JPH04188322A - Software specification information extracting method - Google Patents

Abstract

PURPOSE:To improve the efficiency of work to understand the specification of existing software by storing and outputting the information of a searching origin data item, that is, a generating destination data item and the information opposite to it of a searching destination data item, that is, a generating origin data item by relating them to each other. CONSTITUTION:In addition to the case that a program is analyzed and a value is only transferred to a data item contained in the program, in the case that the value is substituted by calculation, the data item to be the generating origin is searched as including the case of the intermediation of an intermediate data item. In this case, the information of the searching origin data item, that is, the generating destination data item and the information opposite to it of the searching destination data item, that is, the generating origin data item are stored and outputted by being related to each other. Accordingly, the data item to be the generating origin becomes capable of being automatically detected as including the case that the value is calculated and generated. Thus, the improvement of the efficiency of the work to understand the specification of the existing software at the time of reusing the existing software can be attained.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to improving software productivity, and in particular provides a software specification information extraction method suitable for understanding the specifications of existing software when reusing already developed software. Regarding.

[Conventional technology]

Examples of reusing existing software include cases in which the specifications of existing software are reused to develop new software, and cases in which existing software is preserved such as changes and function extensions.

In such work, it is first necessary to understand the specifications of existing software. Particularly in the case of office processing systems, it is important to identify data items contained in forms, screens, or mask files that users are directly aware of in their work, and to understand the generation relationships between these data items.
Important for understanding existing software.

This kind of information should originally be described in the business-level specifications of existing software, and if it is, it would be easy to understand the above information based on the specifications. It is.

However, business-level specifications for existing software may not exist, and even if they do exist, the content often does not match the software that is actually running, or the information described above is not included. , it was difficult to process tasks efficiently.

One possible way to solve this problem is to analyze downstream software products such as source programs that match current operations and extract data items that are of concern to operations and related information between data items. .

Conventional techniques related to this method include a method of analyzing source programs, job control parameters, etc., and detecting data transfer relationships between data items included in files, forms, and the like.

[Problem to be solved by the invention]

The above-mentioned conventional technology detects the transfer relationship between data items within existing software, and is a powerful technology that supports understanding the flow of data within existing software.

However, with the above-mentioned conventional technology, it is possible to detect and output the source data item whose value is transferred as is for the data item included in the software, but when the value is calculated and generated, it is difficult to It was not possible to detect and output the data item.

Furthermore, it was not possible to detect and output information on how to generate a value from the source data item.

Also, it detects that a certain generation destination data item is generated by a different generation method depending on the case, adds that information, and outputs the generation source data item information and generation method information from the generation source data item. I couldn't.

Also, if it is detected that a certain destination data item is generated using a different generation method depending on the case, condition information for using that generation method is detected, and that information is added to the source data item or the source data item. It was not possible to output the generation method information from .

Also, among files or tables defined in different programs, data items that share a physical location with each other are detected for data items in the same file or table, and this data item is also used as a target. It has not been possible to detect the generation source data item, or in addition to this, detect and output information on the generation method from the generation source data item across multiple programs.

Also, among files defined in different programs, physically identical files are automatically detected by analysis of job control parameters or command procedures, and among the data items belonging to the files, the physical locations are mutually shared. It is possible to search for the generation data item across multiple programs, or to detect and output generation method information from the generation data item, using the data item as a target. There wasn't.

Furthermore, it is not possible to facilitate understanding by converting the names of data items in the detected generation items or generation method information in accordance with standards, and then storing and outputting them.

Furthermore, by automatically or semi-automatically generating a new program using the generation method information, it has not been possible to streamline the creation of new rough software using the specifications of existing software.

The first object of the present invention is to create a data item that is a generation destination for a data item included in software, not only when the value is directly transferred from the generation destination data item, but also when the value is calculated and generated. The objective is to make it possible to detect and output information, and to streamline the process of understanding the specifications of existing software when reusing existing software.

A second object of the present invention is to make it possible to search for a generation destination data item for a data item included in software, as well as to detect and output information on how to generate a value from the generation destination data item. The goal is to streamline the process of understanding the specifications of existing software when reusing it.

A third object of the present invention is to detect that a certain generation destination data item is generated by different generation methods depending on the case,
By adding this information, it is possible to output the generation destination data item and generation method information from the generation destination data item, which in turn makes it possible to understand the specifications of existing software when reusing existing software. It's all about efficiency.

A fourth object of the present invention is that when it is detected that a certain generation destination data item is generated by a different generation method depending on the case, condition information for using that generation method is detected, and that information is added to the above-mentioned method. It is possible to output generation destination data items and generation method information from generation data items,
Furthermore, the goal is to streamline the work of understanding the specifications of existing software when reusing it.

A fifth object of the present invention is to detect data items that share a physical location among files or tables that are physically the same among files or tables defined in different programs, and to This makes it possible to search for, detect, and output generation method information from the source data item and destination data item across multiple programs, even for data items, and even reuse existing software. The goal is to streamline the process of understanding the specifications of existing software.

A sixth object of the present invention is to automatically detect physically identical files among files defined in another program by analyzing job control parameters or command procedures, and to extract data belonging to the files. Among the items, detect data items that share physical locations with each other,
Targeting this data item, it is possible to search for, detect, and output the generation destination data item and generation method information from the generation destination data item across multiple programs. The goal is to streamline the process of understanding the specifications of existing software when reusing it.

A seventh object of the present invention is to facilitate understanding of the detected generation items or generation method information by converting the names of data items therein in accordance with standards and storing and outputting the information. The objective is to streamline the process of understanding the specifications of existing software when reusing it.

An eighth object of the present invention is to make it possible to automatically or semi-automatically generate a new program using the detected generation method information, and to develop new software by reusing existing software. The aim is to make the process more efficient.

[Means to solve the problem]

In order to achieve the first objective of item 2, the software specification information extraction method of the present invention analyzes a program included in target software and transfers or calculates values for data items included in the program. Another data item to be substituted is searched for and detected, including through intermediate data items, and the information of the search destination data item (i.e., the generation data item) and the corresponding search destination data item (i.e., the generation data item) is detected. ) is stored or output in association with each other.

In order to achieve the second objective, the software specification information extraction method of the present invention searches for the generation source data item of the value for the generation destination data item in the software specification information extraction method described above, and also searches for the generation source data item of the value. It is characterized by searching and detecting information on how to generate a value from an item, and storing or outputting it in association with the generation source data item information.

In addition, the data items belonging to external files, lists, screens, tables, etc. that are the input and output of the target program can be used as generation destination data items, or in addition, as generation source data items, and the generation source data can be generated using the method described above. It is characterized by searching and detecting the generation source data item or, in addition to this, the generation method information of the value from the generation source data item for the item, and storing or outputting it.

