JPH0981423A - Dividing type order organized file device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate the necessity of consciously deleting garbage and to relieve the burden on a user by periodically performing a garbage deletion processing for a file of garbage existence. SOLUTION: In accordance with the instruction by a user, a PAM file edition processing part 1-1 performs edition processings such as the addition, the update and the deletion of the member in the file of a PAM file part 1-4 storing the file that the user instructs. In the case of the update and deletion of the member, queuings to a garbage file queue part 1-3 are performed for the file names for which these processings are performed. A garbage deletion part 1-2 refers to the garbage file queue part 1-3 for every fixed time interval and performs the deletion processing of garbage for the PAM file part 1-4 corresponding to the file name for which the queuing to the part 1-3 is performed. Thus, the PAM file part 1-4 periodically performs the deletion processing of garbage for only the file for which the queuing is performed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a split type sequential knitting file device, and more particularly to a split type sequential knitting file device capable of efficiently deleting garbage at regular intervals.

[0002]

2. Description of the Related Art A divided sequential file is suitable for accommodating a large number of relatively small files as members, such as a program library, and is often used.

That is, a divided type sequential file is composed of a plurality of members each in the form of a sequential file, and the members are accommodated in the member substance section in the order of creation, and the members are named with the member name as a key. The storage location is recorded in the directory section.

The addition, update, and deletion of members for such a divided sequential file will be described.

When a new member is added to a divided sequential file, the directory part is referred to and accommodated in the member entity part in the order of creation from the end of the last existing member. At the same time, the name of the added member and its storage location are registered in the directory section.

When updating the contents of existing members,
First, the members whose contents have been updated are stored in the member entity part from the end of the existing members by referring to the directory part, and the corresponding member name in the directory part is stored based on the storage position. Change the old storage location recorded in to the new storage location. Do not change the member name. For this reason, the old member entity before the update, which is located in the old storage position of the member entity part, remains in the file as invalid data that is never read.

When deleting an existing member, simply delete the member name and its storage location in the directory section.
For the member entity to be deleted in the member entity section,
Like the old member entity that has been updated, the invalid data that will never be read remains in the file as invalid data.

As described above, when updating and deleting the members of the divided sequential file, the previously registered member entity becomes invalid, the member area remains as garbage, and the memory is effectively used. There was a problem that I could not do it.

Therefore, conventionally, each time a member is added or updated, the garbage in the divided sequential file is usually deleted as necessary so that the user does not have any inconvenience. there were.

Further, in order to solve the burden on the user, when updating a member, the member to be updated is first erased, and then the members remaining in the file are condensed (refilling from the head of the member entity portion). Japanese Laid-Open Patent Publication No. 58-163054 discloses a method in which garbage is deleted by adding the members whose contents have been updated.

[0011]

In the conventional garbage deleting method of the divided sequential file system, the user must always keep track of the garbage state of the divided sequential file system. However, there is a drawback that it imposes a heavy burden on the user.

Further, in the division type sequential file system garbage deletion method described in Japanese Patent Laid-Open No. 58-163054, which is carried out each time updating is performed, the inconvenience due to the garbage generated by the member deletion at the time of adding members cannot be avoided.
Even when a member is added, the condensation work of the file member has to be performed, and there is a problem that the performance of the processing at the time of member update and addition is deteriorated.

The object of the present invention is to store the files that have been subjected to the member update and deletion processing, and to perform the garbage deletion processing for these files at regular intervals so that the memory can be efficiently used in a divided order. It is to provide a knitting file device.

[0014]

According to a first aspect of the present invention, there is provided a split type sequential knitting file device, a file storing means for storing a plurality of split type sequential knitting files, and a split type sequential knitting file stored in the file storing means. A file edit processing means for updating and deleting and adding a member, a garbage file storing means for storing an identifier of a file which is updated or deleted by the file editing processing means and has a garbage, and the garbage file. And a garbage deleting means for periodically deleting the garbage of the file corresponding to the file identifier stored in the storage means.

The split type forward knitting file device of the second invention is the split type forward knitting file device of the first invention,
The garbage file storage means is characterized in that the identifier of a file having a garbage which has been updated or deleted is stored as a queue.

[0016]

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

FIG. 1 is a block diagram showing an embodiment of a split type sequential knitting file device of the present invention.

