JP7010538B2 - File management device, file management method, and program - Google Patents

Description

The present invention relates to a file management device for managing data in a file system, a file management method, and a program for realizing these.

Traditionally, in the field of computers, file systems have been introduced to manage data stored in storage devices. Normally, a file system is provided as one of its functions by an operating system (OS: Operating System).

By the way, if all the data output by the computer is stored in the storage device as it is, the data size in the file system may increase and the storage capacity of the storage device may be insufficient. Therefore, for example, Patent Document 1 proposes a file system capable of performing data compression.

Specifically, in the file system disclosed in Patent Document 1, first, an exclusive OR of the input data and the model data is calculated, and the calculated exclusive OR is used as a difference, and further, from the difference. Calculate the dissimilarity. In addition, a plurality of model data are prepared, and the dissimilarity is calculated for each model data. Then, in the file system disclosed in Patent Document 1, when the calculated dissimilarity satisfies a certain condition, the difference, the name of the input data, and the name of the corresponding model data are used. , Dissimilarity is associated and stored.

In this way, the input data is stored not in the data itself but in a format using the difference from the model data. Therefore, according to the file system disclosed in Patent Document 1, the stored data is compressed. Therefore, the increase in data size is suppressed. Further, the file system disclosed in Patent Document 1 stores the input data as one of the new model data in an uncompressed manner when there is no model data in which the dissimilarity satisfies a certain condition.

Japanese Unexamined Patent Publication No. 2006-65524

However, in the file system disclosed in Patent Document 1, it is necessary to execute a logical operation for each model data to calculate the exclusive OR of the model data and the input data. Therefore, in this file system, the time required from the start to the end of the compression process becomes long, which in turn takes too much time to register the data. Further, since the number of model data increases as the computer operates, it is difficult to shorten the time required for the compression process in the file system disclosed in Patent Document 1.

An example of an object of the present invention is to provide a file management device, a file management method, and a program that can solve the above-mentioned problems and reduce the time required for data registration while compressing data in a file system. To provide.

In order to achieve the above object, the file management device in one aspect of the present invention is
A summary information creation unit that analyzes the input file and creates summary information indicating the contents of the input file.
A similarity calculation unit that calculates the similarity with the input file for each registration file by comparing the summary information of the input file with the summary information of the already registered registration file.
A difference calculation unit that calculates the difference between the registration file with the highest degree of similarity and the input file, and
A registration unit that registers the calculated difference and similar file information that identifies the registration file having the highest degree of similarity as the input file.
It is characterized by having.

Further, in order to achieve the above object, the file management method in one aspect of the present invention is:
(A) A step of analyzing an input file and creating summary information indicating the contents of the input file.
(B) A step of comparing the summary information of the input file with the summary information of the already registered registration file and calculating the similarity with the input file for each registration file.
(C) A step for calculating the difference between the registration file having the highest degree of similarity and the input file, and
(D) A step of registering the calculated difference and the similar file information for specifying the registration file having the highest degree of similarity as the input file.
It is characterized by having.

Further, in order to achieve the above object, the program in one aspect of the present invention is:
On the computer
(A) A step of analyzing an input file and creating summary information indicating the contents of the input file.
(B) A step of comparing the summary information of the input file with the summary information of the already registered registration file and calculating the similarity with the input file for each registration file.
(C) A step for calculating the difference between the registration file having the highest degree of similarity and the input file, and
(D) A step of registering the calculated difference and the similar file information for specifying the registration file having the highest degree of similarity as the input file.
It is characterized by executing.

As described above, according to the present invention, in the file system, it is possible to reduce the time required for data registration while reducing the data compression.

FIG. 1 is a block diagram showing a schematic configuration of a file management device according to an embodiment of the present invention. FIG. 2 is a block diagram showing a specific configuration of the file management device according to the embodiment of the present invention. FIG. 3 is a diagram showing an example of a file registered in a database by a tree structure in the present embodiment. FIG. 4 shows an example of summary information when the input file and the registration file are document files. FIG. 5 shows an example of summary information when the input file and the registration file are source files. FIG. 6 is a flow chart showing an operation of the file management device at the time of input file registration processing according to the embodiment of the present invention. FIG. 7 is a flow chart showing an operation of the file management device at the time of file reading processing according to the embodiment of the present invention. FIG. 8 is a flow chart showing the operation of the file management device at the time of file deletion processing according to the embodiment of the present invention. FIG. 9 is a block diagram showing an example of a computer that realizes the file management device according to the embodiment of the present invention.

(Embodiment)
Hereinafter, the file management device, the file management method, and the program according to the embodiment of the present invention will be described with reference to FIGS. 1 to 9.

[Device configuration]
First, the schematic configuration of the file management apparatus according to the present embodiment will be described with reference to FIG. FIG. 1 is a block diagram showing a schematic configuration of a file management device according to an embodiment of the present invention.