In addition, the relative physical position within the record of the generated data item to be searched for or in addition to this, the intermediate data item is calculated, and another data item that shares the physical position with the data item is detected, and the above-mentioned process is performed. The method searches for and detects the source data item or information on how to generate a value from the source data item in addition to the source data item, including other data items, and stores or outputs the source data item. It is characterized by

Further, detecting still another data item that shares a physical location with a data item that is a generation source of another data item that shares a physical location with the generated light data item or intermediate data item, and by the method described above, Also targeting the other data item, search and detect the generation source data item for the generation destination data item of the search destination, or information on how to generate a value from the generation source data item in addition to this, and store or output it. It is characterized by

In addition, the relationship between the physical position and length of another data item that shares a physical position with the generated light data item or intermediate data item, and the data item that is the generation source of the other data item and this data item. Compare the physical position/length relationship of further data items that share the same physical position, and determine which of the corresponding further data items does not have a data generation relationship with the generated light data item or the intermediate data item. Except for the data item, the method described above searches for and detects the source data item for the search destination generation data item or information on how to generate a value from the generation source data item, and stores or outputs it. It is characterized by

In order to achieve the third objective, the software specification information extraction method of the present invention includes a method in which, in the above method, a certain generated light data item or intermediate data item is generated by a different generation method depending on the case. It is characterized in that it detects and adds the information, and stores or outputs a generated light data item for the source data item or information on how to generate a value from the generated data item in addition to the generated light data item.

In order to achieve the fourth objective, the software specification information extraction method of the present invention provides that when it is detected in the above method that the generated light data item or the intermediate data item is generated by a different generation method depending on the case, Detects condition information under which the generation method is used, adds the condition information, and stores or outputs the generation light data item for the generation destination data item or, in addition to this, the generation method information for the value from the generation source data item. Features.

In order to achieve the fifth objective, the method for extracting rough 1 to wear specification information of the present invention allows mutual physical locations of files or tables that are physically the same among files or tables defined in separate programs. Detect a shared data item, and use the method described above to generate a source data item for a destination data item, or a generated data item in addition to the source data item, using the method described above. It is characterized by searching and detecting value generation method information, and storing or outputting it.

In order to achieve the sixth object, the software specification information extraction method of the present invention automatically detects physically identical files among files defined in another program by analyzing job control parameters or command procedures, and Among data items belonging to a file, a data item that shares a physical location with each other is detected, and by using the method described above, the generation data item for the generation destination data item is generated across multiple programs. , or in addition to this, information on how to generate a value from a generation destination data item is searched and detected, and stored or output.

In order to achieve the seventh objective, the software specification information extraction method of the present invention replaces the data item name of the generation destination data item detected by the above method or the generation destination data item in the generation method information with an existing data item. It is characterized in that it is converted in accordance with the standard based on the correspondence information between the data item name and the standard data item name, and then stored and output.

In order to achieve the eighth object, the automatic software generation method of the present invention is characterized in that a new program is automatically or semi-automatically generated using the data item generation method information accumulated in the above method. .

[Effect]

In order to achieve the first object, the present invention analyzes a program and, in addition to another data item to which a value is transferred to a data item included in the program, also creates another data item to be substituted by calculation. Search including the case via intermediate data items, search destination data item name, generation destination data item -
The information of the search destination data item part or the generation destination data item corresponding thereto is stored or output in association with each other. Therefore, it is possible to detect the data item to be generated, including when a value is calculated and generated, which in turn streamlines the work of understanding the specifications of existing software when reusing existing software. There is a particular thing.

In order to achieve the second object, the software specification information extraction method of the present invention includes, in the software specification information extraction method described above, searching for a generation destination data item and detecting and storing information on a value generation method; It is stored or output in association with the generation destination data item information. Therefore, it is possible to detect and output information on a value generation method, and furthermore, the purpose is to make the work of understanding the specifications of existing software more efficient when reusing existing software.

In order to achieve the third objective, the software specification information extraction method of the present invention detects that a certain generation source data item or intermediate data item is generated by a different generation method depending on the case, and adds that information. , stores or outputs information on the generation destination data item for the generation destination data item and the method of generating a value from the generation destination data item. Therefore, it is possible to easily identify that the value of a generated data item is generated in different ways depending on the case, and furthermore, to streamline the work of understanding the specifications of existing software when reusing existing software. It is in.

In order to achieve the fourth objective, the software specification information extraction method of the present invention, when it is detected that the generation source data item or intermediate data item is generated by a different generation method depending on the case, determines whether the generation method is used. The condition information is added to the generation source data item for the generation destination data item, or in addition to this, the generation method information of the value from the generation destination data item is stored or output.

Therefore, it is easy to detect in which cases and by what generation method the value of a data item is obtained, and this also makes it easier to understand the specifications of existing software when reusing existing software. It is about becoming.

In order to achieve the fifth objective, the software specification information extraction method of the present invention uses files or tables that are physically the same among files or tables defined in separate programs to share physical locations with each other. Detects a data item, and searches for a destination data item for the destination data item across multiple programs, or detects information on how to generate a value from the destination data item, using the data item as a target. and store or output. Therefore, it becomes easy to detect the generation source data item and information on how to generate a value from that data item across multiple programs, which in turn makes it easier to understand the specifications of existing software when reusing it. You can streamline your work.

In order to achieve the sixth objective, the software specification information extraction method of the present invention automatically detects physically identical files among files defined in another program by analyzing job control parameters or command procedures. Among the data items belonging to the file, it is possible to detect data items that share physical locations with each other, and to search for the source data item for the destination data item across multiple programs that also target the data item.・Detection, or in addition to this, detect and store or output information on how to generate a value from a generation source data item.

Therefore, it is possible to search for generation source data items across multiple programs and to detect information on how to generate values from the generated data items more efficiently. The work of understanding the specifications can be made more efficient.

In order to achieve the seventh objective, the software specification information extraction method of the present invention extracts the data item name of the generation source data item detected by the above method or the generation source data item in the generation method information from existing data. Based on the correspondence information between the item and the standard data item name, it is automatically converted and stored in accordance with the standard. Therefore, for a certain data item, it is possible to efficiently obtain information on the generation source data item and generation method using the standard data item name, which in turn makes it possible to improve the efficiency of existing software when reusing existing software. The work of understanding specifications can be made more efficient.

In order to achieve the eighth object, the automatic software generation method of the present invention automatically or semi-automatically generates a new program using the data item generation method information accumulated in the above method. Therefore, it becomes easy to develop new software by reusing the specifications of the data item generation method used in existing software as is, and it is possible to efficiently develop new software by reusing existing data. I can do it.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a diagram showing a processing procedure in a first embodiment of the present invention, and FIG. 2 is a diagram showing a hardware environment of this embodiment.

