JPH0540616A - Method for generating program specifications - Google Patents

Abstract

PURPOSE:To produce the specification information that is easily understood and has the contents satisfactory for correction and reuse from an existing program with the use of a computer. CONSTITUTION:A program is analyzed (102) and the data specification information is produced (103). The program processing pattern information showing the skeletal structure of the processing procedure is produced (104) together with production of the data generating formula information showing a process where each output data item is produced from the input data items (105). The input data checking type information is produced to show the conditions relating to the input data items and the operations which are carried out according to the propriety of the conditions (106). Then the decision table information is produced to show each condition and the operations carried out according to the propriety of these conditions (107). Based on these pieces of information, the program specification information is edited and outputted (108). It is also possible to automatically produce the program after those preceding information are changed as necessary.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to software maintenance and reuse, and more particularly to a technique for automatically creating specification information of an existing program by a computer.

[0002]

2. Description of the Related Art It is often necessary to partially modify an existing program or extend its function.
Also, existing programs are often used to develop new programs. In such work, first, it is necessary to fully understand the specifications of the existing program.
Especially in the case of an office processing system, regarding the input / output data items that are included in the form, screen, or master file and are directly related to the work of the system user, the attributes of each data item and their It is extremely important to understand the relationship between data items, such as the process by which one data item is derived from another data item.

Originally, such information should be described in the specifications of the existing program, and if it is described, the above information can be easily obtained from the specifications. However, in reality, the program specification does not exist, or even if it exists, the program specification is not modified when the program is changed. There are many cases where we do not, and in such cases, we must spend a great deal of effort just to understand the contents of the program.

For the purpose of reducing such labor, there have already been proposed some methods of analyzing existing programs by a computer and generating specification information of the programs, as illustrated below.

As a first conventional technique, Japanese Patent Laid-Open No. 62-89
The method described in No. 137 “Program Interactive Method” is
It analyzes a program, converts the contents of the Procedure Division into a graphic representation, and creates information about the data handled by the program.

As a second conventional technique, Japanese Patent Laid-Open No. 2-112
The method described in No. 024 “Method for estimating program processing pattern” is based on the procedure part of the program, and includes a program processing pattern representing a skeletal structure of the processing procedure,
The outline of the processing (for example, file conversion processing, tabulation processing, etc.) is determined.

As a third prior art, Japanese Patent Application No. Hei 2-316
The method proposed by No. 134 “Software Specification Information Extraction Method” is based on the procedure division of a program, data generation formula information indicating the process of deriving an output data item from an input data item, and a condition regarding the input data item. Data check type information representing the operation depending on it is created.

[0008]

Each of the above-mentioned conventional techniques analyzes an existing program and generates specification information thereof, and is a powerful technique for assisting a computer in understanding the contents of the existing program. Is. However, although the information obtained by the first conventional technique includes sufficient program specification information in terms of quantity, it is extremely redundant and unorganized. Then, in order to clearly understand the contents of the program, , Still requires a lot of effort. The information obtained by the second conventional technique is
It is only a small part of the program specification information, and it is impossible to grasp the entire contents of the program from it. Furthermore, the information obtained by the third conventional technique is not sufficient to modify or reuse the program.

A first object of the present invention is to have a sufficient content for modifying or reusing a program from an existing program, and also for allowing a user to easily understand the content of the program. It is to create various program specification information using a computer.

A second object of the present invention is a part of the created program specification information, in particular, program processing pattern information and information relating to input / output data such as data specification information, data generation formula information, data check formula information. And so that they can be easily used for subsequent generation of specification information of other programs and generation of new programs.

A third object of the present invention is to automatically generate another program based on the same specification or a new program having a part of the specification by utilizing the created program specification information. To do.

[0012]

In order to achieve the first object, the first feature of the present invention is to represent the skeletal structure of a processing procedure from program analysis information obtained by analyzing an existing program. A step of extracting a program processing pattern and creating program processing pattern information representing the pattern; a step of creating data specification information including at least one of file specification information and record specification information from the program analysis information; Data generation formula information indicating the process of generating each output data item from the input data item, which is obtained from the analysis information by reversing the generation procedure of each output data item, and the condition including the input data item in the condition. Conditions for each input data item and the conditions A step of creating at least one of input data check formula information representing an operation performed according to whether or not, and at least the identification information of the data specification information, the program processing pattern information, the data generation formula information, and the input data check formula. Editing one or both of the information as program specification information.