The file management device 10 in the present embodiment shown in FIG. 1 is a device for managing files in a file system. As shown in FIG. 1, the file management device 10 includes a summary information creation unit 11, a similarity calculation unit 12, a difference calculation unit 13, and a registration unit 14.

When a file is input, the summary information creation unit 11 analyzes this file (hereinafter referred to as "input file") and creates summary information indicating the contents of the analyzed input file. First, the similarity calculation unit 12 compares the summary information of the input file with the summary information of the registered file already registered, and calculates the similarity with the input file for each registered file.

The difference calculation unit 13 calculates the difference between the registration file and the input file having the highest degree of similarity.
First, the registration unit 14 creates similar file information that identifies the registration file having the highest degree of similarity. Next, the registration unit 14 registers the difference calculated by the difference calculation unit 13 and the created similar file information as an input file.

As described above, in the present embodiment, since the similarity between the summary information of the input file and the summary information of the registration file is used instead of the similarity between the files, the registration file similar to the input file can be specified at high speed. Then, the registration of the input file is completed by the registration of the difference and the similar file information. Therefore, according to the present embodiment, in the file system, it is possible to reduce the time required for data registration while reducing the data compression.

Subsequently, the configuration of the file management device in the present embodiment will be described more specifically with reference to FIGS. 2 to 5. FIG. 2 is a block diagram showing a specific configuration of the file management device according to the embodiment of the present invention.

As shown in FIG. 2, in the present embodiment, the file management device 10 is built on the file server 20 that provides the file system. Specifically, the file management device 10 is constructed by executing the program in the present embodiment on the operating system of the file server 20.

Further, as shown in FIG. 2, the file server 20 includes a database 21 in which various files are registered. Further, the file server 20 is connected to the terminal device 30 via a network 40 such as a LAN (Local Area Network). By accessing the file server 20 via the terminal device 30, the user can register a new file in the database 21, read and update the registered file, and the like. The file server 20 may be realized by a dedicated server computer, or may be realized by any terminal device 30.

Further, as shown in FIG. 2, the file management device 10 includes a file operation unit 15 in addition to the summary information creation unit 11, the similarity calculation unit 12, the difference calculation unit 13, and the registration unit 14 described above. There is. The file operation unit 15 executes a file operation such as reading a registered file and deleting a registered file.

In the present embodiment, the summary information creation unit 11 creates summary information by using an optimum method according to the type of the input file. For example, assume that the input file is a document file output from an application program for creating a document. In this case, the summary information creation unit 11 can analyze the structure of the sentence in the input file, extract a part on the head side of the sentence, and create summary information using the extracted part.

Further, it is assumed that the input file is a source file. In this case, the summary information creation unit 11 can extract the function definition from the input file and create summary information using the extracted function definition. Further, it is assumed that the input file is an image file. In this case, the summary information creation unit 11 can create a thumbnail image in which the size of the image is reduced from the original size as the summary information.

In the present embodiment, the similarity calculation unit 12 first obtains the difference between the summary information of the input file and the summary information of the registration file. Next, the similarity calculation unit 12 uses the difference between the summary information, and sets a value indicating how similar the summary information of the input file and the summary information of the registration file are within the range of 0 to 1 as the similarity. Calculate with.

Specifically, when the input file is composed of text such as a document file or a source file, the similarity calculation unit 12 can calculate the similarity as follows. That is, the similarity calculation unit 12 sets the size of the summary information of the input file to _SA , the size of the summary information of the registration file to _SB , and further sets the size of the difference between the two to size ( _diff ( _SA , SB). )), For example, the similarity can be calculated by the following equation 1. Note that max ( _SA , _SB ) represents the larger of the sizes _SA and _SB . Further, the similarity calculation unit 12 can also calculate the similarity by another formula according to the type of the file.

(Number 1)
Similarity = 1 _- size (diff ( _SA , SB)) / max ( _SA , _SB )

In the present embodiment, when the input file is composed of text such as a document file or a source file, the difference calculation unit 13 determines the content between the input file and the registration file as a difference. Calculate the difference. Further, the difference can be calculated by using a general character string difference extraction algorithm.

When the input file is image data, the difference calculation unit 13 compares (subtracts) the pixels of the input file and the pixels of the registered file in octet units of each RGB, and obtains the difference between the pixels. In addition, the total of the obtained differences is taken as the difference between the registered file and the input file. When the image file contains the α component, the difference calculation unit 13 calculates the difference for each pixel using all the elements of RGBA instead of RGB.

First, as described above, the registration unit 14 creates similar file information for the registration file having the highest degree of similarity to the input file. Then, in the present embodiment, the registration unit 14 registers the difference and the similar file information in the database 21 as an input file. Further, the registration unit 14 manages the input file and the registration file by a tree structure.

As shown in FIG. 3, when the input file is registered by the difference and the similar file information, the registration unit 14 sets the registration file specified by the similar file information as the parent node of the input file.

FIG. 3 is a diagram showing an example of a file registered in a database by a tree structure in the present embodiment. In the example of FIG. 3, since the files B and C have the highest degree of similarity to the file A, they are registered by the difference and the similar file information that identifies the file A.

