JP2000099512A - Document managing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To permit a user to easily describe an application depending conversion rule for extracting partial structure required for an application, to selectively extract only partial structure required for the application and to hold/manage it through the use of the application depending conversion rule and a required system incorporated conversion rule. SOLUTION: An application depending conversion rule for extracting partial structure required at every application by a system incorporated conversion rule and the combination of the system incorporated conversion rules for extracting partial structure from a unified inner format document held in a conversion rule holding part 12 is held. When the registered document is received from a document registration part 11, a document format averaging part 13 converts the document into an inner format. A partial structure extraction part 14 applies the system incorporated conversion rule held by the conversion rule holding part 12 to the document converted into the inner format and the corresponding application depending conversion rule is applied so as to extract partial structure. Then, a document data bolding part 16 relates partial structure with the document and the inner format document so as to hold them.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a document management system for storing documents of various formats and document types and extracting a necessary partial structure from the document in response to a user request.

[0002]

2. Description of the Related Art A basic function of a document management system is to store a large amount of electronic documents and return document data in response to a request from a user or an application. SGML
(Standard Generalized Markup Language; ISO8879 / 19
86; JIS X 4151/1992), a document management system that manages structured documents can access even smaller granularities, that is, in units of document elements. Structured documents are usually created according to the definition of a structure called a document type, and if this document type is appropriately defined according to the purpose of use, documents created according to the document type are effectively reused. Documents that do not follow this document type cannot be reused. Also, this document type cannot be used for different purposes.

Also, a document created by a word processor or the like, which is not a structured document (hereinafter, referred to as a format-oriented document)
Even if the format is made public, there is no structure definition corresponding to the document type of the structured document, and in most cases, font size and type are assigned to each element. And even if there is a structural definition, it is defined only as a very commonly used element regardless of the application, such as a chapter, section, paragraph, or bullet point. Therefore, for format-oriented documents,
The structure could not be used positively, and many of them enabled full-text search only for the text part.

In response to a request to treat a format-oriented document as a document having a logical structure, conventionally, format-oriented document data has been converted into a structured document based on a certain rule and can be used. For example, JP-A-8-1905
No. 42, "Document Database Management Apparatus and Method" holds a rule for generating a logical structure in a database from a logical structure of documents of various formats to be stored in the database, Based on this rule, the logical structure is converted when storing a document.
As a result, a format-oriented document having only a simple logical structure can be stored in a database as a result of conversion into a document having a logical structure according to the purpose of use, and can be reused. Also, when storing a structured document in a database, since it can be converted to another logical structure according to the purpose of use and stored, a document having various different document types has a unified logical structure. It is possible to treat as.

[0005]

However, in the above-described conventional method, a conversion rule or a conversion program depending on the purpose of use is converted into a format for a format-oriented document, and a document type for a structured document. You must be familiar with it and create it for each format or document type. For this reason, the greater the number of types of formats or document types, the greater the human cost. Also, when building a new application, a different purpose of use arises. Therefore, a conversion program from all document formats or document types to a structure to be newly used must have a structure to be used by the application. I had to create it.

The present invention has been made in view of the above-described conventional circumstances, and selectively uses application-dependent conversion rules and necessary system-incorporated conversion rules to selectively select only a partial structure required for an application. It is an object of the present invention to provide a document management system that can extract, hold, and manage. Further, according to the present invention, a user can easily describe an application-dependent conversion rule for extracting a partial structure required for an application without being aware of a difference in structure between different formats and document types of a document. The purpose is to provide a document management system.

[0007]

In a document management system according to the present invention, a conversion rule holding means includes a system built-in conversion rule for extracting a partial structure from a uniform internal format of a document, and a system built-in conversion rule. An application-dependent conversion rule for extracting a partial structure required for each application by a combination of conversion rules is held. When a document to be registered in the document management system is received from the document registration unit, the document format homogenization unit receives the document. To a unified internal format. When an application that uses a partial structure such as a table of contents or a heading is activated from the document, the partial structure extracting unit stores the document in the internal format according to the application in the conversion rule holding unit. The document data holding unit applies the built-in system-based conversion rule and further applies the corresponding application-dependent conversion rule to extract the partial structure, and the document data holding unit converts the partial structure into a document received by the document registration unit and a document in the internal format. Relate and retain. Therefore,
In accordance with the use in various applications, the partial structure required for each application is retained and managed using application-dependent conversion rules and the necessary system-incorporated conversion rules, and the use of the partial structure required for each application is maintained. Can respond quickly.

Here, as a preferred embodiment of the present invention, the document registration means receives the format name or document type together with the document, and the partial structure extraction means incorporates a system incorporated from the format name or document type received by the document registration means. A type conversion rule is specified, the specified system built-in type conversion rule is applied to the document converted into the internal format, and a corresponding application-dependent conversion rule is applied to extract a partial structure. The extraction processing of the partial structure is performed by expanding the condition description of the application-dependent conversion rule based on the condition description of the system-incorporated conversion rule referred to by the application-dependent conversion rule.

