data item information storage unit, 2104 is a new and old data item related information storage unit, 2105 Is a file record information standardization unit, 2106 is a standardized file record information storage unit, 2107
Is a data replacement unit in the program, and 2108 is a standardized program storage unit.

The data item standardization unit 2101 has a function different from that of the data item standardization unit 105 in FIG. That is, a standard name and attribute are given to each data item in the existing file record information, and the given standard name and attribute are stored in the standard data item information storage unit 2103 in the data dictionary 2102 while eliminating duplication. The related information between the standard data item information in the standard data item information storage unit 2103 and the corresponding data item information in the existing file record information in the existing file record information storage unit 103 is stored in the new and old data item related information storage unit 2104. Accumulate in

When assigning a standard name and an attribute, the same standard name and the same attribute are assigned to data items which are in different file records but have the same contents.

Also in this embodiment, the original name, attribute, affiliation file name, and program name of each data item are displayed on the display device 1 using the screen display method shown in FIG. Then, the standardized data name and the standard attribute are given from the input device 2, and the related data estimated to have the same contents are obtained from the data item related information storage unit 104, displayed simultaneously, and given the same standard name and standard attribute. Support.

In this embodiment, the source programs stored in the existing program storage unit 101 (see FIGS. 3 and 4)
Is analyzed by the existing program analysis unit 102, the file record definitions 301 to 303 in FIG. 3 are extracted and stored in the existing file record information storage unit 103, and the transfer and comparison of data items therein are analyzed. The related information of the data item estimated to have the same content is stored in the data item related information storage unit 10.
After storing the standard data item information in the data item information storage unit 2103 by the data item standardization unit 2101,
The new and old data item related information storage unit 2104 stores the new and old data item related information. The file record definition
Existing file record definition information created from 301 is the fifth
As shown in the figure.

The standard data item information is shown in FIG. 22, and the new and old data item related information is shown in FIG.

In FIG. 22, reference numeral 2201 denotes a standardized data name field for storing a standard name defined by the data item information storage unit 2103 for the name of each data item in the existing file record in the existing file record information storage unit 103. , 2202
Is a standard attribute column for storing standard attributes similarly defined for each data item in the existing file record. Reference numeral 2203 denotes a repetition column. The content of the repetition column 2203 is the number of repetitions of a data item which is a kind of attribute. In the present embodiment, the number of repetitions of the preceding data item is maintained. Reference numeral 2204 denotes a standardized data name, reference numeral 2206 denotes repetition, and reference numeral 2205 denotes a standard attribute.

For example, “customer code” of the standardized data name 2204 and “9 (6)” of the standard attribute 2205 in the item number “3” are the third
Data item name 309 in the file record definition 301 in the figure
“I-KOKYAKU-CODE”, attribute 312 “S9 (6)”, and data item name 310 “K-
In FIG. 7, a unified standardized data name and a standard attribute are given to the CODE "and the attribute 313" 9 (6) "using the dialogue screen shown in FIG. "Ten"
Are obtained from the attribute 308 “OCCURS10” in FIG.

In FIG. 23, reference numeral 2301 denotes a standardized data name column for registering the same content as the content of the standardized data name column 2201 in FIG. 22, reference numeral 2302 denotes a program name column 2303, and a file name column 2303.
4 and an original data item name column 2305. As a result, the program name having the data item before standardization for each standardized data name, the program name to which the standardized data item belongs, and the original data item name are registered.

For example, the "customer code" of the standardized data name 2306 of the item number "3" corresponds to the standardized data name 2204 "customer code" of the item number "3" in FIG. 22, and the program "MASTUPD"
This indicates that the two data items "I-KOKYAKU-CODE" in the "Order File" in "C" and "K-CODE" in the "Customer Master" are unified and standardized data items. .

In this manner, the present embodiment provides data dictionary creation support.

Next, a method of converting existing software into a standard format using the data dictionary 2102 of the present embodiment will be described.

The file record information standardization unit 2105 in FIG. 21 detects the standardized data item name from the new and old data item related information storage unit 2104 for the existing data item in each existing file definition information in the existing file record information storage unit 103. . Further, the attribute (including repetition) of the standardized data item is detected from the data item information, and the standardized file record definition information in which the name and the attribute of the existing data item are converted into the standard format is created. It is stored in the record definition information storage unit 2106.

In this embodiment, the existing file record definition information shown in FIG. 5 is converted into the contents of the standard file record definition information shown in FIG.