In the program processing pattern information creating step, the program control structure obtained directly from the program analysis information is converted into a program control structure consisting only of repetitive, selective and sequential processes, and then the program processing pattern is created. You may extract. In this step, the program processing pattern information may be determined by matching the extracted program processing pattern with the program processing pattern registered in the program processing pattern library.

Further, even if each conditional process in the program is analyzed and decision table information representing each condition and the operation performed depending on the success or failure of the condition is created and added to the program specification information. Good. In this case, duplicate information may be removed from at least two of the data generation formula information, the input data check formula information, and the decision table information by the dialog with the user.

To achieve the second object, the present invention provides
As a second feature, further, at least one of registration of created program processing pattern information in the program processing pattern library, registration of data specification information in the data specification library, and data generation formula information and input data check formula information. Are registered in the data item dictionary individually or in combination. To register the program processing pattern,
It may be performed only when the same pattern as the program processing pattern extracted from the program being analyzed is not registered in the program processing pattern library.

In order to achieve the third object, the present invention provides
As a third feature, further, at least one of the program processing pattern information, the data specification information, the data generation formula information, the input data check formula information, and the decision table information is changed according to the user's request, After that, the program is automatically generated based on the various information.

[0017]

According to the first feature of the present invention, the program processing pattern information indicating the skeletal structure of the processing procedure, the data specification information including at least one of the file specification information and the record specification information, and each output data item are stored. At least one of data generation formula information indicating a process of generating from an input data item and input data check formula information indicating a condition regarding each input data item and an operation performed according to the success or failure of the condition is created, and at least One or both of the identification information of the data specification information, the program processing pattern information, and the data generation formula information and the input data check formula information are edited as the program specification information, and the user uses these information. be able to. Further, decision table information indicating each condition and an operation performed depending on the success or failure of the condition can be added to the program specification information. Therefore, it is possible to quickly obtain, from the existing program, information that has sufficient contents for modifying and reusing the program and that allows the user to easily understand the contents of the program.

In addition, if conversion to a program control structure consisting only of iterative, selective, and sequential processing is performed, GO
Since it is not necessary to prepare the program processing pattern library having a wide variety of program control structures including TO instructions in the program processing pattern library, the program processing pattern library can be small in size and the program processing pattern can be easily identified. become. Further, if the duplicate information is removed from the data generation formula information, the input data check formula information, and the decision table information, it becomes easier to understand the program specifications.

According to the second feature of the present invention, the created program processing pattern information, data specification information, data generation formula information, input data check formula information, etc. are accumulated in the system, and therefore these information are stored. Can be easily used for subsequent generation of specification information of other programs and generation of new programs. If the program processing pattern is registered only when the same pattern is not registered, the redundancy of the program processing pattern library can be prevented and its use becomes easy.

According to the third feature of the present invention, the information representing the specifications of the existing program is automatically created and changed according to the needs of the user. A program is automatically generated according to the specification as it is, or the specification after the change if changed. Therefore, another program based on the same specifications as the existing program, or a new program having a part of the specifications different from that of the existing program can be automatically generated.

[0021]

DESCRIPTION OF THE PREFERRED EMBODIMENT FIG. 2 illustrates an example computer system suitable for carrying out the embodiments described below. The input device 201 is a device for the user to input an instruction to the system, and is, for example, a keyboard. Main memory 202
Is an internal storage device for temporarily storing data, programs, and the like. The output device 203 is a device that outputs information generated by the system in a visible form, and is, for example, a display or a printer. The CPU 204 is a processor including a computing device and a control device. External storage device 20
A storage device 5 stores information for a long period of time, and is, for example, a magnetic disk device or a magnetic tape device.

FIG. 1 is a flow chart showing the first embodiment of the present invention.

In the program input step 101, the source program for which the specification information is created is input from the external storage device 205 to the main memory 202. Then, the program name of this program is extracted and stored in a predetermined area of the main memory 202. 3 and 4 respectively show a data definition part and a procedure part of an example of a source program written in the COBOL language.

The program syntax analysis step 102 analyzes the input source program to create data definition part analysis information and procedure part analysis information.