In a preferred embodiment of the present invention, the conversion rule holding means holds a definition of a partial structure that can be extracted by a system built-in type conversion rule, and converts the definition of the partial structure held by the conversion rule holding means. A substructure definition reference means for presenting to the user; a conversion rule registration means for registering, in the conversion rule holding means, an application-dependent conversion rule created by the user with reference to the substructure definition presented by the substructure definition reference means; Is further provided. This allows the user to refer to the presented partial structure definition without having to be aware of the difference in structure between different formats and document types of the document, and to perform application-dependent conversion to extract the partial structure required for the application. Rules can be easily described and registered with the system for future use.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A document management system according to the present invention will be specifically described based on embodiments. FIG. 1 shows one use form of the document management system 1. The document management system 1 is connected to a local area network N and is accessed from many client systems 2. The local area network N is connected to a wide area network or a corporate intranet via a dedicated line, and can be accessed remotely. FIG. 2 shows an example of the internal configuration of the document management system 1. The document management system 1 includes a document registration unit 11, a conversion rule holding unit 12, a document format homogenization unit 13, a partial structure extraction unit 14, a document search. Unit 15, document data holding unit 1
6, a conversion rule registration unit 17, and a partial structure definition reference unit 18. In the present example, these functional units are configured by executing a predetermined document management program using computer hardware resources.

The document registration unit 11 has a communication function via the network N. When a registration request is received from the client 2, the document registration unit 11 receives the document data passed from the client 2, and converts the original document data into the document data. Holder 1
6 and the document format homogenization unit 13
To the original document data and the format name or document type attached to the original document data. The document format homogenization unit 13 converts document data of different formats and document types into a unified internal format, registers this in the document data holding unit 16 for later use, and extracts the partial structure. To the partial structure extraction unit 14.

The conversion rule holding unit 12 is composed of a readable and writable memory, and includes a system built-in conversion rule for extracting a desired partial structure from a unified internal format and a conversion rule defined for each application (hereinafter referred to as an application). In this example, the definition of a partial structure that can be extracted by a system-incorporated conversion rule presented to the user as described later is stored. The conversion rule holding unit 12 is called from the partial structure extraction unit 14 as needed at the time of document registration, and outputs necessary conversion rules. When creating an application-dependent conversion rule, the conversion rule holding unit 12 is called from the conversion rule registration unit 17 and outputs the necessary conversion rules. The structure definition is output to the partial structure definition reference unit 18.

The partial structure extracting unit 14 extracts, from the conversion rule holding unit 12, an application-dependent conversion rule and a system-incorporated conversion rule used by the application-dependent conversion rule, according to the currently executed application. The system-incorporated conversion rule is applied, the application-dependent conversion rule is applied to the application result, and the extracted desired partial structure is registered in the document data holding unit 16 in association with the original document data. . That is, a partial structure corresponding to each application is registered in the document data holding unit 16 in association with the original document data.

When a search request is received from a user or a client, the document search unit 15 gives the request to the document data holding unit 16 and returns original document data or a partial structure that matches the search request as a result. . The document data holding unit 16 is composed of a readable and writable memory, and holds original document data, document data expressed in a unified internal format, and extracted partial structures in association with each other.

The conversion rule registration unit 17 receives an application-dependent conversion rule created by the user and registers it in the conversion rule holding unit 12 when the user constructs a new application for using the partial structure. The partial structure definition reference unit 18 constitutes an interface having a display screen. In order to assist a user in creating a new application-dependent conversion rule, the partial structure definition reference unit 18 has already registered in the conversion rule holding unit 12 during the above registration. The definition of the substructure that can be extracted by the registered system built-in conversion rule is presented to the user.

As described above, the conversion rules are classified into two types, system-based conversion rules and application-dependent conversion rules, according to their levels. The system-incorporated conversion rule is a conversion rule registered in the system in advance to extract a partial structure, and the application-dependent conversion rule is a conversion rule for extracting a partial structure suitable for an application. Therefore, the system-incorporated conversion rules cannot be changed by anyone except the system administrator, and the application-dependent conversion rules can be changed by the application programmer or a person authorized by the person according to the purpose.

That is, according to the present invention, the application programmer refers to only the definition of the substructure obtained from the system built-in type conversion rule, and describes the application-dependent conversion rule by combining one or more of these substructure definitions. This eliminates the need for application programmers to know the format and document type of the original document data in order to create application-dependent conversion rules, thereby reducing the human cost of writing conversion rules. .

FIG. 3 shows the relationship between document data and conversion rules. Reference numeral 21 in the figure denotes a word processor A document in a unified internal format obtained as a result of performing the format homogenization process on the document data of the word processor A by the document format homogenization unit 13. Here, in this example, the document format of the word processor A basically does not have a hierarchical structure.
It is assumed that there is only a style specification. Similarly, reference numeral 22 in the figure denotes a word processor B document in a unified internal format obtained as a result of performing the format homogenization process on the document data of the word processor B by the document format homogenization unit 13. Here, in this example, it is assumed that the document format of the word processor B is a format that can hold a general hierarchical structure of chapters, sections, and paragraphs in advance.

