JP4547954B2 - Server device - Google Patents

Description

  The present invention relates to a server device that provides a document to a client side device such as a web server.

  2. Description of the Related Art Conventionally, when a single document is generated by a large number of users, a program that holds a history of revisions made by each user is known (Non-Patent Document 1).

Note that Non-Patent Document 2 describes the Lisp language that manages list data using hash values, and Patent Document 1 describes data management that uses hash values.
Exmedia, "Super Illustrated (R) Word 2000 for Windows (R) Application", first edition, August 10, 2002, pages 152 to 153 E. Goto, "Monocopy and Associate Algorithms in an Extended Lisp", Report TR 74-3, May 1974, University of Tokyo U.S. Pat. No. 5,977,792

  However, in the above conventional programs and the like, the user's revision operation is constantly monitored and the contents of the revision operation are sequentially recorded. For example, when the document held on the server device is revised, the revision operation is performed. In order to be able to provide a document with revision history, it is necessary to generate a revision operation history on the client side and upload it to the server device side along with the document. There is a problem that the environment is limited to an environment that can generate revision history.

  The present invention has been made in view of the above circumstances, and provides a server device that can provide a document with a revision history without generating a history of revision operations (change operations for documents) on the user side. One of the purposes.

  The present invention for solving the problems of the above-described conventional example is a server device that is connected to a database that holds at least one document, and is for a pre-change document to be processed among the documents. Receiving the changed document obtained by performing the changing operation, difference information generating means for generating difference information between the pre-change document and the changed document, and the change based on the difference information Dividing means for dividing the subsequent document into data fragments for each type of change operation from the pre-change document, and based on the information on the difference, the content of the data fragment obtained by the division is the same as above For data fragments included in the pre-change document, the data fragment Is characterized by comprising registration means for registering the data fragments in the database in association with information identifying at least a portion of the pre-change documents that contain portions corresponding to the instrument, the.

  As described above, since the content of the change operation is specified from the difference information, it is possible to provide a document with a revision history without requiring the user to generate a history of the change operation for the document.

  In addition, the registration unit is related to the additional change operation for the data fragment that is not included in the pre-change document among the data fragments obtained by the dividing unit and is added to the post-change document. The data fragment may be registered in the database in association with the operation relation information.

  As a result, a change operation such as adding a new description to the document is specified, and a document with a revision history can be provided without the user having to generate a history of the change operation for the document.

  Further, the registration means includes, among the data fragments obtained by the dividing means, the contents included in the pre-change document, and for data fragments deleted in the post-change document, operation relation information related to the change operation of the deletion The data fragment may be registered in the database in association with.

  Thus, a change operation such as deleting a part of the document is specified, and a document with a revision history can be provided without requiring the user to generate a history of the change operation on the document.

  Here, the operation-related information may include information for specifying a user who has performed the change operation and information regarding the date and time when the change operation was performed.

  As a result, it is possible to provide information specifying the user who performed the change operation and the date and time.

  The data processing apparatus further includes a providing unit that provides the client apparatus with the data fragment obtained by dividing the document, and the providing unit performs a change operation related to the data fragment for each data fragment obtained by dividing the document. Different types of information may be associated with each other and provided to the client device side.

  Thereby, a document with a revision history can be provided.

  The present invention for solving the problems of the conventional example is a method for controlling a server device connected to a database holding at least one document so as to be processed among the documents. A step of accepting a post-change document obtained by performing a change operation on the pre-change document, a step of generating difference information between the pre-change document and the post-change document, and based on the difference information , Dividing the post-change document into data fragments for each type of change operation from the pre-change document, and based on the difference information, the content of the data fragments obtained by the division is For the data fragment contained in the document before change, It is characterized by comprising a step of registering the data fragments in the database in association with information identifying at least a portion of the pre-change documents that contain portions corresponding to data fragments,.

  As described above, since the content of the change operation is specified from the difference information, it is possible to provide a document with a revision history without requiring the user to generate a history of the change operation for the document.

  Furthermore, the present invention for solving the problems of the above-described conventional example is a program for controlling a server device that is connected to a database holding at least one document, and is a processing target among the documents. Based on the procedure for accepting the changed document obtained by performing the change operation on the pre-change document, the procedure for generating the difference information between the pre-change document and the changed document, and the difference information, The procedure of dividing the post-change document into data fragments for each type of change operation from the pre-change document, and the content of the data fragments obtained by the division based on the information of the difference is the same as the above For data fragments included in the pre-change document, the data It is characterized in that to perform the procedure for registering the data fragments in the database in association with information identifying at least a portion of the pre-change documents that contain portions corresponding to the fragment, to the server device.

