JP4521413B2 - Database management system and program - Google Patents

Description

  The present invention relates to a database management system that manages a database that stores a document (an electronic document) and an index used to search for the document, and particularly occurs when a document registered in the database is updated. The present invention relates to a database management system and program suitable for document re-registration processing.

  In recent years, for example, in an office, a database management system (DBMS) for registering and managing various documents (digitized documents) handled in the office in a database (DB) that can be searched for the purpose of information search is constructed. It is becoming. In such a system, when the content of a document registered in the database is updated, the updated document is re-registered in the database.

  Documents handled in the office, for example, documents created with word processing software of a personal computer (office documents), are updated in units of documents. For this reason, when a document registered in the database is updated, information registration in the database is generally re-registered in document units.

By the way, in order to enable, for example, full-text search of a document registered in the database, it is necessary to extract words included in the document and register the extracted words as an index in the database (for example, patent literature). 1 and 2). Therefore, when a document registered in the database is updated, the following series of operations (1) Delete the registered document (2) Delete the index of the deleted document (3) Full text of the re-registered document Extract words used as index for search (scan for re-registered documents)
(4) Index creation (5) Document registration is required.

In addition, when updating a part of a document, for example, when updating by designating only a specific structure of a structured document such as an XML document, only the part is updated (partial update). When a partial update occurs, only the changed part is re-registered, not the entire document. For this reason, when the partial update of the document increases, the registered document in the database is divided into a plurality of areas.
JP 2004-206473 A JP 2006-172268 A

  As described above, when a document registered in the database is updated, the re-registration of the document to the database is generally performed for each document, and the above five operations are necessary for the re-registration. It is. The most time-consuming process in the re-registration process is an index process (index process). The larger the document size, the longer the index process takes. That is, the larger the document size, the longer the re-registration process.

  The present invention has been made in consideration of the above circumstances, and its purpose is to perform registration in the database by narrowing down the data of locations where changes have occurred to the original registered document among the updated documents, An object of the present invention is to provide a database management system and program capable of reducing the cost of index creation and shortening the time required for re-registration processing.

  According to one aspect of the present invention, there is provided a database management system for managing a database in which a document and an index used for searching the document are stored. In this system, when an original document corresponding to a document to be registered in the database has already been registered in the database, a part of the document to be registered where a change has occurred to the original document is recorded. Difference extraction means for extracting data as difference data, index update means for updating the index of only the portion where the change has occurred, difference registration means for registering only the difference data extracted by the difference extraction means in the database Linking means for setting a link to the difference data registered in the database by the previous difference registration means in the original document.

  According to the present invention, the cost of index creation is reduced by registering in the database only the data of the part where the change has occurred in the original registered document among the updated documents, and it is necessary for the re-registration process. Time can be shortened.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a block diagram showing a hardware configuration of a client-server system according to an embodiment of the present invention. The client-server system mainly includes a database server (database server computer) 10 and a plurality of client terminals. The plurality of client terminals include a client terminal 20. On the client terminal 20, an application (application program) that uses the database server 10 operates. A plurality of client terminals including the client terminal 20 are connected to the database server 10 via a network 30 such as a local area network (LAN). In FIG. 1, client terminals other than the client terminal 20 are omitted.

  The database server 10 is connected to an external storage device 40 such as a hard disk drive. The external storage device 40 stores a database management program 41 and a database 42.

  The database management program 41 is used for management of the database 42 by the database server 10 and search processing based on a search request from a client terminal. The database 42 includes a document part 421 that stores a plurality of documents to be searched, and an index part 422 that stores an index (index data) used to search for a document stored in the document part 421. .

  In the present embodiment, the database management system 50 is realized by the database server 10. FIG. 2 is a block diagram mainly showing a functional configuration of the database management system 50. The database management system 50 includes a search unit 51, a difference extraction unit 52, an index processing unit 53, a registration unit 54, and a document search unit 55.

  When a document registration request is given from the client terminal 20 to the database management system 50 (database server 10), the search unit 51 searches for an already registered document corresponding to the requested document in the database 42 (inside the document unit 421). ) When the registered document exists in the database 42, that is, when the document is re-registered, the difference extraction unit 52 compares the requested document with the registered document, and calculates difference (changed part) data (difference data). ).