The business specification information extraction method of this embodiment will be explained with reference to FIG. First, a job control program (hereinafter referred to as JCL) that describes the execution order of programs to be processed stored in an external storage device and information on files used during execution is input and analyzed (step 010
1).

Figure 3 shows an example of JCL, and Figure 4 shows an example of JCL.
An example of CL analysis information is shown below. rJCL analysis of Figure 1” (
Step 0101) analyzes the JCL shown in FIG.
The analysis information shown in FIG. 4 is created. For example, rPRGoloJ (0301) shown in Figure 3 is the execution program name, rsYsOl, defined in the rEXEC statement.
oJ (0302) is the name given to "rDD sentence",
Indicates the name for associating the file entity name indicated by rsHOHIN, MASTERJ (0303) with rPRGoloJ (0301). The analysis results for each element are shown in the example in Figure 4 with the program name rPRGO10J (04
01), DD name rsYsO101J (0402),
and file entity name rsHOHIN, MASTER
J (0403) and stored in the memory.

Note that the "selection" column is left blank in the JCL analysis (step 0101).

Next, specify the business data shown in the example in Figure 1 (Step 0
102). An example of the business data specification screen shown in FIG. 5 is shown. Specifying business data (0102) displays the business data selection designation screen shown in FIG. 5 from the JCL analysis information stored in the memory on the output device, and the user selects and instructs the file entity name that is business data. As a result, a selection designation symbol is set in the "selection" column shown in FIG. 4 in which the corresponding data body hair is stored. For example, the program name rPRGo10J (04,01
), I) D name rsYs0101J (04-02).

File entity name rsHOHIN, MASTERJ
(04-03) are 0501 and 0 respectively in the example of Figure 5.
Output as 502°0503. For example, the user may enter an r I to indicate that the file entity name is input to or output from the system as business data.
], J (0504,) or "01" (0505) indicating output. The specified selection indicator is the file entity name, program name,
It is stored in the "selection" column of the JCL analysis information shown in the example of FIG. 4 that can be identified by the DD name (0404°0405).

Next, for example, a COB OL source program is analyzed based on the program name in the JCL analysis information (step 0103). In this embodiment, the program analyzed in step 0103 will be explained below using the C0BOL language as an example. Figure 6 is an example of a C0BOL source program, Figure 7 is an example of file definition analysis information of source analysis information, Figure 8 is an example of data definition analysis information of source analysis information, and Figure 9 is an example of data definition analysis information of source analysis information. An example of the analysis information of the procedure part of the source analysis information is shown below. Program analysis shown in the example in Figure 1 (
Step 0103) inputs a COB OL source program with the same name as the program name in the JCL analysis information shown in Figure 4 from an external storage device, and uses the analyzed information to create a file definition (Figure 7) and a data definition (Figure 8). For example, it is stored in the memory every two procedures (FIG. 9).

For example, a COBOL source program having the program name "rPRGO20" shown in FIG. 4 is input from an external storage device (as shown in the example of FIG. 5).

rPRGO2oJ (06o1) in FIG. 5 indicates the program name, and the program name is divided into the program name column (0701) shown in FIG. 7 and the program name column (0701) shown in FIG.
0801), program name 1fjB (09
01) respectively. rSELEC shown in Figure 6
The TJ sentence rI Nu-F I LEJ (0602) is
This is the file name used within the program, and the analysis result is rsYsOlolJ in the same -rslELEcTJ statement.
Along with the rD DJ name indicated by (0603), it is stored on the same line in the "File Name" column (0703) and the "rDD Name" column (0702) of the analysis result shown in FIG. Further, "selection" information is obtained from the JCL analysis information having the same information as the program name and DD name analyzed, and is stored in the "selection" column of the program analysis information. For example, the program name rPRG○20J・ (0706), D shown in FIG.
The D name rsYsO30J (0707) corresponds to the program name (0406) and DD name (04,07) in Figure 4, and the information "01" in the "Selection" column (04,05) in the relevant line
is stored in the "selection" column (0708) in FIG. 7th
The "Record Name" column (0704-) shown in the figure is obtained from the name of the "01" level data item defined subordinate to the file definition that can be identified by rFDJ in FIG. For example, the file name r I N identified by rFDJ in FIG.
]-F I T-E J (0604), record name subordinate to the file rI Nu-RECJ (06
05), information with the program name rPRGO20J and file name rIN1-FILEJ is detected from the file definition information shown in the example in Figure 7, and the same line (070
9) Store the record name rIN1-RECJ. 8th
The analysis information of the data definition section shown in the figure stores analysis information of all data items defined within the program. Each data item stores the name of the record item to which the data item is subordinate, the file definition to which the record item is subordinate, and the name of the program to which the file definition is subordinate. Identifies the file name as blank if there are no dependent files. It also stores information indicating the area occupied by the definition of the file item in terms of the "position" indicating the distance from the beginning of the record and the "length" of the area. For example, the data item name rSURY○J (06
The definition information 07) is stored in the memory as analysis information (0802) in FIG. In addition, the "equivalent number" column is
Leave blank.

The analysis results of the procedure section shown in the example of FIG. 6 include output item names and generation formulas for generating output items regarding data movement and data generation by operations.

The generated items are stored in memory as analysis information.

For example, the analysis result of the formula (0609) shown in the example in Fig. 6 has the output item name rURI-GAKUJ (0902) and the generated content r (W-T, ANKA*5URYO) -N
EBIKIJ (0903) stores the starting line number of the procedure line as "165" (09o4) in the memory.

Creation of file equivalence information in the example shown in Figure 1 (step 0104)
) detects pairs of program name and DD name with JCL analysis information amount and file entity name, and for each, detects the same program name and DD name in the file definition information analysis information.
The information having the name is detected, and the information including the file name and record name in the information is added with the same number, for example, and stored in the memory as file equivalence information. FIG. 10 shows an example of file equivalence information. For example, the file entity name rTRANsO2, DATAJ (040
8,0409) is rPRGOlo.
J (0410) and rPRGO20J (0411), each with a DD name rsys030J (0410)
, rsYsOlo, + (0411). The program name.

Based on the combination of DD names, information having the same combination of program name and DD name is detected in the file definition analysis information shown in the example of FIG. In the example in Figure 7, 0710.07
09 is applicable, and the program name, file name,
and the record name in the “Program name” column (1) in Figure 10.
002).

“File name J column (1003), r record name” column (1
004), and the same number "1" is added to the "equivalent number" column (100, 1).

For example, the equivalence number is 1 for each detection of the same file entity name.
Count each time.

Creation of equivalent item information shown in the example of Figure 1 (step 0105)
) in the data definition analysis information for all data items subordinate to the record with the same equivalence number in the file equivalence information.
Data items having the same position and length are detected, and the same number, for example, is stored in the memory as an equivalent number. For example, regarding the information (1005, 1006) with the equivalence number "1" shown in the example of FIG. 10, in the data definition analysis information shown in the example of FIG. , “File name J.