  As described above, since the content of the change operation is specified from the difference information, it is possible to provide a document with a revision history without requiring the user to generate a history of the change operation for the document.

  Embodiments of the present invention will be described with reference to the drawings. The server device according to the present embodiment can be implemented as, for example, the web server 1, and includes a control unit 11, a storage unit 12, and a communication unit 13 as shown in FIG. And connected via a network. The database server 2 may be directly connected without going through a network.

  The control unit 11 of the web server 1 operates in accordance with a program stored in the storage unit 12 and basically responds to a request received from the client device 3 via the network, and a document related to the request. Is transmitted to the requesting client apparatus 3. The control unit 11 sets the provided document as a pre-change document, accepts the post-change document that has been subjected to the change process, and associates the pre-change document with the post-change document (history management). Process). Further, in the present embodiment, when providing the document, the control unit 11 performs a process (information providing process) for providing information related to the pre-change document associated with the document. These history management processing and information provision processing will be described in detail later.

  The storage unit 12 includes a hard disk device that stores a program executed by the control unit 11 and a memory device that stores data generated during the execution of the control unit 11. The communication unit 13 transmits data via a network in accordance with an instruction from the control unit 11. Further, the communication unit 13 receives data coming via the network and outputs the data to the control unit 11.

  In the present embodiment, the database server 2 holds a document provided by the web server 1. The document stored in the database server 2 is expressed as a sequence of fragments. For example, if a document contains the string "This is an example of a document," this string is split into two fragments: a fragment "This is a document" and a fragment "An example." The database server 2 stores fragment data including a character string of each fragment and a document expressed as document data d = [f1, f2] that defines the fragment string. Here, f1, f2, etc. are pointers for specifying the fragments. The database server 2 stores operation relation information including user name and time information as header data.

  Here, the fragment data stored in the database server 2 can further display a pointer for specifying header data including time information of the user and the operation time related to the operation of generating and deleting each fragment data, and the fragment can be displayed. It is associated with a flag (visible flag) indicating whether or not the current state is in a state. In the present embodiment, these document data, fragment data, and header data are managed in a data format as shown in FIG.

  This data format includes a data portion (D) including any one of document data, fragment data, and header data, and a pointer portion (P). Here, the database server 2 calculates a feature value such as a hash value based on each data represented in this data format and stores it in association with each data (FIG. 3), and the hash value as this feature value is a pointer. In the following, a case where the pointer is a hash value will be described as an example. In addition, when using a hash value, it is set as the data length (for example, about 20 bytes) which can identify each data mutually.

  That is, conceptually, as shown in FIG. 4, it is assumed that the document d1 is obtained by associating the document data [f1, f2] with the hash value h, and the hash value is “d1”. The fragment corresponding to the hash value f1 is obtained by associating the character string “This is a document” as fragment data with the hash value h, and the fragment corresponding to the hash value f2 is the character string “ It is an example "and the hash value h is associated. Initially, the visibility flag corresponding to these fragment data is set to a value (for example, “True”) indicating that “display is possible”.

  Further, the operation relation information corresponding to the hash value h includes the user name “aaaa” and the time “t” at which the operation was performed, and this operation relation information has a “nothing” state as a pointer. The “nil” that represents is associated. In addition, in the example here, when a document “This is an example of a document” is newly registered, the data associated with the hash value h may be generated using the entire character string as fragment data. (In this case, the document data is a hash value [f] for the fragment). Here, for convenience of the following description, a case where the fragment is divided into two is illustrated.

  The client device 3 may be a general personal computer. A browser program as a web client is installed in this personal computer.

  Here, the history management process and the information provision process executed by the control unit 11 will be described. The program for these processes executed by the control unit 11 may be implemented as a program that is activated via a CGI (Common Gateway Interface), for example.

  First, the history management process will be described. The control unit 11 instructs the client device 3 to provide the document to be changed (information for specifying the storage location of the target document (URL (Uniform Resource Locator) or the like). Is received), a web page for inputting the name of the user who performs the instruction is distributed. When the user name is received here, the control unit 11 takes out the document subjected to the change operation from the database server 2 and distributes it as an HTML (Hyper Text Markup Language) document. In addition, it is good also as authenticating using a user name here.

