JPH06282424A - Automatic generating device for program specifications - Google Patents

Abstract

PURPOSE:To automatically generate program specifications by providing a remark information extracting means, an input/output information extracting means, a symbol figure generating means, a character string information generating means, and an information positioning means. CONSTITUTION:When the kind of data discriminated in a source program analyzing processing 1.1 is a comment, the data are extracted as the comment in a processing 1.2 and when input/output information, the input/output information is extracted in a processing 1.3. A next processing 1.4 reads information on the symbol figure corresponding to the extracted input/output information out of an input/output definition files 1.10 and processing 1.5 reads information on a corresponding formal name out of a term definition data base 1.11. Then a processing 1.6 determines positions on the specifications as to the pieces of information of the processings 1.2, 1.4, and 1.5 and a processing 1.7 determines the position of an arrow. A final processing 1.8 generates data on the specifications on the basis of various information obtained by the respective processing by referring to the contents of specification original data 1.12 and symbol figure data 1.13.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a program design document used for developing and maintaining software in a computer system and reconstructing an existing system.

[0002]

2. Description of the Related Art When maintaining software in a computer system or rebuilding the software based on an existing computer system, first, a deep understanding of the hardware and software configuration and operation of the existing system is required. There must be.

Generally, software executed by a computer system is a machine language program obtained by compiling a source program described in a predetermined programming language. Therefore, if the contents of the source program can be understood, the functions and operations of the computer system can be understood. However, since the description content and structure of a normal software language is far from the natural language used by humans,
It is difficult to understand the contents of the source program. Even the person who actually created the source program forgets the contents over time, so it is difficult to understand the source program later.

Therefore, in a company or the like, a programmer creates a program design document in which the functions and operations of a computer system are described in a format that is easy for humans to understand, and the programmer creates the program design document separately from the source program. It is desired that the person who performs the reconstruction can understand the function and operation of the computer system by referring to the program design document.

However, since the work of creating a program design document is heavy on the engineer, at present, only the outline is described in the program design document, or when the program is modified, the contents of the program design document are modified. Is often omitted, so the program design document may not be useful. Further, even when the design document is created at the program development stage, the work of modifying the design document with respect to the modification of the source program during the debugging of the program imposes a heavy burden on the engineer.

A technique for automatically creating a flow chart to help understand a program is disclosed in, for example, Japanese Patent Laid-Open No. Hei 3
-130828 and Japanese Patent Laid-Open No. 4-245327. Further, Japanese Patent Laid-Open No. 4-74220 proposes a technique for creating a document.

[0007]

However, it is not possible to automatically create a program design document by the conventional technique. Also, since the description contents of the flow chart are not in natural language, it is difficult to understand them as in the case of the source program.

[0008] Therefore, it is an object of the present invention to enable automatic creation of a program design document.

[0009]

In order to solve the above-mentioned problems, the automatic program design document creating apparatus of the present invention is a software
Annotation information extracting means (1.2) for reading the contents of the conversion source data (1.9) such as a text program and extracting the annotation information contained therein; included in the conversion source data , An input / output information extraction means (1.3) for extracting information about input and output; a symbol graphic generation means (1) for generating information of a symbol graphic associated with the input / output information extraction means from the information extracted by the input / output information extraction means. . 4); Character string information generating means (1.) for generating character string information associated with the information extracted by the input / output information extracting means.
5); and the annotation information extracted by the annotation information extracting means,
The symbol graphic information generated by the symbol graphic generation means and the character string information generated by the character string information generation means,
Each of them is provided with an information positioning means (1.6) which is assigned to a predetermined position according to the contents of the conversion source data.

In a preferred embodiment, the input / output information extraction means is a specific sub-routine contained in the conversion source data.
Input / output processing detection means (4.2) for detecting a chin call instruction, function, or procedure; name of input / output processing, input / output classification of input / output processing, type of input / output information of input / output processing, and input An input / output definition file means (1.10) in which information on the number of parameters passed to the output processing is held in advance; and an input detected by the input / output processing detection means with reference to the information of the input / output definition file means. Including identification means (4.3 to 4.6) for identifying input / output information of output processing. The symbols shown in the above parentheses are shown by reference to the reference numerals of corresponding elements in Examples described later. However, the constituent elements of the present invention are not limited to the specific elements in the embodiments.