FIG. 4 shows a representation 21 of the document of the word processor A in a uniform internal format, and FIG.
Of the document 22 in a unified internal format. In this example, the unified internal format is XML (Extensible Markup Language 1.0; W3C Reco
mmendation 10-Feb-98). XML is one of the formats of structured documents. However, a flexible description system for expressing as a unified format while retaining the information of the original document, such as having no document type and freely defining attributes, is available. providing.

The unified internal format document 21 of the word processor A shown in FIG.
OC "element. An “original” attribute indicating the original document format of the document is set in the “DOC” element. The “DOC” element has only a sequence of “PARA” elements as its contents. Each "PARA" element has a "size" attribute that indicates the font size, a "type" attribute that indicates the font type (such as bold or italic),
"Alig" for alignment such as centering and right alignment
n ”attribute is set. Thus, the unified internal format corresponding to the document of the word processor A is a flat (having no hierarchical structure) format in which the same elements “PARA” distinguished only by the format information are arranged.

The document of the word processor B shown in FIG. 5 has a "DOC" element following the "XML" declaration in the first line. The "DOC" element contains "orig" indicating the original document format of the document.
inal "attribute is set. The "DOC" element has a "HEAD" element representing a title or a headline, a "PARA" element representing a paragraph, and a "SECT" element representing a chapter as its contents. The "SECT" element starts with "HEAD", which indicates a heading
Element, followed by a sequence of "PARA" elements.
Thus, the unified internal format corresponding to the word processor B document is a format having a general hierarchical structure such as chapters, headings, and paragraphs. Although not shown in FIG. 5, the “SECT” element may further include a “SECT” element in the content. In this case, the first "SECT" element represents a chapter and the second "SECT" element represents a clause.

In FIG. 3, reference numerals 23 to 26 denote conversion rules incorporated in the system. In this example, 23 is a word processor A
It is assumed that the conversion rule is a conversion rule for extracting a “headline” from the unified internal format 21 corresponding to the document. 24
Is a conversion rule for extracting “headings” from the unified internal format 22 corresponding to the word processor B document. It is assumed that reference numeral 25 denotes a conversion rule for extracting a partial structure of "itemized item" from a unified internal format corresponding to the documents of the word processors A and B in common. Reference numeral 26 denotes a conversion rule for extracting a “title” from a unified internal format corresponding to the documents of the word processors A and B in common.

FIG. 6 shows an example of a system built-in type conversion rule 23 for extracting "headings" from the unified internal format 21 corresponding to the word processor A document. Here, the grammar of the conversion rule will be described with reference to FIG. A conversion rule is a connection between a prime rule and a structure generator. An elementary rule consists of a pair of a condition and a destination element name. When each element of the document expressed in the unified internal format satisfies the condition of the elementary rule, the element is converted to the element of the conversion destination element name. Thereafter, when an element X satisfies the condition of the elementary rule Y, “element X is elementary rule Y
Matches. " If there is no destination element, the destination element name is represented by "null".

The conditions include "SOURCE" indicating the condition of the element name as the conversion source, "ATT" indicating the condition regarding the attribute of the element as the conversion source, and "TEXT" indicating the text content of the element as the conversion source. By any number of logical operators (and
Or not). The structure generator connects elementary rules to each other, and defines a connection relationship between an element that matches one elementary rule and an element that matches the other elementary rule.

The structure generator includes "SEQ", "REP", "CH
There are "O", "OPT" and "OPTREP". “SEQ” is an element that matches the elementary rule at the higher level (the left side in FIG. 6), and the sequence of elements at the lower level is different from the elementary rule of the sequence at the lower level of “SEQ”. Indicates that matches must be made in that order. “REP” means that the element below the elementary rule that matches
Occurs one or more times, indicating that all elements must match the elementary rules under "REP". "CH
O "for the element that matches the higher prime rule,
Indicates that the element below it must match any one of the elementary rules in the sequence of elementary rules below “CHO”. “OPT” indicates that, for an element that matches the higher-order elementary rule, there is no element below that element, or if there is an element, it is necessary to match the elementary rule that is lower than “OPT”. “OPTREP” means that the element below the element that matches the elementary rule above it appears 0 or more times, and all the elements
Indicates that the elementary rule below "P" must be matched.