  Specifically, when the document conceptually shown in FIG. 4 (has a hash value of d1) is the target of the change operation, the control unit 11 reads the document from the database server 2. As described above, this document includes a set of hash values for fragments as document data. The control unit 11 sequentially reads out the hash values included in the set, acquires fragments corresponding to the hash values from the database server 2, concatenates them, and sends them to the client device 3 in an editable state. Send.

  That is, in this example, since the hash values f1 and f2 to the fragment are included as the document data, the control unit 11 sequentially acquires the fragments corresponding to the hash value from the database server 2. Here, since the character strings “This is a document” and “an example” are included in the fragments corresponding to the hash values f1 and f2, these are concatenated to obtain “this is an example of a document”. Then, in order to make this character string editable, the control unit 11 sets this character string as a default character string in, for example, an HTML text area (TEXTAREA) tag, and generates an HTML document including the HTML description part. Then, it is transmitted to the client device 3.

  Further, when the change result of the transmitted character string is transmitted from the client device 3, the control unit 11 changes the change result obtained by concatenating the change result (the document after change) and the character string included in the fragment data. Compare with the previous document to generate difference information.

  Here, the difference information may be generated using, for example, a diff command widely used in UNIX (registered trademark). However, in the case of a general diff command, the difference is in line units. There are known diff commands for generating differences in word units (states separated for each word using morphological analysis processing) and diff commands for generating differences in character units. May be.

  For example, when using diff to generate character-by-character differences, if you change “This is an example of a document” as shown above to “This is an example of a document”, the output result of the difference is A state as shown in FIG. 5 can be obtained. Of the changed parts, “+” is added in front of “La” newly added in the changed document, and “−” is added in front of “One” deleted in the changed document. "Symbol.

  The control unit 11 sets each line of the output as a new fragment. That is, a new fragment is determined by dividing into (1) an added part, (2) a deleted part, and (3) an unaltered part, each of which is a continuous part corresponding to a different change operation. Here, it is separated into five fragments of “this”, “ra”, “is a document”, “one”, and “example”. For a fragment that has not been changed, the hash value of the fragment before the change is associated. Here, “this” and “is a document” correspond to “these are a document” (hash value: f1) before the change, and therefore, fragment data in which the character string “this” is associated with the hash value f1, Fragment data in which the character string “has a document” and hash value f1 are associated with each other is generated, hash values for these fragment data are calculated (referred to as f′1 and f′3), and associated with each hash value. The fragment data is registered in the database server 2. That is, the data fragment part without change is registered in the database server 2 in association with information for identifying at least a part of the pre-change document including the part corresponding to the data fragment.

  It should be noted that the information regarding the change operation may be further associated with the fragment relating to the additionally written / deleted data. For example, a “+” symbol may be added before “ra”.

  In addition, when there is a changed part such as “R”, the name of the user who made the change (the name of the user who was first acquired) and the date and time information when the changed document was received are combined. Operation-related information is generated, “nil” is associated with this to make header data, a hash value of this header data is further calculated (here, h2), and the hash value h2 and header data are associated with each other in the database. Register with server 2.

  On the other hand, the control unit 11 associates the character string “ra” with the hash value (h2) related to the operation relation information including the name of the user who changed the character string, and adds new fragment data for the additional portion. The hash value (f′2) of the fragment data is calculated, and the fragment data and the hash value are associated with each other and registered in the database server 2. It is to be noted that the character string “1” which is the deleted portion is also associated with the hash value (h2) related to the operation relation information including the user name who has changed the character string to obtain new fragment data, and the hash of this fragment data A value (referred to as f′4) is calculated, and the fragment data and the hash value are associated with each other and registered in the database server 2. For the deleted fragment data, the visibility flag is set to a value (“false”) indicating that it cannot be displayed.

  In addition, since there is no change in the “example” part, fragment data is generated in association with the hash value f2 of the original fragment, and a hash value (referred to as f′5) is calculated for this fragment data. The fragment data and the hash value are associated with each other and registered in the database server 2.

  The control unit 11 generates document data in which these f′1 to f′5 are arranged, and generates data in which the hash value d1 for the original document data is related to the document data. The hash value for this data is calculated (denoted as d2), and the calculated hash value and data are associated with each other and registered in the database server 2.

  Accordingly, the relationship of data corresponding to FIG. 4 is as shown in FIG. 6 (note that the name of the user who made the change is “bbbb” and the date and time related to the change is “t ′”).