  The index processing unit 53 performs index creation / update processing for the index unit 422 in the database 42. The index processing unit 53 includes an index update unit 531 and an index creation unit 532. The index update unit 531 updates the index for the difference data extracted by the difference extraction unit 52. The index creating unit 532 creates an index of the new document in the index unit 422 of the database 42 when the already registered document corresponding to the requested document does not exist in the database 42, that is, when a new document is registered.

The registration unit 54 performs registration processing for the requested document. The registration unit 54 includes a difference registration unit 541, a linking unit 542, an original text registration unit 543, and a document registration unit 544.
The difference registration unit 541 registers the extracted difference data in the document unit 421 of the database 42. The linking unit 542 links the difference data registered in the document unit 421 and the document (original document) already registered in the document unit 421. In addition, when the requested document is registered in the document unit 421 by the original document registration unit 543 as the original document document when the document is re-registered, the link unit 542 adds the original document document to the original document linked with the difference data Link to.

  The original text registration unit 543 registers a document (requested document) from which difference data from the already registered document (original document) is extracted in the document unit 421 as an original text document. The document registration unit 544 registers a new document in the document unit 421.

  When a document search request is given from the client terminal 20 to the database management system 50 (database server 10), the document search unit 55 searches the index unit 422 in the database 42 for documents that match the requested search condition. Search using it. Note that the search unit 51 may have the function of the document search unit 55.

  In the present embodiment, the search unit 51, the difference extraction unit 52, the index processing unit 53, the registration unit 54, and the document search unit 55 use the database management program 41 stored in the external storage device 40 by the database server 10 of FIG. It is realized by reading into a memory (not shown) in the server 10 and executing it. The program 41 can be stored in advance in a computer-readable storage medium such as a compact disk or ROM and can be distributed. Further, this program 41 may be downloaded to the database server 10 via the network 30.

Next, the document registration process applied in the above embodiment will be described with reference to the flowchart of FIG.
First, it is assumed that document registration is requested from the client terminal 20 to the database management system 50 (database server 10) via the network 30. A request given from the client terminal 20 to the database management system 50 is received by a request management unit (not shown) in the system 50. As in this example, in the case of a document registration request, the request is passed to the search unit 51.

  When receiving the document registration request from the client terminal 20, the search unit 51 searches for a document registered in the document unit 421 in the database 42 (step S1). In general, a document registered in the document unit 421 is given an ID such as a file name, and a document having an ID that matches the requested document ID is searched. Based on the search result, the search unit 51 determines whether a document having the same ID as the requested document is already registered in the document unit 421 (step S2).

  If a document having the same ID as the requested document is not registered in the document unit 421, the search unit 51 determines that registration of a new document is requested and stores it in the index creation unit 532 in the index processing unit 53. Pass control. In response to this, the index creating unit 532 creates an index (for example, a word index) necessary for searching the document by performing a morphological analysis on the requested document (document to be registered), and the index unit of the database 42. Registered in 422 (step S3). The registered index includes a link (representing link information) to a registration target document including a word (vocabulary) used to create the index. Here, for simplification of explanation, a word index is created, but an N-gram index suitable for full-text search may be created.

  In step S3, the index creating unit 532 embeds (sets) a link (reverse link) from the requested document to the registered index. Then, the document registration unit 544 registers the requested document in which the reverse link to the registered index is embedded in the document unit 421 (step S4).

  An example of the state of the document part 421 and the index part 422 in the database 42 at this time is shown in FIG. In the example of FIG. 4, a document 400 having a content “high-speed search DB processing system” is registered in the document unit 421. Documents other than the document 400 are also registered in the document part 421, but are omitted in FIG. On the other hand, the index unit 422 includes an index relating to each word including the words “high speed”, “search”, “DB”, “processing”, and “system”, that is, a word including “high speed”, “search”, “DB”, “processing”, and “system”. For each (character string), an index indicating a document including the word is registered.

  Each of the indexes related to the words “high speed”, “search”, “DB”, “processing”, and “system” is a link from the index (a rectangular frame hatched in FIG. 4) to the document 400 including the word indicated by the index. (Link information representing) 401, 402, 403, 404, 405. In FIG. 4, links 401, 402, 403, 404, and 405 are represented by arrows, and a base end and a tip end of the arrows indicate a link source and a link destination, respectively.