For example, “I” of data item name 506 of item number “3” in FIG.
-KOKYAKU-CODE "matches the standardized data name from the used location column of the new and old data item related information shown in Fig. 23 because the data belongs to the program" MASTUPD "and the file" Order file ". And finds that the standardized data name 2306 “customer code” of item number “3” is a standard data name.Furthermore, based on the data item information in FIG. It is detected that the standard attribute for “customer code” is “9 (6)”, and the standardized data name is changed to the standard data item name 801 “customer code”, as shown in item number “8” in FIG. The standard file record definition information in which the standard attribute is set to the standard attribute 802 “9 (6)” is created, and other data item contents are created in the same manner.

The in-program data replacement unit 2107 replaces the contents of the file record definition information in the existing source program in the existing program storage unit 101 with the corresponding standardized file record definition information in the standardized file record information storage unit 2106. At the same time, for each data item in the file record definition information, the reference location of the data item name in the source program is replaced with the corresponding standard data name. At this time, if necessary, it may be modified with a file name or the like so as to be unique.

FIGS. 24 and 24 show examples in which the source program shown in FIGS. 3 and 4 is replaced by the in-program data replacement unit 2107 using the standardized file record definition information in the standardized file record information storage unit 2106.
As shown in Figure 25.

In this embodiment, the file record definition 301 shown in FIG.
24 is based on the standardized file record definition information in the standardized file record information storage unit 2106 (see FIG. 8).
It is replaced by the standardized file record definition information 2401 in the figure. Also, the file record definitions 302 and 303 are similarly replaced with standardized file record definition information 2402 and 2403, respectively.

The data item reference positions 314 and 315 in FIG. 3 and 402 to 409 in FIG. 4 are all standardized because the corresponding data items are defined in any of the existing file record definition units 301 to 303. File record definition 2401 ~
Reference to the corresponding standard data item in 2403,
It is replaced as shown in 2404, 2405 in FIG. 24 and 2501-2508 in FIG. Of these, the “customer code” of 2404, 2501, 2502 and the “product code” of 2405, 2505, 2506 have the same data name, and therefore have unique file names (“order file”, “order file”). It has been converted to be qualified with "customer master" or "article master".

As described above, according to the present embodiment, it is possible to efficiently create a dictionary in which standardized data item names are registered, and use the created dictionary to use standard file code definition information or standard file code definition information in a program. Standardization and unification of data names and attributes can be performed efficiently.

In addition to the present embodiment, for example, a method in which an existing program is analyzed and only a data dictionary is created may be considered. According to this method, a part of the standardization and unification work of the data in the existing software can be made more efficient. Further, new software for sharing data using the created dictionary can be efficiently developed.

Furthermore, a method of creating standardized file record definition information from the created data dictionary information, or additionally, only converting data definition locations and reference locations in existing programs into standard names and attributes. Is also conceivable. With this method,
Part of the work of standardizing data in existing software can be made more efficient.

Further, in the sixth embodiment, when the data items in one file are defined with different names by the program in the same data item of the program, it can be estimated that those data items are the same. Can not. In order to solve this, it is conceivable to expand the functions of the existing program analysis unit 102 in FIG. 21 as in the database design support methods in the first to fifth embodiments. That is,
When extracting existing file record information, calculate the relative position and length of each data item from the beginning of the record, and from the same file record information extracted from another program, the data item with the same relative position and length And a function of detecting the data item and storing it in the data item related information storage unit 2104.

Thereby, the data item standardization unit 2101 can simultaneously present the file data items having the same substance to the user, and thus can easily give the same standard name and attribute to the same data item.

Next, in a software environment that allows a program to refer to the same file by different file names, by associating the same entity file, the entity among the data items between the same entity files used in different programs is Describes how to support the standardization of the same.

FIG. 26 is an explanatory diagram showing functions and processing procedures of a data dictionary creation / use support system in a seventh embodiment of the software standardization method of the present invention.

In the present embodiment, an existing software analysis unit 1602 of the second embodiment (FIG. 16) is provided instead of the existing program analysis unit 102 of the sixth embodiment (FIG. 26). A similar job control parameter storage unit 1601 is provided.

As in the second embodiment, the job control parameter storage unit 1601 stores a job control parameter or a command process that associates a physical name of a file assigned to each file entity with a logical name of a file used in a program. Accumulate information about Jure.

