JPH01114936A - File system - Google Patents

Abstract

PURPOSE:To simply and rapidly update an individual index file by the main work of a host device in file updating processing such as the deletion or modification of an already registered document by forming an updating history list in a file device. CONSTITUTION:In case of updating a document file 7 and an index file 9 by the registration of a new document and the deletion or modification of an already registered document in the file device 100, a file controller 4 enters the updated record in the updating history list 19 in a magnetic disk device simultaneously with updating. A host device 200 periodically requests the sending of updated records following the final serial number when the records exist in the history list 19. After receiving the updated records, the device 200 inspects an optical disk number 23, a document number 24, a processing sort 25, and accessory information 26 entered in the history list 19, determines information necessary for updating its index file 18 at the time of updating the file and requests the sending of the information to the device 100. Consequently, the file 18 can be simply and rapidly updated by the main work of the device 200.

Description

[Detailed Description of the Invention] Industrial Application Field This invention relates to a file system in which a file device that organizes and stores document information, etc. on an optical disk, etc., and a host device (computer) that uses the information of this file device are connected via a network. The present invention relates to a system, and particularly relates to a technique for revising an index file provided on a host device side to match a file device for searching file information.

2. Description of the Related Art As a conventional example of this type of file system, FIG. 4 shows a schematic configuration of a document file system.

A file device 100 and a host device 200 are connected to a line 3 via communication adapters 1 and 2, respectively. Typically, multiple host devices are connected to this network.

The file device 100 includes a file controller 4 that performs overall control of this device, a display 5 and a keyboard 6 that are workstations, and a document file (master file) 7 in which image information of a large amount of documents is filed in an optical disk device. , a directory 8 in which the management information of each document registered in the document file 7 is organized and recorded in a magnetic disk device, and search information such as keywords for each document registered in the document file 7 is stored in the magnetic disk device. A directory 8 that has been organized and recorded, an index file 9 that has search information such as keywords for each document registered in the document file 7 that has been organized and recorded on a magnetic disk device, and a search using this index file 9. Serve the search processor 10 that executes the process. Each document registered in the document file 7 is given a serial document number. The register 8 records information indicating where in the document file 7 the document number is stored (address indicating the sector of the track of the optical disc number).

The search processor IO examines the index file 9 and outputs the document number of the document corresponding to the specified keyword and search condition as a search result. The file device 100 also includes terminals such as a printer 11 that prints out document information, and an image scanner 12 that reads document information of a document.

The host device 200 has a CPU 13, ROM 15, and RAM 1.
5. A general computer equipped with a display 16 and a keyboard 17, which is connected to a line 3 through a communication adapter 2 in order to use the information in the file device 100, and which also has an independent index file 18 on a magnetic disk device.
is being created. This index file 18 is essentially the same as the index file 9 of the file device lOO, and searches for keywords, etc. of each document registered in the document file 7 so that the document number of the corresponding document can be retrieved from keywords and search conditions. This is an edited version of the information. However, the contents of the index file 18 do not need to be exactly the same as the index file 9. Index file 9 of file device 100
This is the original, so to speak, and the edited content is versatile.

The index file 18 of the host device 200 may be edited to suit the user's purpose of using the information in the document file 7.

The host device 200 searches using its own index file 18, and once the document number of the desired document is determined, it transmits the document number to the file device 1oo and retrieves the corresponding document information (entity information) from the document file 7. have it transferred. By independently performing search processing on the host device 200, the load on the 7-aisle device 100 is reduced, and waiting time is reduced in a network with a large number of host devices (for example, Japanese Patent Laid-Open No. 116159/1983).

Problems to be Solved by the Invention In the file system described above, the contents of the document file 7 are usually updated over time. That is, new documents are registered one after another in the document file 7, and processes such as deleting or revising already registered documents are also performed. Of course, as the document file 7 is updated, the index file 9 is also updated.

There is no problem since the file device 100 proactively updates the document file 7 and the index file 9, but if the index file 18 on the host device 200 connected to the network is not constantly updated with new contents, the file device 100 pieces of information cannot be fully utilized.

Therefore, each time the file device 100 performs a file update process, the file device 100 updates each host device 200.
The first method that can be considered is to notify the user of the updated contents and have the index file 18 updated. However, according to this method, the file device 100 must have this function, resulting in a very expensive device. Furthermore, if the frequency of file updates is high and the number of host devices is large, updating the files of the entire system takes a lot of time and may interfere with other operations. Furthermore, the host device 200 must have the ability to perform predetermined processing in response to update notifications from the file device 100, which poses a major constraint on host devices joining the system.

For these reasons, this method has not been adopted in reality.

Actually, from the host device 200 side to the file device 10
A method is used in which the contents of 0 are checked at any time, and when an updated part is found, the index file 18 is updated accordingly. According to this method, the file device 10
0 can be handled with just a simple passive function of sending information requested by the host device 200, making it a very inexpensive device.

Conventionally, when updating the index file 18 of the host device 200 using the above method, there have been the following problems.