  In FIG. 4, the tips of links 401, 402, 403, 404, and 405 (representing arrows) indicate the positions of the words “fast”, “search”, “DB”, “process”, and “system” in the document 400. Such an expression facilitates understanding that the links 401, 402, 403, 404, and 405 are set based on the words “fast”, “search”, “DB”, “process”, and “system” in the document 400. It has been applied for. The links 401, 402, 403, 404, and 405 are used to search the document 400 in a search process using the corresponding words “high speed”, “search”, “DB”, “process”, and “system” as keywords. Just pointing is enough. As described above, the links 401, 402, 403, 404, and 405 indicate the document 400. For convenience of explanation, the links 401, 402, 403, 404, and 405 are expressed as indicating the words “high speed”, “search”, “DB”, “process”, and “system” in the document 400. Sometimes.

  On the other hand, in the document 400, reverse links 411, 412, 413, 414, and 415 to the indexes related to the words “high speed”, “search”, “DB”, “processing”, and “system” included in the document 400 are indicated by arrows. It is represented. The base ends of the reverse links 411, 412, 413, 414, and 415 (representing arrows) are located in the word 400 “Fast”, “Search”, “DB”, “Process”, and “System” in the document 400. Such expressions are also set based on the words “high speed”, “search”, “DB”, “process”, and “system” in the document 400, as in the case of the previous links 401, 402, 403, 404, and 405. It has been applied to facilitate understanding. The reverse links 411, 412, 413, 414, and 415 are indexes created from the document 400 based on the words “high speed”, “search”, “DB”, “processing”, and “system” (links 401, 402, and 400). 403, 404, 405) can only be traced.

  In FIG. 4, among the indexes related to the words “high speed”, “DB”, “processing”, and “system”, from a rectangular frame not hatched to a document (a document other than the document 400) including the word indicated by the index. This link and its reverse link are omitted.

  Next, it is assumed that document registration is requested again from the client terminal 20 to the database management system 50 (database server 10). At this time, it is assumed that the document part 421 and the index part 422 in the database 42 are in the state shown in FIG. In addition, a request from the client terminal 20 is registered for a document with the content “high-speed search XML processing system” in which a document with the content “high-speed search DB processing system” (that is, the document 400) is updated in units of documents. Suppose it is a request.

  When receiving the document registration request from the client terminal 20, the search unit 51 of the database management system 50 searches for a document registered in the document unit 421 in the database 42 (step S1). Based on the search result, the search unit 51 determines whether a document having the same ID as the requested document is already registered in the document unit 421 (step S2).

  If a document having the same ID as that of the requested document is registered in the document unit 421, the search unit 51 assumes that registration (re-registration) of the updated document corresponding to the already registered document is requested. Control is passed to the extraction unit 52. In response to this, the difference extraction unit 52 extracts a registered document (that is, a document before update) from the document unit 421 (step S5). The difference extraction unit 52 generates difference data (difference data) between the updated document (re-registered document) requested from the client terminal 20 and the registered document extracted from the document unit 421, for example, in units of words (vocabulary). Extract (step S6).

  Here, the document (document 400) having the content “fast search database processing system” is updated to the document having the content “fast search XML processing system”. That is, “DB” in the “fast search DB processing system” is changed to “XML”. In this case, in the updated document document, data of a portion (changed portion) where the original registered document (original document) 400 has changed, that is, “XML” is stored in “DB” in the document 400. Extracted as corresponding difference data.

  The index update unit 531 in the index processing unit 53 deletes the index (that is, the link 403 to the document 400) indicating the word “DB” in the document 400 corresponding to the extracted difference data “XML” from the index unit 422. (Step S7). In step S7, the index update unit 531 deletes the reverse link 413 to the index of the word “DB” from the document 400 and adds specific information (deletion mark) indicating deletion to the word “DB” in the document 400. (Step S7). In step S <b> 7, the index update unit 531 further creates an index (index data) related to the extracted difference data (word after change) “XML” and adds it to the index unit 422. The created index has a link to a document (original document) 400 including a word (word before change) “DB” at a change location corresponding to difference data (word after change) “XML”. In this way, the index update unit 531 updates only the index data related to the changed portion from which the difference data is extracted by the difference extraction unit 52 (step S7).

  When executing the step S7, the index update unit 531 passes control to the difference registration unit 541 in the registration unit 54. Then, the difference registration unit 541 registers only the difference data, that is, the word “XML” in the document unit 421 (step S8).