FIG. 5 shows an example of the data definition part analysis information obtained by the analysis of the data definition part shown in FIG. Figure 3
The data definition part 301 in FIG. 5 is sequentially analyzed for each item, and data definition part analysis information including an identifier, a file name, a record name, a level number, an item name, and an attribute is created as shown in FIG. And stored in a predetermined area in the main memory 202. From these information items it is possible to identify each data item in the program and to recognize their interrelationships. For example, from the description of the data item 302 in FIG. 3, the level number 501 and the item name 502 in FIG.
And the attribute 503 are extracted. Also, the data item 302
Is included in the file name 303 and the record name 304, the file name 303 and the record name 304 in FIG. 3 are registered as the file name 504 and the record name 505 in FIG. 5, respectively. Since the data item 302 is a data item included in the file definition, “F” is entered as the identifier 506. Other identifiers include "W" (for example, 507) indicating a data item in the work place section and "L" (for example, 508) indicating a data item in the link section.

FIG. 6 shows an example of procedure division analysis information obtained by the analysis of the procedure division shown in FIG. The various instructions in the procedure section 401 shown in FIG. 4 are sequentially analyzed in units of instructions, and the procedure section analysis information including the item number, parent item number, instruction name, and instruction content as shown in FIG. It is created and stored in a predetermined area in the main memory 202. From these information items, it is possible to identify each command and recognize their mutual relationship. For example, the instruction name 602 and the instruction content 603 in FIG. 6 are extracted from the description of the instruction 407 in FIG. Further, since the consecutive item number 601 is generated and registered, and the instruction 407 is included in the instruction 405, the value of the item number 605 of the instruction 606 in FIG. 6 corresponding to the instruction 405 in FIG. No. 604 is registered.

In the data specification information extraction step 103, file specification information and record specification information are created from the data definition section analysis information shown in FIG. 5 and the procedure section analysis information shown in FIG. The program name stored in the input step 101 is added and registered in the data specification library in the external storage device 205, and each file name in the file specification information and each record name in the record specification information are assigned. , And are stored in the main memory 202 as a file specification information name and a record specification information name, respectively. Further, the level number, the data name and the attribute of each data item in the work place section and each data item in the link section are stored in the main memory 202.

FIG. 7 shows an example of the contents of the data specification information created from the data definition analysis information shown in FIG. 5 and the procedure analysis information shown in FIG. 6 and registered in the data specification library. In creating the file specification information, for example, the file specification information having the file name 706 is registered in the data specification information shown in FIG. 7 based on the identifier 514 and the file name 515 of the data item 513 shown in FIG. To be done. Also, the OPEN in the procedure division analysis information shown in FIG.
The input / output type of this file is determined from the command and registered in the file specification information. In the illustrated example, the input / output type of the file with the file name 515 “IN-FILE” in FIG.
Of the contents of 611, "INPUT (IN-FILE)"
Input / output type 707
Register the "input file" as. Then, the file name 706 in the file specification information is stored in a predetermined area in the main memory 202 as the file specification information name.

Next, in creating the record specification information,
For example, the identifier 510 of the data item 509 shown in FIG.
Based on the file name 511 and the record name 512, the record specification information having the file name 701 and the record name 705 in FIG. 7 is registered, and the level number 7
02, a data name 703, and an attribute 704 as shown in FIG.
The level number, the item name, and the attribute in the data item 509 are registered. The record name 705 in this record specification information is stored in a predetermined area in the main memory 202 as the record specification information name.

In the program processing pattern information extraction step 104, an input command, an output command, a GOTO command, and a command encompassing these commands are extracted from the procedure division analysis information, and the process is composed of only these commands. Create a program processing pattern that represents the skeletal structure of the procedure.
Then, this is collated with a group of program processing patterns prepared in advance in the program processing pattern library in the external storage device 205, and if there is a match, the name of the program processing pattern is assigned to a predetermined area in the main memory 202. Remember. If there is no matching registered pattern, the user is requested to input an appropriate program processing pattern name and it is stored.