[0011]

According to the present invention, the annotation information extracting means extracts the annotation information (comment sentence or the like) from the conversion source data (source program or the like), and the input / output information extracting means extracts the conversion source data. Information about inputs and outputs from the data. Also,
The symbol graphic generation means generates information of the symbol graphic associated with it from the information extracted by the input / output information extraction means, and the character string information generation means associates it with the information extracted by the input / output information extraction means. Generate string information that can be Then, the information positioning means respectively converts the annotation information extracted by the annotation information extracting means, the symbol graphic information generated by the symbol graphic generating means, and the character string information generated by the character string information generating means into the conversion source data. −
It is assigned to a predetermined position according to the contents of the data.

In this way, the program design document automatically generated is the annotation information associated with each process in the source program, the symbol graphic information indicating the contents of input and output, and the input and output. It includes character string information indicating the contents of the. Generally, the annotation information in the source program is a sentence explaining the contents of each process. Also, many processes include input and output operations. Therefore, from the content of the generated program design document, the content of the process of each step in the program, the information input to the process (for example, the name and type of the input file) and the information output from the process (for example, the output You can know the name and type of the file.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows the functional configuration of an automatic program design document creating apparatus according to an embodiment. This automatic program design document creating apparatus is realized by, for example, a computer system such as a workstation and an application program of this apparatus executed on the computer system. The source program 1.9 and the input / output definition file 1.
10, term definition data base 1.11, design document base paper data 1.12, and symbol graphic data 1.13 are prepared in advance and held on a hard disk, for example.

An example of contents of the source program 1.9 is shown in FIG. 9, an example of contents of the input / output definition file 1.10 is shown in FIG. 11, and contents of the term definition database 1.11 are shown. An example is shown in FIG. 20, an example of the contents of the design document base paper data 1.12 is shown in FIG. 18, and an example of the contents of the symbol graphic data 1.13 is shown in FIG. 19 shows an example of the display form of the program design document generated by this device. The source program 1.9 shown in FIG.
It is written in the programming language of ORTRAN77.

An outline of each process shown in FIG. 1 will be described first. In the source program analysis process 1.1, the data of each line of the source program 1.9 is read and the data is read.
Identify the type of data. Data type is comment (comment)
If it is, the process proceeds to process 1.2 and the data is extracted as a comment. When the type of data is input / output information, the process proceeds to processing 1.3 and the input / output information is extracted from the data. At this time, the contents of the input / output definition file 1.10 are referenced. In the next processing 1.4,
Information on the symbol graphic corresponding to the extracted input / output information,
Read from input / output definition file 1.10. Further, in the next processing 1.5, information of the official name (Japanese) corresponding to the extracted input / output information is searched from the term definition database 1.11, which is a dictionary, and read.

In process 1.6, the comment information extracted in process 1.2, the symbol graphic information obtained in process 1.4, and the formal name information obtained in process 1.5 are generated. Determine the position on the design document to be used. In the next process 1.7, the positions of the arrows (indicating the connection between the input and the process and the process and the output) described on the generated design document are determined.

In the final processing 1.8, based on various information obtained by the above processing, and referring to the contents of the design document base paper data 1.12 and the symbol graphic data 1.13,
Generate design document data. The contents of the design document can be displayed in a form as shown in FIG. 19 by reading the data of the design document generated in this way with, for example, a word processor.

Next, the details of the specific processing will be described. The contents of the process 1.1 of FIG. 1 are shown in FIG. 2, the contents of the process 1.2 of FIG. 1 are shown in FIG. 3, and the processes 1.3 and 1.
4 shows the contents of FIG. 4, and the contents of processing 1.5 of FIG. 1 are shown in FIG.
1, the contents of the process 1.6 of FIG. 1 are shown in FIG. 6, the contents of the process 1.7 of FIG. 1 are shown in FIG. 7, and the contents of the process 1.8 of FIG. 1 are shown in FIG.

First, a description will be given with reference to FIG. In the first step 2.1, the source program file 1.9
Open and make it readable. In the next step 2.2, 1 is stored in the block No register and the block flag is cleared.

In step 2.3, one line of data is read from the source program file 1.9 and stored in the memory M.
Store in AIN. Each data string delimited by a line feed code is one line. For example, in the second line of the source program shown in FIG. 9, the data string "C 1. Read data from file" is read and stored in the memory MAIN.