Here, specific examples of processing by the summary information creation unit 11 and the similarity calculation unit 12 will be described with reference to FIGS. 4 and 5. FIG. 4 shows an example of summary information when the input file and the registration file are document files. FIG. 5 shows an example of summary information when the input file and the registration file are source files. In FIGS. 4 and 5, the file A is a file that has already been registered.

In the example of FIG. 4, the input file and the registration file are document files, for example, files having an extension of "txt", "docx", "xml", "md", or the like. The input file is file B, and the registration file is file A.

When the content of the input file is compressed, such as a file having docx as an extension, the summary information creation unit 11 first decompresses the content of the input file. Next, the summary information creation unit 11 extracts the sentence title and the chapter structure. At this time, if the input file is a file with xml in the extension, the tag structure (tag name only) is extracted.

Further, the summary information creation unit 11 extracts the first few lines of the input file as text. This is because the outline etc. is often written at the beginning of the document file. In addition, since the chapter structure of the text is almost the same in the document file created according to the template, it may be judged that they are similar even though the contents are different.

Subsequently, as shown in FIG. 4, the summary information creation unit 11 combines the extracted sentence title and chapter structure and the first few lines into one, and uses this as summary information. If the size of the obtained summary information is too large, the summary information creation unit 11 sets it to a constant size.

The similarity calculation unit 12 first extracts the difference between the summary information of the file A and the summary information of the file B. In the example of FIG. 4, the difference is 0.1 KB. The difference is obtained by using a general character string difference extraction algorithm. Then, the similarity calculation unit 12 applies the obtained difference to the above number 1 to set the similarity between the file A and the file B to 0.9 (= 1- (0.1 KB / 1.0 KB)). Is calculated.

In the example of FIG. 5, the input file and the registration file are source files, for example, files such as c, cpp (c ++), and java.

The summary information creation unit 11 extracts a class definition, a global function definition, or the like from an input file, and collects the extracted definitions into summary information. Even in the example of FIG. 5, if the size of the obtained summary information is too large, the summary information creation unit 11 sets it to a constant size.

The similarity calculation unit 12 also obtains the difference between the summary information of the file A and the summary information of the file B in the example of FIG. The difference is 0.1 KB. The difference is also obtained in the example of FIG. 5 by using a general character string difference extraction algorithm.

Then, the similarity calculation unit 12 applies the obtained difference to the above number 1 even in the example of FIG. 5, and sets the similarity between the file A and the file B to 0.875 (= 1- (0.1 KB). /0.8KB))).

Further, although not shown in FIGS. 4 and 5, when the input file and the registration file are image files, the summary information creation unit 11 compresses the image file as described above to reduce the resolution. Create a thumbnail image of, and use the created thumbnail image as summary information.

In this case, the similarity calculation unit 12 compares each pixel and the pixel at the same position of the thumbnail image of the registered file for each pixel of the thumbnail image of the input file in octet units of each RGB, and makes a difference. To calculate. When the image file contains the α component, the similarity calculation unit 12 calculates the difference using all the elements of RGBA instead of RGB.

Then, the similarity calculation unit 12 obtains a binary logarithm (0 to 7) from the absolute value of the difference (−128 to 127) of the octets for each pixel, and uses this as the pixel octet similarity value. Further, the similarity calculation unit 12 divides the total value of the pixel octet similarity values of the pixel components (RGB or RGBA) by the number of elements (3 for RGB, 4 for RGBA) to obtain the pixel similarity value. do. After that, the similarity calculation unit 12 divides the total value of the pixel similarity values of all the pixels by the number of pixels × 7, and subtracts the obtained value from 1 to obtain the similarity. The similarity calculation method is not particularly limited, and a similarity calculation algorithm other than the above-mentioned example may be used.

When the image size of the input file and the image file of the registered file are different from each other, the similarity calculation unit 12 can calculate the similarity as 0 (zero) because there is no point in taking a difference.

Specifically, in the thumbnail image, it is assumed that the number of pixels in the horizontal direction is 64, the number of pixels in the vertical direction is 64, and the component of the pixel is RGB. Further, it is assumed that 400 pixels (= 20 × 20) of the two thumbnail images are different, and the absolute value of the difference between the thumbnail images (difference for each RGB octet of each different pixel) is 40.

In this case, the pixel octet similarity value of the difference of 400 pixels is calculated as the following equation 2, and as a result, the pixel similarity value is calculated as the following equation 3. Since the other pixels (64 × 64-400 = 3696 pixels) are equal, the pixel similarity value is 0. From these results, the similarity is calculated as the following equation 4. In the following number 2, round () indicates rounding, and lg indicates a binary logarithm (logarithm with base 2).