Therefore, the system-incorporated conversion rule (conversion rule for extracting “headings” from the unified internal format 21 corresponding to the word processor A document) shown in FIG. 6 has the following contents. First, the elementary rule at the top has a condition that the element name of the conversion source is “DOC”, and the element of the conversion destination is “null”, that is, there is no element of the conversion destination. Next, there is a structure generator "CHO" under the structure generator "REP", and there are two elementary rules under the structure generator "CHO". The first elementary rule indicates that an element whose attribute “type” has a value of “bold” matches, and that this element is converted to “heading”. The second elementary rule indicates that an element that does not satisfy the condition that the value of the attribute “type” is “bold” matches by the negation operator, and indicates that there is no element to be converted. The structure generator "REP" and its subordinate "CHO" allow one of the elements
It is shown that there is a sequence of repetitions more than twice, and each element in it always matches one of the above two elementary rules, and only the element that matches the former elementary rule is converted to "heading" Have been. That is, according to the system built-in type conversion rule shown in FIG. 6, from the unified internal format document of the word processor A shown in FIG. Is taken out as a “heading”.

FIG. 7 shows an example of a conversion rule 24 for extracting "headings" from the unified internal format 22 corresponding to the word processor B document. Note that the meaning of this conversion rule is clear from the description of the conversion rule described above, and the description is omitted here. That is, FIG.
According to the system built-in conversion rule shown in FIG. 5, the unified internal format document of word processor B shown in FIG.
The “HEAD” element below the “SECT” element is extracted as a “heading”.

FIG. 8 shows common items from the unified internal format corresponding to the word processor A document and the unified internal format corresponding to the word processor B document.
An example of a conversion rule 25 for taking out an expression is shown.
Here, the condition “TEXT” regarding the text content of the elementary rule, which is not included in the conversion rules shown in FIGS. 6 and 7, will be described. The condition regarding the text content is described in accordance with a general regular expression grammar. In the conversion rule shown in FIG. 8, there are two elementary rules having a condition on the text content, but the condition on the text content of one elementary rule is a negative condition of the condition on the text content of the other elementary rule. The regular expression pattern itself is the same. In the conversion rule shown in FIG. 8, "."
, Or any string that begins with "(" and ")" at the beginning of a line, or any string that begins with "[" and "]" at the beginning of a line, Or the element which comes with a number at the beginning of the line and followed by "."
Converted to elements, other elements are not converted.

FIG. 9 shows an example of a conversion rule 26 for extracting "title" from the unified internal format corresponding to the word processor A document and the unified internal format corresponding to the word processor B document. Note that the meaning of this conversion rule is clear from the description of the conversion rule described above, and the description is omitted here. However, in the figure, the condition "[]" indicates that any element unconditionally matches this elementary rule.

In the present embodiment, the above-described system-incorporated conversion rules 23 to 26 are stored in the conversion rule storage unit 12 in association with the names and document types of applicable document formats. That is, when a document format name is given, a system-incorporated conversion rule applicable to the document format is specified, and is passed from the conversion rule holding unit 12 to the partial structure extraction unit 14.

Next, in FIG. 3, reference numerals 27 to 29 denote definitions of partial structures obtained as a result of applying the system built-in type conversion rules 23 to 26. Reference numeral 27 denotes a rule 23 for extracting “headings” from the unified internal format 21 corresponding to the word processor A document or “headings” from the unified internal format corresponding to the word processor B document.
Is a “heading” element obtained as a result of applying the rule 24 for extracting the “heading”. 28 is a unified internal format 21 corresponding to the document of word processor A or word processor B.
Is an “item” element obtained as a result of applying a rule 25 for extracting “itemized items” in common from the unified internal format 22 corresponding to the document. 29 is
Unified internal format 2 for Word Processor A documents
This is a “title” element obtained as a result of applying a rule 26 for extracting a “title” from the unified internal format 22 corresponding to the document 1 or word processor B in common.

Further, in FIG. 3, reference numeral 30 denotes an application-dependent conversion rule for extracting a partial structure requested by the application 31, and a required partial structure is extracted from the document by a combination of the system built-in conversion rules 23 to 26. In this example, the application-dependent conversion rule 30 is created by the user with reference to the definition 27 of the partial structure (that is, the “heading” element) and the definition 28 of the partial structure (that is, the “item” element). The application-dependent conversion rule 30 is held in the conversion rule holding unit 12 via the conversion rule registration unit 17. The details of the application-dependent conversion rule 30 will be described later.

In the following, processing for creating an application-dependent conversion rule, registering a document, executing an application, and registering a new application will be described in order. here,
The application of this example searches for a certain technical document from the document data holding unit 16, and sequentially retrieves and displays documents referred to by the document as reference documents from the document data holding unit 16 in response to a user request. I do.

<Conversion Rule Creation> In the process of creating an application-dependent conversion rule, the application programmer issues a request from the conversion rule registration unit 17 to the document management system 1 to refer to the partial structure definition registered in the system. Is started by In accordance with this request, the document management system 1 sets the
Then, the partial structure definition is extracted from the conversion rule holding unit 12 via the subroutine, and the result is presented to the programmer (user) from the partial structure definition reference unit 18. Then, the application programmer defines a desired partial structure (here, a reference list) by referring to the presented partial structure definition, and performs a conversion for generating this partial structure definition from the combination of the presented partial structure definitions. A rule (that is, an application-dependent conversion rule for extracting a reference) is described and registered in the conversion rule holding unit 12 via the conversion rule registration unit 17.

