JPH08320785A - Data maintenance tool automatic generation method - Google Patents

Abstract

(57) [Abstract] [Purpose] To provide a method for automatically generating a data maintenance tool that is highly reliable and appropriately converts data types and displays them in a form that is easy for humans to understand. [Structure] A structure of a data structure description unit 10 of a predetermined programming language is analyzed, the analysis result is translated, and the translation result and a user interface unit 4 prepared in advance.
By compiling / linking 0 with 0, a data maintenance rule for referring to, updating, etc. is generated for a file having a record according to the format of the programming language.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a field of manufacturing a program by directly coding a programming language,
Automatic data maintenance tool generation method for direct analysis of data structure and data type in data structure description part, especially program design / manufacturing / inspection according to waterfall model, with source list coding in the middle The present invention relates to those used in such a form.

[0002]

2. Description of the Related Art Hereinafter, JP-A-4-276835 (hereinafter referred to as "first prior art") and JP-A-3-244
The conventional technique will be described with reference to Japanese Patent Publication No. 043 (hereinafter referred to as the second conventional technique).

In the first prior art, in the unit test case generation, the data type classification of the input / output interface parameters of the program into two groups is performed based on the method by the black box test format which does not depend on the source program. When a data definition statement of the input interface is input, the type definition table is provided, and according to the classification according to the table, from the data definition statements belonging to one of the groups, all the data value combinations of the parameter data value areas are tested. From the data definition statement belonging to the other group, the data value area of the parameter is further divided into the data value areas of the input interface as the test case. How to generate program unit test data that is generated It has been proposed.

In the second conventional technique, an item definition analyzing means of a data part of a source program that defines a plurality of types of items in the same data area, and an instruction to generate test data for the plurality of types of items. A first input means for executing the test data, a generation instruction storage means for storing information on the test data generation instruction, a second input means for designating the generation order and output frequency of the test data for each item definition, and a test A generation number information storage unit that stores information on the generation order and output number of data, and a test data generation unit that generates test data are provided. There has been proposed a test data generation device configured to generate a series of stored test data.

[0005]

In the upstream process of program design, the data structure (record layout) is designed,
This definition is clearly stored as a document, and after the outline design, detailed design, and manufacturing of the program are performed, enter the unit verification or combination / comprehensive verification stage to create test data for verification or to copy a part of it. I sometimes wrote a new program to do operations (patch data, etc.). In addition, a different data access tool was sometimes created for each different data structure.

That is, the data structure (record layout) defined in the upstream process of the program design may be inconsistent with the actual program at the time of verification because the data structure is modified during manufacturing in the actual manufacturing process. A tool created based on such a document causes confusion during verification, and this problem becomes remarkable especially in the manufacturing process using the waterfall model. However, the above-mentioned first and second conventional techniques solve this problem. Can not.

Further, if a data operation program is created for each data structure, it takes a lot of man-hours and the quality itself of the data operation program becomes a problem. On the other hand, a general-purpose data maintenance tool is not always easy for humans to read, such as expressing the file contents in hexadecimal notation. In other words, there is an increasing need for a non-general purpose maintenance tool suitable for the program.

The object of the present invention is to make highly accurate data easily, to make detailed data at the manufacturing stage and unit test stage, to improve productivity, and to re-automatically re-tool the tool as the program is updated. It is to provide a data maintenance tool for generating and improving maintainability, and disclosing the information of the program data structure description part to be tested to the user.

[0009]

According to the present invention, a structure of a data structure description part of a predetermined programming language is analyzed, the analysis result is translated, and the translation result and a user interface part prepared in advance are analyzed. By compiling / linking, a data maintenance tool automatic generation method characterized by generating a data maintenance rule for referring to, updating, etc. for a file having a record according to the format of the programming language can be obtained.

Further, according to the present invention, the translation result is suitable for the input section, which is a program for storing in a proper format in a predetermined position in the record area, and the data item in the selected record area according to the data type. It is translated under the three rules of the output section, which is a program for extracting in various formats, and the attribute description section, which is an attribute information table such as the name, type, and redefinition of data items in the selected record area. A data maintenance tool automatic generation method characterized by

Further, according to the present invention, the user interface section uses the information of the attribute description section to display the data in the record area on the front face using the output section. And the input made to the data item selected by the user of the data maintenance tool is checked according to the data type of the item, and the data input part to store in the record area and the real of the specified real file are stored. There is provided a data maintenance tool automatic generation method comprising a record and a file access unit for operating data in a record area.