FIG. 8 shows a control structure in which the contents of the procedure division analysis information shown in FIG. 6 are expressed by a PAD diagram, and FIG.
Shows the PAD schematic representation of the program processing pattern obtained by compressing the control structure shown in FIG. FIG. 9 corresponds to FIG.
From the input instructions 801 and 804, the output instruction 803, and the instruction 802 including the input instruction 804 and the output instruction 803.
Corresponds to the extracted one. That is, the input instruction 901 in FIG. 9 corresponds to the input instruction 801 in FIG. 8, the input instruction 904 corresponds to the input instruction 804, the output instruction 903 corresponds to the output instruction 803, and the output instruction 903 and the input instruction 904. Iterative instruction 902 that subsumes

FIG. 10 shows an example of a program processing pattern library prepared in advance in the external storage device 205. This library includes a program processing pattern name, processing contents, and a program processing pattern.
Since the program processing pattern shown in FIG. 9 matches the program processing pattern 1001 in the program processing pattern library shown in FIG. 10, its program processing pattern name 1002 is stored in a predetermined area in the main memory 202. Remembered.

The data generation formula information extraction step 105
The output data item of the program is searched from the data specification information registered in the data specification library in the external storage device 205, and then the procedure division analysis information is checked to reverse the generation procedure of each output data item. Following the direction, each source data item name included in the generation formula of each output data item name is replaced with the generation formula. By repeating such replacement, the generation expression of each output data item name is finally converted into an expression including only the input data item name as the generation source data item, and each output data item is converted from the input data item. Data generation formula information indicating a generation process is created and stored in a predetermined area in the main memory 202. FIG. 11 shows a flowchart of this processing.

FIG. 12 shows an example of data generation formula information created from the procedure division analysis information shown in FIG. 6 and the data specification information shown in FIG. Below, the output data item "URI-G
The process of creating the data generation formula information for "AKU" will be described.

First, the data specification library in the external storage device 205 is accessed, the data specification information (FIG. 7) is checked, and from the file specification information, the name of the file whose input / output type is "output file""OUT-FIL
"E" is detected, and then the data name in the record specification information having this file name is checked. As a result of this investigation, the output data item name "URI-GAKU" is detected. The detected output data item name "URI-GAKU"
Is registered as a generation result item 1201 of the data generation formula information, as shown in FIG.

Next, when the "command content" column of the procedure division analysis information (FIG. 6) is examined, a command 607 having this "URI-GAKU" as a generation result item is found. In this command, the generating expression of "URI-GAKU" is "(W-
TANKA-NEBIKI) * SURYO ". Therefore, the data specification information is checked again, the file name “IN-FILE” whose input / output type is “input file” is detected from the file specification information, and the record specification information having this file name is read, Each item of the generation formula, that is, the name of the generation source data item is collated with the data name in the record specification information. As a result, of these source data items, the data item "W-TA
It turns out that "NKA" is not an input data item.

Therefore, when the instruction that brings the data item "W-TANKA" is searched for in the procedure division analysis information shown in FIG. 6, the instruction 608 is detected. Whether "TANKA" corresponding to the generation source data item in this command is a data item in the input file is checked based on the record specification information of the input file in the same output order as the above, and as a result, " It turns out that "TANKA" is an input data item. Therefore, "WT
An expression in which "ANKA" is replaced with "TANKA", that is, "(TANKA-NEBIKI) * SURYO"
Is stored as the generation source item 1202 of the data generation formula information, as shown in FIG.

Data check type information extraction step 106
Is a conditional instruction (for example, IF instruction, EVALUAT) having the input data item name in the conditional expression from the procedure division analysis information.
E command), and for each applicable conditional command, data check formula information representing the relation between the conditional formula and the operation performed depending on the success or failure thereof is created and stored in a predetermined area in the main memory 205. To do.

FIG. 13 shows an example of data check formula information created from the procedure division analysis information shown in FIG. The process of creating the data check formula information for the instruction of item number "14" in FIG. 6 will be described below.

Examining the procedure division analysis information shown in FIG.
Conditional instruction 602 is found. In the instruction 602, the data item targeted by the conditional expression is “SURYO”. Therefore, when the data specification library in the external storage device 205 is accessed and the record specification information of the input file is checked, "SURYO" is the record of the input file.
It is present as a record name in the de-specification information, and it is thus found that "SURYO" is an input data item. Therefore, "SURYO" is registered as the data item name 1301 of the data check expression information shown in FIG. 13, and the condition "IS NUMERI" is registered.
“C” is registered as the condition 1302. Furthermore, when the subsequent instruction is examined, the conditional expression "SURYO IS NUMER
If "IC" is true, the action is "CONTINUE", and if it is false, the action is "MOVE".
"ERR01" TO ERR-CODE ". Therefore, as the identifier 1303, "Y" representing "true"
And the corresponding action 1304 is “C
"ONTINUE" is registered, "N" representing "false" is entered as the identifier 1305, and "MOVE" ERR01 "TO is specified as the corresponding operation 1306.
ERR-CODE "is registered.