  Next, the information providing process will be described. The control unit 11 receives the input of the hash value related to the document as an instruction to request the document from the client device 3, and operates as follows. Data related to the input hash value is acquired from the database server 2, a hash value of a fragment as document data included in the data is acquired, and fragment data corresponding to the hash value of each fragment is further acquired from the database server 2. Acquired, concatenated character strings included in each piece of fragment data, and provides them to the client device 3.

  Specifically, when the data shown in FIG. 4 is stored in the database server 2, when a request for a document having the hash value d1 is accepted from the client device 3, the data corresponding to the hash value d1 is received from the database server 2. Read out. Here, since the data includes hash values f1 and f2 for the fragments as document data, the control unit 11 sequentially acquires fragments corresponding to the hash values from the database server 2. Here, since the character strings “This is a document” and “an example” are included in the fragments corresponding to the hash values f1 and f2, these are concatenated to obtain “this is an example of a document”. Then, the control unit 11 generates an HTML document including this character string and transmits it to the client device 3.

  At this time, the control unit 11 reads the hash value h included in each data (here, fragment data), obtains the header data associated with the hash value h from the database server 2, and the header User name and date / time information included in the data may be included in the HTML document. In this case, for example, when the mouse cursor is overlaid on the character string portion corresponding to each fragment data on the client device 3 side, a special tag (SPAN Tag).

  That is, the control unit 11 repeats the process of reading the data associated with the hash value from the database server 2 until the hash value included in the acquired data becomes nil, and when the hash value becomes nil, the hash value By reading the information associated with nil, it is possible to acquire operation related information (information included in the header data) such as the user name and date.

  Further, when the data as shown in FIG. 6 is stored in the database server 2 and a request for a document having the hash value d2 is accepted from the client device 3, the data corresponding to the hash value d2 is received from the database server 2. Read out. Here, since the data includes the hash values f′1 to f′5 as the document data, the control unit 11 sequentially obtains the fragments corresponding to these hash values from the database server 2. Here, the strings corresponding to the hash values f′1 to f′5 include the character strings “this”, “ra”, “is a document”, “one”, and “is an example”. This is an example of a document. Then, the control unit 11 generates an HTML document including this character string and transmits it to the client device 3.

  One characteristic of the present embodiment is that the control unit 11 generates this HTML document by associating different information for each type of change operation regarding each fragment. Specifically, in the present embodiment, the control unit 11 traces d2 → d1 → h to extract header data corresponding to the hash value d2 of the document, and the user name aaaa that is the original creator of the document, and The date and time t is acquired.

  On the other hand, for the fragment corresponding to the hash value f′1, in order to obtain the header data related to the character string “this”, the associated hash value f1 is read, and the data corresponding to the hash value f1 is read from the database server 2 ( Fragment data) is read, and the hash value included in it is h instead of nil, so further data (header data) corresponding to the hash value h is obtained, the user name “aaaa”, etc. is obtained, For example, it is related as the title attribute of the SPAN tag.

  The fragment corresponding to the hash value f′2 is a character string “R” added, and the hash value h2 included in the fragment includes the user name “bbbb” as header data. Since it is associated, the user name “bbbb” or the like is acquired, and the character string “ra” is highlighted (for example, bold, underline) using the INS tag etc. As the attribute.

  Further, for the fragment corresponding to the hash value f′4, since the visible flag associated with the character string “one” is set to a value indicating “not displayable”, the character string “ “1” is highlighted by the second method (for example, the display is canceled by a strike line). Further, since the hash value h2 included in the fragment corresponding to the hash value f'4 is associated with the header data including the user name "bbbb", the user name "bbbb" is acquired. The user name is associated as the attribute.

  The generated HTML document is as shown in FIG. 7, for example. When the client device 3 that has received the HTML document shown in FIG. 7 displays the HTML document, it depends on the function of the browser. In general, the state as shown in FIG. 8 is obtained, and an underline is drawn in the additional writing portion, and an “X” mark is superimposed on the deletion portion or a strike-through line is drawn. Also, depending on the browser function, if the user moves the mouse cursor over a part of these strings that is surrounded by SPAN tags, such as “this”, “is a document”, or “example”, “aaa, t ”and the like, information on the associated user name and modification date are provided. Similarly, when the mouse cursor is over the added part or the deleted part, information associated with the CITE or DATETIME attribute is provided, such as “bbbb, t ′”.

  As described above, according to the present embodiment, it is possible to provide a document with a change history without the user having to generate a history of change operations for the document.

