JPH0496151A - Random access system for sequential organization file - Google Patents

Abstract

PURPOSE:To enable random input/output processing from a user program to be performed by generating an intermediate file with attribute fitted in the file definition of the user program temporarily by comparing file attribute information on an auxiliary storage medium with the file definition of the user program. CONSTITUTION:A file attribute recognizing means 21 obtains the file attribute information of an actually allocated file 31 before the file opening processing of a data processor 20 is started up for the open instruction 11 of the user program 10, and it is compared with the file definition of the program 10 by a file attribute comparison means 22. When it is judged that data conversion between them can be performed, the intermediate file 32 with attribute according to the file definition of the program 10 is generated temporarily by an intermediate file change means 23A, and the data conversion from the actually allocated file 31 to the intermediate file 32 is performed, and the file opening processing 25 on the intermediate file 32 is executed. Thereby, the random input/output processing by the user program can be performed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a random access method for sequential files, and in particular, the present invention relates to a random access method for sequential files, and in particular, access from a user program whose file definition is indexed sequential organization or relative organization to an auxiliary storage medium such as a magnetic tape. This invention relates to a random access method for sequential files.

[Conventional technology]

Random input/output processing is possible for indexed sequential files that have an index section consisting only of specific key field parts in data records, and for relative organization files that position data records using relative sequence numbers within the file as keys. Due to its nature, it is constructed on a large-capacity storage medium such as a magnetic disk device.

Conventionally, when there is a shortage of file space on a magnetic disk device under system management, a service program or tool is used to convert an indexed sequential organization or relative organization file to a sequential organization file, and an auxiliary storage medium such as magnetic tape is used. The file area is secured by copying or saving the file. If you need the files saved on magnetic tape, perform the reverse file conversion using the service program or tool mentioned above.
Processing was being performed to reload the file area on the magnetic disk device.

[Problem to be solved by the invention]

Files with index sequential organization or relative organization that allow random input/output processing must be constructed on large-capacity violet storage media such as magnetic disks. However, unless the file is constantly processed by a program, it is wasteful to permanently store it on a magnetic disk device under system management, so it is better to store it on an auxiliary storage medium such as magnetic tape. become. In order to use the data in such a file with a program that performs random access, you must start the conversion service program, create a randomly accessible file on the magnetic disk device, and copy the data. There are certain drawbacks.

An object of the present invention is to provide a random access method for sequential files that enables random input/output processing from a user program to sequential files on a magnetic tape.

[Means to solve the problem]

The random access method for sequential files of the present invention includes a file attribute recognition means for reading file attribute information on an auxiliary storage medium allocated by the system, a file definition of a user program that attempts to access the file, and the file attribute information. and a file attribute comparison means for determining whether or not mutual data conversion is possible when the former is a sequential file and the latter is not, and when data conversion is possible, attributes that match the file definition of the user program. Temporarily create an intermediate file to convert and copy the data in the actual sequential file, perform file input/output processing by the user program in the intermediate file, and if necessary, convert the data in the intermediate file to the actual sequential file. intermediate file converting means for deleting the intermediate file after converting and copying it again into an organized file; and an error notification means for notifying an abnormality when a logical problem occurs during data conversion by the intermediate file converting means. It is configured.

〔Example〕

Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.

The feature of the present invention basically lies in file open processing. First, an overview will be explained with reference to FIG. In response to the open command 11 of the user program 10, before the file open processing 25 of the data processing device 20 is started, the file 3 actually allocated by the file attribute recognition means 21
1 is obtained and compared with the file definition of the user program 10 by the file attribute comparing means 22. As a result, if the actually allocated file 31 is a sequential file and the file definition of the user program is indexed sequential organization or relative organization that allows random input/output, it is determined whether data can be converted between them. If conversion is possible, the intermediate file 3 with attributes according to the file definition of the user program 10 is created by the intermediate file conversion means 23A.
2 is temporarily created, data is converted from the actually allocated file 31 to the intermediate file 32, and file open processing 25 is executed on the intermediate file 32. This eliminates the attribute mismatch between the file definition of the user program 10 and the actually allocated file 31, and enables the user program to perform random input/output processing on the file allocated to the magnetic tape medium.

Further, when the data processing is completed and a close command 13 from the user program 10 is received, the data processing device 20 performs file close processing 27 and then closes the processed file 3.
0 is the intermediate file 32, and when writing and updating are performed on that file, the intermediate file conversion means 2
3B performs data conversion from the intermediate file 32 to the actually allocated file 31 in the opposite manner to that at the time of opening.

The data processing device 20, which is basically O8, performs the above processing.
By performing this internal processing, the user program 10 appears to have executed random input/output processing on the actually allocated file 31. Note that if an input/output error occurs during data conversion processing by the intermediate file conversion means 23A, 23B, the error notification means 24A, 24B
transfers control to the user program and notifies the user program of the error.

FIG. 2 is a flowchart of processing when the user program 10 executes the open command 11. As mentioned above, in response to an open request for the data processing bag W20, the file attribute recognition means 21 is first activated and checks the device type, file format, file organization, record format, maximum logical record length, block length, etc. of the actually allocated file 31. The file attribute information of the file is read (step 41). next,
Step 4: Control is transferred to the file attribute comparison means 22 to determine whether the actually allocated file 31 is a sequential file.
2), whether or not the file definition specified in the user program 10 is sequential organization; step 43), whether the file definition is index sequential organization or relative organization and the file 3 actually allocated;
1 (step 44). As a result, if the actually allocated file 31 is a sequential file and the file definition is indexed sequential organization or relative organization and file conversion is possible,
Control is passed to the intermediate file conversion means 23A. If this condition is not met, the conventional file open processing 25 is executed directly at step 48.

