JP3577480B2 - Reverse engineering equipment - Google Patents

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a reverse engineering tool for grasping an accurate file layout of a data set in an application system operated by using a computer for each program configuring the application system.
[0002]
[Prior art]
Many enterprises, not only private enterprises but also central and local governments, have built and operated a large number of computer systems (application business systems) of enormous scale in order to improve business efficiency and service levels. I have.
[0003]
These application systems are required to be constantly updated (changed) due to various needs that occur continuously and changes in circumstances surrounding the business entity.
[0004]
However, especially in a company that has owned and maintained a computer system for a long time, a great deal, and most of the labor, has been devoted to updating the programs that constitute the application system. When a program is updated, it is generally strongly desired that documents, such as specifications, be updated synchronously in anticipation that further maintenance work can be smoothly performed in the future. However, the current situation of many companies is that the update of specifications (such as documents) cannot easily catch up with the update of application systems.
[0005]
Updating the application system almost always involves updating the layout of the file (data set). In spite of the high frequency of updating the file layout, it is hard to say that many companies reflect the content in a timely manner in the file layout specification (design document). Even for companies that are kept up to date, considerable maintenance man-hours are spent.
[0006]
The file layout is an indispensable document for the person in charge of development / maintenance of the application system. However, the above makes it difficult to share the latest file layout among many developers / maintenance personnel.
[0007]
[Problems to be solved by the invention]
SUMMARY OF THE INVENTION An object of the present invention is to provide a reverse engineering tool for accurately and easily grasping the latest file layout of a real file that actually exists in the development and maintenance work of an application business system.
[0008]
[Means for Solving the Problems]
The present invention has been made to achieve the above object. The reverse engineering device according to claim 1 according to the present invention,
This is a reverse engineering device including a computer that reads and uses data stored in a repository database. In the reverse engineering equipment,
The repository database is
First table storage means in which device name data, volume name data, and data set name data are associated with job name data, step numbers in the job, and DD numbers in the step, and
Second table storage means in which the job name data, the step number in the job, the DD number in the step, and the DD name data are stored in association with each other;
Third table storage means in which job name data, Step numbers in the job, and program name data are stored in association with each other;
A fourth table storage in which the program name data, the DD name data, and the appearance start number of the item related to the DD in the program and the appearance end number of the item related to the DD in the program are related and stored. Means,
Fifth table storage means in which parent (calling side) program name data and child (calling destination) program name data are associated and stored;
Sixth table storage means in which the program name data, the appearance order number of the item declared in the program, the item name data, the level number, the attribute data, the offset value, and the number of bytes are associated and stored. And
The first table storage, the second table storage, and the third table storage are generated by analyzing a job control language source code, which is one of definition data, and are stored in a fourth table storage. The means, the fifth table storage means, and the sixth table storage means are generated by analyzing a program source code which is one of definition data,
In addition, the reverse engineering device
Input means for inputting device name data, volume name data, and data set name data that are data specifying a data set;
The first table storage unit is searched to derive the job name data, the step number in the job, and the DD number in the step from the device name data, volume name data, and data set name data input from the input unit. First deriving means for performing
A second deriving unit that retrieves the second table storage unit and derives the DD name data from the job name data, the Step number in the job, and the DD number in the Step derived by the first deriving unit; ,
A third deriving unit that retrieves the third table storage unit and derives program name data from the job name data derived by the first deriving unit and the Step number in the job;
Fourth extraction means for searching the fifth table storage means, applying the program name data derived by the third derivation means to the parent program name, and extracting child program name data of child programs under the parent program name;
Searching the fourth table storage means, the program name data derived by the third derivation means or the child program name data extracted by the fourth extraction means, and the DD name derived by the second derivation means Fifth deriving means for deriving, from the data, an appearance start number of an item related to DD in the program and an appearance end number of an item related to DD in the program;
By searching the sixth table storage means, the program name data derived by the third derivation means or the child program name data extracted by the fourth extraction means and the program derived by the fifth derivation means From the appearance start number of the item related to the DD and the appearance end number of the item related to the DD in the program, the appearance order number of the item declared in the program, the item name data, the level number, and the attribute data , An offset value and the number of bytes, a sixth obtaining unit,
Editing for editing and outputting the file layout from the appearance order number of the item declared in the program, the item name data, the level number, the attribute data, the offset value, and the number of bytes acquired by the sixth acquiring means. Output means,
When the device name data, the volume name data, and the data set name data, which are the data specifying the data set, are input from the input means, the file layout of the data set in the program for inputting or outputting the data set is edited and output. Output from
[0010]
A computer program according to claim 2 according to the present invention ,
In the reverse engineering apparatus according to claim 1, based on the apparatus name data, the volume name data, and the data set name data, which are data for specifying a data set,
The edit output unit that outputs the file layout of the data set in the program receives the appearance order number of the item declared in the program, the item name data, the level number, the attribute data, the offset value, and the number of bytes. It is a computer program to create.