(Number 2)
Pixel octet similarity value = round (lg 40) ≒ round (5.32) = 5

(Number 3)
Pixel similarity value = (5 + 5 + 5) / 3 = 5

(Number 4)
Similarity = 1- (5 x 400 + 0 x 3696) / (64 x 64 x 7) ≒ 0.93

[Device operation]
Next, the operation of the file management device 10 in the present embodiment will be described with reference to FIGS. 6 to 8. Further, in the first embodiment, the file management method is implemented by operating the file management device 10. Therefore, the description of the file management method in the present embodiment is replaced with the following description of the operation of the file management device 10.

First, the input file registration process by the file management device 10 will be described with reference to FIG. FIG. 6 is a flow chart showing an operation of the file management device at the time of input file registration processing according to the embodiment of the present invention.

As shown in FIG. 6, first, the summary information creation unit 11 receives a file (input file) input from the terminal device 30 to the file server 20 (step A1). Further, in step A1, if the input file is compressed, the summary information creation unit 11 decompresses it as a preprocessing.

Next, the summary information creation unit 11 creates summary information for the input file (step A2). Specifically, in step A2, the summary information creation unit 11 creates summary information by using the optimum method according to the type of the input file.

Next, the similarity calculation unit 12 compares the summary information of the input file with the summary information of the registration file for each registration file, extracts the difference between the summary information, and uses the extracted difference to input the input file. And the registration file are calculated (step A3).

Next, the difference calculation unit 13 identifies the registration file having the highest degree of similarity, and extracts the difference between the specified registration file and the input file (step A4). When the input file is compressed, the difference calculation unit 13 extracts the difference from the registered file by using the decompressed input file. Further, when the registered file is registered with the difference and the similar file information, the process shown in FIG. 7 described later is executed for the registered file.

Next, the registration unit 14 creates similar file information for the registration file having the highest degree of similarity to the input file (step A5). Specifically, the registration unit 14 creates similar file information using the identification number assigned to the registration file in the tree structure. In the present embodiment, as the identification number, an "i-node number" assigned to a file on a general file system is used. Hereinafter, the identification number is also referred to as an i-node number.

Next, the registration unit 14 registers the difference extracted in step A4 and the similar file information created in step A5 in the database 21 instead of the actual file of the input file (step A6).

In step A6, the registration unit 14 may compress the difference extracted in step A4. Further, the registration unit 14 can also register other information such as a file name as an input file in addition to the difference and similar file information. Further, as shown in FIG. 3, the registration unit 14 manages the input file by a tree structure.

Subsequently, the file reading process by the file management device 10 will be described with reference to FIG. 7. FIG. 7 is a flow chart showing an operation of the file management device at the time of file reading processing according to the embodiment of the present invention.

As shown in FIG. 7, when the terminal device 30 first instructs the file server 20 to read a specific file, the file management device 10 accepts the instruction (the file operation unit 15). Step B1).

Next, the file operation unit 15 determines whether or not the file to be referred is registered as actual data in the database 21 (step B2).

As a result of the determination in step B2, if the file to be referred is not registered as actual data, the file operation unit 15 reads the difference and similar file information registered as the file to be referredent. (Step B3)

Next, the file operation unit 15 identifies the registration file (parent file) specified by the similar file information read in step B3 (step B4). Further, the file operation unit 15 reads the specified registration file. If the registration file specified in step B4 is registered with the difference and similar file information, steps B1 to B4 and later are described for this registration file before the execution of step B5 described later. Steps B5 to B7 are executed.

Next, the file operation unit 15 combines the difference read in step B3 with the parent file specified in step B4 to generate a file to be referred (step B5).

Next, the file operation unit 15 outputs the file generated in step B5 to the file server 20 as a file to be instructed (step B6).

Further, as a result of the determination in step B2 described above, when the file to be instructed is registered as actual data, the file operation unit 15 reads the file to be instructed (step B7).

After the execution of step B6 or step B7, the processing in the file management device 10 is completed, and the read file is transmitted from the file server 20 to the terminal device 30. After that, the file server 20 transmits the file to be instructed to the terminal device 30.

Subsequently, the file deletion process by the file management device 10 will be described with reference to FIG. FIG. 8 is a flow chart showing the operation of the file management device at the time of file deletion processing according to the embodiment of the present invention.

As shown in FIG. 8, when a specific file deletion is first instructed by the terminal device 30 to the file server 20, the file operation unit 15 in the file management device 10 accepts the instruction (step). C1).

Next, the file operation unit 15 deletes the correspondence between the file name of the file to be instructed and the inode number from the directory information created by the file server 20 (step C2).

Next, the file operation unit 15 determines whether or not a file (child file) registered by the similar file information for specifying the referent exists in the file to be instructed (step C3). ..

As a result of the determination in step C3, if the child file exists, the file operation unit 15 does nothing and ends the process. On the other hand, if the child file does not exist as a result of the determination in step C3, the file operation unit 15 determines whether or not the file to be instructed is registered as an actual file (step C4).

As a result of the determination in step C4, if the file to be instructed is registered as an actual file, the file operation unit 15 deletes the actual file (step C5) and ends the process.