Decision table information extraction step 10
7 extracts all conditional instructions from the procedure division analysis information, creates decision table information representing the relationship between one or more conditional expressions and operations in the form of a two-dimensional table for each conditional instruction, and creates it. It is stored in a predetermined area in the main memory 202.

FIG. 14 shows an example of the decision table information created from the procedure division analysis information shown in FIG. The process of creating decision table information for the IF instruction 609 in FIG. 6 will be described below. This instruction has two conditional expressions, namely "NEBIKI-CODE ="
1 "and" SURYO <100 ", and these conditional expressions are connected by the logical operator" AND ".
Therefore, the first conditional expression “NEBIKI-CODE =
"1" is registered as a condition 1401, and the second conditional expression "SURYO <100" is registered as a condition 1402. The operation "COMPUTE URI-GAKU = (W-TA
NNKA-NEBIKI) * SURYO "is operated 1
If it is registered as 403 and it is false, the action "MOVE` ERR02` TO ERR-COD
E "is registered as action 1404.

The relationship between the truth of each condition and the resulting operation is expressed in the form of a two-dimensional table as follows. That is, for each condition, a code "Y" indicating that the condition is true and a code "N" indicating that the condition is false are set so as to cover all combinations of true and false conditions. Fill in the fake part 1405. Also, for each operation, the code "x" indicating that it is performed is
The execution unit 1406 is filled in at the position corresponding to the display of the true / false combination of both conditions corresponding to the case where it should be performed. For example, a position 1407 in which “Y” is entered in both the conditions 1401 and 1402 and a position 1408 in which “x” is entered in the operation 1403 performed when both conditions are true are at the same horizontal axis position. ..

The program specification output step 108 is
The above-mentioned program input step 101, data specification information extraction step 103, program processing pattern information extraction step 104, data generation formula information extraction step 10
5, the data check type information extraction step 106 and the decision table information extraction step 107, the various information stored in the predetermined area of the main memory 202 is edited as program specification information, and the edited information is edited. External storage device 2
05, and outputs as a visible program specification through the output device 203.

FIG. 15 shows an example of the program specifications generated from the source programs shown in FIGS.
Program ID 1501 is the program input step 1
01 is obtained from the program name stored in the main memory and stored in the main memory. The program processing pattern name 1402 is determined in the program processing pattern information extraction step 104. Data specification information extraction step 103, obtained from information stored in the main memory, and data generation formula information 1404
Is obtained from the information created in the data generation expression information extraction step 105 and stored in the main memory, and the data check expression information 1405 is created in the data check expression information extraction step 105. , And decision table information 1406 is obtained from the information created in decision table information extraction step 106 and stored in main memory.

The program processing pattern information extraction step 104 in this embodiment may be modified so that the collation with the program processing pattern library is omitted and the user gives the program processing pattern name each time.

FIG. 16 is a flow chart showing the second embodiment of the present invention. This embodiment is the same as the embodiment shown in FIG. 1 except that a data item dictionary registration step 1601 is added and the content of the output information of the program specification output step 1602 is slightly different. ..

The data item dictionary registration step 1601
Data generation formula information and data check formula information created in the data generation formula information extraction step 105, the data check formula information extraction step 106, and the decision table information extraction step 107 and stored in the main memory, respectively. , And the decision table information, by interacting with the user, duplicate information is removed according to the user's judgment, and the data generation formula information and data check formula information from which the duplication has been eliminated in this way are edited as a data item dictionary. Then, the data is stored in the external storage device 205.

In FIG. 17, regarding the data generation formula information, the data check formula information, and the decision table information shown in FIGS. 12, 13 and 14, respectively, overlapping information among them is output device 203 (display in this example). ), And when the user has instructed the selection,
The example of a screen is shown. In the figure, between the decision table information 1701 and the data generation formula information 1702, “SU
Although the condition and the operation regarding “RYO” are duplicated, the fact that there is no generation expression that overlaps with the operation is displayed in the data generation expression information 1703, and further, which of the overlapping information is adopted and which is deleted. A message 1704 requesting the user to select one is displayed. In this example, the user guesses the intention of the creator of this program, determines that the process 408 shown in FIG. 4 is a data check, and
In the selection table 1705 of the decision table information, enter "D" indicating deletion, and select the selection column 17 of the data check formula information.
In “06”, “R” representing the adoption is input. If the contents of the operation in the decision table information or the data check formula information includes the generation formula of the output data item, and the information that overlaps with it also exists in the data generation formula information, it is also displayed and the user discards it. Make a choice.