  In the description of the present embodiment, the user name and the date / time related to the change are included as information to be included in the header data. However, the present invention is not limited to this, and other information such as a changer's comment may be included. Although the web server 1 and the database server 2 are described as separate devices here, they may be integrated devices. Furthermore, although the web server has been described here as an example, the server device of the present invention may be another server device, such as a general file server.

  The processing executed by the control unit 11 described in the above embodiment will be described specifically by dividing it into already-read determination processing, document update processing, and document display processing. First, the read determination process will be described. In the already-read determination process, it is determined whether or not the user has viewed the target fragment f in the past. That is, the control unit 11 acquires an array F of fragments that the user has viewed in the past (registered as a cookie for each user, for example, in a client device on the user side), and starts the processing illustrated in FIG. Then, it is checked whether or not the fragment of interest f is included in the fragment sequence F (S1). If the fragment of interest f is included in the fragment array F (if Yes), the value “true” representing “read” is output and the process is terminated (S2).

  If the target fragment f is not included in the fragment array F (if No), then it is checked whether or not there is a hash value (reference) associated with the target fragment f (S3). If not (No), a value “false” indicating “not read” is output and the process is terminated (S4).

  On the other hand, if there is a reference in the process S3 (if Yes), the fragment specified by the reference is set as a new target fragment f (replace f with a reference; S5), and the process returns to the process S1 to continue the process.

  Further, when the document is updated, the control unit 11 starts the processing shown in FIG. 10, and in the original document, dod is the original document, hnew is the information indicating the updater and update date, and tnew is the updated text. The fragment set to be visible is taken out, and the text in the taken out fragment is concatenated to generate the text tod before change (S11).

  Then, a difference array D between "told" and "tnew" is generated (S12). Here, the elements of the difference array D include (1) an added part, (2) a deleted part, and (3) an information element that specifies an unaltered part. "] Is added to generate data such as [+, text], and deleted parts are marked with"-"to generate data such as [-, text]. For the part that is not, the data such as [nil, text] is generated with nil without adding a sign.

  Next, the control unit 11 obtains a fragment array Fold constituting the original document dold (S13). This can be acquired from the database server 2. Then, a common subdivision Fnew between Fold and the difference array D is generated (S14). Here, the common subdivision is as follows. That is, the original document “dold” is “this is an example of a document”, and the fragment array includes two fragments of a fragment f1 “this is” and a fragment f2 “an example of a document”. (Fold = [f1, f2]).

Here, when the text tnew after the change is “these are examples of documents”, the difference array D between the text “told” before the change “this is an example of documents” is
[nil, this]
[+, Ra]
[nil, is the document]
[-,one]
[nil, example]
These five elements are included.

Here, when the minimum division including all the division positions of the original fragment array Fold and the difference array D is generated as a common subdivision, in Fold, “this is” and “document” are divided. So, including the division at that position,
[nil, this]
[+, Ra]
[nil]
[nil, document]
[-,one]
[nil, example]
The six elements are included.

In addition, when the original fragment array Fold is divided into “this is a document” and “an example”, the common subdivision with the difference array D is
[nil, this]
[+, Ra]
[nil, is the document]
[-,one]
[nil, example]
These five elements are included.

  The control unit 11 generates the document dnew including the common subdivision Fnew thus generated in the data unit and the hash value (reference) of the original document in the reference unit (S15), and ends the process.

  Further, the control unit 11 starts the process shown in FIG. 11 as the document display process. Here, the document dold that the user has already viewed (read) is found from the specified document dnew (from the information registered as a cookie or the like on the user side while following the reference in order from dnew) Find the documents you ’ve viewed).

  Then, the fragment array Fold is extracted from dold, and the fragment array Fnew is extracted from dnew (S21). Then, it is checked whether there is an unprocessed fragment in Fnew (S22). Initially all fragments are unprocessed. In this process S22, if there is no unprocessed fragment (if No), the process is terminated. If there is an unprocessed fragment (if Yes), one of the unprocessed fragments is taken out as the target fragment f (S23). Then, the header h of the target fragment f is extracted (S24). Next, it is determined whether the target fragment f is included in Fold (an array of fragments browsed by the user in the past) and has already been browsed (S25). In the determination in step S25, the already-read determination process shown in FIG. 9 is performed. In the process S25, when it is determined that the target fragment f has been read (already viewed) with respect to Fold (in the case of Yes), it is determined whether or not the target fragment f is visible (S26). If it is not visible (if No), the process proceeds to step S22 to continue the process.

  On the other hand, if the fragment of interest f is visible in processing S26 (if Yes), the header h is added to the text of the fragment of interest f and output (provided as an attribute of the span tag surrounding the text) (S27). The process proceeds to S22.

