JPH0264822A - Error detecting method for file processing - Google Patents

Abstract

PURPOSE:To facilitate the detection of errors by permitting a compiler to automatically generate an objective program which detects a processing on a unopened file and issues an error message by means of designation. CONSTITUTION:A program is added which identifies processings at every analysis of one instruction when there is prescribed designation, inspects the flag of a file concerned before an access processing when it is the designation of access with respect to the file and outputs the prescribed error message when the flag is in the display of an unopened state. When designation is given to the compiler as necessary, the compiler detects the processing related to the unopened file, and the objective program issuing the error message is automatically generated. Thus, the cause of the errors is easily retrieved since the adequate error message is outputted at the time of access to the unopened file if the program is executed.

Description

[Detailed Description of the Invention] [Summary] Regarding compiler processing for detecting programming errors related to file processing in a computer, a purpose program that detects processing related to unopened files and issues an error message is specified. The purpose of the file processing error detection method is to automatically generate a file processing error detection method that is automatically generated by a compiler to facilitate error detection, and when there is a predetermined specification in the compiler process that translates a computer source program and generates a target program, For each file specified in the source program, a predetermined flag is set to indicate whether it is in an open or non-open state, and the target program generated from the source program that specifies opening and closing of the file is set. After the open and close processing, add a program to display the flag as open state and non-open state, respectively, and add a program to the target program generated from the source program that specifies access to the file, before the access processing. A program is added that checks the flag and outputs a predetermined error message when the flag indicates a non-oven state.

[Industrial application field]

The present invention relates to compiler processing for detecting programming errors related to file processing in a computer.

Since programs for office processing and the like generally use files frequently, there are many programming errors related to file processing, and even if an abnormality is detected during execution of such errors, it is not easy to search for the cause.

[Prior art and problems to be solved by the invention] FIG. 3 is an example of a program written in, for example, the well-known C0BOL language.

rDAT, which defines a file, known in the C0BOL language
A DIVISION J (7) rFILE 5E
A file called FILE-^ is defined in CTTONJ. As is well known, the description here is FILE-
A consists of records called A-REC, and there are 4 records.
It shows that the field consists of a B field consisting of a digit code, a C field consisting of a 5 digit character code, and a FILLER field consisting of a 71 digit character code.

rPROcEDURE DIVISIONJ In the procedure described above, rOPEN 0UTPUT F
ILE-AJ is a program instruction statement (hereinafter simply referred to as an instruction) that specifies the open processing of FILE-A, and the compiler generates a target program that executes the so-called file open processing in response to this instruction. .

This purpose program associates FILE-A with the actual file, sets control information to enable access, and allocates a record area for input/output with that file (i.e., its record area (obtain a pointer that points to). Here, the structure of the record area is A-, which is described in the file definition above.
Make sure to follow the REC record definition.

Continued < rMOVE 1234 TO8J etc.
The target program generated from the OVE instruction moves data to the specified field of the record area pointed to by the pointer prepared above, and executes the next rWRIT.
The object program generated from the lli A-RECJ command writes the contents of the record area to the opened file as FILE-A.

In this example program, then rcLOsEFILE-A
Since there is a J instruction, the compiler responds to this instruction by
When the above processing is completed, a purpose program for closing the file FILE-A is generated, and this close processing purpose program performs the record area allocation and F
Dissolve the connection between ILE-A and the actual file.

Therefore, when the target program generated from the next rDISPLAY CJ command, that is, the command that specifies that the contents of the C field of the record area be output to the console, etc., accesses the storage area pointed to by the pointer as the record area, Since the pointer value already has meaningless content, it is not guaranteed that the result is correct and will generally result in an abnormal condition.

In other words, the reason why the rDIsPLAY CJ instruction is placed in this step is a programming error of accessing the record area of a file that has not been opened, but if this error was overlooked during the source program check stage. In this case, the existence of an error is noticed only when the generated target program is executed and some abnormal execution result is obtained.

However, since the abnormal result generally does not directly indicate an error in file processing, there is a problem in that it is not always easy to search for the cause of the abnormal result.

According to the present invention, a compiler automatically generates a program that detects processing related to unopened files and issues an error message, and operates together with this detection program during execution to facilitate file processing error detection. Aimed at detection methods.

[Means to solve the problem]

FIG. 1 is a process flowchart showing the configuration of the present invention.

The figure shows the flow of processing for inserting a file processing error detection program in a compiler, and 1 to 7 are processing steps.

[For production]

In the process of a compiler that translates a computer source program to generate a target program, each time an l instruction is analyzed, if there is a predetermined designation identified in process step 1 of FIG. 1, process steps 2 to 3 are performed. In step 5 and step 6, after the open and close processing, the process is specified in the target program to be generated from the source program. For each file, a flag is set to indicate whether it is in an open state or a non-open state, and a program is added to display the flag as an open state or a non-open state, respectively.