  The linking unit 542 sends a link (document link) from the word “DB” marked with a deletion mark in the document 400 (original document) to the difference data “XML” registered by the difference registration unit 541. (Step S9). In step S <b> 9, the linking unit 542 embeds a reverse link to the index data related to the difference data “XML” in the document 400 linked by the document link with the difference data “XML” registered in the document unit 421.

  As described above, in this embodiment, even when updating is performed in document units, the original document (registered document) already registered in the document unit 421 and the updated document (document to be re-registered) are updated. The difference is extracted, and only the difference data is registered in the document unit 421, and the index is created / updated only for the changed portion from which the difference data is extracted. For this reason, the number of changes is limited compared to the prior art that required index processing for the entire updated document (document to be re-registered), and the time required for index processing is shortened, resulting in high-speed document registration processing. (Re-registration processing) can be realized. FIG. 4 shows an example of a very small document 400 for the sake of simplicity, but this effect is even greater for large documents.

  FIG. 5 shows the state of the document part 421 and the index part 422 immediately after the difference data “XML” is registered in the state shown in FIG. In the example of FIG. 5, a deletion mark indicated by a symbol x is attached to the word “DB” in the document (original document) 400, and a document link 431 is set from the word “DB” to the difference data “XML”. ing. The index for pointing to a document including the word “XML” includes a link 406 to a document (original document) 400 in which a document link 431 is set between the word “XML”. On the other hand, a reverse link 416 to the index related to the word “XML” is embedded in the document (original document) 400. Further, as apparent from comparison with FIG. 4, the link 403 and the reverse link 413 are deleted.

  Thereafter, as a result of further updating the document having the content “high-speed search XML processing system” (that is, the updated document 400) in units of documents, the client terminal 20 sends the “high-speed search XML processing system” to the database management system 50 (database server 10). It is assumed that a registration request for a document (update document) with the content “search XML management system” is given. In this case, the word “management” is extracted as difference data for “processing” in the document 400 (step S6). As a result, only “management” is registered in the document part 421, and only “management” and “processing” are respectively created and updated.

  FIG. 6 shows the state of the document part 421 and the index part 422 immediately after the difference data “management” is registered in the state shown in FIG. In the example of FIG. 6, a deletion mark “×” is added to the word “processing” in the document 400, and a document link 432 is set from the word “processing” to the difference data “management”. The index for pointing to a document including the word “management” includes a link 407 to the document 400 in which a document link 432 is set between the word “management”. On the other hand, a reverse link 417 to an index related to the word “management” is embedded in the document 400.

  Now, as shown in FIG. 6, there are document links between the words “DB” and “processing” with deletion marks in the document 400 and the words (difference data) “XML” and “management”, respectively. Assume that a search for a document including at least one of the words “fast”, “search”, “XML”, “management”, and “system” is requested in a state where 431 and 432 are set. This request is passed to the document search unit 55 by the request management unit. Then, the document search unit 55 searches the document unit 421 for a document including at least one of the words “fast”, “search”, “XML”, “management”, and “system” based on the index registered in the index unit 422. To do. By this search, a document group including the document 400 is searched.

  In document retrieval in a system such as a document management system, document units are often retrieved. Therefore, for example, when the document 400 is searched, the following document extraction process including character string extraction is performed.

  First, the character string “fast search” in the document 400 is extracted. Next, following the character string “fast search”, the document link 431 is traced from the word “DB” to which the deletion mark is attached, whereby the updated word “XML” for the word “DB” is extracted. Subsequently, the updated word “management” for the word “processing” is extracted by following the document link 432 from the word “processing” followed by the word “DB”. Then, the word “system” following the word “processing” is extracted. In other words, the updated document content “Fast Search XML Management System” is retrieved. In the document retrieval process described above, a process for removing the reverse link to the index set in the document 400 is performed.

  As is apparent from FIG. 6, when the registration of the difference data is repeated, the document links from the document (original document) 400 to the difference data increase. However, in a document search process in which a document having a large number of document links to difference data is extracted, there is a cost for extracting the difference data based on the document link when the document is extracted.

  Therefore, in the present embodiment, a mechanism is employed in which the contents of the updated document can be easily extracted even if the registration of difference data is repeated. Hereinafter, this mechanism will be described with reference to the above-described flowchart of FIG.