  If it is determined in step S25 that the target fragment f has not been read with respect to Fold (has not been browsed yet) (No), it is next determined whether the target fragment f is visible. Judgment is made (S28), if it is not visible (if No), a header h is given to the text of the fragment of interest f, emphasis modification is performed to indicate that it has been deleted, and output (a del tag surrounding the text is set) (Attributes and header information are added) (S29), and the process proceeds to S22.

  On the other hand, if the fragment of interest f is visible (if Yes) in step S28, a header h is added to the text of the fragment of interest f, and emphasis modification is performed to indicate that it has been added (the ins tag surrounding the text is output) Set and assign header information as the attribute) (S30), and the process proceeds to S22.

  As a result, the document can be presented in a state as shown in FIG. 8 according to the change status since the user referred to it in the past.

It is a block diagram showing the structure of the server apparatus which concerns on embodiment of this invention, and its network connection example. It is explanatory drawing showing an example of the data structure managed within a database server. It is explanatory drawing showing an example of the table which a database server manages. It is explanatory drawing showing the example of the relationship between the data managed within a database server. It is explanatory drawing showing the example which produced | generated the difference of the document. It is explanatory drawing showing another example of the relationship between the data managed within a database server. It is explanatory drawing showing the example of a part of HTML document which a server apparatus produces | generates. It is explanatory drawing showing the example which displayed the HTML document provided from a server apparatus. It is a flowchart figure showing the example of the program which concerns on the Example of this invention. It is a flowchart figure showing the example of the program which concerns on the Example of this invention. It is a flowchart figure showing the example of the program which concerns on the Example of this invention.

Explanation of symbols

1 web server, 2 database server, 3 client device, 11 control unit, 12 storage unit, 13 communication unit.

Claims (7)

A server device connected to a database that holds at least one document,
An accepting unit for accepting a post-change document obtained by performing a change operation on a pre-change document to be processed among the documents;
A registration unit that divides the pre-change document into a plurality of pre-change character strings, and registers a fragment in which the hash value of the operation relation information is associated with each pre-change character string in the database as the pre-change fragment;
Dividing means for dividing the post- change document into post-change character strings for each type of change operation from the pre-change document;
Including
Said registration means, out of the after-string, the changes after a character string, the content of which is included as is to the pre-change document, hash of the before-fragment containing the portion corresponding to the character string after the change A server device , wherein a fragment associated with a value is generated as a modified fragment, and the modified fragment is registered in the database. The server device according to claim 1,
The registration means includes the additional character string added to the post-change document, the content of which is not included in the pre-change document among the post-change character strings obtained by the dividing means. A server device , wherein a fragment associated with a hash value of operation related information related to a change operation is generated as a post-change fragment, and the generated post-change fragment is registered in the database. The server device according to claim 1 or 2, wherein
Said registration means, out of the pre-change string, the content of which is included in the pre-change documents for deleted before-string after change document, hash value of the operation related information related to the deletion of the change operation the association was fragments, generated as post-change fragment, the server apparatus characterized by registering the changes after fragment thus generated in the database. The server device according to claim 2 or 3, wherein
The server apparatus according to claim 1, wherein the operation-related information includes at least one of information for identifying a user who has performed the changing operation and information regarding a date and time when the changing operation is performed. The server device according to any one of claims 1 to 4, wherein
Providing means for providing a fragment obtained by dividing the document to the client device side;
The server device, wherein the providing means associates different information for each type of change operation related to the fragment for each fragment obtained by dividing the document, and provides the information to the client device side.   The server device according to any one of claims 1 to 5,
  The pre-change document is expressed as a set of hash values of the pre-change document, and is registered in the database in association with a hash value of operation relation information indicating a user who has operated the pre-change document. A server device as a feature. A computer connected to a database holding at least one document,
An accepting unit for accepting a post-change document obtained by performing a change operation on a pre-change document to be processed among the documents;
A registration unit that divides the pre-change document into a plurality of pre-change character strings, and registers a fragment in which the hash value of the operation relation information is associated with each pre-change character string in the database as the pre-change fragment;
Dividing means for dividing the post- change document into post-change character strings for each type of change operation from the pre-change document;
Function as
Examples registration means, out of the after-string, the changes after a character string, the content of which is included as is to the pre-change document hash of the before-fragment containing the portion corresponding to the character string after the change A program that generates a fragment associated with a value as a modified fragment and registers the modified fragment in the database.

Images