In order to realize the application in this example, it is necessary to extract a reference list from documents in various formats. Therefore, the application programmer defines the structure of the reference list. The definition of the reference list is defined as one or more "reference" elements arranged as subordinate elements of the "reference list" element. A conversion rule for generating such a structure definition from the proposed combination of partial structure definitions, that is, an application-dependent conversion rule will be written. This conversion rule must first include "References" or "R
"heading" element that contains only "eferences", convert the element above it into a "bibliography list" element, and add any characters between "[]" and "()" following the "heading" element If there is a bulleted "item" element that starts with a column, it will be extracted until there is no more arrangement, and converted to a "literature" element. FIG. 10 shows an example of an application-dependent conversion rule for extracting a reference.

<Document Registration> At the time of document registration, a registration request is sent from the client 2 via the network N together with the document data and the corresponding document format name or document type. When the document management system 1 receives the request, the document data and the name or document type of the document format are passed to the document registration unit 11. The document registration unit 11 registers the document data in the document data holding unit 16 and sends the document data and the name or document type of the document format to the document format homogenization unit 13.

The document format homogenizing unit 13 converts the document data into a uniform document format in the system, that is, a uniform internal format. At this time,
A conversion program is executed for each format of the input document. The conversion program converts the information of the original document as much as possible without losing it. The reason for this is to prevent the input document from being used for other purposes if it is not converted how to use the input document. In this embodiment, XML is used for the unified internal format as described above. In this example, it is assumed that the document of the word processor A is registered. Therefore, the result of the conversion into the unified internal format by the document format homogenization unit 13 is as shown in FIG.

Next, the document data expressed in the unified internal format is registered in the document data holding unit 16 and sent to the partial structure extracting unit 14, where the conversion rule holding unit 1
2 is applied. That is, among the conversion rules stored in the conversion rule storage unit 12 described above, first, a conversion rule (application-dependent conversion rule) for extracting a partial structure requested by an application is extracted, and the conversion rule is converted into a system-incorporated conversion rule. If the rules are referred to, the system-incorporated conversion rules are taken out and applied in order, and finally the application-dependent conversion rules are applied to determine whether a desired partial structure exists. As a result of applying these series of conversion rules, if a desired partial structure is extracted,
The partial structure is associated with the original document data and the expression in the unified internal format, and the document data holding unit 16
Is held.

The conversion rule application processing will be described in detail with reference to the flowcharts shown in FIGS. First, the partial structure extraction unit 14 converts the conversion rule holding unit 1
2, one application-dependent conversion rule corresponding to the application currently being executed is extracted (step S1). In this example, a conversion rule for extracting the reference list shown in FIG. 10 is extracted. Next, the partial structure extraction unit 14 applies the extracted application-dependent conversion rule (Step S2). Then, it is determined whether all application-dependent conversion rules have been applied (step S3). If application-dependent conversion rules still remain, the process returns to step S1 to continue, and all application-dependent conversion rules have been applied. By the way, the process ends.

The application-dependent conversion rule application process (step S2) is performed according to the procedure shown in FIG. First, for the application-dependent conversion rule extracted in step S1, the partial structure extraction unit 14 specifies a system-incorporated conversion rule referred to by the conversion rule (step S11). This can be realized by searching for a source element name specified as a condition in each elementary rule of an application-dependent conversion rule, and searching for a system built-in type conversion rule having the source element name as a structure definition. .

For example, when the application-dependent conversion rule shown in FIG. 10 is extracted, two elements of “heading” and “item” are used as the conversion source element names in the elementary rule. Therefore, as shown in FIG.
From the system built-in conversion rules 23 and 24 that provide a “heading” element as a structure definition, ie, conversion rules for extracting “headings” from the unified internal format of the word processor A document, and from the unified internal format of the word processing B document Conversion rule for extracting "heading"
Two are first identified, and then a system-incorporated conversion rule 25 that provides "items" as a structure definition, i.e., to extract "bulleted items" from the unified internal format of the word processing A and word processing B documents. Is specified.

Next, the partial structure extraction unit 14 specifies the original format of the document to be registered (step S12). In this example, the document of word processor A is registered, and the attribute of the “DOC” element is read from the document shown in FIG. 4 to specify that the original document format is that of word processor A. Next, the partial structure extraction unit 14 specifies a system built-in type conversion rule applicable to the format of the original document (step S13). Here, since the document of word processor A is registered, as shown in FIG.
One is specified.

