JP2015197810A - Server device, information processing method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform deletion processing of a folder without making processing of a side for calling a web service complicated.SOLUTION: A server device includes reception means for receiving an instruction to delete a folder, and deletion means for limiting access to the folder of the lowermost layer of the folder receiving the deletion instruction in the case of receiving the instruction to delete the folder by the reception means, executing the deletion of the folder of the lowermost layer in the case that the limitation of the access to the folder of the lowermost layer is successful, limiting access to a folder of the next lowermost layer in the case that the deletion of the folder of the lowermost layer is successful, and repeating processing to the folder of the next lowermost layer without executing the deletion of the folder of the lowermost layer and executing the deletion of the folder receiving the instruction to delete in the case that the limitation of the access to the folder of the lowermost layer is not successful.

Description

  The present invention relates to a server device, an information processing method, and a program.

  Conventionally, as a file management mechanism for managing electronic files, a mechanism for organizing and managing files in a hierarchical structure (tree structure) called a folder or directory is used. In such a mechanism, there is a problem that processing time is required when a folder containing many files is deleted from the file management system. The reason for the processing time is that if another process is operating the file in the folder you are trying to delete, it must be deleted after the process is completed because the file is locked exclusively. Because. As a countermeasure, there is a technique such as Patent Document 1 that shortens the time by repeating the process of ending the deletion order of locked files.

  On the other hand, with the development of Web technology in recent years, system cooperation in which a plurality of Web services provide functions in cooperation with each other is performed instead of being completed with only one Web service. In order to simplify the cooperation of a plurality of Web services, an increasing number of Web services employ a simple HTTP interface called Representational State Transfer (REST). The Web service using the REST includes all necessary information in a single call unit, and the Web server that provides the Web service has a feature that the session state is not stored (state management is not performed).

JP2013-109630A

  However, when a Web service using REST is provided by a file management system, there are the following problems. Assume that the folder can be deleted by a single call from the Web service. Even if processing is performed from what can be processed as in patent literature, the side that calls the Web service cannot predict the processing time of other processing that locks the file or the number of other processing that locks, so that Unable to predict folder deletion time. For this reason, there is a problem that it is difficult for the side calling the Web service to set the communication timeout time. As a countermeasure, there is a method in which the file management system side separately provides a web service call for accepting processing and a web service call for responding to the processing result. However, as a result, in order to know the result, the caller of the Web service must perform the deletion processing request and obtain the results multiple times separately, even if it is a folder deletion of a small amount of files that should end in a short time. become. As a result, the processing on the side that calls the Web service becomes complicated.

  Accordingly, the present invention provides a server device that performs file management based on a folder hierarchical structure, and includes a receiving unit that receives a folder deletion instruction, and a folder deletion instruction when the reception unit receives a folder deletion instruction. The access to the lowermost folder of the folder that has received the instruction is restricted, and when the access to the lowermost folder is successfully restricted, the lowermost folder is deleted, and the lowermost folder is deleted. If the deletion is successful, the access to the next lowest folder is restricted. If the access restriction to the lowest folder is not successful, the deletion of the lowest folder is not executed, and the next Delete means for repeatedly executing the process of deleting the folder in the lowest layer and executing the delete process of the folder that has received the delete instruction.

  According to the present invention, it is possible to delete a folder without complicating the processing on the side that calls the Web service.

It is a figure which shows an example of the system configuration | structure of a document management system. It is a figure which shows an example of the hardware constitutions of a storage server. It is a figure which shows an example of the software configuration of a storage server. It is a figure which shows an example of the table of folder data. It is a figure which shows an example of the table of file data. It is a figure which shows an example of the folder management table during deletion. It is a flowchart which shows an example of a folder deletion process. It is a flowchart which shows an example of a recursive process.

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

<First Embodiment>
A first embodiment will be described. In the embodiment, a process for deleting a folder will be described as an example of a document management system.