In addition, the existing software analysis unit 1602 associates files having different names in different programs by analyzing job control parameters or command procedure information together with program analysis, as in the second embodiment. Further, using the above-described method, the same data item in the file record information indicating the same entity is associated and stored in the data item related information storage unit 2104.

As in the database design support methods in the first to fifth embodiments, it is assumed that data items having similar or matching names are assumed to have the same contents and presented to the user, and the same standard name is determined by the user. And methods to assist in providing attributes. In this case, in the existing program analysis unit 102 in FIG. 21, among the data items in each file record information extracted from the existing program, those having the same or similar names are associated and stored in the data item related information storage unit 104. .

Similarly, a method is also conceivable in which, among the data items in each file record extracted from the existing program, a data item whose attribute matches or is similar to the data name is detected and stored in the data item related information storage unit 104.

Similarly, when the related data is detected by the existing program analysis unit 102, the certainty that the related data has the same content is obtained, and the certainty information is stored in the data item related information storage unit 104, and the data item standardization is performed. The unit 2101 may also present a method of presenting the name of related data and certainty information to the user. Thereby, the user can easily determine whether or not the related data has the same content, and can give a correct data name and attribute. Note that, instead of the certainty information, a method of presenting a basis for determining that the related data has the same content or a method of presenting the certainty information and the basis thereof may be considered.

As described above, the data dictionary creation and utilization methods in the sixth and seventh embodiments use the file record definition information extracted from the existing program in the dictionary creation operation.

Next, a data dictionary creation and utilization method that does not use file record definition information extracted from an existing program in the dictionary creation work will be described.

FIG. 27 is an explanatory diagram showing functions and processing procedures of a data dictionary creation and utilization system in an eighth embodiment of the software standardization method of the present invention. FIG. 28 is a diagram showing data created by the existing program analysis unit in FIG. It is an item related information example figure.

In FIG. 27, reference numeral 2701 denotes an existing program analysis unit, 2702
Denotes a data item related information storage unit, and 2703 denotes a data item standardization unit.

The existing program analysis unit 2701 of the present embodiment analyzes the existing program in the same manner as the existing program analysis unit 102, accumulates the existing file record definition information in the existing file record information accumulation unit 104, and presumes that it has the same contents. Information related to the data items is stored in the data item related information storage unit 2702. It should be noted that data item attributes and repetition information are added to the stored data item related information. An example of this data item related information is shown in FIG. 28, and is different from the example shown in FIG. 6 in that a repetition column and an attribute column are added in addition to a data item name, a file name, and a belonging program name.

Further, the data item standardization unit 2703 has a function different from that of the data item standardization unit 2101 shown in FIG. That is, a standard name and an attribute are given to the data item information in the data item related information storage unit 2702 using the display device 1 and the input device 2, and the given standard name and attribute are eliminated from the data dictionary 2102 by eliminating duplication. In the data item information storage unit 2103, the standard data item information in the data item information storage unit 2103 and the corresponding data item information in the data item related information storage unit 2702 (in FIG. The information indicating the relationship with the left or right half) is stored in the new / old data item related information storage unit 2104. When the standard name and the attribute are given, the same standard name is given to data items in different file records and having the same contents.

The other functions are the same as those shown in FIG.

In addition to the present embodiment, for example, a support method having only a function until a data dictionary is created can be considered. It is also conceivable that the above-mentioned alternative method is applied to the existing program analysis unit 2701.

In this way, the existing software can be maintained,
The efficiency of the reuse operation can be further improved. Also,
Since the file records are not divided or integrated, the conversion work of the existing program can be easily performed according to the data conversion.

〔The invention's effect〕

According to the present invention, when converting existing software to a database, operations such as extraction of data items to be included in the database, normalization of data, generation of a logical schema, etc.
Using information on existing software products, such as programs, can be supported effectively and consistently.

When converting the data in the existing software to the standard format, such as extracting data items to be included in the data dictionary, standardizing the data items and registering them in the data dictionary, standardizing the data of the existing software using this data item, etc. Work can be supported effectively and consistently using information of existing software products such as programs.