Next, the partial structure extraction unit 14 specifies a system built-in type conversion rule to be actually applied (step S14). This calculates the intersection of the set of system built-in conversion rules (23, 24, and 25) specified in step S11 and the set of system built-in conversion rules (23, 25, and 26) specified in step S13. It is realized by doing. As a result, the conversion rules actually applied to the system are 23 and 2
It becomes two of five. Next, the partial structure extraction unit 14 applies these system built-in type conversion rules (step S15). Here, the two system-incorporated conversion rules 23 and 25 specified in step S14 are sequentially applied.
That is, a conversion rule for extracting “headings” from the unified internal format of the word processor A document is applied. FIG. 13 shows the result of applying the system-incorporated conversion rule for extracting the “headline” shown in FIG. 6 to the expression of the word processor A document shown in FIG. 4 in the unified internal format.

Then, the partial structure extraction unit 14 determines whether or not the system-incorporated conversion rule to be applied remains (step S16), and if it still remains, returns to step S15 and applies it. Here, the system built-in conversion rules 25, ie, word processor A and word processor B
Since there is a conversion rule for extracting the "itemized item" from the unified internal format of the document,
Return to step S15 to apply this. FIG. 14 shows the result of applying the system-incorporated conversion rule for extracting the “itemized item” shown in FIG. 8 to the expression in the unified internal format of the document of the word processor A shown in FIG.

Next, when the application of all the system-incorporated conversion rules is completed, the partial structure extraction unit 14 sequentially applies the elementary rules of the application-dependent conversion rules (step S17). Here, first, the elementary rule 101 shown in FIG. 10 is applied. The elementary rule 101 matches any element. Next, the partial structure extraction unit 14 determines whether all elementary rules have been applied (step S18). Here, since the elementary rule 102 remains, the process returns to step S17 and is applied. In the elementary rule 102, an element whose conversion source element name is "heading" and whose text content includes "Reference" or "References" matches.

In step S18, it is determined again whether or not all elementary rules have been applied. Since elementary rules 103 remain, the flow returns to step S17 to apply these. In the elementary rule 103, an element having a text content whose conversion source element name is "item" and whose arbitrary character string starts with "[" and "]" or "(" and ")" is enclosed. Matches. It is again determined in step S18 whether all elementary rules have been applied, and since application of all elementary rules has been completed, a partial structure is extracted (step S19). FIG. 15 shows a partial structure extracted from the unified internal format of the document of the word processor A in this manner. As described above, when extracting the partial structure by applying the application-dependent conversion rule, the minimum necessary system built-in conversion rule is specified and applied according to the original format of the target document. Partial structures can be efficiently extracted.

<Application Execution> At the time of application execution, the user sends a document search request to the document management system 1 from the client 2 to the document management system 1 via the network N. The document management system 1 sends this request to the document search unit 15, and the document search unit 15 searches the document data holding unit 16 for original document data that satisfies the request, and presents it to the user. Then, when the user wants to refer to the reference of the document, a search request for the reference is sent to the document search unit 15, and the document search unit 15 performs the above-described partial structure extraction processing from the original document data. Using the contents of each "document" element of the extracted "reference document list" element as a search key, another original document data held in the document data holding unit 16 is searched. Presents it to the user.

<Registration of New Application> Next, a process for constructing a new application different from the above application will be described. Here, in this example, it is assumed that an application programmer considers constructing an application for newly creating a table of contents of a document. First, the application programmer issues a request to the document management system 1 to refer to the partial structure definition registered in the system. The document management system 1 that has received this extracts the partial structure definition from the conversion rule holding unit 12 via the partial structure definition reference unit 18 and presents it to the application programmer.

Next, the application programmer:
Refer to the provided definition of the substructure, and
(Here the table of contents) is defined. To realize this application, it is necessary to extract a table of contents from documents in various formats. Therefore, the application programmer defines the structure of the table of contents. For example, the definition of the table of contents is such that the "title" element is 1
And one or more “heading” elements are defined. A conversion rule for generating such a structure definition from the proposed combination of partial structure definitions, that is, an application-dependent conversion rule will be written.
This conversion rule will take out the "title" element first, and then take out the "heading" element. FIG. 16 shows an example of the application-dependent conversion rule. The application programmer registers the conversion rule for extracting the table of contents created in this manner in the conversion rule holding unit 12 via the conversion rule registration unit 17.

When a new application-dependent conversion rule for extracting a table of contents is registered, the partial structure extraction unit 1
4 retrieves a document group expressed in a unified internal format from the document data holding unit 16, sequentially applies application-dependent conversion rules and system-incorporated conversion rules according to the procedures shown in FIGS. The structure is extracted and held in the document data holding unit 16. FIG. 17 shows a state where a new application 33 and a new application-dependent conversion rule 32 are registered with respect to the relationship between the document data and the conversion rule shown in FIG. Application-dependent conversion rule 3
2 refers to the partial structure definition 27 of the “heading” element and the partial structure definition 29 of the “title” element as the partial structure definition. According to the procedure shown in FIG. 12, the system built-in conversion rules 23 and 2
6. After the system built-in conversion rules 24 and 26 are applied to the document of the word processor B, the application-dependent conversion rule 32 (shown in FIG. 16) is applied to extract a desired partial structure, that is, a table of contents. You.