Combination of “Record Name” (rPRGo 10゜0
UT-FILE, TR-RECJ or rPRGO2
0, INI-FILE, INI-RECJ) with the same "position" and "length" (for example, r15J, r5J of 0803°0802)
), since the same area on the same file entity is used, the data is the same, so the same number (
For example, 0803.0802 rlOJ) is the "equivalent number"
Add to column. The "equivalence number" is counted for each data definition analysis information of the same value.

Creation of output item related information shown in the example in Figure 1 (Step 01)
06) sequentially detects generation expressions and data items related to generation based on procedure division analysis information for data items belonging to files and records that have "selection" information indicating output business data in file definition analysis information. , is stored in memory as output item related information. FIG. 11 shows output item related information. For example, information having "selection" information indicating output business data is detected from the file definition analysis information shown in the example of FIG. In the example in Figure 7, roll (
0708) is applicable, and rPRGO20J (
0706), rOUT-FILEJ (0711), rO
UT-RECJ (0712) as “program name”,
At least one data item in the "file name" and "record name" is detected from the data definition analysis information. 8th
In the example shown in the figure, it corresponds to 0804, the program name of the information is rPRGO20J, and the data item name is rURI-GA.
Detects information that has KUJ in the "program name" and "output item name" of the procedure division analysis information. In the example of FIG. 9, rU
RI-GAKUj (0902) corresponds to "Program name J (090'7), r output item name" (090
2), r generation source item” (0903).

The "line number" (0904) is added to the level and stored in the memory as output item related information. In the example of FIG. 1-1, ]-],, OL corresponds to this. Next, for each data item name in the generation source item, information having the program name, the data item as the program name, and the output item name is detected from the procedure division analysis information. For example, the data item r W-TANKAj in the generation source item is set as "output item name", the program name rPRGO20J is set as "program name j", and the information it has is detected from the procedure division solution analysis information. In the example of FIG. 9, 0908 corresponds to this, the level is increased by 1 in the sense of going backwards up the generation path, and it is stored in the output item related information together with the detected information. In the example of FIG. 11, 11o2 corresponds.

Next, go back up the generation path in the same way as above, but since the generation source item r T A N K A J of 1102 is input information in the program name rPRGO20J, set the generation source item as the output item name. The information cannot be detected from the procedure division parsing information. If the data item cannot be detected from the procedure division analysis information, the data item is detected using the data definition analysis information, and information with the same "equivalence number" as the "equivalence number" in the detected information is detected from the data definition analysis information. do. In the example of FIG. 8, information 0805 corresponds, and the equivalent number “1” in the information corresponds to
1" is detected from the data definition analysis information. Information 0806 corresponds to the program name rPRG in the information.
OIOJ, item name rGAKUJ as “program name”,
Detects the information in "output item name" from the procedure division analysis information. In the example of FIG. 9, 0909 is applicable, and the level is increased by 1.''2 It is stored in the output item related information together with the detected information.

In the example of FIG. 11, 1103 corresponds. Next, the generation path is reversed in the same way as described above, but the generation source item rnI of 1103 is
Since N-TANKAJ is input information in the program name rPRGo10J, information that uses the generation source item as the output item name cannot be detected from the procedure division analysis information, and
Since the data item does not have an "equivalence number" in the data definition analysis information (0807), the process of retracing the generation path of the data item is completed, and the unprocessed "generation source item" in the output item related information is For the data item, the generation path is reversed in the same manner as described above, and the detected information is stored in the output item related information.

Output of item related information shown in the example of Fig. 1 (step 0107)
) edits the output item related information and outputs it to the output device. FIG. 12 shows an example of item-related information output to the output device. For example, among the information shown in the example of FIG. 11, the information whose level is "1" indicates the generation content of the data item subordinate to the business data indicating output, and the information with the largest level number among the information subordinate to the information Since this indicates that the generation source item is the end of the generation path, each data item in the generation item is replaced with the end data item of the generation path of each data item and output to the output device. 1 blue report 1] - The terminal data item of "W-TANKAJ" in the data item in the generator item of 〇1 is the generator item rH engineering in information 1102 with level "3" J, the data item is program name rPR, GO10J, item name rH
Information with "IN-TANKA" is detected from the data definition analysis information.

In the example of FIG. 8, "selection" information corresponding to information 0807 and selected as business data is obtained from the file definition analysis information having the "program name", "file name", and "record name" of the information. be able to. In the example in Figure 7,
The selection information "Work 2" of information o713 corresponds to this, and as a result, data item name rW-TANKAJ in the generation source item of information 1101 shown in FIG. 11 is changed to data item name rHIN-.
1”ANKA” and r1, which is replaced with r:I2J and output to the output device, indicating business data selection information.
202). Next, level "1" information of the output item related information and information subordinate to this information are output to the output device.

According to this embodiment, by specifying a file that has data necessary for a business, the generation path and generation formula of each data item in the file are made clear, and the maintenance tendency of existing programs and business information are made clear. This has the effect of making it easier to support system reconstruction and data analysis by extracting data item generation formulas.

FIG. 13 shows the processing procedure of the second embodiment of the present invention. In addition to the first embodiment shown in the processing procedure of FIG. By detecting data items that share a common target position and detecting source data items using these data items as targets, the detection accuracy of source data items, etc. is improved; (2) when a certain source data item is The feature is that it detects that different generation methods are used depending on the case.

The processing procedure of this embodiment is almost the same as that in FIG. 1, but the program analysis step 0103 in FIG. The step has been changed to information output step 1304. Other steps are the same as in the first embodiment.

Program analysis step 1301 differs from step 0103 of the first embodiment in that, in analyzing the procedure section of the program, it only analyzes the executable statement and extracts the generated light item whose value is to be assigned to the output data item and its generation method. Rather, by analyzing a conditional statement such as an IF statement that is executed before the executable statement, condition information for which the generation method is used is extracted and set in the procedure division analysis information.

Equivalent item information creation step 1302 is different from step 1302 of the first embodiment, in which data items that are physically assigned to the same position are determined by definition or redefinition based on the hierarchical relationship in the data structure, and data that has the same value is determined. Determine the item,
Set the equivalent data item number information in the data definition analysis information.

Item-related information output step 1304 differs from step 0107 of the first embodiment in that condition information for obtaining generation method information for the source data item based on procedure division analysis information is combined with condition information of procedure division analysis information and data definition. Obtained from the analysis information and set it in the output item related information.

Item related information output step 1304 differs from step 0107 of the first embodiment in that the condition information in the output item related information is added to the generation formula information and output.

Hereinafter, this embodiment will be explained in detail using an example.