Therefore, the number of man-hours for these operations can be significantly reduced.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram showing functions and processing procedures of a database generation support system in a first embodiment of a software standardization method of the present invention, and FIG. 2 is a diagram of a database generation support system to which the database design support method of the present invention is applied. FIG. 3 is a diagram showing an example of a data definition section of a source program stored in the existing program storage section of FIG. 1, and FIG. 4 is an example of a procedure section of a source program stored in the existing program storage section of FIG. FIG. 5, FIG. 5 is an example of existing file record information created from the file record definition of the order file of FIG. 3, FIG. 6 is an example of data item related information generated from the source program of FIG. 3 and FIG. Seventh
FIG. 8 is a diagram showing an example of a display screen at the time of data standardization in the first embodiment of the present invention. FIG. 8 is a diagram showing an example of standardized file record information created by converting the existing file record information shown in FIG. 5, and FIG. FIG. 10 is an explanatory diagram showing functions and processing procedures of a normalized relation creating unit in FIG. 10, FIG. 10 is an example diagram of data relation information extracted by the relation extracting unit in FIG. 9,
FIG. 11 is an explanatory diagram showing a method of creating first normal type relation information in the first embodiment of the present invention, and FIG. 12 is a diagram showing a method of generating third normal type relation information in the first embodiment of the present invention. Illustrative diagram shown, FIG. 13 is an example diagram of normal relation information created from the data relation information shown in FIG. 10, FIG. 14 is an example diagram of integrated relation information created from the normal form relation information shown in FIG. 13, Fig. 15 shows 14
FIG. 16 is an example diagram of logical schema information created based on the integrated relation information shown in FIG. 16, FIG. 16 is an explanatory diagram showing functions and processing procedures of a database generation support system in a second embodiment of the software standardization method of the present invention, and FIG. Explanatory diagram showing functions and processing procedures of a file design support system in a third embodiment of the software standardization method of the present invention,
FIG. 18 is a diagram showing an example of new file record information in the third embodiment of the software standardization method of the present invention, and FIG. 19 is a diagram showing functions and processing procedures of a database generation support system in the fourth embodiment of the software standardization method of the present invention. FIG. 20 is an explanatory view showing functions and processing procedures of a database generation support system in a fifth embodiment of the software standardization method of the present invention, and FIG. 21 is a sixth view of the software standardization method of the present invention. FIG. 22 is an explanatory diagram showing functions and processing procedures of the data dictionary creation / use support system in the embodiment. FIG. 22 is an example diagram of standard data item information created by the data item standardization unit of FIG. 21, and FIG.
FIG. 24 is an example diagram of new and old data related information created by the data item standardization unit of FIG. 24, FIG. 24 is an example diagram of the data definition unit of the standardized source program created by the in-program data replacement unit of FIG. 21, and FIG. Example diagram of procedure section of standardized source program created by data replacement section in program
FIG. 26 is an explanatory diagram showing functions and processing procedures of a data dictionary creation / use support system according to a seventh embodiment of the software standardization method of the present invention. FIG. 27 is a diagram showing data in the eighth embodiment of the software standardization method of the present invention. FIG. 28 is an explanatory diagram showing functions and processing procedures of the dictionary creation / use support system. FIG. 28 is an example diagram of data item related information created by the existing program analysis unit in FIG. 1: display device, 2: input device, 3: memory, 4: CPU, 5: external storage device, 101: existing program storage unit, 102: existing program analysis unit, 103: existing file record information storage unit, 104: data Item related information storage unit, 105: data item standardization unit, 106: standardized file record information storage unit, 107: normalized relation creation unit, 108: logical schema information creation unit, 109: database logical schema information storage unit, 301 to 303 : File record definition, 304: File name, 305: Level number, 306, 309, 310, 31
4,315: Data item name, 307, 308, 312, 313: Attribute, 311: Program name, 401: Condition part of procedure part, 402 to 409: Data item name, 501: File name, 502: Level, 503, 506: Data item name, 504, 507: Attribute, 505: Repeat, 601,602: Data item name,
603,605: File name, 606: Affiliated program name, 801, 803, 80
5: Standard data item name, 802: Standard attribute, 804: File name, 9
01: Relation extraction unit, 902: Relation information storage unit, 9
03: Relation normalization part, 904: Relation integration part, 90
5: Integrated relation storage unit, 1001, 1006 to 1008: Item number, 100
2,1005: Relation name, 1003: Relation contents, 1004:
Original file name, 1101,1102: Relation name, 1301-130
7: normal relation information, 1401-1404: integrated relation information, 1405: relation name, 1406: integrated relation content, 1407: original file name, 1501: database name, 150
2: Logical schema information, 1503: Attributes, 1601: Job control parameter storage unit, 1602: Existing software analysis unit, 1701: New file record information generation unit, 1702: New file record information storage unit, 1901: Data dictionary information generation unit , 1902: Data dictionary storage unit, 2001: Database physical information creation unit, 2002: Database physical information storage unit, 2101: Data item standardization unit, 2102: Data dictionary, 2103: Data item information storage unit, 2104: New and old data item related Information storage unit, 2
105: File record information standardization section, 2106: Standardized file record information storage section, 2107: In-program data replacement section, 2108: Standardized program storage section, 2201: Standardized data name section, 2202: Standard attribute section, 2203: Repeat section, 2204: Standardized data name, 2205: Standard attribute, 2206: Repetition, 2301: Standardized data name column, 2302: Used location column, 2303: Program name column, 2304: File name column, 2305: Original data item name column, 2401 2403: standardized file record definition information, 2404, 2405, 2501 to 2508: standardized data item name, 2701: existing program analysis unit, 2702: data item related information storage unit, 2703: data item standardization unit.