[0012]
The recording medium according to claim 3 according to the present invention,
A recording medium for recording the computer program according to claim 2.
[0013]
BEST MODE FOR CARRYING OUT THE INVENTION
In the present invention, a repository generated by reverse engineering in application system development and maintenance is used. Here, "reverse engineering" is to determine the components of the system and their interrelationships and re-express the system in another form or generate a high-level system representation. Further, the “repository” here is a database of information describing the object of use or operation in the application system, and a meta-data about data items, programs, system resources, and various definition bodies in the application system. It is a database of information.
[0014]
The “repository created by reverse engineering” in the present invention is a known technology. As described below, the repository is preferably constructed according to a relational database model (see FIGS. 4 and 6).
[0015]
FIG. 9 is a block diagram schematically illustrating generation of the repository 14 according to the present invention. The repository 14 according to the present invention includes a program 2, a JCL (Job Control Language; Job Control Language) 4, a COPY clause 6, a DB (database) definition unit 8, and a current component, which are components of a subsystem (application system). Various definition bodies such as the management data 10 are used as original data. The repository 14 is formed by an analysis process for organizing and mapping these interrelationships.
[0016]
The above-mentioned various definition bodies constitute a data set on a recording medium. Therefore, the above analysis processing is performed using the file group of various definition bodies as an input file. This analysis processing can be executed in a computer such as a personal computer, a workstation, a mainframe, or the like. The computer that performs the analysis processing is not particularly limited.
[0017]
Assuming that the latest definition file group is an input file, the latest components are naturally mapped in the application system components and their interrelationships. Therefore, if a system is created that automatically creates documents such as system specifications, design documents, and requirement documents based on the repository without manual intervention, it is ensured that the documents are kept up-to-date. .
[0018]
The types of the various definition bodies are not particularly limited. For example, the programming language for the source program may be COBOL, FORTRAN, PL / I, C language, or the like. In the following, the description will proceed with the programming language being COBOL and the processing mode being batch processing (mainly).
[0019]
<< Preferred embodiment >>
A preferred embodiment of a reverse engineering tool for grasping the file layout of each data set (file) input / output in the application system using the above repository will be described. The reverse engineering tool is operated on a computer such as a personal computer, a workstation, and a mainframe.
[0020]
In grasping the file layout of an actual file (hereinafter, referred to as an actual file), the source program alone is not sufficient, and the JCL is indispensable. In a program, usually, a virtual file name (that is, a file name valid only in the program) is described without directly describing a real file name, and the virtual file name is associated with the real file name in JCL. Because it is.
[0021]
If the virtual file name in the program can be grasped, the file layout description portion in the input / output record definition portion and the variable declaration portion can be grasped via the description portion of the virtual file name in the program. FIG. 2 is a schematic diagram simply showing how variable declarations (related to file layout) are traced from an actual file.
[0022]
In FIG. 2, a simple configuration example of the job control language (JCL) is shown on the left, and a simple configuration example of the COBOL program is shown on the right. From the left part of FIG. 2, the real file “A01” is used (that is, input or output) as the virtual file “* A” in the “PGM1” program (or the subprogram called by the “PGM1” program). (The subprogram will be described later in detail).
[0023]
Check the "PGM1" program. If the portion related to the virtual file “* A” is extracted from the source code of the “PGM1” program, the file layout of the real file “A” viewed from the “PGM1” program can be determined.
[0024]
In the “PGM1” program, from the input / output (virtual) file name use declaration related to “* A”,
・ "Input / output instruction" part,
-It will be possible to search (extract) the "variable declaration" part. From the "variable declaration" part, the file layout is uniquely determined and can be displayed on, for example, a display screen of a personal computer (see FIG. 1). Here, the “variable declaration” portion is related to the data item specified by the “input / output instruction”. Therefore, if it is a COBOL program,
(A) Data items that are directly used in READ and WRITE statements and are declared in WORKING STORAGE SECTION.