On the other hand, as a result of the determination in step C4, if the file to be instructed is not registered as an actual file, the file operation unit 15 selects the registered file having the highest degree of similarity with the difference registered as the instructed file. The specified similar file information (= i-node number) is deleted from the database 21 (step C6).

Next, the file operation unit 15 determines whether or not the parent file of the instructed file has been deleted (step C7). Specifically, the file operation unit 15 determines whether or not the correspondence between the file name of the parent file and the inode number is deleted from the directory information.

If the parent file has not been deleted as a result of the determination in step C7, the file operation unit 15 ends the process. On the other hand, if the parent file is deleted as a result of the determination in step C7, the file operation unit 15 changes the processing target to the parent file (step C8), and the file operation unit 15 again performs steps C3 to C8. Execute.

[Effect in the embodiment]
As described above, in the present embodiment, the registration file most similar to the input file is specified from the similarity between the summary information of the input file and the summary information of the registration file, not from the similarity between the files. Therefore, the process of specifying the registration file most similar to the input file can be speeded up. The registration of the input file can be completed by registering the difference and the similar file information, and the difference can be further compressed. Therefore, according to the present embodiment, in the file system, it is possible to reduce the time required for data registration while reducing the data compression.

[Modification example]
Subsequently, a modification of the present embodiment will be described. In this modification, in order to efficiently identify the registered file having the highest similarity with the input file, the similarity calculation unit 12 calculates the similarity according to the following steps.

First, the similarity calculation unit 12 identifies the file directly under the tree (see FIG. 3) that manages the registered files, and calculates the similarity between the input file and the specified file. The file directly under the root of the tree is a file that is not composed of the difference and the similar file information, and in the example of FIG. 3, the files A and D correspond to each other.

Subsequently, the similarity calculation unit 12 identifies the file having the highest similarity among the files directly under the root of the tree, and searches for child files in the identified file. If a child file exists, the similarity calculation unit 12 calculates the similarity between the child file directly under the file directly under the root of the tree and the input file.

As a result of the calculation, if the similarity of the file directly under the root of the tree is higher than the similarity of the child files directly under it, the similarity calculation unit 12 is most similar to the file directly under the root of the tree. Make it a high-degree file and end the process.

On the other hand, if the similarity of the child files directly under the tree root is higher than the similarity of the files directly under the root of the tree, the similarity calculation unit 12 may use the child files directly under the child file directly under the tree root. Search the file and calculate the similarity. That is, the similarity calculation unit 12 further searches the child files and calculates the similarity as long as the similarity of the child files is high.

This search and calculation of the similarity are performed until the similarity of the parent file is higher than the similarity of the child files directly under it, or until the child files no longer exist. As described above, in this modification, the similarity calculation unit 12 traces the branch having the higher similarity from the root of the tree toward the leaf node, and detects the registration file having the highest similarity with the input file. do.

Specifically, in the example of FIG. 3, the similarity calculation unit 12 first calculates the similarity between the file A and the file D with the input file. Then, when the similarity of the file D is higher than that of the file A, the similarity calculation unit 12 calculates the similarity between the input file and the child file E of the file D. In this case, if the similarity of the file D is higher than the similarity of the child file E, the similarity calculation unit 12 ends the process.

On the other hand, when the similarity of the child file E is higher than the similarity of the file D, the similarity calculation unit 12 further calculates the similarity between the grandchild file F directly under the file E and the input file. .. Then, the similarity calculation unit 12 compares the similarity between the child file E and the grandchild file F, and depending on the result of the comparison, searches for a lower file and calculates the similarity. ..

[Application example]
Here, a case where the input file is an archive file such as a zip file will be described.

<At the time of registration>
When an archive file is input as an input file, the file management device 10 decompresses the internal files of the archive file and executes steps A2 to A6 shown in FIG. 6 for each file constituting the archive file. do.

That is, in the archive file, each file is registered by the difference and the similar file information. For this reason, the file management device 10 can also handle the archive file itself like a directory in the file system provided by the file server 20.

<When reading>
When the reading of the archive file is instructed, the file management device 10 executes steps B2 to B6 shown in FIG. 7 for each file constituting the internal file group of the archive file. As a result, the original archive file is restored and transmitted to the user's terminal device 30.

Further, in the present embodiment, the input file is a file in a format in which another file such as an image can be pasted in the file, such as a file having doxx or pptx as an extension, and another file is pasted in the file. If so, the file management device 10 can handle this input file in the same manner as the above archive file.

[program]
The program in the present embodiment may be any program that causes a computer to execute steps A1 to A6 shown in FIG. 6, steps B1 to B6 shown in FIG. 7, and steps C1 to C8 shown in FIG. By installing and executing this program on a computer, the file management device 10 and the file management method according to the present embodiment can be realized. In this case, the computer processor functions as a summary information creation unit 11, a similarity calculation unit 12, a difference calculation unit 13, a registration unit 14, and a file operation unit 15 to perform processing.