JCL analysis step 0101 is the same as in the first embodiment.
Analyze CL (job control parameters) and create JCL analysis information. FIG. 14 is an example of JCL to be analyzed. This JCL is almost the same as the JCL illustrated in FIG. 3, but the part shown in 1401 - program PRGO15
The specified part of the execution and use file has been added, and the DSN (file entity name) as shown in 1402 has changed from rTRANsO2, DATAJ to rTRANsl.
o.

The difference is that it has been changed to DATAJ. An example of JCL analysis information created by analyzing the JCL in Figure 14 is shown in Figure 15.
As shown in the figure. This JCL analysis information also has 1501 parameters added and 150 parameters compared to the JCL analysis information in Figure 4.
2, the file entity name is [TRANsO2,DAT
AJ has been changed to rTRANslo and DATAJ.

In the business data specification step 010.2, as in the first embodiment, the business data selection specification screen displays the file entity name in the JCL based on the JCL analysis information, and selects the file containing the target generation source data item. , the file containing the data item to be detected as the generation source is specified, and the specified information is stored in the dry weight selection information column in the JCL analysis information.

FIG. 16 shows an example of the business data selection screen in this example.

The display and specified input on this screen are almost the same as in Figure 5.
The program rPROGO10 as shown in 1601
Files rTRANsO2, DATAJ and rTRANS used by J.

DATAJ is added and the file "rTRANSO2,,DATA" used by "PROG020" is rTR
The only difference is that ANsOlo and DATAJ are changed and displayed.

The program analysis step 1302 is almost the same as step 1301 of the first embodiment, although there is a difference that it is entered earlier, and the program analysis step 1302 is almost the same as step 1301 of the first embodiment.
File definition analysis information, data definition analysis information 9, and procedure division analysis information are created by analyzing the OL source program and JCL analysis information.

FIG. 17 shows an example of a target COBOL source program, which is almost the same as the example shown in FIG. 6, but the part 1701 is changed.

FIG. 18 is an example of file definition information obtained by analyzing three programs including the program in FIG. 17 and the JCL analysis information in FIG. 15. This is the same as the file definition information shown in FIG. 4, but a definition part 1801 of the file used by PRGO 15 is added.

FIG. 19 similarly shows an example of data definition analysis information created by analyzing three programs including the program in FIG. 17. However, in this step, the equivalent number column of this information is not set. This information is also the same as the data definition analysis information in FIG. 8, but the data definition section analysis information in PROGO 15 at 1901 is added.

FIG. 20 similarly shows an example of procedure section analysis information created by analyzing three programs including the program in FIG. 17. Unlike the procedure division analysis information in FIG. 19, the analysis information in FIG. 20 is a condition column indicating the conditions to which the generation method information is applied when the generation method information in the generation source item column is generated under different conditions depending on the case. has been added. In 2001, the output item rURI-GAKUJ with number]65 is r
NEBIKI C0DE= '1' If J conditions are met, rW-TANKA*5UURYO)-NEB
It is shown that it is generated by the IKIJ generation method. Also, in 2002, the same output items r UR D-G A, KU J were changed to “NEBIK■-CODE”.
= '2' If the J condition is met, rW-TANK
A-NEB IKI)*5UURY○" generation method.

File equivalence information creation step 0104 is a file defined by a different program based on JCL analysis information and file analysis information as in FIG.

Create file equivalence information indicating which files and records in the records represent the same entity. 1st
FIG. 21 shows an example of file equivalence information created from the JCL analysis information illustrated in FIG. 5 and the file definition analysis information illustrated in FIGS. 1-8.

This file equivalence information indicates the equivalence relationship between files and records defined in the three target programs by matching the numbers in the equivalence number column of this information. For example 21
01 and 2102 create the record "TR-R" of the file "○UT-FILEJ" of the program rPRGolOJ.
FC” and the file rI of the program rPRGo 15J
The record rTR-4ECJ of I'J-FILEJ is shown to have the same value by the same equivalence number 1.

Similar to the equivalence item information creation step 0105 of the first embodiment, the equivalence item information creation step 1302 in FIG. 13 sets equivalence numbers to data items representing the same entity among the data items in the data definition analysis information. . However, unlike the first embodiment, by definition or redefinition based on the hierarchical relationship in the data structure, data items that are physically assigned to the same position are determined and data items that have the same value are determined, and data definition analysis is performed. Set the equivalent data item number information in the information.

FIG. 19 shows data definition analysis information to which equivalence information is added for items that are equivalent based on the file equivalence item information of FIG. 21. For example, file rIN1 of 1904-
-FILE" data item rIN1-RECJ is the file rOUT-FIL of 1901 with the same equivalent number 2.
It is shown that the value is the same as the data item "○UT-RECJ" of EJ. Also, in 19o5, regarding the data item rIN1-RFC,1, which is the same as 1904, (15, 5) indicates that the data item 15 pie 1 ~ From the eyes to 5 by 1,
Equivalence number] 0 causes the same file r I of 1907
N1. - rSURY○ of FILEJ" and 15 bytes of data item rOU'1-RECJ of file 1902 "0UT-FILEJ" indicate that it is equivalent to 3 pi.

The output item related information creation step 1304 in FIG. 13 is similar to the output item related information creation step 0107 of the first embodiment, and includes generation method information and data definition analysis information between items in units of executable statements in the procedure division analysis information. From each output data item in the output file specified in the file definition parsing information, based on the data item equivalence information in the file definition parsing information, the value of that data item is The generation path of is sequentially searched in the direction of the generation source and recorded in the output item related information. Also, unlike 0107, regarding condition information, the generation path of the data item value included in the condition information is searched.

Figure 22 shows "0U" in the output file specified in Figure 18.
For the data items in record rOUT-RECJ in T-FI LEJ, the procedure division analysis information in Figure 20 and the first
An example of output item related information created using the data definition section analysis information in Figure 9 is shown. The description content of this information is almost the same as that in FIG. 11 of the first embodiment, and a condition column is added. For example, 2201 is output data item rUR by level 1.
The output record targeted by I-GAKUJ is “○UT-
Indicates that it belongs to R'ECJ, and r
When NEBIKI-CODE='1' J, r(W
It is shown that the value is generated by the generation method of -TANKA*5UURY○)-NEBIKIJ. 22
01 is the data item rW-T targeted by level 2
This indicates that ANKAJ relates to the data item rW-TANKAJ included in the generation method information 2201, and also indicates that the value is generated from rTANKAJ. 2203 is a data item rTAN in which 8 bytes from the 20th byte of the target data item rOUT-RECJ are included in the generation method information of 2201 due to level 3.
It shows that it is the same value as KAJ, and rIN-REC
It is shown that from the 20th byte of J, 8 pies 1 to 8 are generated. 2204 indicates that the target data item rGAKUJ is equal to the 20th byte to 8 bytes of the data item rIN-RECJ included in the generation method information of 2203, and rHIN-T
It's pathetic that it will be created by ANKAJ. rHI
Since N-TANKAJ is an intentional data entry included in the target input record (1806 in FIG. 18), no further search for the generation source is performed for this data item. In this way, starting from the generation method of the data item belonging to the target output file, the generation source of the data item in the generation method is searched until reaching the data item in the target output file. Furthermore, regarding the data items in the condition information, the generation source data item is searched until the target input file is reached and recorded in the output item related information.