[0012]

According to the present invention, a tool for referencing / updating data (file) can be automatically generated by directly using the data structure description part in the description language of the target program. . Note that the source list of a program is more reliable as the data of the program than the documents. In particular, the existence of the program to be verified means that the data structure description part already exists, and obtaining the data maintenance tool directly from that part leads from the manufacturing of the program to testing. Up to, even if there is any specification change to the data structure on the way, it means that the most reliable maintenance tool can be obtained immediately.

Direct analysis of the data structure makes it possible to obtain complete information about the structure and type of the data. By preparing a user interface that can use this information, it is possible to make the data corresponding to each item accessible in a format that is most easy for humans to see. For example, when an item defined as a 1-byte numerical value is displayed in decimal, or when this item is changed, it is possible to detect a numerical value less than 0 and 256 or more or a character input as an error on the spot.

[0014]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described with reference to FIG. Here, a concrete application example of this tool automatic generation method in the program development in the COBOL language will be described. In the simplest form, this method is a data structure description part (hereinafter referred to as a record layout) 10
File access tool 5 corresponding to
This means that 0 is generated.

In FIG. 1, a record layout 10 as an input and a record layout analysis unit 20 for performing a minimum parsing for knowing a data structure from a COBOL language source are provided. Upon receiving the analysis result 30 from the record layout analysis unit 20, the analysis result is translated according to a predetermined rule. Here, specifically, the translation result based on the following three rules is obtained.

The input unit 31 functions as a first rule,
It is a program for storing input data in an appropriate format at a predetermined position in a record area according to a data type. The output unit 32 is a program that functions as the second rule and extracts the data items in the selected record area in an appropriate format according to the data type. The attribute description unit 33 functions as a third rule, and is an attribute information table such as the name, type, repetition, and redefinition of the data item in the selected record area. The input unit 31 and the output unit 3
The second program does not have to be the same as the program to be analyzed. However, it is assumed that the user interface unit 40 described below is also output in the COBOL language because it is easy to prepare and the description is simple.

The user interface section 40 uses the input section 31, output section 32, and attribute description section 33 provided to refer to and update the record data. The user interface unit 40 includes a data list display unit 401, a data input unit 411, and a file access unit 420. It should be noted that it is possible, depending on circumstances, to independently enhance what kind of enhanced function is added, even after the establishment of this mechanism.

The data list display section 401 is a mechanism for displaying the data in the record area on the screen in an appropriate format by using the output section 32 while using the information of the attribute description section. The data input unit 411 is a mechanism for storing the input made to the item selected by the tool user in the record area after performing a check according to the data type of the item. The file access unit 420 is a group of file / record access functions such as record read, write, and delete, and is a mechanism that causes the actual record of the specified actual file to act on the data in the record area. Specifically, there are reading 421 for transferring data from the actual record to the record area and writing 422 on the contrary. It is desirable to cover as much as possible the file / record operations provided by the language system, such as the key search 423 and the record deletion 424. A user interface unit 40 prepared in advance and a record layout analysis unit 20
The target file access tool 50 is generated by combining (linking) the input unit 31, the output unit 32, and the attribute description unit 33, which are the analysis results of 1.

One user interface section 40 is manufactured in advance. In this state, the mechanism can obtain the file layout tool 10 corresponding to the record layout as the output, with the record layout 10 as the input. User interface unit 40
In addition, it is possible to add a test data automatic composition mechanism that creates data of multiple records with a certain pattern, and a search mechanism that refers to the necessary records with the above functions of the target language. For example, it is possible to perform a key search for any item in the sequential file. Various methods are conceivable for the test data automatic synthesis mechanism and the search mechanism.

On a system having a dynamic link mechanism, it is desirable that the user interface section 40 be connected by a dynamic link. This is because the capacity of the tool is small and the user interface unit 40 can be independently strengthened later.

A supplementary description will be given of the attribute description section 33 in the case of the COBOL language. This is also a specific caution when actually using this method to construct a system. Making the description of the data structure (record layout 10) a list of items that are the units of actual input / output will be referred to as "expand" here. The simplest example of expansion is that when a repeating item is defined as “10 times repetition of item A”, this is defined as 10 independent A (1), A (2), ..., A (10). It is to be treated as an item. 1000
For record layouts that include repetitions,
Sometimes you want to focus on the first or last few items. The following method can be used to limit the number of expansions.

First, the record layout analysis result 3
There is a method of including a description expressing a repeating structure in 0. In this case, the user interface unit 40 has a mechanism that allows the specification of arbitrary repeated items while interpreting this description, and the creation of the user interface unit 40 becomes rather difficult.