The intermediate file conversion means 23A is the user program 10
An intermediate file 32 with attributes according to the file definition is created as a temporary file, and the file to be opened is changed from the actually allocated file 31 to the intermediate file 32 (step 45). If the access request to the file of the user program 10 is for an initial load of data (step 46), the file open process 25 is executed as is (step 48), but if the access request is for a data read or update process , the data in the actually allocated file 31 is transferred to the intermediate file 3.
After copying with attribute conversion (step 47), file open processing is performed. In addition, if an error occurs in the process of this data copying process, the error notification means 2
Control is transferred to 4A to notify the user program 10 of the abnormality (step 49).

After performing the above processing, the user program 1
File input/output processing 26 is executed using input/output command 12 of 0.

FIG. 3 is a flowchart of processing when the user program 10 executes the close command 13. Data processing device 20
After executing file close processing 27 (step 5
1) The processed file 30 is the intermediate file 32
step 52), further updating the data,
Determine whether deletions, additions, etc. have been made (Step 5)
3).

If an update, deletion, or addition is made to the intermediate file 32, the file 31 to which the result is actually assigned
In order to reflect the data, the intermediate file conversion means 23B is called, and data is copied from the intermediate file 32 to the actually allocated file 31, contrary to the open processing (step 54), and after the copy is completed, the intermediate file 32 is deleted. (Step 55). Note that if an error occurs during the conversion process, an error notification process (step 56) is performed.

FIG. 4 is a flowchart showing details of the processing of the intermediate file conversion means when a file is opened, and FIG. 5 is a flowchart showing details of the processing of the intermediate file conversion means when a file is closed.

When opening the file, the actually allocated file 31
(step 61) and records are copied to the intermediate file 32, which is an indexed sequential or relative organization file (step 65). However, if the record read from the sequential file is a deleted record (step 63), if the intermediate file 32 is an index sequential file, it is not output to the intermediate file 32 and step 6
1 (step 67), the next record is read from the actually allocated file 31, and if it is a relative organization file, it is written to the intermediate file 32 as a deletion record corresponding to the relative key (step 68). Note that when reading the actually allocated file 31 or the intermediate file 32
If an error occurs when writing (step 64.
In step 66), control is passed to the error notification means 24A (step 69), and when the actually allocated file 31 is read and an EOF state is reached (step 62), this process ends.

When closing a file, the intermediate file conversion means 23B is called only when a record is written, updated, deleted, etc. to the intermediate file 32 in the file input/output processing 26 of the user program 10, and is the opposite of when opening the file. Records are copied from the intermediate file 32 to the actually allocated file 31 (steps 71 and 75).
>. Basically, the process is the same as when opening a file, and if the intermediate file 32 is a relative file, a deleted record may be read (step 73), so the deleted record is written to the actually allocated file 31 (
In step 77), if an error is detected during input/output of each file (steps 74 and 76), control is passed to the error notification means 24B.In step 78), intermediate file 3
If you try to read a record with a key greater than the maximum record key of 2, an EOF state will occur (step 7).
2) End the process.

With the processing method explained above, even if a sequential file is allocated to a magnetic tape device, for example, to a user program that defines an index*mm file relative file and performs random access processing, the program will not be interrupted. It becomes possible to execute random access processing.

〔Effect of the invention〕

As explained in detail above, the present invention is capable of processing indexed sequential organization and relative organization files, which can originally only be constructed on large-capacity storage media such as magnetic disk devices, by allocating them to auxiliary storage media such as magnetic tape. This makes it possible for a user program that performs random input/output processing to process files that have been evacuated and stored on magnetic tape in magnetic tape format, thereby interrupting system processing and redirecting files to the file area of a magnetic disk device, etc. This has the effect of eliminating the need to reload data.

10... User program, 11...
Open instruction, 12...--I/O instruction, 13...
---Close command, 20, data processing device, 21, file 22, file attribute comparison means, 23A, 23B, intermediate file conversion Means, 24A, 24B... Error notification means, 25... File open processing, 26.
...File input/output processing, 27...File close processing, 30...--File, 31.
...Actually allocated file, 32-...
・Intermediate file.

Claims (1)

[Claims] A file attribute recognition means for reading file attribute information on an auxiliary storage medium allocated by the system compares the file definition of a user program attempting to access the file with the file attribute information, and determines whether the former is a sequential file and the latter is a sequential file. In other cases, a file attribute comparison means is used to determine whether mutual data conversion is possible, and when data conversion is possible, an intermediate file with attributes matching the file definition of the user program is temporarily created and the actual The data in the sequential file is converted and copied, file input/output processing by the user program is performed in the intermediate file, and if necessary, the data in the intermediate file is converted and copied again into an actual sequential file, and then the intermediate file is A random access method for a sequential file, characterized by comprising intermediate file converting means for deleting, and error notification means for notifying an abnormality when a logical failure occurs during data conversion by the intermediate file converting means. .