Similar to the item-related information creation step 1304 of the first embodiment, the item-related information creation step 1304 in FIG. As a relation, generation method information is created and output/displayed. However, 13
Unlike 04, condition information to which the generation method is applied is also created and output displayed as a relationship with data items in the input file.

FIG. 23 shows an example of output/display information created from the output item related information shown in FIG. 22. □ For example, the generation formula information 2301 is created based on the item-related information below 2201 in Figure 22, and is based on the data item rUR'I-GAKU in the target output file.
J is the data item rHIN-TANKAJ and rHIN-8UJ in the input file "Engineering 2" (file INFILE of program PRGOIO in FIG. 18).
and rNEBIKIJ (HIN-TANKA:
I2*HIN-8U: l2)-NEBIKI: I
It has been shown that it can be obtained by the production method of 2J. The condition to which this generation method is applied as condition information is the data item rNEB IKI-CODEJ in input file ■2.
rNEBIKI−
CODE: I2='1' Indicated by J.

2302, 2303, 2304, 2305, etc. correspond to the contents below 2202 in FIG. 22, and are examples of individual item-related information in the program searched to obtain the generation method information of 1301. Condition information rNEBIKI-CODE:
Similarly, for I2='1' J, the generation method information in the program searched to obtain the condition information is displayed from the item related information.

In addition to the first embodiment shown in the processing procedure shown in FIG. Since it is possible to detect a shared data item and detect the generation source data item using this data item as a target, it is possible to improve the detection accuracy of the generation source data item, etc. The feature is that it is possible to detect and output that different generation methods are used for data items depending on the case.6 FIG. 24 shows the processing procedure of the third embodiment of the present invention.

In the example shown in FIG. 1, generation information for data items subordinate to specified business data is created, but in this embodiment, a new program is created from the generation information of business data. show.

In this embodiment, step 010 of the embodiment shown in the example of FIG.
6 to step 2401, and step 2402. Step 2403. Step 2404゜Step 240
This is the same as the embodiment shown in FIG. 1 except that 5 is added.

In this embodiment, in step 2401 shown in the example of FIG. 24, in addition to step 0106 of the embodiment shown in the example of FIG. Set the same equivalence number in the data definition analysis information. FIG. 25 shows an example of data definition analysis information in this embodiment. For example, in FIG. 11 of the embodiment shown in the example of FIG. 1, information 1102 is data item r
W -TANKAJ indicates that the data value of data item rTANKAJ is being moved, and data item "
W-TANKAJ rTANKAJ let, indicates that they have the same data value. The "equivalent number" shown in the example of FIG. 25 is such that when the data values of data items are the same, the same number is added, so the same number is added to the data items (2501 and 2502). Similarly, from the information 1103 in FIG.
It turns out that ANKAJ is also the same value, and the equivalence number "11" shown in FIG. 25 is added (r2503). If an abnormal "equivalent number" has already been set, set a new [equivalent number] to each "equivalent number" and the "equivalent number".
By replacing all "equivalent numbers" with
Save the equivalence of configured data items.

The standard name setting (step 2402) in Figure 24 uniquely sets data item names among data items with the same data value based on the data definition analysis information created up to the output item related information creation (step 2401). By setting, data item names are standardized. Figure 26 is an example of the standard base setting screen displayed on the output device, Figure 27 is an example of setting the standard data name in the data definition analysis information and storing it in memory,
Figure 8 shows an example of related information between the standard data name and the generation source item. For example, the target data item rsURYOJ in FIG.
(2601) edits and outputs the attribute of information 25o4 in FIG. 25 to the output device (2602).

Since the equivalence number is “10”, the equivalence number is “10”
(2505 in Figure 25).

2506) is edited and output to the output device as an equivalent data item (2603 in FIG. 26). When the user specifies a standard data name (2604), the standard data name is stored in the memory as the standard data name of the target data item and all of the equivalent data items. In the example of FIG. 27, the target data item rsURYOJ (2701) and the data items with equivalence number "10" (2702, 2703)
Each standard data name is rURIAGE-8URY○
”. In addition, the level "1" shown in the example of FIG.
The generation formulas or data items of data items subordinate to the business data are also replaced with standard data names and stored in the memory as related information between the standard data names and standard generated light items.

rURI-8UJrSURYOJ of the generated information 1104 are both standard data names rURIAGE-3UUR
As YOJ, the standardized generation information (2801) in Figure 28
to be memorized. Similarly, standard data names are set for other data items, and the same standard data names are set for data items having the same equivalent number.

Next, in file record specification creation (step 2403) in FIG. 24, a new file/record specification is created using standardized data names. FIG. 29 shows an example of a file/record specification defined using standard data names. Creation of program specifications shown in Figure 24 (step 240)
4. ) defines a program specification that refers to a file/record specification defined using a standard data name. FIG. 30 shows an example of a program specification.

Next, create the program shown in Figure 24 (step 2405)
Now, step 2403. File record specification defined in step 2404.

A program written in, for example, the C0BOL language is generated based on the program specifications and standardized generation information. Third
Figure 1 shows an example of a newly generated program. for example,
Program name rsPRG of the program specification in Figure 30
20J (3001), program I in Figure 31
Contains DrsPRG20J (3101), +' CO
B Generates an OL language program name definition line. The file definition part (3102) of the newly created program is the file definition part (3003 in Figure 30) of the program specification.
Generate from and record definition part (3103 in Figure 31)
is generated from the record definition part of the file/record specification having the same specification name as the file specification name in the file definition section (3003) in FIG. For example, the third
The second file with the file specification name rTFILEOIJ of figure 0
The file/record specification shown in the example in Figure 9 is Specification 2.
901 is applicable, and the record definition defined in the specification is stored in the file/record definition section (31
04). Similar generation is performed for other file/record definition sections. The procedure section of a newly created program is classified by the pattern name defined in the program specifications shown in Figure 30, and is created in advance for the main flow of the procedure section, extracted from the procedures stored in the external storage device, and each output It is generated along with the standard generation expression for each data item. For example, the pattern name rc shown in FIG.
ONVOIJ (3002) performs the main part of the procedure "Repeat reading records from input '117 file 1 and input file 2 and outputting them to output file 1 until there are no more records in input file 1 or input file 2." Classification name rcONVoI on external storage device
It is accumulated as J.