After the duplication information that the user has determined to be unnecessary is removed in this way, the data generation formula information and the data check formula information are edited as a data item dictionary and stored in the external storage device 205. FIG. 18 shows an example of the contents of the data item dictionary. For example, the data item name 1707 in FIG. 17 is registered as the data item name 1801 in FIG. 18, and the condition 1708 is the data check expression 1802.
Is registered as the identifier 1709, the identifier 1709 is registered as the identifier 1803, and the operation 1710 is registered as the operation 1804. The attribute can be obtained from the record specification information in the data specification library or the data definition part analysis information.

Program specification output step 1602
1 in the same manner as in the embodiment shown in FIG.
The program specification as illustrated in 9 is output. However, the data generation formula information 190 output in this step
1, data check formula information 1902 and decision table information 1903 are data item dictionary registration step 16
01 removes redundant information.

Since the data generation formula information and the data check formula information are registered in the data item dictionary, it is possible to create a program specification that does not include such information.

According to this embodiment, since the data generation formula information and the data check formula information are registered in the data item dictionary, these information can be used to create a new program using the same or similar data items. Moreover, since there is no duplication of this information, it is much easier to understand and use.

FIG. 20 is a flow chart showing the third embodiment of the present invention. This embodiment is the same as the embodiment shown in FIG. 1 except that the processing in the program processing pattern information extraction step 2001 is different.

A program processing pattern information extraction step 2001 creates a program processing pattern, collates it with a program processing pattern library, and if there is a match, stores the program processing pattern name in the main memory. Up to this point, there is no difference from the program processing pattern information extraction step 104 of the embodiment shown in FIG. If there is no matching program processing pattern in the program processing pattern library, the created program processing pattern is newly registered in the program processing pattern library in this embodiment. For that purpose, the name and processing contents of the already registered program processing pattern are displayed for reference, and the user inputs the name and processing content of the program processing pattern to be newly registered. Prompt to.

FIG. 21 shows an example of an input screen for registering a new program processing pattern. As shown in the figure, the registered program processing pattern name 2101 and
The processing contents 2102 are read from the external storage device 205 and displayed on the output device 203 for reference.
Further, a message prompting the user to input the name of the program processing pattern to be newly registered and the processing content is displayed. The user refers to the registered pattern name 2101 and the processing content 2102 that are displayed, and devises and inputs a name 2103 that does not overlap with the registered pattern name and a token 2104 that briefly describes the processing content. As a result, together with the input program processing pattern name 2103 and processing content 2104, the program processing pattern of the program currently being analyzed is newly registered in the program processing pattern library in the external storage device 205. The inputted program processing pattern name is stored in a predetermined area in the main memory 202 as the program processing pattern name of the program currently being analyzed.

According to this embodiment, new program processing patterns are added to the program processing pattern library one after another so that the program processing patterns can be reused. Therefore, the performance of the program specification creating system is automatically improved as it is used. To do.

However, instead of registering only the unregistered program processing patterns in the program processing pattern library in this way, all the program processing patterns obtained in the process of creating the program specification information may be registered.

FIG. 22 is a flow chart showing the fourth embodiment of the present invention. This embodiment is different from the embodiment shown in FIG. 1 in that the program specification information changing step 220.
1 and a program generation step 2202 are added, and other steps are the same as those in the embodiment shown in FIG.

Program specification information change step 2201
Changes the program specification information and the data specification information created in the previous steps by an interactive method as needed by the user.

The program generation step 2202 receives the program specification information and the data specification information which have passed through the program specification information change step 2201 as input, and a known program generation tool (for example, Hitachi review, Vol. 65, No. 7). ,
July 1983, pp. 53-58, "Development of COBOL program generation system" DSL "with data-based specifications as input"), and the generated program is stored in the external storage device 205. Accumulate in.