(System configuration)
FIG. 1 is a diagram illustrating an example of a system configuration of a document management system. As shown in FIG. 1, the document management system includes a storage server 110, a document management Web server 120, and a PC terminal 130. The storage server 110, the document management Web server 120, and the PC terminal 130 are connected via the LAN 100.
The storage server 110 is a back-end server for managing files and folders according to an HTTP protocol instruction from the document management Web server 120, and performs file management based on a folder hierarchical structure. The document management Web server 120 is a front-end server for providing the PC terminal 130 with HTML and JavaScript (registered trademark) for configuring a user interface for managing electronic documents. The document management Web server 120 inputs and outputs files using the Web service provided by the storage server 110 when reading and writing document data as a file based on a request from the PC terminal 130. That is, the storage server 110 is an example of a file management system. The document management Web server 120 is an example of a side that calls a Web service related to the file management system. The PC terminal 130 is a terminal device for a user who uses the system to create a document or view a document. In the present embodiment, it is assumed that the PC terminal 130 executes a Web browser that communicates with the document management Web server 120 as an application. However, in the system according to the present embodiment, the number of components is not limited. Also, the network configuration may be an Internet environment instead of a LAN.

(Hardware configuration)
FIG. 2 is a diagram illustrating an example of a hardware configuration of the storage server 110.
As shown in FIG. 2, in the storage server 110, a CPU 201, a RAM 202, a ROM 203, an input device 204, a network I / F 205, an HDD 206, and a display device 207 are connected via a system bus 208 so as to communicate with each other.
The ROM 203 or HDD 206 stores programs such as an operating system and applications. The CPU 201 is a central processing unit, and reads the program from the ROM 203 or HDD 206 onto the RAM 202 and executes it as necessary. As a result, the software configuration of the storage server 110 and the processing of the flowchart described later are realized. The CPU 201 also controls communication with other devices on the network connected via the network I / F 205.
The document management Web server 120 and the PC terminal 130 also have the same hardware configuration as the storage server 110. That is, the function of each device is realized by the CPU of each device executing processing based on a program stored in the HDD or the like of each device.

(Software configuration)
FIG. 3 is a diagram illustrating an example of a software configuration of the storage server 110.
The WEB service unit 301 receives an HTTP request from the document management Web server 120 and returns a processing result.
The business logic unit 302 controls processing related to file operations and controls folder deletion processing described later.
The asynchronous processing unit 303 is a processing unit for performing control processing separately from the request / response processing from the document management Web server 120. The asynchronous processing unit 303 periodically makes an inquiry to the database unit 304 and instructs the business logic unit 302 to perform processing if there is an operation to be processed.
The database unit 304 is a processing unit for realizing a relational database that can be controlled by SQL. In the present embodiment, the database unit 304 manages data such as files and folders using a table described later.

(Table structure)
Next, how folders and files are managed in the database unit 304 in this embodiment will be described.
The storage server 110 of this embodiment forms a tree by hierarchizing folders, and manages so that each folder can contain zero or more folders and files. Note that a folder at a higher level when viewed from a specific file or folder is called a parent folder, and a folder at a lower level when viewed from a specific folder is called a child folder. In addition, files and folders have names composed of character strings, and in the Web service provided by the storage server 110, files and folders can be specified using the names.

FIG. 4 is a diagram illustrating an example of a folder data table stored in the database unit 304. Columns 401 to 405 indicate table columns.
The column of the folder ID 401 indicates an ID that uniquely identifies a folder managed by the storage server 110 in the database.
The column of parent folder ID 402 indicates the folder ID of the parent folder that is a folder in the upper hierarchy of the folder. When the parent folder ID is empty, it indicates that the folder is the root folder which is the highest hierarchy.
The column of the folder path 403 indicates an absolute path in which the folder name that is the name of the folder is connected in a hierarchy from the root folder. In the present embodiment, the delimiter between layers is expressed by a separator character of “/”.
The column of size 404 indicates the total size of files included immediately below the folder. The value of the column of the size 404 is the same value as that obtained by acquiring a record whose parent folder ID matches this folder ID from the file data table described later and taking the total of the file sizes. When the storage server 110 calculates the file size every time the size of the folder is referred from the document management Web server 120, the performance is affected. Therefore, it is assumed that the storage server 110 calculates the total value of the file sizes when registering or deleting a document, and stores it in a folder record.
The column of update date and time 405 indicates the date and time when a folder or file directly under this folder was added / deleted.
Lines 406 to 410 are specific examples of folder data. A line 406 represents a root folder whose folder name is “FolA”. Lines 407 and 409 represent “FolB” and “FolD”, which are child folders of “FolA”, respectively. A row 408 represents a child folder “FolC” of “FolB”. A line 410 represents a child folder “FolE” of “FolD”.