In step 2.4, it is checked whether the end of the source program file 1.9 is detected. When the end is detected, the process proceeds to step 6.1 in FIG. 6, and if not, the process proceeds to the next step 2.5.

In step 2.5, the data of one line held in the memory MAIN is referred to, and the first character is "C".
Check whether it is the character code of. FORTRAN7
In the programming language of 7, when the first character of each line is "C", the data of the entire line including it is
It means a comment. The contents of the comment line have no effect on the actual contents of the program, but in general, in order to facilitate understanding of the source program, the explanation of the processing executed in that part is described as a comment. It That is, in step 2.5, it is checked whether or not one line of the read source program is a comment.

If the comment line is detected in step 2.5, the process proceeds to step 3.1 of FIG. 3, otherwise, in the next step 2.6, the block flag is cleared and step 2. Proceed to 7.

In step 2.7, the contents of the memory MAIN which holds the data of one line of the read source program are referred to, and it is checked whether or not the character string "CALL" exists in it. Find out. In the FORTRAN77 programming language, the character string "CALL" in the instruction code is assigned to the instruction for calling the subroutine.

For example, 4 of the source program shown in FIG.
"CALL SGREC (FILE1, ...
"" Calls a subroutine assigned the name "SGREC", and at that time, "FILE1, ..."
As a parameter to the subroutine.

When "CALL" is detected in step 2.7, the process proceeds to step 4.1 in FIG. 4, otherwise, the process goes through step 2.8 and returns to step 2.3. The above process is executed for all the lines of the source program. Next, description will be made with reference to FIG. In step 3.1, it is checked whether the second character stored in the memory MAIN is a blank (space) code,
If it is a blank code, the process proceeds to the next step 3.2, and it is checked whether or not the second and subsequent character strings held in the memory MAIN are all blank codes. If there is a character code other than a blank, then go to step 3.3.

In step 3.3, the first character string (for example, "C" and a blank character) is deleted from the data of one line held in the memory MAIN, and the remaining data is processed information (comment). Content)).

At the next step 3.4, the block flag is referred to, and when it is cleared to 0, the value of the block No register is incremented (+1) at step 3.5, and the block is blocked at step 3.6. The flag is set, and 1 is set in the row number register in step 3.7.
When the block flag is set to 1 in step 3.4, the value of the row number register is incremented in step 3.9.

In step 3.8, the processing information (contents of the comment), the contents of the block number register, the contents of the line number register, and the page number extracted in step 3.3 are stored in the memory.

By the processing of FIG. 3, the contents of the comment in the source program are extracted by being divided into blocks as shown in FIG. 9, for example.

Next, the processing shown in FIG. 4 will be described. In step 4.3, open the input / output definition file 1.10,
Make it readable. In the next step 4.4, read all the contents of the input / output definition file 1.10.
The read data group is stored in a predetermined memory array.

The contents of the input / output definition file 1.10 are shown in FIG.
Is a group of text format character string data as shown in 1.
It consists of a number of lines separated by line feed codes. The string on each line contains four parameters separated by a colon ":". For example, the character strings on the first line in FIG. 11 are “SGREC”, “INPUT”, and “SY”.
It includes four parameters of MBL01 "and" 1 ". These four parameters are each
It is associated with the chin name, the input / output classification, the symbol figure name, and the number of parameters passed to the subroutine. All these parameters are in step 4.4,
Each row is divided and stored in independent memory areas.

In the next step 4.5, the memory MAIN
For one line of data stored in file 1.
Identify whether any of the subroutine names (SUB) registered in 10 is included.
That is, "SGREC", "SPEC", "SGIT"
The character strings “M”, “SPITM”, ...
It is retrieved from N data.

In step 4.5, if any of the registered subroutine names is detected from the data of one line stored in the memory MAIN, the input / output definition file 1.10 , An input / output section (“INPUT” or “OUTPUT” associated with the subroutine name.
T ”), symbolic graphic name (“ SYMBL01 ”,“ SY
The values of MBL02 "," SYMBL03 ", etc.) and the block No. register are stored in step 4.6.

That is, each time a registered subroutine is found in each line of the source program data, the block number at which it is found, the input / output classification of the subroutine, and the relevant The symbol graphic names associated with the subroutines are stored respectively. Then, the process proceeds to step 5.1 in FIG.

Next, description will be made with reference to FIG. In step 5.1, the detected subroutine name (for example, "SGRE" in the data of one line held in the memory MAIN).
Detect the left parenthesis '(' at the position following "C").