The result of applying the system built-in type conversion rule 23 shown in FIG. 6 to the word processor A document shown in FIG. 4 (FIG. 13) is compared with the system built-in type conversion rule 26 shown in FIG. FIG. 18 shows the result of application.
19 shows the result of applying the application-dependent conversion rule 32 (FIG. 16) to the document shown in FIG. Also,
FIG. 2 shows the result of applying the system built-in type conversion rule 24 shown in FIG. 7 to the word processor B document shown in FIG.
FIG. 21 shows the result of applying the system built-in type conversion rule 26 shown in FIG. 9 to the document shown in FIG.
FIG. 22 shows the result of applying the application-dependent conversion rule 32 (FIG. 16) to the document shown in FIG.

As described above, a conversion rule for extracting a document element which will be used by a plurality of applications is described and provided for each format, so that a partial structure can be extracted by applying an application-dependent conversion rule. , The minimum necessary system built-in conversion rules are applied, and the partial structure required by the application can be efficiently extracted. Further, when creating an application-dependent conversion rule, it is possible to describe the conversion rule without being aware of the difference in the document format. In the above example, an example of a so-called format-oriented document such as a word processor A or a word processor B has been described as a target document, but this may be a structured document. When the target is a structured document, the conversion to the unified internal format by the document format homogenization unit 13 is not necessary, and the format of the structured document may be handled as it is as the unified internal format.

Next, another embodiment that enables the conversion rule to be applied more efficiently will be described. In the example of the reference list extraction described in the above-described embodiment, when the application-dependent conversion rule is applied, the conversion rule embedded in the system to be used (the conversion rule for extracting the “heading” and the “bulleted item” in the above embodiment) Conversion rule to extract "). However, the actual heading needed to extract the bibliography is "References"
Or, only the headings that contain "References" in the textual content, and the only required bulleted items are those corresponding to the reference list. That is, since unnecessary “heading” elements and “item” elements are extracted, in this example, application-dependent elements such as “heading” elements and “item” elements that satisfy the restrictions of the application-dependent conversion rule are extracted. Conversion rules are expanded.

First, for each elementary rule of the application-dependent conversion rule for extracting the reference list shown in FIG. 10, a search is made for one that requires a “heading” element and an “item” element as the conversion source element name. In FIG. 10, the elementary rule 102 requests the “heading” element as the source element name, and the elementary rule 103 requests the “item” element as the source element name. Therefore, for each format of the original document,
The corresponding system built-in conversion rules are extracted, and each elementary rule is expanded. Specifically, it is assumed that the document of the word processor A has been subjected to partial structure extraction. At this time, for the application-dependent conversion rule for extracting the reference list shown in FIG. 10, a system-incorporated conversion rule for extracting the “heading” element shown in FIG.
The system built-in type conversion rule for extracting the “item” element shown in FIG. 8 is applied.

In the conversion rule shown in FIG. 6, [ATT: type = "bold"] is set as a condition for conversion to the "heading" element. Therefore, the condition [SOURCE: heading] for the conversion source element name of the elementary rule 102 is expanded to [ATT: type = "bold"]. In addition, in the conversion rule shown in FIG. 8, as a condition for converting to an “item” element, [(TEXT: “^ ·”) v (TEXT: “^ \ [.
+ \] ") v (TEXT:" ^ \ (. + \) ") v (TEXT:" ^ [0-9] + \. ")]. Change the condition [SOURCE: item] for the element name to [(TEXT: "^ ・") v (TEXT:
"^ \ [. + \]") v (TEXT: "^ \ (. + \)") v (TEXT: "^ [0-9] +
23 shows the conversion rules expanded in this manner.However, in FIG. 23, the conditions for conversion to the “item” element are omitted as a result of the logical operation as a result of the logical operation. There is. By directly applying the developed conversion rules to the unified internal format corresponding to the document of the word processor A, it is possible to extract the “heading” element and “bulleted item” element unnecessary for the application. It is possible to extract only the necessary partial structure without the need.

Next, it is assumed that the document of the word processor B has been subjected to partial structure extraction. At this time, for the application-dependent conversion rule for extracting the reference list shown in FIG. 10, the system-incorporated conversion rule for extracting the "heading" element shown in FIG. The system built-in type conversion rule for extracting is applied. In the conversion rule shown in FIG. 7, [SOURCE: HEAD] is set as a condition for conversion to a “heading” element. Therefore, the condition [SOURC
Expand [E: Heading] to [SOURCE: HEAD]. Also, in the conversion rule shown in FIG. 8, [(TEXT: "^.") V (TEXT: "^ \ [. + \]") V (TEXT: "^ \ ( . +
\) ") v (TEXT:" ^ [0-9] + \. ")] is set, so the condition [SOURC
E: Item] is changed to [(TEXT: "^ ・") v (TEXT: "^ \ [. + \]") V (TEX
T: expands to "^ \ (. + \)") V (TEXT: "^ [0-9] + \.")]. FIG. 24 shows the conversion rules developed in this way.
However, in FIG. 24, the condition to be converted to the “item” element is omitted as a result of the logical operation. By directly applying the developed conversion rules to the unified internal format corresponding to the document of the word processor B, it is possible to extract the "heading" element and the "bulleted item" element unnecessary for the application. It is possible to extract only the necessary partial structure without the need.