Among file update processes, registration of new documents can be easily handled. The host device 200 stores its own index file 1
8 contains information about document numbers up to, and check the register 8 of the file device 100 from time to time, and if a new document number increases, Document file 7 for necessary information about the document
and from the index file 9, and updates its own index file 18 according to that information.

However, it is extremely difficult to handle update operations such as deletion and revision of already registered documents.

In the above case, it would have been sufficient to check whether the document number in Registry 8 has increased, but deletion or revision of already registered documents cannot be known unless all the contents of Registry 8 are investigated in detail. This takes a considerable amount of time. This may not be a big problem when the file size is small, but as the file size increases, searching the register 8 becomes a heavy burden, and the number of searches decreases, resulting in the contents of the index file 18 being delayed. It ends up.

This invention was devised in view of the above-mentioned conventional problems, and its purpose is to easily and quickly create a unique index by the host device's own initiative, even when it comes to file update processes such as deleting or revising already registered documents. The purpose is to provide a file system that allows files to be updated.

Means for Solving the Problems Therefore, in the present invention, each time the file device updates the master file and the index file, the update contents are sequentially recorded in a history book, while the host device records the updated contents in the history book. is checked at any time, and if new updated contents are recorded there, necessary information is extracted from the master file and the index file of the file device to update its own index file accordingly.

Function: Although the above-mentioned file device requires the work of keeping the above-mentioned history record when updating a file, this is a simple task in terms of both software and hardware. By looking at the history list, the host device can immediately know whether a new file has been updated. Furthermore, in order to update the index file of the host device in accordance with the file update, it is also possible to know from the history record what information needs to be retrieved from the file device.

Embodiment FIG. 1 shows a schematic configuration of a document file system according to an embodiment of the present invention. What is different in the drawing from the conventional system shown in FIG.
The only point is that 9 was provided. While the contents of the history register 19 and its related operations and functions, which will be described in detail below, are the features of the present invention, the basic operation as a document file system is the same as the conventional one, so a description of the same parts will be omitted. do.

When updating the υ document file 7 and index file 9 in the file device 100 Vc by registering a new document and deleting or revising an already registered document, the file controller 4 simultaneously updates the history book 19 in the magnetic disk device.
An update record is added in the format shown in Figure 2. The update record includes the serial number 21 assigned to each code, the date and time of the update process, the document number of the document that was updated, the optical disk number U on which it is stored, and the type of update process. The process type 5 shown in FIG. Such update records are entered in the history book 19 in serial number order.

The flowchart in FIG. 3 shows the procedure for the host device 200 to update its own index file 18.

The CPU 13 of the host device 200 stores up to which serial number update records in the history record 19 have already been processed. Let n be the last processed serial number. The host device 200 is the file device 1
Periodically request 00 to send any update records after serial number n in the history record 19 (
Step 31). If the file device 100 has a corresponding update record, it will send it, and if it does not, it will reply to that effect.

When the update record is sent from the file device 100, the host device 200 checks the optical disk number 24, document number 24,
Information necessary to consider processing type 5/attached information addition and update its own index file 18 in response to the file update (information on document file 7 and index file 9)
File device 1
00 (step 33). File device 10
0 reads the requested information from the document file 7 and the index file 9 and sends it to the host device 200, which receives it and updates the index file 18 (step 34).

Effects of the Invention As explained in detail above, in the file system according to the present invention, the host device can immediately know whether or not it is necessary to update its own index file by looking at the update history record of the file device. Furthermore, it is easy to know from the history list what information about the file device is needed to update its own index file. Therefore, by issuing a request to the file device and having it send the necessary information to the host device,
Then you can update the index file. The processing during this time is simply a matter of the file device sending out the information requested by the host device, which is one of the basic operations of the file device. For file devices, the work of keeping a history record increases, but this is not very sophisticated and can be easily performed.

Finally, although the increase in the burden on the system as a whole is extremely small, updating of the index file of the host device can be processed efficiently and rationally.

[Brief explanation of the drawing]

Figure 1 is a schematic configuration diagram of a document file system according to an embodiment of the present invention, Figure 2 is a conceptual diagram of update records entered in a history book in the embodiment, and Figure 3 is an index diagram of a host device in the embodiment. FIG. 4 is a flowchart of file updating processing and is a schematic diagram of a conventional document file system. 7...Document file (master file A), 9...
Index file a3), 18... Index file 0.19
...Name of agent in resume Patent attorney Toshio Nakao and 1 other person 1I
I21 100c7 Eye 2 Shij hanging load) 1g2 Mouth 3 Figure 4/θO() Eye λ Waste trap 1)

Claims (1)

[Claims] A file device having a master file A and an index file B for searching information of the master file A, and a host device having a unique index file C for searching information of the master file A are connected via a network, and the above In a system in which a host device requests the file device to retrieve the entity of information searched using index file C from master file A, the file device retrieves the information from master file A and index file B each time master file A and index file B are updated. The host device has means for sequentially recording update contents in a history record, and the host device checks the history record from time to time, and if new update contents are recorded therein, the host device records the master file A and index file B accordingly. A file system characterized by having means for extracting necessary information from a file and updating its own index file C.