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-1-134624 (JP, A) JP-A-2-83731 (JP, A) "The IPSJ 36th Annual Convention (early 1988)" Lecture Papers "P. 967-968 (58) Field surveyed (Int. Cl. ⁶ , DB name) G06F 9/06

Claims (12)

(57) [Claims] 1. Extracting data items having the same name but different names with the same contents in a file record of existing software and displaying them on a display device, and inputting the respective names of the data items having the same names. The standard name input through the device is converted into a standard name and displayed on the display device. Based on the interactive operation of the input device and the display device, the existing software in the system that supports standardization of each data in the software is A first standardization method, comprising: generating relation information indicating the subordination relation of each data item having a subordination relation in an expression using the converted standard name and displaying the relation information on the display device; The relation information is divided based on the division designation input through the device, and the data item ( A second step of generating normalized relation information including data items dependent on the key data item and displaying the same on the display device; A third step of integrating the contents of the existing software, giving a name input through the input device to generate integrated relation information, and normalizing and integrating the file records of the existing software. Characterized software standardization method. 2. The software standardization method according to claim 1, wherein said second step detects a data item repeated in said relation information, and divides said repeated data item to divide said data item. Generating standardized relation information. 3. The software standardization method according to claim 1, wherein the third step includes a step of eliminating duplication of each data item constituting the integrated relation information. A software standardization method characterized by including: 4. The software standardization method according to claim 1, wherein said second step includes a step of converting a data item including a specific character string into a data item name in said relation information. Detecting and displaying as a candidate for the key data item via the display device. 5. The software standardization method according to claim 1, wherein each name of the data items having the same and different names is converted into the standard name via the input device. Converting the attributes of each of the data items into standard attribute information input from the input device; and, based on the standard attribute information, converting data items in the integrated relation information into databases corresponding to the respective subordination relationships. Generating logical schema information of the software. 6. The software standardization method according to claim 1, wherein an existing program is analyzed, and the names, attributes, and structure information of a plurality of data items are extracted from the existing program. Extracting the file record information including, extracting each data item and attribute of the file record information, extracting the data item based on the synonym equivalent and converting it to a standard name, and performing each of the steps based on the converted standard name A software standardization method characterized by performing processing in a computer. 7. The software standardization method according to claim 1, wherein an existing program is analyzed, and the data items having the same name are identified based on the transfer / comparison relation of each data item. A software standardization method characterized by performing extraction. 8. The software standardization method according to claim 1, wherein an existing program is analyzed, and the file record of the data item in the file record referred to by a plurality of programs is analyzed. A software standardization method, comprising calculating a relative position and a length from the beginning of the data item and extracting data items having the same relative position and length as the data items having the same synonym. 9. The software standardization method according to claim 1, wherein a job control parameter or a command process for associating a logical name of a file and a physical name assigned to each file entity is provided. Software for analyzing any one of the data files, detecting and associating the same file referenced by different names from a plurality of programs, and extracting the data item having the same name and the same name from the associated file. Wear standardization method. 10. A software dictionary according to claim 5, wherein the data dictionary used for replacing the data item of the existing software by eliminating duplication from the data item of the logical schema information of the generated database. A software standardization method, comprising the step of: 11. The software standardization method according to any one of claims 1 to 9, wherein duplicates of the extracted data items having the same name are excluded from each data item of the existing software. Creating a data dictionary for use in replacing data items of existing software with the standard names. 12. The software standardization method according to claim 10 or 11, wherein the data item converted into the standard name is associated with the standard name in the data dictionary. A software standardization method characterized by registering related information including a name of a software.