In the next step 5.2, step 5.1
From the character string data following the left parenthesis '('
Information on all parameters of the subroutine (INF)
Respectively. For example, in the fourth line of the source program shown in FIG. 14, "CALL SGREC (FIL
"FILE1," ... from the character string "E1, ..."
.. is extracted as the information INF.

In the next step 5.3, step 5.2.
For each of the information INFs extracted in step 1, the official name (Japanese name) associated with it is defined in the term definition database.
Search from 1.1.1. As shown in FIG. 20, in the term definition database 1.11, a large number of pieces of information on the English keyword and the official name (Japanese) associated therewith are registered. Therefore, for example, when the extracted information INF is "FILE1", the character string information of "file 1" is generated. This character string information is stored in the next step 5.4 in association with the subroutine name and block number (see FIG. 14).

Next, description will be made with reference to FIG. In this processing, the data as shown in FIG. 14 generated by the processing so far is processed. First, in step 6.1, 1 is set in the page No register and the page flag is cleared. In the next step 6.2, it is identified whether the position determination has been completed for the generated data of all blocks. If not completed, the process proceeds to the next step 6.3, and if completed, the process proceeds to step 7.1 of FIG.

In step 6.3, the page flag is checked. When this is cleared, the procedure proceeds to the next step 6.4, and otherwise, the procedure proceeds to step 6.16.

In step 6.4, the total number of lines of data assigned to the block number currently being processed is obtained. In the next step 6.5, the position of the last line of the information of the block is calculated from the total number of lines obtained in step 6.4, and in the next step 6.6, the position of the last line is calculated from the position of the last line. It is checked whether all the text information (in the central portion of FIG. 15) of (1) can fit on one sheet of paper. If it fits, then go to step 6.7, and if it goes out, go to step 6.12.

In step 6.7, the total number of symbol graphics included in all the data assigned to the block number currently being processed is calculated for each of the input and the output. Then, in the next step 6.8, the position of the final line of the symbol graphic information (the graphic in the left column and the right column in FIG. 15) of the block is calculated from the total number of symbol graphics for each input and output. In the next step 6.9, it is checked from the position of the last line of the input and output obtained in step 6.8 whether all the symbol graphic information of the block fits on one sheet. When it fits, the process proceeds to step 6.10, and when it protrudes, the process proceeds to step 6.13.

In step 6.10, the page flag is set, and in the next step 6.11, the block No. is incremented (+1) and the process returns to step 6.2 to move to the next block data processing.

When it is determined in step 6.6 that the information in the block is out of the paper, in step 6.12
Update the page number (for text information) of the line that protrudes. For example, when the previous page number is 1,
Set page number to 2. Similarly, when it is determined in step 6.9 that the input / output information in the block is out of the paper, in step 6.13, the page number (for symbol graphic) of the line that has overflowed is updated. And the next step 6.1
At 4, the page flag is cleared. Continued Step 6.15
Then, the text information page No. and the symbol figure page
The result of adding 1 to the larger value of the page number is stored as the page number of the entire block.

In step 6.16, the larger of the position (Y coordinate) of the last line of the text and the final position (Y coordinate) of the symbol graphic regarding the previous block is set to Ymax. In step 6.17, the margin value is added to Ymax to update it.

In step 6.18, the total number of lines of text information of the block being processed is calculated, and the next step 6.19.
Then, the final position of the text information of the block being processed is
It is calculated from the final position Ymax of the previous block and the total number of rows in the block being processed. In the next step 6.20, it is checked from the final position of the text information of the block being processed whether or not all the text information of the block (the central portion in FIG. 15) fits on one sheet of paper. If it fits, then step 6.
21. If it goes out, go to step 6.26.

In step 6.21, the total number of symbol graphics included in all the data assigned to the block number currently being processed is calculated for each of the input and the output.
Then, in the next step 6.22, the position of the final line of the symbol graphic information (the graphic in the left column and the right column in FIG. 15) of the block is calculated from the total number of symbol graphics for each input and output. Next step 6.2
In step 3, it is checked from the position of the last line of the input and output obtained in step 6.22 whether or not all the symbol graphic information of the block fits on one sheet of paper. When it fits, the process proceeds to step 6.24, and when it protrudes, the process proceeds to step 6.26.