Further, in the present embodiment, the database 21 stores the data files constituting the database 21 in a storage device such as a hard disk provided in the computer, or connects the recording medium in which the data files are stored to the computer. It can be realized by mounting it on a reading device.

Further, the program in the present embodiment may be executed by a computer system constructed by a plurality of computers. In this case, for example, each computer may function as any one of the summary information creation unit 11, the similarity calculation unit 12, the difference calculation unit 13, the registration unit 14, and the file operation unit 15. Further, the database 21 may be built on a computer different from the computer that executes the program in the present embodiment.

Here, a computer that realizes the file management device 10 by executing the program according to the present embodiment will be described with reference to FIG. FIG. 9 is a block diagram showing an example of a computer that realizes the file management device according to the embodiment of the present invention.

As shown in FIG. 9, the computer 110 includes a CPU (Central Processing Unit) 111, a main memory 112, a storage device 113, an input interface 114, a display controller 115, a data reader / writer 116, and a communication interface 117. And. Each of these parts is connected to each other via a bus 121 so as to be capable of data communication. The computer 110 may include a GPU (Graphics Processing Unit) or an FPGA (Field-Programmable Gate Array) in addition to the CPU 111 or in place of the CPU 111.

The CPU 111 expands the programs (codes) of the present embodiment stored in the storage device 113 into the main memory 112 and executes them in a predetermined order to perform various operations. The main memory 112 is typically a volatile storage device such as a DRAM (Dynamic Random Access Memory). Further, the program in the present embodiment is provided in a state of being stored in a computer-readable recording medium 120. The program in the present embodiment may be distributed on the Internet connected via the communication interface 117.

Further, specific examples of the storage device 113 include a semiconductor storage device such as a flash memory in addition to a hard disk drive. The input interface 114 mediates data transmission between the CPU 111 and an input device 118 such as a keyboard and mouse. The display controller 115 is connected to the display device 119 and controls the display on the display device 119.

The data reader / writer 116 mediates the data transmission between the CPU 111 and the recording medium 120, reads the program from the recording medium 120, and writes the processing result in the computer 110 to the recording medium 120. The communication interface 117 mediates data transmission between the CPU 111 and another computer.

Specific examples of the recording medium 120 include a general-purpose semiconductor storage device such as CF (Compact Flash (registered trademark)) and SD (Secure Digital), a magnetic recording medium such as a flexible disk, or a CD-. Examples include optical recording media such as ROM (Compact Disk Read Only Memory).

The file management device 10 in the present embodiment can also be realized by using the hardware corresponding to each part instead of the computer in which the program is installed. Further, the file management device 10 may be partially realized by a program and the rest may be realized by hardware.

A part or all of the above-described embodiments can be expressed by the following descriptions (Appendix 1) to (Appendix 15), but the present invention is not limited to the following description.

(Appendix 1)
A summary information creation unit that analyzes the input file and creates summary information indicating the contents of the input file.
A similarity calculation unit that calculates the similarity with the input file for each registration file by comparing the summary information of the input file with the summary information of the already registered registration file.
A difference calculation unit that calculates the difference between the registration file with the highest degree of similarity and the input file, and
A registration unit that registers the calculated difference and similar file information that identifies the registration file having the highest degree of similarity as the input file.
A file management device characterized by being equipped with.

(Appendix 2)
The file management device described in Appendix 1
When the registration unit manages the input file and the registration file by a tree structure and the input file is registered by the difference and the similar file information, the input file is referred to as the similar file information. The file directly under the registration file specified by
A file management device characterized by that.

(Appendix 3)
The file management device according to Appendix 1 or 2.
When the input file is a document file output from an application program for creating a document,
The summary information creation unit analyzes the composition of the sentence in the input file, extracts a part of the beginning side of the sentence, and creates the summary information using the extracted part.
The similarity calculation unit identifies a difference between the summary information of the input file and the summary information of the registration file that has already been registered, and uses the size of the specified difference to register the input file and the registration. Calculate the similarity with the file,
A file management device characterized by that.

(Appendix 4)
The file management device according to Appendix 1 or 2.
When the input file is a source file,
The summary information creation unit extracts the function definition from the input file, creates the summary information using the extracted function definition, and creates the summary information.
The similarity calculation unit identifies a difference between the summary information of the input file and the summary information of the registered file that has already been registered, and uses the size of the specified difference to indicate the input file and the above. Calculate the similarity with the registration file,
A file management device characterized by that.

(Appendix 5)
The file management device according to Appendix 1 or 2.
When the input file is an image file,
The summary information creation unit creates a thumbnail image in which the size of the image is reduced from the original size as the summary information.
The similarity calculation unit identifies the difference between the thumbnail image of the input file and the thumbnail image of the registered file that has already been registered, and uses the pixel similarity obtained from the specified difference to input the input. Calculate the similarity between the file and the registration file,
A file management device characterized by that.