As shown in FIG. 1, the split type forward knitting file apparatus of the present embodiment has a split type forward knitting file (Par).
Titted Access Method Fi
le, hereinafter abbreviated as PAM file), a PAM file edit processing section 1-1 for adding, updating, and deleting.
A member that is invalidated at regular intervals, that is, a garbage removing unit 1-2 that performs garbage deletion processing, and a file that has been updated and deleted, that is, a garbage file that stores a file with garbage. Queue section 1-3 and a plurality of PAM file sections 1-4
It is comprised including.

The PAM file section 1-4 stores a member entity section 1-4-2 that stores the entity of the member stored in this file, the name of the stored member and the name of the member. It is configured to include a directory section 1-4-1 which stores a start address S and a final address E of the member entity section 1-4-2.

PAM of PAM file edit processing section 1-1
The addition, update and deletion of members to the file unit 1-4 will be described.

For adding a new member to the PAM file section 1-4, refer to the directory section 1-4-1.
The members are accommodated in the member actual part 1-4-2 in the order of creation from the last member of the existing members. At the same time, the names of the added members and S and E indicating their storage positions are registered in the directory unit 1-4-1.

When updating the contents of existing members,
First, the members whose contents have been updated are searched for in the directory section 1-4.
-1 is stored in the member entity part 1-2-2 from the end of the last member among the existing members with reference to -1, and the corresponding member in the directory part 1-4-1 is based on the storage position. Change the old storage location recorded in the name to the new storage location. Do not change the member name. Therefore, the member entity part 1
-The old member entity before the update in the old storage position of -4-2 is
The invalid data that will never be read remains in the file. The file name for which the member has been updated is the garbage file cue unit 1-3.
Sent to and queued.

When deleting an existing member, only the member name of the directory section 1-4-1 and its storage position are deleted, and the member entity to be deleted in the member entity section 1-4-2 is Like the old member entity that has been updated, the invalid data that will never be read remains in the file as invalid data. In addition,
The file name from which the member has been deleted is sent to the garbage file queue unit 1-3 for queuing.

The garbage file cue unit 1-3 is a PA
The member supplied from the M-file edit processing unit 1-1 queues the updated or deleted file name, and removes the deleted file name of garbage from the queue.

The garbage removing unit 1-2 is activated at regular intervals, and the garbage file cue unit 1-3 is referred to, and the directory unit 1 of the PAM file unit 1-4 corresponding to the file name stored therein is referred to. -4-1,
Refill members and remove garbage.

FIG. 2 is a flowchart showing the operation of the editing process of the PAM file editing processing section 1-1, and FIG. 3 is the garbage removing section 1
2 is a flowchart showing an operation of a garbage deletion process of -2.

The update processing operation and the garbage deletion processing operation of this embodiment will be described with reference to FIGS.

First, with reference to FIG. 1, a general operation of the present embodiment will be described.

According to the user's instruction, the PAM file edit processing section 1-1 performs edit processing such as addition, update and deletion of members to the file in the PAM file section 1-4 storing the file instructed by the user. In the case of updating and deleting members, the file names subjected to these processes are queued in the garbage file queue unit 1-3.

The garbage removing unit 1-at regular time intervals
2 refers to the garbage file cue section 1-3, and the PAM corresponding to the file name queued to this
Garbage deletion processing is performed on the file units 1-4. In this way, the PAM file unit 1-4 performs the garbage deletion processing only for the queued files at regular intervals, without imposing a burden on the user and in a shorter time than before. Garbage can be deleted, and PAM file section 1-
Therefore, the memory for storing 4 can be efficiently used.

Initially, as shown in FIG. 2, the PAM file unit 1-4 does not have a garbage area for n member entities having member names 1 to n in the member entity unit 1-4-2. , And each storage position can be known by the directory unit 1-4-1 in which the start address S and the end address E of the storage position are stored corresponding to each member name. it can.

Next, the member update processing operation during the edit processing operation of the PAM file edit processing section 1-1 will be described in detail with reference to FIG.

The PAM file edit processing section 1-1 stores the file specified by the user in the PAM file section 1-4.
Is activated by the update request (step 1). As an example of the update request, a request to update the member 3 of the file a to the new member 3 whose content has been updated will be described.

Next, the PAM file edit processing section 1-1
Refers to the final address of each member of the directory portion 1-4-1, and sets a new member 3 having the updated content from the member having the largest address value (in this case, the member n). It is stored in the position of the double frame 4-2 (step 2).

Next, the storage position corresponding to the member 3 in the directory section 1-4-1 is rewritten to the start address S and the final address E which are the storage positions of the new member 3 stored in step 2 (corresponding to the old address). The position indicated by the dotted arrow is shown, and the position corresponding to the new address is shown by the thick arrow) (step 3).