(B) data items declared in FILE SECTION
Are output.
[0025]
The program name specified in the job control language is the main program. By the way, in many cases, file use (input / output) is not performed by the main program but by a subprogram called by the main program. In the reverse engineering tool according to the present invention, as will be described later, information (table) for associating the main program with the sub-program is referred to, so that even if the target program is a sub-program, it can be extracted without omission.
[0026]
It can be easily assumed that there are a plurality of programs that use one real file. By tracing a plurality of programs each using one real file, a plurality of (different) file layouts for one real file viewed from the plurality of programs are derived.
[0027]
Next, a procedure for creating a file layout for one actual file using a repository (table) created from the JCL and a program and outputting the created file layout to a computer screen will be described.
[0028]
FIG. 3 is an example of a template of the source code of the job control language. FIG. 4 is a repository (table) created from the source code of FIG. (1) is a table of “Relationship between job and dataset” (tblDataset_JOB).
・ Volume name,
・ Data set name,
・ Job name,
Step number in the job (the step number),
-DD number in Step (the order of DD),
The data about the file is stored as its contents. The respective item names are “Unit”, “Volume”, “DSName”, “Job”, “StepNum”, and “DDNum”, as shown in FIG.
[0029]
Here, as is well known, “DD” is “Dataset Definition (data set definition)”, and immediately before “DD” in the DD line in the JCL, the above “virtual file” name in the program is used. Is described.
[0030]
FIG. 4B is a table of “DD information” (tblDD),
・ Job name,
Step number in the job (the step number),
-DD number in Step (the order of DD),
・ Data set name,
・ DD name,
The data about the file is stored as its contents. The respective item names are “Job”, “StepNum”, “DDNum”, “DSName”, and “DD”, as shown in FIG.
[0031]
FIG. 4C is a table of “Relation between job and step” (tblSTEP).
・ Job name,
Step number in the job (the step number),
·program name,
The data about the file is stored as its contents. The respective item names are “Job”, “StepNum”, and “PGM”, as shown in FIG. (1), (2), and (3) are preferably created in accordance with the relational database model.
[0032]
FIG. 5 is an example of a program code model described in COBOL. FIG. 6 is a repository (table) created from the program code of FIG. (1) is a table of “relation between program and DD” (tblEXTMODDD_LY),
・ Program name (external module name),
・ DD name (from the ASSIGN clause of the SELECT statement),
An appearance number (start) of an item related to the DD in the program,
An appearance number (end) of an item related to the DD in the program;
The data about the file is stored as its contents. The respective item names are “EXTMOD”, “DD”, “FromSeq”, and “ToSeq”, as shown in FIG.
[0033]
FIG. 6B is a table of “items declared in the program” (tblLocalItem),
・ Program name (external module name),
・ Sequential numbers in the order of appearance of items declared in the program,
・ Level number,
・ Item name (symbol name),
·attribute,
・ Offset value,
・ Number of bytes,
The data about the file is stored as its contents. The respective item names are “EXTMOD”, “Seq”, “LevelNum”, “Item”, “Attr”, “Offset”, and “Byte”, as shown in FIG. 6 (2).
[0034]
FIG. 6 (3) is a table of “program parent-child relation” (tblRELEXTMOD),
・ Parent (calling side) program name (external module name),
-Child (callee) program name (external module name),
The data about the file is stored as its contents. The respective item names are “P_EXTMOD” and “C_EXTMOD” as shown in FIG. 6 (3). In (1), (2), and (3) of FIG. 6, it is desirable that all of them be created in accordance with the relational database model.
[0035]
In the “relation between program and DD” (tblEXTMODDD_LY) table in FIG. 6A, “appearance number (start) (FromSeq) of item related to DD in program” and “item number related to DD in program” The “appearance number (end) (ToSeq)” is determined as follows.
[0036]
First, when a READ instruction of a file related to a DD name exists in a program, the tblLocalItem table is used for an item name of an FD statement related to an input file name or an item name described after an INTO clause of a READ statement. refer. The above-mentioned item name is determined based on the magnitude relationship between the “level number (LevelNum)” in the corresponding row in which the item name is stored and the “level number (LevelNum)” stored in the preceding and following rows (particularly, the subsequent row). Is determined, and “ToSeq” (maximum value) is determined and stored in the corresponding row of the “relationship between program and DD (tblEXTMODDD_LY)” table.