In step 6.24, the block number is incremented, in the next step 6.25, the page flag is set, and the process returns to step 6.2. Step 6.
At 26, the page number is incremented, the page flag is cleared, and the process returns to step 6.2.

By the processing of FIG. 6, for example, as shown in FIG. 15, the text information and the symbol graphic information of each block are properly positioned. Of course, the processing of FIG. 6 merely determines the position of each information, and the actual graphic information and the like are generated by the subsequent processing.

Next, description will be made with reference to FIG. This processing is to determine the position of the arrow line connecting the input symbol graphic and the text information frame and the arrow line connecting the text information frame and the output symbol graphic as shown in FIG. FIG. 14 obtained by the above processing
The position of each arrow is determined based on the information as shown in.

The first step 7.1 is to identify for each symbol graphic whether it is an input or an output. That is, the input / output character string associated with it is "INPU
If it is "T", it is regarded as an input, and if it is "OUTPUT", it is regarded as an output. In case of input, proceed to step 7.2,
If it is an output, go to step 7.8.

In step 7.2, the coordinates of the left end of each arrow line are obtained, and in step 7.3, the coordinates of the right end of each arrow line are obtained. In the next step 7.4, it is identified whether or not the coordinates of the right end of the arrow line are within the text information area of the block. If it is outside the area, in step 7.6, the coordinates of the right end of the arrow are set to Y in the last line of the text information area of the block.
Replace with coordinates. By this replacement, for example, FIG.
Even when a large number of symbolic figures are associated with the text area of one block as shown in FIG. 6, the connection relationship between them can be appropriately indicated by arrows.

In step 7.8, the coordinates of the left end of each arrow line are obtained, and in step 7.9, the coordinates of the right end of each arrow line are obtained. In the next step 7.10, it is identified whether or not the coordinate of the left end of the arrow is within the text information area of the block.
If it is outside the area, in step 7.11, the coordinate of the left end of the arrow is replaced with the Y coordinate of the last line of the text information area of the block.

At the final step 7.7, the obtained coordinates of the left end and the right end of each arrow line are stored in a predetermined memory.

Next, description will be made with reference to FIG. In step 8.1, the page number is set to 1 and the block number is set.
Set 1 to. In the next step 8.2, the design document base paper data 1.12 (see FIG. 18) is read and stored in the memory. The data of the design document is created based on this data.

In step 8.3, for each block, the length of the text frame in the Y-axis direction is calculated from the number of lines in the block. In the next step 8.4, the block number is checked, and if it is 1, the process proceeds to step 8.5, and if not, the process proceeds to step 8.9.

In steps 8.5 and 8.9, the starting point coordinates and the ending point coordinates of the text frame of each block are obtained. Then, in the next step 8.6, each step 8.
Information of a text frame (rectangular figure information) having the start point coordinates and the end point coordinates generated in 5 or 8.9 is generated, and the information is inserted into the raw paper data on the memory. In the next step 8.8, the character code string of the text information is generated so that the text information (comment sentence extracted in 1.2) of the block is placed in the text frame generated in step 8.6. And the information of the coordinates of the position where it is arranged are inserted into the raw paper data on the memory.

In step 8.10, for each symbol graphic, the coordinates of the starting point and the ending point of the position where it is arranged are obtained. These coordinates are determined by the coordinates at which the block is arranged and the order in which the symbol graphic in the block appears.

In step 8.11, the pattern information of the symbol graphic is read from the symbol graphic data 1.13 and stored in the memory. That is, the code of the symbol graphic information generated in the process 1.4 (for example, SYMBOL0
From 1), the pattern information (for example, refer to FIG. 17) of the figure associated with it is read. The correspondence between the code of the symbol graphic information and the actual pattern information is as shown in FIG.

Since the coordinates of this pattern information are initially initial values, in the next step 8.12, the coordinates of the pattern information are calculated based on the coordinates of the start point and end point determined in 8.10. To the coordinates where it is actually placed.

At step 8.13, the pattern information of each symbol graphic is inserted into the raw paper data on the memory.
Further, in the next step 8.14, the pattern information corresponding to the arrow information generated in the processing 1.7 is generated and inserted into the raw paper data on the memory.

When a plurality of symbolic figures are associated with one block, steps 8.10 to 8.15.
The above process is repeatedly executed, and pattern information is generated for each symbol graphic. When the processing of one block is completed, the process returns from step 8.16 to 8.3, and the above processing is repeatedly executed. When the processing of all blocks in the page is completed, the process returns from step 8.17 to 8.2 and the processing of the next page is executed.