According to this embodiment, the user can change the existing program only by changing the program specification without touching the program itself.
It is very easy to modify or reuse existing programs.
In addition, the existing program can be changed to a program having the same specifications that matches the applied program generation tool.

FIG. 23 is a flow chart showing the fifth embodiment of the present invention. In this embodiment, a program structure conversion step 2301 is added to the embodiment shown in FIG. 1, and other steps are the same as those in the embodiment shown in FIG.

The program structure conversion step 2301
Known restructuring tools (eg Nikkei Computer 1
986.4.14, No. 119, pp. 81-90, "Restructuring Tool C to Reduce Program Maintenance and Development Costs"
SP / 2 ”), the program control structure (eg, FIG. 8) obtained from the procedure division analysis information
Converts to a program control structure that does not include statements and consists only of iterative, selective, and sequential processing. Therefore, the program processing pattern obtained by the program processing pattern information extraction step 104 also includes only repeated, selected, and sequential processing.

According to this embodiment, the program processing pattern
Since it is not necessary to prepare a large number of program processing pattern information including GOTO instructions in the program library, the scale of the program processing pattern library is small, and the collation of the program processing patterns is easy. Become.

[0066]

According to the first feature of the present invention, the program processing pattern information representing the skeletal structure of the processing procedure,
Data generation formula information indicating the process of generating each output data item from the input data item, input data check formula information indicating the condition related to each input data item and the operation performed depending on the success or failure of the condition, and further, if necessary, Decision table information representing each condition and an operation performed depending on the success or failure of the condition is created, and from this information, program specification information in a format easily understandable by the user is edited. Therefore, it has enough contents to modify or reuse the program from the existing program, and
Information that allows the user to easily understand the contents of the program can be quickly obtained.

In addition, if conversion to a program control structure consisting only of iterative processing, selection processing and sequential processing is performed, GO
Since it is not necessary to prepare the program processing patterns of various program control structures including TO commands in the program processing pattern library, the scale of the program processing pattern library is small and the program processing patterns can be easily identified. become. Further, if the duplicate information is removed from the data generation formula information, the input data check formula information, and the decision table information, it becomes easier to understand the program specifications.

According to the second feature of the present invention, in addition to the above effects, the created program processing pattern information, data specification information, data generation formula information, input data check formula information, etc. are accumulated in the system. Therefore, these pieces of information can be easily used for subsequent creation of specification information of other programs and creation of new programs. If the program processing pattern is registered only when the same pattern is not registered, the redundancy of the program processing pattern library can be prevented and its use becomes easy.

According to the third feature of the present invention, in addition to the above effect, the information indicating the specifications of the existing program is changed according to the need of the user, and if the user does not change the original information, A program is automatically generated according to the specification as it is, or the specification after the change if changed. Therefore, another program based on the same specifications as the existing program, or a new program having a part of the specifications different from that of the existing program can be automatically generated.

In summary, according to the present invention, the information that is sufficient for modifying or reusing a program from an existing program, and that allows the user to easily understand the content of the program. Can be obtained quickly, and they can be stored for future use, and some of them can be modified as needed to automatically generate a program based on the modified specifications. You can also do it. Therefore, it is possible to efficiently understand the contents of the existing program, and it is also possible to efficiently develop a new program by using the extracted data generation method or check method. Excellent effects such as easy maintenance and reuse of software can be expected.

[Brief description of drawings]

FIG. 1 is a flow chart of a first embodiment of the present invention.

FIG. 2 is a block diagram of a computer system suitable for implementing the present invention.

FIG. 3 is a diagram showing an example of a data definition part of an existing program.

FIG. 4 is a diagram showing an example of a procedure section of an existing program.

FIG. 5 is a diagram showing an example of data definition part analysis information.

FIG. 6 is a diagram showing an example of procedure division analysis information.

FIG. 7 is a diagram showing an example of contents of a data specification library.

8 is a PAD diagram showing a program control structure of a procedure division shown in FIG.

9 is a PAD diagram showing a program processing pattern of the program control structure shown in FIG. 8. FIG.

FIG. 10 is a diagram showing an example of the contents of a program processing pattern library.

FIG. 11 is a flowchart of a data generation formula information extraction step.

FIG. 12 is a diagram showing an example of data generation formula information.

FIG. 13 is a diagram showing an example of data check type information.

FIG. 14 is a diagram showing an example of decision table information.