However, the generation formula for the data item in the record of output file 1 is not stored because the file specification name definition in the program specification is different.

The data name in the record definition of the file/record specification (2902 in Figure 29) that corresponds to the file specification name rTMAsTERJ that corresponds to output file 1 in the example of Figure 30, for example rURIAGE-GAKUJ (29
The standardized output item name shown in the example of FIG. 28 having the same data name as 03) is detected from the standard generation information. In the example of FIG. 28, information 2802 corresponds, and the standardized generation item of the data item is generated as the data item generation expression (3105) of the example of FIG. 31 as the generation expression of the data item.

According to this embodiment, by setting a standard data name, it is possible to set a standard data item generation method, a new program with high maintainability can be generated semi-automatically, and the efficiency of program creation is improved.

〔Effect of the invention〕

Since the present invention has the processing procedure specified above, it produces the effects described below.

Analyzes a program and searches for data items included in the program that are the source of generation, not only when values are simply transferred, but also when assigned by calculation, including when using intermediate data items. , the information of the search source data item - that is, the generation source data item - and the corresponding information of the search destination data item, that is, the generation data item can be stored or output in association with each other, so the value can be calculated and generated. This makes it possible to automatically detect the data item that is the source of the data, including when the data item is being generated, which in turn makes it possible to streamline the process of understanding the specifications of existing software when reusing it.

In addition, in the above method, information on how to generate a value from the source data item can be detected and stored in addition to searching for the source data item, and stored or outputted in association with the source data item information. Therefore, it becomes possible to automatically detect and output information on how to generate a value, which in turn makes it possible to streamline the work of understanding the specifications of existing software when reusing existing software.

In addition, in the above method, data items included in external files, two form screens, lists, etc. that are input/output of the target program are the above-mentioned generation data items, or in addition, data items from the generation source data items. Since value generation method information can be detected and stored or output, it is possible to automatically detect and output generation source information and generation method information for input/output data items that are of concern in business. This makes it possible to streamline the process of understanding the specifications of existing software when reusing existing software.

In addition, it calculates the relative physical position within the record of the generated data item that is the search source, or in addition to this, the intermediate data item, detects another data item that shares the physical position with the data item, and performs the classification. Targeting data items as well, it becomes possible to search and detect the source data item for the destination data item, or in addition to this, how to generate a value from the source data item, and store or output it. Therefore, it is possible to improve the detection accuracy of the generation source data item, etc., and furthermore, when reusing existing software, it is possible to streamline the work of understanding the specifications of the existing software.

In addition, the target data item or intermediate data item may also be targeted at another data item that shares a physical location with the data item that is the generation source of another data item that shares a physical location with the intermediate data item. Since it becomes possible to search and detect the generation source data item for an item or, in addition, information on the generation method of the data item, it is possible to further improve the detection accuracy of the generation source data item and generation method, and by extension When reusing software, it is possible to streamline the work of understanding the specifications of existing software.

In addition, the relationship between the physical position and length of another data item that shares a physical location with the generation source data item or intermediate data item, and the data item that is the generation source of the other data item and this data item. Compare the physical position and length relationship of further data items that share the physical position, and determine the data generation relationship with the original generation source data item or intermediate data item among the corresponding further data items. Excluding data items that do not have data items, it is possible to search and detect the source data item for the destination data item or, in addition, the generation method from the source data item. You can raise the degree further,
Furthermore, when reusing existing software, the work of understanding the specifications of existing software can be made more efficient.

Further, in the above method, it is detected that a certain generation destination data item or intermediate data item is generated by a different generation method depending on the case, and the generation source data item or generation method information is stored or generated by adding that information. This makes it possible to easily identify that the value of a destination data item may be generated in different ways, which in turn makes it easier when reusing existing software.
The work of understanding the specifications of existing software can be made more efficient.

In addition, in the above method, if it is detected that the generation source data item or intermediate data item is generated by a different generation method depending on the case, condition information under which the generation method is used is detected, and the condition information is added. This allows you to store or output information about the source data item or, in addition, how to generate a value from the source data item, so you can easily determine which generation method will be used to obtain the value of the data item in what situations. This makes it easier to detect whether or not the existing software exists, which in turn makes it possible to streamline the work of understanding the specifications of existing software when reusing existing software.

In addition, in the above method, files or tables that are physically the same among files or tables defined in different programs
Data items that share physical locations with each other are detected in the table, and using the method described above across multiple programs, the data items that share the physical location with each other are detected, and in addition to or in addition to the source data item for the destination data item. Since the generation method information from the generation source data item can be searched and detected, and stored or output, it is possible to detect the generation source data item and the generation method information of the value from the data item across multiple programs. This makes it easier to understand the specifications of existing software when reusing existing software.

In addition, in the above method, physically identical files are automatically detected among files defined in different programs by analyzing job control parameters or command procedures, and data items belonging to the files are automatically detected. By detecting data items that share a physical location and across multiple programs that also target these data items, the above method can be used to detect information about the source data item and how to generate a value from the source data item. , detection information can be stored or output, so it is possible to more efficiently detect generation source data items that span multiple programs and how to generate values from the generation source data items. When reusing existing software, it is possible to streamline the work of understanding the specifications of existing software.

In addition, the data item name of the generation source data item detected by the above method or the generation source data item in the generation method information is automatically changed to a standard name based on the correspondence information between the existing data item and the standard data item name. Since it is possible to store and output data by converting it into a standard data item name, information on the generation source and generation method for a certain data item can be obtained using a standard data item name. The work of understanding the specifications can be made more efficient.

In addition, since a new program can be automatically or semi-automatically generated using the data item generation method information accumulated in the above method, business specifications related to data item generation used in existing software can be automatically or semi-automatically generated. It becomes easier to develop new software by reusing existing software as is, and it is possible to streamline the development of new software by reusing existing software.