  When executing step S9, the linking unit 542 activates the original text registration unit 543. The original text registration unit 543 registers the document itself requested from the client terminal 20 in the document part 421 of the database 42 as an original text document (step S10). Then, the linking unit 542 sets a link (original document document link) from the original document linked with the difference data to the original document document in the original document (step S11). Thereby, the document registration process according to the flowchart of FIG. 3 ends.

  FIG. 7 shows the state of the document part 421 immediately after registration of the original document and setting of the link from the original document to the original document in the state shown in FIG. In the example of FIG. 7, when a registration request for a document having the content “fast search XML management system” is given and the document registration process according to the flowchart of FIG. 3 is executed up to step S11, the document is a source document. The document part 421 is registered as 430. In the original document 400, a link (original document link) 440 to the original document 430 (that is, an updated document corresponding to the original document 400) is set. In the present embodiment, when the document (original document) 400 is searched, the contents of the original document 430 can be extracted at high speed based on the link 440 as described below.

  Now, as a result of a request for searching for a document including at least one of the words “high speed”, “search”, “XML”, “management”, and “system”, the registration unit 54 searches for a document group including the document (original document) 400. It shall be assumed. Here, when the document (original document) 400 is searched, the document search unit 55 extracts the content of the original document 430 based on the link 440 set in the document (original document) 400.

  As described above, according to the present embodiment, the contents of the document reflecting the difference data, that is, the contents of the original document document can be extracted at high speed without following the link from the retrieved document (original document) to the difference data. Can do. In addition, unlike the original document, the reverse link to the index is not set in the original document document, so that it is not necessary to remove the reverse link when extracting the contents of the original document document, and the document can be extracted at a higher speed.

  Here, in the case of a document that is frequently updated (re-registered), the updated document is registered in the document unit 421 as an original document every time re-registration is requested. The storage area requires a large capacity. Therefore, it is preferable to compress the original document and register it in the document part 421.

[Modification]
Next, a modification of the above embodiment will be described.
The document registration process applied in the above embodiment is premised on the case where the document is updated in units of documents. However, the registration process similar to that in the above embodiment can also be applied when partial updating of a document is performed.

  In the case of partial update of the document, the difference extraction unit 52 uses the update data of the partial update portion as it is as difference data. For this reason, in this modification, the time which the difference extraction part 52 requires for the difference extraction is shortened compared with the said embodiment. However, in the case of partial update of a document, there is no re-registration document (update document) that can be registered in the document part 421 as an original document document unlike update in document units, so the original document registration unit 543 performs the following processing. There is a need to do.

  In the case of a partial update of a document, the original text registration unit 543 extracts from the document unit 421 an original text document linked with a document (original document) in which the partial update has occurred by an original text document link. The original text registration unit 543 performs an update process (partial update process) that reflects the content of the partial update (updated part) in the original text document extracted from the document part 421. In the case of the first partial update for a registered document, there is no original document document linked to the original document. In such a case, the original text registration unit 543 extracts the original document from the document unit 421, and reflects the updated portion in the extracted original document.

  The original text registration unit 543 registers a document (original text document or original document) in which the updated portion is reflected in the document part 421 as a new original text document. When the original document is searched, this new original document becomes a target for document retrieval.

  As described above, in the modification of the above embodiment, even in the case of partial update, a document that reflects the content (updated part) of the partial update is acquired using the registered original document or original document, and the update is performed. By registering the document in which the part is reflected in the document part 421 as the original document, it is possible to prevent the already registered document from being divided by the partial update. At the time of document retrieval, by extracting the original document document in which the updated portion is reflected, it is possible to speed up the document retrieval.

  Moreover, in the said embodiment, the detection of a difference (change location) is performed per word (vocabulary). However, in the case of a document having a structure such as an XML (Extensible Markup Language) document in which a hierarchical structure is described using tags (that is, a structured document), if a difference is detected in units of words, There are many changes, and the document linking process becomes complicated. Therefore, in consideration of the fact that the unit of update (change) in a structured document is often the unit of an element described using a tag, in the registration process for the structured document, the unit of the element is used. A configuration in which a difference is detected may be applied.

  FIG. 8 shows an example of an XML document that is a representative of a structured document. The XML document shown in FIG. 8 includes an element (name element) 81. The name element 81 includes a <name> tag and a character string (text) that is the content (value) of the element 81 “fast search DB processing system”. If the XML document shown in FIG. 8 is updated, registration of the XML document in which “DB” in the character string “Fast Search DB Processing System” included in the name element 81 is changed to “XML” is requested. When it is done, the difference extraction part 52 should just extract the name element 81 as a difference (change location).