[0037]
When the WRITE instruction of the file related to the DD name exists in the program, the item name of the FD statement related to the output file name or the item name described after the FROM clause of the WRITE statement is similar to the above. "FromSeq" (minimum value) and "ToSeq" (maximum value) are determined.
[0038]
FIG. 10 shows a sample COBOL program (MACTH-KOUSHIN) and a part of the tblLocalItem table. Regarding "DD name 1" (DD name) of "MACTH-KOUSHIN" (program name) in FIG.
(FromSeq, ToSeq) = (1, 1), (6, 12)
It turns out that it becomes.
[0039]
FIG. 8 shows, from a principle aspect, a procedure in which the reverse engineering tool according to the preferred embodiment of the present invention creates a file layout for one actual file. FIG. 7 is a diagram for explaining the procedure in FIG. 8 using the above-described tables (FIGS. 4 and 6). The procedure of FIG. 8 is shown below.
[0040]
S02: Information of a real file is input. The information is “Unit (device name)”, “Volume (volume name)”, and “DSName (data set name)”.
[0041]
S04: “Job (job name)”, “StepNum (Step number in job)”, “DDNum (DD number in Step)” from the actual file information in S02 based on “Relationship between job and dataset (tblDataset_JOB)” table ) "Is derived.
[0042]
S06: The corresponding row in the “DD information (tblDD)” table is derived from the above “Job”, “StepNum”, and “DDNum” ((A) to (A) in FIG. 7). Similarly, the corresponding line in the “relation between job and step (tblSTEP)” table is derived from the above “Job” and “StepNum” ((a) to (c) in FIG. 7).
[0043]
S08: The related DD name (DD) is determined from the data included in the corresponding row of the above-mentioned “DD information (tblDD)” table. Similarly, the related program name (PGM) is found from the data included in the relevant row of the above-mentioned "relation between job and step (tblSTEP)" table. With reference to the "DD name (DD)" and "program name (PGM)" as keys, the "reference between program and DD (tblEXTMODDD_LY)" table is referred to ((a) (c) to (d) in FIG. 7).
[0044]
S10: When referring to the above-mentioned "relationship between program and DD (tblEXTMODDD_LY)" table, use the "program parent-child relation (tblRELEXTMOD)" table to change "program name (PGM)" to "parent program name (P_EXTMODOD). )), And all child programs (sub-programs) under the program are also extracted. Then, a key is further created with all of the "child program name (C_EXTMOD)" and the above-mentioned "DD name (DD)", and the "program-DD association (tblEXTMODODD_LY)" table is referenced (FIG. 7). (A) (o) to (d)).
[0045]
S12: The program name (including the sub-program) and the start item number and end item number of the layout associated with the DD name declared in the program name are derived from the extraction result by the reference.
[0046]
S14: The data items corresponding to the above-mentioned start item number to end item number are edited and output as a file layout from the “items declared in the program (tblLocalItem)” table ((FIG. 7: C) to (d)). When creating a file layout, information on a level number, an attribute of a data item, an offset value, and the number of bytes is required, and these are obtained from the tblLocalItem table.
[0047]
FIG. 1 is an output example of a file layout on a computer display device. As shown in FIG. 1, on the screen,
・ UNIT,
・ VOLUME,
·data set,
-It is set so that the used external module is input as condition data. On this screen, a file layout is created and output in accordance with the procedure (principle) of FIG.
[0048]
As explained earlier, if the programming language is COBOL,
(A) Data items that are directly used in READ and WRITE statements and are declared in WORKING STORAGE SECTION.
(B) data items declared in FILE SECTION
Are output.
[0049]
≫Other embodiments≫
The preferred embodiment according to the present invention has been described above, but the present invention is not limited to the above embodiment. If the relationship between the file and the program is clear, a mechanism is established for a program in a language other than COBOL by specifying the actual file and the program in the same procedure and grasping the file layout viewed from the program. It is possible. Further, a similar mechanism can be constructed not only in batch processing but also in a system related to online processing.
[0050]
【The invention's effect】
By using the present invention, the latest file layout of an actual file that actually exists can be accurately and easily grasped for each program in the development and maintenance work of an application business system.
[Brief description of the drawings]
FIG. 1 is an example of output of a file layout on a computer display device.