FIG. 15 is a diagram showing an example of program specification information.

FIG. 16 is a flowchart of the second embodiment of the present invention.

FIG. 17 is a diagram showing an example of an interactive screen for eliminating duplication of information.

FIG. 18 is a diagram showing an example of the contents of a data item dictionary.

FIG. 19 is a diagram showing an example of program specification information from which duplication of information is removed.

FIG. 20 is a flowchart of the third embodiment of the present invention.

FIG. 21 is a diagram showing an example of an input screen for newly registering a program processing pattern.

FIG. 22 is a flowchart of the fourth embodiment of the present invention.

FIG. 23 is a flowchart of the fifth embodiment of the present invention.

[Explanation of symbols]

102 ... Step of analyzing existing program 103 ... Step of creating data specification information 104 ... Step of creating program processing pattern information 105 ... Data generation expression Step for creating information 106 ... Step for creating data check expression information 107 ... Step for creating decision table information 108 ... Step for editing program specification information 1502 ... Program processing pattern name 1503 ... Data specification information identification information 1504 ... Data generation formula information 1505 ... Input data check formula information 1506 ... Decision table information 1601 ... Duplicate Step of removing information and registering data item dictionary 2001 ... New registration and program of program processing pattern Ram processing pattern information creation step 2201 ... Changing program specification information 2202 ... Program generating step 2301 ... Program control structure conversion step

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Hirofumi Danno 1099, Ozenji, Aso-ku, Kawasaki-shi, Kanagawa, Ltd. Inside the Hitachi, Ltd. Systems Development Laboratory (72) Inventor Tsukuda, Gunji 1099, Ozen-ji, Aso-ku, Kawasaki, Kanagawa Hitachi Systems Development Laboratory

Claims (10)

[Claims] 1. A computer system comprising at least a processing device, a storage device, an input device and an output device, a step of analyzing an existing program to create program analysis information, and a skeleton of a processing procedure from the program analysis information. Extracting a program processing pattern representing a structure and creating program processing pattern information representing the pattern; and creating data specification information including at least one of file specification information and record specification information from the program analysis information. , The data generation formula information indicating the process of generating each output data item from the input data item, which is obtained from the program analysis information by reversing the generation procedure of each output data item, and the input data item is in the condition By the analysis of the conditional processing included in A step of creating at least one of input data check-type information representing a condition relating to each input data item and an operation performed depending on the success or failure of the condition, and at least identification information of the data specification information and the program processing pattern. And a step of editing one or both of the information, the data generation formula information and the input data check formula information as program specification information. 2. The method according to claim 1, wherein the program processing pattern information creating step includes a program control structure obtained directly from the program analysis information, which is a program consisting only of repetitive, selective and sequential processes. A method for generating a program specification, comprising: converting into a control structure; and extracting the program processing pattern from the converted program control structure. 3. The method according to claim 1, wherein the program processing pattern information creating step comprises:
Furthermore, a step of collating the extracted program processing pattern with a group of program processing patterns already registered in a program processing pattern library in the computer system, and determining program processing pattern information of the program according to the result of the collation. A method for generating a program specification, the method including: 4. The method according to claim 3, wherein the program processing pattern information creating step further comprises:
A program specification generation method comprising the step of registering the program processing pattern information in the program processing pattern library. 5. The program specification generation method according to claim 4, wherein the registration of the program processing pattern information is performed only when no match is detected in the collation. 6. The method according to claim 1, further comprising the step of registering the data specification information in a data specification library in the computer system. 7. The method according to any one of claims 1 to 6, further including at least one of the data generation formula information and the input data check formula information in a data item dictionary in the computer system. A program specification generation method comprising a step of registering. 8. The method according to any one of claims 1 to 7, further comprising: analyzing each conditional process to represent each condition and an operation performed depending on success or failure of the condition. And a step of creating the program specification generation method, wherein the editing step edits the data specification information including the decision table information. 9. The method according to claim 8, further comprising: extracting duplicate information from at least two of the data generation formula information, the input data check formula information, and the decision table information, and the duplicate information. Removing one of the two by interacting with the user. 10. The method according to claim 8, further comprising at least one of the program processing pattern information, data specification information, data generation formula information, input data check formula information, and decision table information.
A method of creating a program, which comprises a step of changing the user's request according to the user's request and a step of automatically generating a program based on the various information after that.