FIG. 5 is a diagram illustrating an example of a file data table stored in the database unit 304. Columns 501 to 505 indicate table columns.
The column of the file ID 501 indicates an ID that uniquely identifies the file managed by the storage server 110 in the database.
The parent folder ID 502 column indicates the folder ID of the parent folder of the file. The CPU 201 can determine in which folder the file is included by referring to the record in the folder table having this folder ID.
The column of the file name 503 indicates the name of the file.
The column of size 504 indicates the size of the file.
The column of update date and time 505 indicates the date and time when this file was added / updated.
A column of the file data 506 indicates binary data of the file.
Lines 507 to 511 are specific examples of file data. A row 507 indicates a file “FileA” having the folder “FolB” as a parent folder. Others are the same.

FIG. 6 is a diagram illustrating an example of the deleting folder management table. The deleting folder management table registers a folder being deleted as a record.
The column of the deleting folder path 601 indicates a folder path in which folder deletion is being executed. Note that the table setting is such that the value in the column 601 does not enter the same value as other records. Therefore, the same folder path cannot be set for two records.
The column of the start date / time 602 indicates the time when the folder deletion is started.
Lines 603 and 604 are specific examples of records. In the example of FIG. 6, it is shown that the deletion process of “FolA” and the child folder “FolR” of “FolQ” is being executed.
The business logic unit 302 confirms this table when adding / updating / deleting files and adding folders. As a result of the confirmation, if the target upper folder or itself is registered, the business logic unit 302 determines that the target file or folder may be deleted because the upper folder is being deleted. When the business logic unit 302 determines that there is a possibility, the WEB service unit 301 sends an HTTP status code “503” (an error indicating that the server or service cannot be used) to the document management Web server 120. Response).

(Folder deletion process)
Next, folder deletion information processing according to the present embodiment will be described.
First, a user of the system uses a Web browser of the PC terminal 130 to display a Web UI provided by the document management Web server 120, and instructs a folder deletion process using the Web UI. In response to this, the document management Web server 120 calls the Web service of the storage server 110. In the present embodiment, the document management Web server 120 calls an HTTP “DELETE” method with respect to the URL “http: // {host name of the storage server 110} / SS / Folder / FolA”. As a result, the document management Web server 120 instructs the storage server 110 to delete the folder with the absolute path “FolA”. “SS” is information for identifying the file management application of the storage server. Also, “Folder” in the URL is information for identifying that the process is related to a folder.
When called from the document management Web server 120, the CPU 201 of the storage server 110 executes folder deletion processing. First, the WEB service unit 301 analyzes the received communication using the HTTP protocol, and if it is a folder deletion instruction, calls the folder deletion process of the business logic unit 302.