  In addition, this invention is not limited to the said embodiment or its modification example as it is, A component can be deform | transformed and embodied in the range which does not deviate from the summary in an implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment or its modification. For example, you may delete a some component from all the components shown by embodiment or its modification.

The block diagram which shows the hardware constitutions of the client-server system which concerns on one Embodiment of this invention. The block diagram which mainly shows a function structure of the database management system shown by FIG. 6 is a flowchart showing a procedure of document registration processing applied in the embodiment. The figure which shows an example of the state of the document part and index part in a database. The figure which shows the state of a document part and an index part immediately after difference data "XML" was registered in the state shown in FIG. The figure which shows the state of a document part and an index part immediately after difference data "management" is registered in the state shown in FIG. The figure which shows the state of the document part immediately after registration of an original text document and the setting of the link from an original document to the said original text document were performed in the state shown in FIG. The figure which shows an example of the XML document by which difference detection is performed per element.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Database server, 20 ... Client terminal, 30 ... Network, 40 ... External storage device, 41 ... Database management program, 42 ... Database, 50 ... Database management system, 51 ... Search part, 52 ... Difference extraction part, 53 ... Index Processing unit 54... Registration unit 55 55 Document search unit 531 Index update unit 532 Index creation unit 541 Difference registration unit 542 Linking unit 543 Original document registration unit 544 Document registration unit

Claims (4)

In a database management system for managing a structured document in which a hierarchical structure is described using tags and a database in which an index used to search the structured document is stored,
If a structured document to be registered in the database and an original structured document corresponding to the structured document to be registered including the part where the change has occurred is already registered in the database, the registered document A difference extraction unit that extracts, as difference data, an element in which a change has occurred in the original structured document among the structured documents to be processed;
Index update means for updating only the index of the element where the change has occurred;
Difference registration means for registering only the difference data extracted by the difference extraction means in the database;
Source text registration means for registering the structured document to be registered in the database as a source text structured document;
The reverse link to the previous SL links and index on the difference data to the difference data registered in the database, with linking means for setting the original structured document in accordance with the registration in the database of the difference data there are, a link to the original structured document used to access the original structured document corresponding to the structured document of those said original when said original structured document is retrieved, the original structured Linking means for setting the original structured document according to the registration of the document in the database ,
When the partial extraction of the original structured document registered in the database or the original structured document linked to the original structured document occurs, the difference extraction unit Using the update data of the element including
The original text registration means takes out the original structured document or the original text structured document in which the partial update has occurred, and reflects the content of the partial update in the original structured document or the original text structured document. And register a new source text structured document reflecting the contents of the partial update in the database.
Database management system which is characterized a call.   The index update means deletes an index indicating the original structured document created based on an element corresponding to the extracted difference data in the original structured document from the database, The database management system according to claim 1, wherein an index indicating the structured document to be registered is created and registered in the database. A database server computer that manages a database storing a structured document in which a hierarchical structure is described using tags and an index used to search the structured document;
If a structured document to be registered in the database and an original structured document corresponding to the structured document to be registered including the part where the change has occurred is already registered in the database, the registered document A step of extracting, as difference data, an element in which a portion where a change has occurred with respect to the original structured document among the structured documents to be performed;
Updating the index of only the element where the location where the change occurred,
Registering only the extracted difference data in the database;
Registering the structured document to be registered in the database as a source text structured document;
Setting a reverse link to a link and the index on the difference data to the difference data registered in the database, the original structured document in accordance with the registration in the database of the difference data,
A link to the original structured document used to access the original structured document said registered in the database corresponding to the original structured document when the structured document before Kimoto is retrieved, Setting the original structured document in the original structured document in response to registration in the database ;
When a partial update occurs to the original structured document registered in the database or a source text structured document linked to the original structured document, update data of an element including a portion where the partial update occurs Directly using the difference data as the difference data;
The original structured document or the original text structured document in which the partial update has occurred is taken out, and the contents of the partial update are reflected in the original structured document or the original text structured document. A program for executing a step of registering a new original text structured document reflecting the contents in the database . In the updating step, an index indicating the original structured document created based on an element corresponding to the extracted difference data in the original structured document is deleted from the database, and the difference data 4. The program according to claim 3, wherein an index pointing to the structured document to be registered is created.

Images