Furthermore, if the processed instruction specifies access to a file, in step 7 a program is executed that checks the flag of the file before the access process and outputs a predetermined error message if the flag indicates a non-open status. Add.

With the above processing method, if you specify it to the compiler when necessary, a target program that detects processing related to unopened files during compilation and issues an error message will be automatically generated, so when you run that program, , an appropriate error message is displayed when accessing an unopened file, making it easier to search for the cause of the error.

〔Example〕

The flag that indicates the open status of a file, which is required by the present invention, is set by the compiler for each file as a control program (called a file management block, etc.) that holds control information for file access, for example. For example, when the focus is off, the non-open state is displayed, and when the focus is on, the open state is displayed.

Further, the compiler of the present invention allows a file processing error detection option to be specified in addition to various known options at the time of compiling, and is accepted at the time of starting the compiler.

FIG. 2 is a diagram illustrating the source program example shown in FIG. 3 by associating the instructions of the source program, the processing of the compiler, and the execution result of the generated file processing error detection program.

The compiler analyzes each instruction of the source program in the same way as in the conventional case, and creates the program example (7) rOP in Figure 3.
When the EN 0UTPUT FILE-Al command is processed in processing step 10 of FIG. 2, the processing of FIG. If so, the processing here is finished and the processing proceeds to the next instruction.

If the file processing error detection option is specified, the process proceeds to step 5 via step 2, and FILE-A
Turn on the flag provided in the file management block of FILE-A (hereinafter referred to as the FILE-A flag), and
Output a target program that displays that LE-A is in an open state.

In reality, the compiler outputs a target program that calls a required program, and configures this program to call, for example, a subroutine that executes the above processing using the address of the file management block of FILE-A as a parameter. As shown in the execution result column of the figure, when this program is executed, the FILE-A flag is turned on.

Next, rMOVE 1234 To BJ et al. (7) M
OVB command, rWRITE A-RfICl command, etc.
When commands for setting and referencing file-related data are processed in processing step 12, the processing shown in FIG. 1 is entered in processing step 13 after each command is processed, and processing steps 1.2.
3 and 4, proceed to processing step 7, identify the flag to FILE-, and output an error message if a non-oven state is displayed due to the flag being turned off.
The message outputs, for example, a target program that calls a program that instructs the instructions of the corresponding source program.

This program is a program configured to call a subroutine that executes the above processing, for example, using the address of the file management block of FILE-A and the address of the required message content as parameters, and is shown in the execution result column of Figure 2. When this program is executed, the FILE-A flag is on in this case, so no error message is output and the process proceeds normally.

Next, in processing step 14 of FIG.
When the LE-AJ instruction is processed, the first
Enter the process shown in the figure, proceed to process step 6 through process step 1.2.3, turn off the FILE-A flag,
Output a target program that calls a program that indicates that FILE-A is not open.

This program is, for example, a program configured to call a subroutine that executes the above processing using the address of the file management block of FILE-A as a parameter, and as shown in the execution result column of FIG. Then, the flag of FILE-A is turned off.

In the next processing step 16 of FIG.
When the J instruction is processed, the process shown in FIG. 1 is entered in step 17, and the same target program as in step 13 is output.

When this program is executed, as shown in the execution result column of Figure 2, since the FILE-A flag was turned off in the above process, an error message is output. For example, the type of error and the relevant instruction are shown. Therefore, the user can easily detect a programming error in attempting to refer to a record area of a file that is not opened by the command.

〔Effect of the invention〕

As is clear from the above description, according to the present invention, in detecting programming errors related to computer file processing, a compiler automatically generates a program for detecting processing related to unopened files and issuing an error message. Therefore, when the program is executed, an error is detected and an appropriate message is output, the cause of the error can be easily searched, and the work efficiency of program development is improved, which has a significant industrial effect.

[Brief explanation of the drawing]

FIG. 1 is a process flow chart showing the configuration of the present invention, FIG. 2 is a detailed explanatory diagram of the present invention, and FIG. 3 is a diagram showing an example of a source program. In the figures, Figure 1 is a process flowchart showing the configuration of the present invention.

Claims (1)

[Claims] In the processing of a compiler that generates a target program by translating a computer source program, if there is a predetermined designation, each file specified in the source program is set in an open state or a non-open state. A predetermined flag is provided to indicate whether the file is to be opened or closed, and after the opening or closing processing is performed, the flag is set to indicate an open state or a non-open state, respectively, in the target program generated from the source program that specifies opening and closing of the file. (1 to 3, 5, 6), and to the target program generated from the source program that specifies access to the file, check the flag before the access processing, and check if the flag is non-zero. A method for detecting file processing errors, comprising: adding a program (4, 7) that outputs a predetermined error message when an open state is displayed.