FIG. 7 is a flowchart illustrating an example of folder deletion processing executed by the business logic unit 302.
In step S701, the business logic unit 302 acquires the start time of the folder deletion process. For example, the business logic unit 302 acquires the time from the timer of the storage server 110 or the like when the folder deletion process is called, and sets it as the start time of the folder deletion process.
In step S702, the business logic unit 302 inquires of the database unit 304, specifies a record whose folder path matches the folder path to be deleted from the folder table, and acquires a folder ID from the record.
In step S703, the business logic unit 302 adds (or records) a folder path to be deleted to the folder management table for deletion. More specifically, before adding a folder path, the business logic unit 302 checks whether the folder path of the upper folder of the target folder path is registered in the deleting folder management table. For example, if the deletion target is “FolA / FolB / FolC”, the business logic unit 302 confirms that “FolA / FolB” and “FolA” are not registered in the deleting folder management table. If registered, the upper level folder is being deleted, and the business logic unit 302 determines that the record registration has failed. As described above, the business logic unit 302 cannot register a record that already has the same folder path. In this case, the business logic unit 302 determines that the record registration has failed because the deletion target folder is being deleted. If the business logic unit 302 determines that the registration has failed, the business logic unit 302 proceeds to step S709; otherwise, the business logic unit 302 proceeds to step S704.

In step S <b> 704, the business logic unit 302 calls a recursive process for deleting the deletion target folder and the child folders and child documents included therein. Then, the business logic unit 302 receives deletion success and deletion failure as a result of the recursive processing. The recursive process will be described with reference to FIG. At the time of calling the recursive process, the business logic unit 302 passes the process start time acquired in step S701 and the folder ID acquired in step S702. As will be described later, in the recursive processing, a response is returned within a predetermined time from the processing start time.
In step S705, the business logic unit 302 deletes the record in the deleting folder management table registered in step S703.
In step S706, the business logic unit 302 determines the result of step S704. If the deletion target folder has been successfully deleted, the business logic unit 302 transitions to step S707, and otherwise transitions to step S710.
In step S707, the business logic unit 302 registers in the database unit 304 information for asynchronous processing for updating the update date and time of the parent folder of the deletion target folder. More specifically, the business logic unit 302 registers information (folder path and update time) necessary for updating the record of the parent folder in the table of the database unit 304 referred to by the asynchronous processing unit 303. As described above, the asynchronous processing unit 303 periodically monitors the database unit 304, and executes record update of the parent folder asynchronously with the folder deletion processing based on the registered information.

The process in step S708 is a process when the deletion of the folder is successful. In step S <b> 708, the business logic unit 302 sends a deletion success response to the WEB service unit 301. Then, the WEB service unit 301 responds to the call source document management Web server 120 with an HTTP status code “200” (meaning success).
The process of step S709 is a process when it overlaps with the folder deletion process operating in parallel. In step S709, the business logic unit 302 responds to the WEB service unit 301 that it is temporarily overloaded. Then, the WEB service unit 301 responds to the caller (requester) document management Web server 120 with an HTTP status code “503” (which means an error indicating that the server or service cannot be used).
The processing in step S710 is processing when the folder cannot be deleted during the folder deletion processing due to duplication with other processing. In step S710, the business logic unit 302 responds to the WEB service unit 301 that the process has collided. Then, the WEB service unit 301 responds to the caller document management Web server 120 with an HTTP status code “409” (which means an error that could not be completed due to a collision with the current state of the resource).

FIG. 8 is a flowchart showing an example of recursion processing called inside the folder deletion processing executed by the business logic unit 302.
When the recursive process is called, the start time and the target folder ID are passed.
In step S801, the business logic unit 302 acquires a folder record list in which the parent folder ID matches the folder ID passed when the recursive process is called from the folder table.
In step S802, the business logic unit 302 repeats the process in step S803 until the number of folder records acquired in step S801 or the difference between the current time and the start time is equal to or greater than a predetermined timeout time.
In step S803, the business logic unit 302 recursively calls the recursive process of FIG. At this time, the business logic unit 302 calls the recursive process by passing the ID and start time of the child folder that has not yet called the delete process.

In step S804, the business logic unit 302 determines the lock waiting time from the remaining time of the difference between the current time and the start time. The business logic unit 302 determines the lock waiting time in stages according to the remaining time. For example, the business logic unit 302 locks if the remaining time is low, such as 10 seconds if the remaining time is 1 minute or more, 2 seconds if it is 1 minute or less and 30 seconds or more, or 0 seconds if it is 15 seconds or less. Decide on a short waiting time.
In step S805, the business logic unit 302 locks the record of the folder ID to be processed in the folder table (restricts access from other (access restriction)). At this time, the business logic unit 302 sets the lock waiting time as the time determined in step S804. The business logic unit 302 transitions to step S806 when the lock is successful, and transitions to step S814 when the lock target is not in the table or when the lock wait time elapses and time-out occurs.
In step S806, the business logic unit 302 acquires a file record list in which the parent folder ID matches the folder ID passed when the recursive process is called from the file table.