(Appendix 6)
(A) A step of analyzing an input file and creating summary information indicating the contents of the input file.
(B) A step of comparing the summary information of the input file with the summary information of the already registered registration file and calculating the similarity with the input file for each registration file.
(C) A step for calculating the difference between the registration file having the highest degree of similarity and the input file, and
(D) A step of registering the calculated difference and the similar file information for specifying the registration file having the highest degree of similarity as the input file.
A file management method characterized by having.

(Appendix 7)
The file management method described in Appendix 6
In the step (d), the input file and the registration file are managed by a tree structure, and when the input file is registered by the difference and the similar file information, the input file is used. The file directly under the registration file specified by the similar file information,
A file management method characterized by that.

(Appendix 8)
The file management method described in Appendix 6 or 7.
When the input file is a document file output from an application program for creating a document,
In the step (a), the composition of the sentence in the input file is analyzed, a part of the beginning side of the sentence is taken out, and the summary information is created by using the taken out part.
In the step (b), the difference between the summary information of the input file and the summary information of the registered file that has already been registered is specified, and the input file and the said are used using the size of the specified difference. Calculate the similarity with the registration file,
A file management method characterized by that.

(Appendix 9)
The file management method described in Appendix 6 or 7.
When the input file is a source file,
In the step (a), the function definition is extracted from the input file, and the summary information is created using the extracted function definition.
In the step (b), the difference between the summary information of the input file and the summary information of the registered file that has already been registered is specified, and the size of the specified difference is used with the input file. Calculate the similarity with the registration file,
A file management method characterized by that.

(Appendix 10)
The file management method described in Appendix 6 or 7.
When the input file is an image file,
In the step (a), as the summary information, a thumbnail image in which the size of the image is reduced from the original size is created.
In the step (b), the difference between the thumbnail image of the input file and the thumbnail image of the registered file already registered is specified, and the pixel similarity obtained from the specified difference is used to describe the difference. Calculate the similarity between the input file and the registration file,
A file management method characterized by that.

(Appendix 11)
On the computer
(A) A step of analyzing an input file and creating summary information indicating the contents of the input file.
(B) A step of comparing the summary information of the input file with the summary information of the already registered registration file and calculating the similarity with the input file for each registration file.
(C) A step for calculating the difference between the registration file having the highest degree of similarity and the input file, and
(D) A step of registering the calculated difference and the similar file information for specifying the registration file having the highest degree of similarity as the input file.
A program characterized by executing.

(Appendix 12)
The program described in Appendix 11
In the step (d), the input file and the registration file are managed by a tree structure, and when the input file is registered by the difference and the similar file information, the input file is used. The file directly under the registration file specified by the similar file information,
A program characterized by that.

(Appendix 13)
The program described in Appendix 11 or 12,
When the input file is a document file output from an application program for creating a document,
In the step (a), the composition of the sentence in the input file is analyzed, a part of the beginning side of the sentence is taken out, and the summary information is created by using the taken out part.
In the step (b), the difference between the summary information of the input file and the summary information of the registered file that has already been registered is specified, and the input file and the said are used using the size of the specified difference. Calculate the similarity with the registration file,
A program characterized by that.

(Appendix 14)
The program described in Appendix 11 or 12,
When the input file is a source file,
In the step (a), the function definition is extracted from the input file, and the summary information is created using the extracted function definition.
In the step (b), the difference between the summary information of the input file and the summary information of the registered file that has already been registered is specified, and the size of the specified difference is used with the input file. Calculate the similarity with the registration file,
A program characterized by that.

(Appendix 15)
The program described in Appendix 11 or 12,
When the input file is an image file,
In the step (a), as the summary information, a thumbnail image in which the size of the image is reduced from the original size is created.
In the step (b), the difference between the thumbnail image of the input file and the thumbnail image of the registered file already registered is specified, and the pixel similarity obtained from the specified difference is used to describe the difference. Calculate the similarity between the input file and the registration file,
A program characterized by that.

As described above, according to the present invention, in the file system, it is possible to reduce the time required for data registration while reducing the data compression. The present invention is useful for various file systems.

10 File management device 11 Summary information creation unit 12 Similarity calculation unit 13 Difference calculation unit 14 Registration unit 15 File operation unit 20 File server 21 Database 30 Terminal equipment 40 Network 110 Computer 111 CPU
112 Main memory 113 Storage device 114 Input interface 115 Display controller 116 Data reader / writer 117 Communication interface 118 Input device 119 Display device 120 Recording medium 121 Bus

Claims (15)