Second, the record layout analysis section 20.
There is a method to limit the number of expansions. In this case, the record layout analysis result 30 may be output one-dimensionally, and if the user interface unit 40 has a mechanism for inputting / outputting to / from each item without being aware of the repeated item. good.

In the form in which the formats of the record layout analysis unit 20 and the analysis result 30 are provided at the same time, and the user interface unit 40 can be designed independently by the user, the manufacturing load of the user interface unit 40 is reduced. The latter method is preferred for the purpose. Hereinafter, description will be made assuming that the analysis result 30 of the record layout is a list of such one-dimensional items (a model that the user interface unit 40 can easily construct).

The group item, unlike the basic item, does not represent the substance of data. Therefore, it may not be preferable for the user interface that the group item is repeatedly expressed on the repeated item. That is, the ratio of basic items in one screen is small, and the listability of actual data is reduced. Therefore, it is preferable that the record layout analysis unit 20 has a mechanism for suppressing the output of the repeated items other than the first appearing group item.

Regarding REDEFINE,
It is preferable to be able to trace the redefined parent-child relationship in the record layout analysis result 30 with a bidirectional pointer.
This is because, when the data of a certain basic item is changed, it is a good user interface to redisplay the affected item in the currently displayed data.
It is good to redisplay all the affected items by following the pointer, but in a system where it can be assumed that the display is fast enough, it is possible to redraw all the information in the current field of view when changing the data. The same effect can be obtained.

As an example, the record layout (format written in DATA-DIVISION) of the COBOL language is analyzed and developed, and the COBOL high productivity language (I
The GUI of the X-Window system is output by outputting the item attribute part, the input description part, and the output description part described in DL-II).
There is a track record of generating dozens of maintenance tools and linking them with the user interface part that was created in advance together with the file handling system. In this method, the record layout analysis result 30
By also outputting in the IDL-II language, the record layout that is the input is captured as it is, and the exchange of item data between the record area and the user interface unit is realized by directly using the COBOL mechanism. For this reason, it was possible to obtain high reliability by reducing the development man-hours without creating a new mechanism for simulating COBOL regarding redefinition of areas and correspondence with repeated areas.

[0028]

According to the present invention, since the user interface section 40 is independent, if the format of the record layout analysis result 30 is determined, the record layout analysis section 2 will be described.
0 can be independently developed, and the user interface unit 40 can be independently operated even when the user requests it.
Can be highly functionalized.

Further, since the tool is automatically generated by directly using a part of the source of the target program, bugs are not mixed in during the tool manufacturing process, and the reliability of the tool is high.

If the program to be inspected exists, it is certain that the data structure description part, which is the part, actually exists as the source. To use it, another document can be used or new data can be created. There is no need to wake up. In other words, it is possible to obtain a tool that perfectly corresponds to the program to be inspected with an extremely short man-hour (nearly zero).

Since the data structure / data type in the data structure description part is directly analyzed, it is possible to construct a natural user interface according to them. In other words, the accuracy of data verification and creation is high because a human-readable format can be used when referring to or updating data. That is, highly reliable verification is possible.

[Brief description of drawings]

FIG. 1 is a diagram showing a flow from a record layout to generation of a file access tool.

[Explanation of symbols]

 10 record layout (data structure description section) 20 record layout analysis section 30 (record layout) analysis result 31 input section 32 output section 33 attribute description section 40 user interface section 401 data list display section 411 data input section 420 file access section 50 File access tool

Claims (3)

[Claims] 1. The programming by analyzing a structure of a data structure description part of a predetermined programming language, translating the analysis result, and compiling / linking the translation result and a user interface part prepared in advance. An automatic data maintenance tool generation method characterized by generating a data maintenance rule for referencing and updating a file having a record according to a language format. 2. The translation result includes an input section, which is a program for storing in a proper format in a predetermined position in a record area, and for extracting a data item in a selected record area in an appropriate format according to a data type. The output part, which is a program, and the names of the data items in the selected record area,
Attribute description part 3 which is an attribute information table for types, redefinition, etc.
The method for automatically generating a data maintenance tool according to claim 1, wherein the method is automatically translated according to one rule. 3. The user interface unit uses the information of the attribute description unit to display the data in the record area on the front surface using the output unit, and the data maintenance tool. The input made to the data item selected by the user is checked according to the data type of the item, and the data input part to store in the record area, the actual record of the specified real file, and the record area 2. The method for automatically generating a data maintenance tool according to claim 1, further comprising a file access unit that causes the data of FIG.