As described above, according to the present invention, it is possible to efficiently extract specification information such as how to generate data items in existing software, and further improve existing software when reusing existing software. This has excellent effects such as increasing the efficiency of understanding software specifications and developing new software by reusing existing software.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the processing procedure in the first embodiment of the present invention, FIG. 2 is a diagram showing the hardware environment of this embodiment, FIG. 3 is a diagram explaining JCL (job control program), Figure 4 shows an example of JCL analysis information, Figure 5 shows an example of a business data specification screen, and Figure 6 shows C0BOL.
An example of a source program written in the language, Figure 7 is an example of analysis information in the file definition section, Figure 8 is an example of analysis information in the data definition section, Figure 9 is an example of analysis information in the procedure section, and Figure 10 is an example of analysis information in the procedure section.
The figure shows an example of information indicating file equivalence relationships, FIG. 11 shows an example of output item related information, FIG. 12 shows an example of item related information output to the output device, and FIG. 13 shows the example of the second embodiment. Figure 14 is a diagram illustrating the processing procedure, Figure 14 is a diagram explaining JCL (job control program), Figure 15 is a diagram showing an example of JCL analysis information, Figure 16 is a diagram showing an example of a business data specification screen, Figure 17 is an example of a source program written in the C0BOL language, Figure 18 is an example of analysis information in the file definition section,
Figure 19 is an example of analysis information of the data definition section, Figure 20 is an example of analysis information of the procedure division, Figure 21 is an example of information indicating file equivalence relationships, and Figure 22 is an example of related information of output items. ,
Figure 23 is an example of item-related information output to the output device,
Figure 4 shows the processing procedure of the third embodiment, Figure 25 shows an example of data definition and section analysis information, Figure 26 shows an example of the standard data name specification screen, and Figure 27 shows the standard data name. An example of the analysis information of the set data definition section, Figure 28 is an example of a standard data item generation formula using a table data name, Figure 29 is an example of file/record specification information specified by a standard data name,
FIG. 30 shows an example of program specification information, and FIG. 31 shows an example of a program generated using standard data names and standard data item generation formulas. 0101...JCL (job control program) analysis step, 0102...Business data specification step,
0103...Program analysis step, 0104...
・File equivalence information creation step, 0105...Equivalent item information creation step, 0106...Output item related information creation step, 0107...Item related information output step, 0301...Program name, 0302...DD
Name, 0303...File entity name, 0401゜040
6...Program name, 0402,0407...DD
Name, 0403,0408,0409...File entity name, 0404,0405...Business data selection ID, 0
410,0411...Pair of program name and DD name, 0
501...Program name, 0502...DD name, 0
503...File entity name, 0504゜0505...
・Business data selection ID, 06o1...Program name,
0602,06'04...File name, 0603...
・DI) name, 0605...Record name, 0607...
・Data item name, 0609...Data generation formula, 070
1゜0706--Program name, 0702,0707
--DD name, 0703,0711...File name,
0704. 0712...Record name, 0705,0708...
Business data selection I0.0709', 0710,0713
...File/record definition analysis information, 0801...
-Program name, 0802-0807...Data item analysis information, 0901,0907...Program name,
o9o2,0905...Output item name, 0903゜09
06...Generated light item, 0904...Number, 0908
. 0909...Procedure division analysis information, 1001...File equivalence number field, 1002...Program name field, 10
03...File name column, 1004...Record name column, 1005.

Claims (1)

[Claims] 1. At least one program is analyzed to transfer or calculate a value to another data item included in the program, including the case through at least one intermediate data item. 1. A method for extracting software specification information, characterized in that information on a generation source data item is detected with respect to information on a generation destination data item by searching, and the information is stored or outputted in association with the information on the generation source data item. 2. In the software specification information extraction method according to claim 1, the generation source data item is searched and information on a method of generating a value from the generation source data item to the generation destination data item is searched and detected; A software specification information extraction method characterized by storing or outputting information in association with data item information. 3. In the software specification information extraction method according to claims 1 and 2, a data item belonging to any one of an external file, list, screen, or table that is an input/output of the target program is a generation source data item or in addition to the generation source data item. Targeting a data item, using the method described above to search for and detect information on a generation data item, or in addition to this, information on how to generate a value from the generation source data item to the generation destination data item, and store or output it. A software specification information extraction method characterized by: 4. In the software specification information extraction method according to claims 1, 2, and 3, by analyzing the data definition information in the program, the generation target data item to be searched for, or in addition to this, the record of the intermediate data item. Calculate the physical relative position within, detect another data item that shares the physical location with the data item, and also target the other data item,
Software characterized by searching and detecting a generation source data item for a search source generation data item or additionally information on how to generate a value from the generation source data item using the above method, and storing or outputting the generated data item. Specification information extraction method. 5. In the method for extracting software specification information according to claim 4, a further data item that shares a physical location with a data item that is a generation source of another data item that shares a physical location with the generation destination data item or intermediate data item. Detects the data item, and uses the method described above to detect the source data item for the search source generation data item, or in addition to this, generates value generation method information from the generation source data item, also targeting the other data item. A software specification information extraction method characterized by searching, detecting, and storing or outputting. 6. In the software specification information extraction method according to claim 5, The relationship between the physical position and length of the data item that is the generation source and another data item that shares the physical location with this data item is compared, and among the corresponding further data items, the original generation data Except for data items that have no data generation relationship with items or intermediate data items, the method described above is used to generate values from the source data item or in addition to the source data item for the search source generation data item. A software specification information extraction method characterized by searching, detecting, and storing or outputting method information. 7. In the method for extracting software specification information according to claims 1, 2, 3, 4, 5, and 6, the above method extracts a generation source data item for a search source generation data item, or in addition to this, a generation source data item. In addition to searching and detecting information on the method of generating a value from a data item, searching and detecting that a certain generation destination data item or intermediate data item is generated by a different generation method depending on the case, and adding that information. A method for extracting software specification information, comprising: accumulating or outputting a generation source data item for a generation destination data item or, in addition to this, information on how to generate a value from the generation source data item. 8. In the software specification information generation method according to claim 7, when it is detected that the generation destination data item or the intermediate data item is generated by a different generation method depending on the case, condition information under which the generation method is used is detected. ,
A software specification information extraction method characterized by adding the information to store or output a generation source data item for a generation destination data item or information on how to generate a value from the generation source data item in addition to this. 9. In the software specification information extraction method according to claims 1, 2, 3, 4, 5, 6, 7, and 8, physically identical file/table data items among files or tables defined in separate programs are Among them, data items that share physical locations with each other are detected, and using the method described above, the data items that are the generation source for the generation data item of the search source or in addition to the generation source data items are detected. A software specification information extraction method characterized by searching and detecting value generation method information across multiple programs, and storing or outputting the information. 10. In the software specification information extraction method according to claim 9, a physically identical file is detected among files defined in another program by analysis of job control parameters or command procedure, and data items belonging to the file are detected. Among them, data items that share physical locations with each other are detected, and using the method described above, the data items that are the generation source for the generation data item of the search source or in addition to the generation source data items are detected. A software specification information extraction method characterized by searching and detecting value generation method information across multiple programs, and storing or outputting the information. 11.Claims 1, 2, 3, 4, 5, 6, 7, 8, 9, 1
In the software specification information extraction method described in 0, the data item name of the detected generation source data item or generation source data item in the generation method information is converted based on the correspondence information between the existing data item and the standard data item name. A method for extracting software specification information, characterized by accumulating or outputting it. 12. Claims 2, 3, 4, 5, 6, 7, 8, 9, 10,
12. An automatic software generation method characterized in that a new program is automatically or semi-automatically generated using data item generation method information accumulated by the software specification information extraction method described in 11.