A summary information creation unit that analyzes the input file and creates summary information indicating the contents of the input file.
A similarity calculation unit that calculates the similarity with the input file for each registration file by comparing the summary information of the input file with the summary information of the already registered registration file.
A difference calculation unit that calculates the difference between the registration file with the highest degree of similarity and the input file, and
A registration unit that registers the calculated difference and similar file information that identifies the registration file having the highest degree of similarity as the input file.
A file management device characterized by being equipped with. The file management device according to claim 1.
When the registration unit manages the input file and the registration file by a tree structure and the input file is registered by the difference and the similar file information, the input file is referred to as the similar file information. The file directly under the registration file specified by
A file management device characterized by that. The file management device according to claim 1 or 2.
When the input file is a document file output from an application program for creating a document,
The summary information creation unit analyzes the composition of the sentence in the input file, extracts a part of the beginning side of the sentence, and creates the summary information using the extracted part.
The similarity calculation unit identifies a difference between the summary information of the input file and the summary information of the registration file that has already been registered, and uses the size of the specified difference to register the input file and the registration. Calculate the similarity with the file,
A file management device characterized by that. The file management device according to claim 1 or 2.
When the input file is a source file,
The summary information creation unit extracts the function definition from the input file, creates the summary information using the extracted function definition, and creates the summary information.
The similarity calculation unit identifies a difference between the summary information of the input file and the summary information of the registered file that has already been registered, and uses the size of the specified difference to indicate the input file and the above. Calculate the similarity with the registration file,
A file management device characterized by that. The file management device according to claim 1 or 2.
When the input file is an image file,
The summary information creation unit creates a thumbnail image in which the size of the image is reduced from the original size as the summary information.
The similarity calculation unit identifies the difference between the thumbnail image of the input file and the thumbnail image of the registered file that has already been registered, and uses the pixel similarity obtained from the specified difference to input the input. Calculate the similarity between the file and the registration file,
A file management device characterized by that. (A) A step of analyzing an input file and creating summary information indicating the contents of the input file.
(B) A step of comparing the summary information of the input file with the summary information of the already registered registration file and calculating the similarity with the input file for each registration file.
(C) A step for calculating the difference between the registration file having the highest degree of similarity and the input file, and
(D) A step of registering the calculated difference and the similar file information for specifying the registration file having the highest degree of similarity as the input file.
A file management method characterized by having. The file management method according to claim 6.
In the step (d), the input file and the registration file are managed by a tree structure, and when the input file is registered by the difference and the similar file information, the input file is used. The file directly under the registration file specified by the similar file information,
A file management method characterized by that. The file management method according to claim 6 or 7.
When the input file is a document file output from an application program for creating a document,
In the step (a), the composition of the sentence in the input file is analyzed, a part of the beginning side of the sentence is taken out, and the summary information is created by using the taken out part.
In the step (b), the difference between the summary information of the input file and the summary information of the registered file that has already been registered is specified, and the input file and the said are used using the size of the specified difference. Calculate the similarity with the registration file,
A file management method characterized by that. The file management method according to claim 6 or 7.
When the input file is a source file,
In the step (a), the function definition is extracted from the input file, and the summary information is created using the extracted function definition.
In the step (b), the difference between the summary information of the input file and the summary information of the registered file that has already been registered is specified, and the size of the specified difference is used with the input file. Calculate the similarity with the registration file,
A file management method characterized by that. The file management method according to claim 6 or 7.
When the input file is an image file,
In the step (a), as the summary information, a thumbnail image in which the size of the image is reduced from the original size is created.
In the step (b), the difference between the thumbnail image of the input file and the thumbnail image of the registered file already registered is specified, and the pixel similarity obtained from the specified difference is used to describe the difference. Calculate the similarity between the input file and the registration file,
A file management method characterized by that. On the computer
(A) A step of analyzing an input file and creating summary information indicating the contents of the input file.
(B) A step of comparing the summary information of the input file with the summary information of the already registered registration file and calculating the similarity with the input file for each registration file.
(C) A step for calculating the difference between the registration file having the highest degree of similarity and the input file, and
(D) A step of registering the calculated difference and the similar file information for specifying the registration file having the highest degree of similarity as the input file.
A program characterized by executing. The program according to claim 11.
In the step (d), the input file and the registration file are managed by a tree structure, and when the input file is registered by the difference and the similar file information, the input file is used. The file directly under the registration file specified by the similar file information,
A program characterized by that. The program according to claim 11 or 12.
When the input file is a document file output from an application program for creating a document,
In the step (a), the composition of the sentence in the input file is analyzed, a part of the beginning side of the sentence is taken out, and the summary information is created by using the taken out part.
In the step (b), the difference between the summary information of the input file and the summary information of the registered file that has already been registered is specified, and the input file and the said are used using the size of the specified difference. Calculate the similarity with the registration file,
A program characterized by that. The program according to claim 11 or 12.
When the input file is a source file,
In the step (a), the function definition is extracted from the input file, and the summary information is created using the extracted function definition.
In the step (b), the difference between the summary information of the input file and the summary information of the registered file that has already been registered is specified, and the size of the specified difference is used with the input file. Calculate the similarity with the registration file,
A program characterized by that. The program according to claim 11 or 12.
When the input file is an image file,
In the step (a), as the summary information, a thumbnail image in which the size of the image is reduced from the original size is created.
In the step (b), the difference between the thumbnail image of the input file and the thumbnail image of the registered file already registered is specified, and the pixel similarity obtained from the specified difference is used to describe the difference. Calculate the similarity between the input file and the registration file,
A program characterized by that.

Images