FIG. 2 is a schematic diagram simply showing how to follow a variable declaration related to a file layout from an actual file name.
FIG. 3 is an example of a template of a source code of a job control language.
FIG. 4 is a part of a repository (table) created from the source code of FIG. 3;
FIG. 5 is an example of a model of a program code described in COBOL.
FIG. 6 is a part of a repository (table) created from the program code of FIG. 5;
FIG. 7 is a diagram for explaining, by using a table example, a procedure in which the reverse engineering tool according to the preferred embodiment of the present invention creates a file layout for one actual file.
FIG. 8 is a flowchart showing a procedure for creating a file layout for one actual file by a reverse engineering tool according to a preferred embodiment of the present invention from a principle aspect.
FIG. 9 is a block diagram schematically illustrating generation of a repository according to the present invention.
FIG. 10 is a sample of a COBOL program (MACTH-KOUSIN) and a part of its tblLocalItem table.
[Explanation of symbols]
2 ... program file, 4 ... JCL file, 6 ... COPY clause file, 8 ... DB definition file, 10 ... other current management data file, 12 ... various definition file, 14 ... Repository.

Claims (3)

In a reverse engineering device consisting of a computer that reads and uses data stored in a repository database,
  The repository database is
First table storage means in which device name data, volume name data, and data set name data are associated with job name data, step numbers in the job, and DD numbers in the step, and
  Second table storage means in which the job name data, the step number in the job, the DD number in the step, and the DD name data are stored in association with each other;
  A third table storage unit in which job name data, Step numbers in the job, and program name data are stored in association with each other;
  A fourth table storage in which the program name data, the DD name data, and the appearance start number of the item related to the DD in the program and the appearance end number of the item related to the DD in the program are related and stored. Means,
  Fifth table storage means in which parent (calling side) program name data and child (calling destination) program name data are associated and stored;
  Sixth table storage means in which the program name data, the appearance order number of the item declared in the program, the item name data, the level number, the attribute data, the offset value, and the number of bytes are associated and stored. And
The first table storage, the second table storage, and the third table storage are generated by analyzing a job control language source code, which is one of definition data, and are stored in a fourth table storage. The means, the fifth table storage means, and the sixth table storage means are generated by analyzing a program source code which is one of the definition data,
  In addition, the reverse engineering device
  Input means for inputting device name data, volume name data, and data set name data that are data specifying a data set;
  The first table storage unit is searched to derive the job name data, the step number in the job, and the DD number in the step from the device name data, volume name data, and data set name data input from the input unit. First deriving means for performing
  A second deriving unit that retrieves the second table storage unit and derives the DD name data from the job name data, the Step number in the job, and the DD number in the Step derived by the first deriving unit; ,
  A third deriving unit that retrieves the third table storage unit and derives program name data from the job name data derived by the first deriving unit and the Step number in the job;
  Fourth extraction means for searching the fifth table storage means, applying the program name data derived by the third derivation means to the parent program name, and extracting child program name data of child programs under the parent program name;
  Searching the fourth table storage means, the program name data derived by the third derivation means or the child program name data extracted by the fourth extraction means, and the DD name derived by the second derivation means Fifth deriving means for deriving, from the data, an appearance start number of an item related to DD in the program and an appearance end number of an item related to DD in the program;
  By searching the sixth table storage means, the program name data derived by the third derivation means or the child program name data extracted by the fourth extraction means and the program derived by the fifth derivation means From the appearance start number of the item related to the DD and the appearance end number of the item related to the DD in the program, the appearance order number of the item declared in the program, the item name data, the level number, and the attribute data Sixth acquisition means for acquiring the offset value and the number of bytes;
  From the appearance order number of the item declared in the program, the item name data, the level number, the attribute data, the offset value, and the number of bytes acquired by the sixth acquiring means, Editing output means for editing and outputting the layout.
  When the device name data, the volume name data, and the data set name data, which are the data specifying the data set, are input from the input means, the file layout of the data set in the program for inputting or outputting the data set is edited and output. Output from
Reverse engineering equipment. In the reverse engineering apparatus according to claim 1, based on the apparatus name data, the volume name data, and the data set name data, which are data for specifying a data set,
The edit output unit that outputs the file layout of the data set in the program receives the appearance order number of the item declared in the program, the item name data, the level number, the attribute data, the offset value, and the number of bytes. Computer program to create. A recording medium for recording the computer program according to claim 2.

Images