In step S807, the business logic unit 302 repeats steps S808 to S810 until the number of file records acquired in step S806 or the difference between the current time and the start time is equal to or greater than a predetermined timeout time.
In step S808, the business logic unit 302 calculates the lock waiting time from the remaining time of the difference between the current time and the start time, as in step S804.
In step S809, the business logic unit 302 locks the record of the file ID to be processed in the file table. At this time, the business logic unit 302 sets the lock waiting time to the time determined in step S808. If the lock is successful, the business logic unit 302 transitions to step S810, and if the lock target is not in the table or if the lock wait time elapses and the time-out occurs, the business logic unit 302 returns to the repeated process of step S807.
In step S810, the business logic unit 302 deletes the file record locked in step S809. Then, the business logic unit 302 returns to the repetitive processing in step S807.

In step S811, the business logic unit 302 acquires a list of child folders and child files in the same manner as in steps S801 and S806, and determines whether all of the child folders and child files have been deleted. The business logic unit 302 transitions to step S812 if there is no folder or file whose parent folder ID is the folder ID passed when the recursive process is called, and otherwise transitions to step S814.
In step S812, the business logic unit 302 deletes the record of the folder ID passed when the recursive process is called from the folder table.
In step S813, the business logic unit 302 performs a commit process on the database unit 304. The changes made in this recursive process are confirmed by this commit process.
The processing in step S814 is processing when the folder cannot be locked in step S805, or when a child folder or child file remains in the folder in step S811. In step S814, the business logic unit 302 determines whether or not one or more child folders or child files have been successfully deleted in this recursive process. If the business logic unit 302 deletes one or more child folders or child files, the business logic unit 302 proceeds to step S815, and otherwise proceeds to step S813.
The process in step S815 is a process when one or more child folders or files are deleted but the deletion target folder itself cannot be deleted, and the update date and time of the deletion target folder needs to be changed. In step S815, the business logic unit 302 registers, in the database unit 304, information for asynchronous processing for updating the update date and time of the deletion target folder, as in step S707. The update date and time of the deletion target folder is an example of attribute information of the deletion target folder.

The above is the folder deletion process in the present embodiment. By this processing, it is possible to realize a function of performing locking from the lowermost folder in a consistent manner, and not performing folder deletion in the case where locking cannot be performed redundantly with other processing. For example, in the example of FIGS. 4 and 5, “FolC” indicated by the record 408 is first deleted by the processing of FIGS. 7 and 8, and locks are performed in the order of “FolC” and “FileA”. Are deleted in the order of “File A” and “Fol C”, and the process is once determined. Then, “FolB” deletion processing is performed. However, when the lock of “FileA” times out, “FileA” and “FolC” are not deleted. Further, if the deletion process of “FolB” does not end within the time, the deletion process below “FolD” which should have been processed after that is not performed, and a response is sent to the caller.
In this way, when it takes a long time to wait for the lock, the deletion process is skipped, so that the storage server 110 can also prevent a timeout of communication with the document management Web server 120. Further, the deletion process itself of the child folder or the like is skipped depending on the elapsed time from the process start time, and a communication timeout can be prevented.
In addition, by using the deletion folder management table in FIG. 6 to temporarily reject the acceptance of a process for a child folder whose folder is being deleted, the probability of duplication of processes can be lowered.

Since the storage server 110 has the above folder deletion function, the document management Web server 120 only needs to call the storage server 110 using a simple HTTP protocol. Then, according to the response, it is possible to know whether or not the deletion was possible. As a result, the document management Web server 120 may determine whether to retry a plurality of times or to continue the deletion by inquiring with the PC terminal 130.
According to the present embodiment, it is possible to provide a folder deletion process in which a folder is deleted in a consistent manner up to the point where processing can be performed within the time-out time on the client side without using state management in the file management system. .