By doing so, the entity of the member 3 previously stored is invalid data which is never read out, that is, it remains as garbage in the file a (this area is stored in the area a). It is shown by hatching).

Next, the PAM file edit processing section 1-1
Stores the file name a for which the update processing has been performed in the garbage file queue unit 1-3 and stores that it has become the target file for the garbage deletion processing, and ends the update processing operation of the member (step 4). . At this time, if the same file name has already been queued, that file name is not queued.

Next, the garbage deletion processing operation of the garbage deletion unit 1-2 will be described in detail with reference to FIG.

As an example of the garbage deletion processing operation, the case of deleting the garbage generated by updating the member 3 of the file a will be described. The PAM file part 1-4 before the deletion process is shown on the lower left side of FIG. 3 (the garbage area is shown by hatching), and the PAM file part 1 after the deletion process is shown.
-4 is shown on the lower right side of FIG.

The garbage removing section 1-2 is periodically activated by a built-in timer to perform the garbage removing processing operation described below, and the timer is started in response to the end of the garbage removing processing operation.

The timer starts the garbage removing processing operation of the garbage removing unit 1-2 after a lapse of a certain time after starting (step 21).

The garbage deleting unit 1-2 reads the file names (files whose members have been updated or deleted) queued in the garbage file cue unit 1-3 one after another (step 22). The garbage deletion processing operation described below is performed.

The garbage removing section 1-2 checks the directory section 1-4-1 of the PAM file section 1-4 and obtains it from the start address of the garbage area, that is, the final address E of the member 2 in FIG. Subsequent member entities (double frame area, members 4 to n, and entity of member 3) are copied from the start address of the garbage area (step 23).

Next, the start address S and the end address E of each member of the directory section 1-4-1 are set in step 23.
The start address S and the final address E of the entity of each member obtained as a result of the copying performed in step S are rewritten (step 24).

Now that the garbage has been deleted, the garbage deletion processing operation is terminated and the timer is started to prepare for the next garbage deletion processing operation (step 25).

If there are several garbage areas and the addresses are not continuous, steps 23 and 24 are performed.
The garbage removal process of is repeatedly performed for each of the several garbage areas.

As described above, the divided sequential knitting file device of the present embodiment does not need to consciously delete the garbage by the user by regularly performing the garbage deleting process of the file having the garbage. The burden on the user can be significantly reduced.

Further, by storing the file names of the files in which the members have been updated and deleted, and referring to this and performing the garbage deleting process only for these files, the garbage deleting process can be efficiently and shortly performed. You can do it in time.

Further, since the garbage deleting process is carried out periodically, there is no influence on the performance of the normal file updating process and the like.

[0050]

As described above, the divided sequential knitting file device of the present invention eliminates the need for the user to consciously delete the garbage by regularly performing the garbage deleting process of the file having the garbage. This has the effect of significantly reducing the burden on the user.

Further, by storing the file names of the files in which the members have been updated and deleted, and referring to this and performing the garbage deleting process only for these files, the garbage deleting process can be efficiently performed. It has the effect that it can be done in time.

Further, since the garbage deleting process is carried out periodically, there is an effect that it does not affect the performance of the file updating process which is a normal operation.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a split type sequential knitting file device of the present invention.

FIG. 2 is a flowchart showing an example of an operation of a PAM file edit processing section 1-1 in the split type sequential file system of the present embodiment.

FIG. 3 is a flowchart showing an example of the operation of the garbage removing unit 1-2 in the split type sequential knitting file device according to the present embodiment.

[Explanation of symbols]

 1-1 PAM file editing processing section 1-2 Garbage deleting section 1-3 Garbage file cue section 1-4 PAM file section 1-4-1 Directory section 1-4-2 Member entity section

Claims (2)

[Claims] 1. A file storage means for storing a plurality of division type sequential files, and a file editing processing means for updating, deleting and adding members constituting the division type sequential files stored in the file storage means. The file edit processing means updates or deletes members to store a file identifier of a file having a garbage, and a garbage file storing means for periodically storing the file identifier corresponding to the file identifier stored in the garbage file storing means. A divided type forward knitting file device including a garbage removing means for selectively deleting. 2. The divided sequential file apparatus according to claim 1, wherein the garbage file storage means stores, as a queue, identifiers of files which have been updated or deleted and which have garbage.