When the processing of all pages is completed, the conversion processing of the design document data (base paper data in which various information is inserted) generated by the above processing is executed in step 8.18. . That is, the data is converted into a data format handled by an individual system such as a word processor or a computer for displaying or printing out the design document, and the design document is converted into a format as shown in FIG. Enable display.

In the above embodiment, FORTRA
The case of processing the source program described in the N77 programming language to create the design document is shown, but by the same processing as this embodiment, other programming languages such as COBOL, C, PASCAL, BASIC, etc. It is also possible to create a design document from the source program described in.

[0065]

As described above, according to the present invention, the annotation information extraction means extracts the annotation information (comment sentence etc.) from the conversion source data (source program etc.) and the input / output information extraction means , Information about input and output is extracted from the conversion source data. Further, the symbol graphic generation means generates information on the symbol graphic associated with the information extracted by the input / output information extraction means, and the character string information generation means,
From the information extracted by the input / output information extraction means, character string information associated with it is generated. Then, the information positioning means includes the annotation information extracted by the annotation information extracting means, the symbol graphic information generated by the symbol graphic generating means, and the character string information generated by the character string information generating means,
Each is assigned to a predetermined position according to the contents of the conversion source data. Therefore, a practical program processing design document
It can be automatically generated from the source program.

[Brief description of drawings]

FIG. 1 is a functional block diagram showing a configuration of an apparatus according to an embodiment.

FIG. 2 is a flowchart showing the contents of processing 1.1 in FIG.
It is

FIG. 3 is a flowchart showing the contents of processing 1.2 in FIG.
It is

FIG. 4 is a flowchart showing the contents of processes 1.3 and 1.4 of FIG.

5 is a flowchart showing the contents of processing 1.5 in FIG.
It is

6 is a flowchart showing the contents of processing 1.6 in FIG.
It is

FIG. 7 is a flowchart showing the contents of processing 1.7 in FIG.
It is

FIG. 8 is a flowchart showing the contents of processing 1.8 in FIG.
It is

FIG. 9 is a memory map showing an example of the relationship between a source program and extracted blocks.

FIG. 10 is a memory map showing an example of a relationship between a source program and extracted processing information (comment sentence) and the like.

FIG. 11 is a memory map showing an example of the contents of an input / output definition file.

FIG. 12 is a map showing the correspondence between the code of the symbol graphic and its pattern information.

FIG. 13 is a memory map showing an example of a relationship between a source program and extracted blocks.

FIG. 14 is a memory map showing an example of a relationship between a source program and generated intermediate information.

FIG. 15 is a plan view showing the coordinates of each area on the generated design document.

FIG. 16 is a plan view showing an example of a generated layout on the design document.

FIG. 17 is a memory map showing the contents of one symbol graphic data (pattern information).

FIG. 18 is a memory map showing the contents of design document base paper data.

FIG. 19 is a plan view showing a relationship between a source program and contents of a generated design document.

FIG. 20 is a memory map showing an example of the contents of a term definition database.

[Explanation of symbols]

 1.9: Source program 1.10: Input / output definition file 1.11: Term definition data base 1.12: Design document base paper data 1.13: Symbol graphic data

Claims (2)

[Claims] 1. An annotation information extracting means for reading the contents of conversion source data such as a source program and extracting the annotation information contained therein; included in the conversion source data.
Input / output information extraction means for extracting information about input and output; symbol graphic generation means for generating information on symbol graphics associated with the information extracted by the input / output information extraction means; the input / output information extraction means Generates character string information associated with it from the extracted information,
Character string information generating means; the annotation information extracted by the annotation information extracting means, the symbol graphic information generated by the symbol graphic generating means, and the character string information generated by the character string information generating means, respectively, as the conversion source. An apparatus for automatically creating a program design document, which comprises an information positioning means, which is assigned to a predetermined position in accordance with the contents of data. 2. The input / output information extracting means is a conversion source data source.
I / O processing detection means for detecting a specific subroutine call instruction, function, or procedure included in the data; name of I / O processing, input / output classification of I / O processing, I / O information of I / O processing Input / output definition file means in which information on the type of the input / output and the number of parameters passed to the input / output processing is held in advance; The automatic program design document creating apparatus according to claim 1, further comprising: an identifying unit that identifies input / output information of the output process.