<Other embodiments>
The present invention supplies a program that realizes one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in a computer of the system or apparatus read and execute the program This process can be realized. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

As described above, according to each of the above-described embodiments, it is possible to provide a file management system capable of performing folder deletion processing while maintaining the consistency of the file management system up to the point where processing can be performed within a certain time. As a result, the document management Web server 120 can determine whether to perform the process again according to the folder deletion result, and can continue the process, thereby simplifying the process of the document management Web server 120.
Therefore, the folder deletion process can be performed without complicating the process on the side that calls the Web service.

Claims (8)

A server device that performs file management with a folder hierarchy,
Receiving means for receiving an instruction to delete the folder;
When an instruction to delete a folder is received by the receiving unit, if access to the lowermost folder of the folder that received the deletion instruction is restricted, and access restriction to the lowermost folder is successful, If the deletion of the lowest layer folder is successful and the deletion of the lowest layer folder is successful, the access to the next lowest layer folder is restricted and the access to the lowest layer folder is successfully restricted. If there is not, delete the lowermost folder, do not execute the deletion, repeat the process to the next lowest folder, the deletion means for executing the deletion process of the folder that received the deletion instruction,
Server device having   The deletion means determines a waiting time for access restriction when restricting access to the lowermost folder, and deletes the lowermost folder when access restriction is successful within the determined waiting time. If the access restriction to the lowest folder is not successful within the determined waiting time, the deletion of the lowest folder is not executed and the access to the next lowest folder is restricted. The server device according to claim 1, wherein the processing is repeated.   The server device according to claim 2, wherein the deletion unit determines the waiting time based on an elapsed time from a start time of a folder deletion process.   When deletion of a folder in a lower hierarchy of the folder that has received the deletion instruction is not successful and attribute information relating to the folder that has received the deletion instruction needs to be updated, processing for deleting the folder by the deletion unit 4. The server device according to claim 1, further comprising update means for updating attribute information relating to the folder that has received the delete instruction asynchronously.   The server apparatus according to claim 4, wherein the update unit updates the update date and time of the folder as attribute information related to the folder that has received the deletion instruction. It further has a recording means for recording information on the folder to be deleted,
The deletion means, when an instruction to delete a folder is received by the receiving means, confirms whether or not the information of the folder in the upper hierarchy of the folder that received the deletion instruction is recorded by the recording means, A response that the request source requesting the deletion instruction of the folder cannot be temporarily processed when information on a folder in a higher hierarchy of the folder that has received the deletion instruction is recorded by the recording unit. The server device according to any one of 1 to 5. An information processing method executed by a server device that performs file management by a folder hierarchical structure,
A receiving step for receiving an instruction to delete the folder;
When an instruction to delete a folder is received by the receiving step, if access to the lowermost folder of the folder that received the deletion instruction is restricted, and access to the lowermost folder is successfully restricted, If the deletion of the lowest layer folder is successful and the deletion of the lowest layer folder is successful, the access to the next lowest layer folder is restricted and the access to the lowest layer folder is successfully restricted. If not, deleting the lowermost folder is not executed, the process of repeating the process to the next lowest folder, and deleting the folder that received the deletion instruction,
An information processing method including: On a computer that manages files using a folder hierarchy,
A receiving step for receiving an instruction to delete the folder;
When an instruction to delete a folder is received by the receiving step, if access to the lowermost folder of the folder that received the deletion instruction is restricted, and access to the lowermost folder is successfully restricted, If the deletion of the lowest layer folder is successful and the deletion of the lowest layer folder is successful, the access to the next lowest layer folder is restricted and the access to the lowest layer folder is successfully restricted. If not, deleting the lowermost folder is not executed, the process of repeating the process to the next lowest folder, and deleting the folder that received the deletion instruction,
A program for running

Images