As described above, by expanding the application-dependent conversion rule with reference to the corresponding system-incorporated conversion rule, only the “heading” element relating to the reference is extracted, and is continued to the “heading” element. Since only the element is checked as the item of the bulleted item, unnecessary "heading" element and "item" element do not need to be extracted, and further efficiency can be achieved.

[0058]

As described above, according to the present invention, conversion for extracting a partial structure required for an application without regard to the difference in structure between documents of different document formats and document types. Rules can be easily described by using a system-incorporated partial structure extraction rule that previously absorbs differences in format and the like. Further, according to the present invention, since only a partial structure required for an application can be selectively extracted, a document structure can be efficiently extracted. Furthermore, by using the constraint condition of the partial structure required by the application, more efficient partial structure extraction processing can be realized.

[Brief description of the drawings]

FIG. 1 is a diagram showing an overall configuration of a system using a document management system according to the present invention.

FIG. 2 is a diagram illustrating an example of a configuration of a document management system.

FIG. 3 is a diagram illustrating the relationship between document data and conversion rules.

FIG. 4 is a diagram showing an example of a unified internal format document of the word processor A.

FIG. 5 is a diagram showing an example of a unified internal format document of a word processor B.

FIG. 6 is a diagram showing an example of a conversion rule for extracting “headings” from a unified internal format document of word processor A.

FIG. 7 is a diagram illustrating an example of a conversion rule for extracting “headings” from a unified internal format document of word processor B.

FIG. 8 is a diagram showing an example of a conversion rule for extracting “itemized item” from a unified internal format document.

FIG. 9 is a diagram illustrating an example of a conversion rule for extracting a “title” from a unified internal format document.

FIG. 10 is a diagram showing an example of an application-dependent conversion rule for extracting a reference list.

FIG. 11 is a flowchart illustrating a procedure of a conversion rule application process.

FIG. 12 is a flowchart illustrating a procedure of a conversion rule application process.

FIG. 13 is a diagram showing a result of extracting “headings” from the unified internal format document of word processor A.

FIG. 14 is a diagram showing a result of extracting “headings” and “items” from the unified internal format document of word processor A.

FIG. 15 is a diagram showing a reference list extracted from the unified internal format document of the word processor A.

FIG. 16 is a diagram showing an example of an application-dependent conversion rule for extracting a table of contents.

FIG. 17 is a diagram illustrating the relationship between document data and conversion rules.

FIG. 18 is a diagram illustrating a result of extracting “headings” and “titles” from the unified internal format document of word processor A.

FIG. 19 is a diagram showing a “table of contents” extracted from the unified internal format document of the word processor A.

FIG. 20 is a diagram illustrating a result of extracting “headings” from the unified internal format document of the word processor B.

FIG. 21 is a diagram illustrating a result of extracting “headings” and “titles” from the unified internal format document of word processor B.

FIG. 22 is a diagram showing a “table of contents” extracted from the unified internal format document of the word processor B.

FIG. 23 is a diagram illustrating an example in which an application-dependent conversion rule is developed for a document of a word processor A.

FIG. 24 is a diagram illustrating an example in which an application-dependent conversion rule is developed for a document of a word processor B.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Document management system 11 ... Document registration part 12 ... Conversion rule holding part 13 ... Document format homogenization part 14 ... Partial structure extraction part 16 ... Document data Holding unit, 17: Conversion rule registration unit, 18: Partial structure definition reference unit,

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 5B009 QA05 QA06 QA09 QA11 SA03 SA12 SA13 SA14 TB03 VA02 5B075 ND03 ND23 ND35 NK02 NK43 NK46 NS01 UU05

Claims (3)

[Claims] In a document management system that stores and manages various types of documents, a document registration unit that receives a document to be registered in the document management system in response to a user request, and a document that is received by the document registration unit are unified. Document format homogenization means for converting to internal formats, system-integrated conversion rules for extracting partial structures from unified internal formats of documents, and parts required for each application by combining system-integrated conversion rules A conversion rule holding unit for holding an application-dependent conversion rule for extracting a structure; and a system built-in type conversion held in the conversion rule holding unit for a document received by the document registration unit and converted into an internal format according to an application. Application that applies rules and further supports A partial structure extraction unit that extracts the partial structure by applying the application-dependent conversion rules, a document received by the document registration unit, a document in a uniform internal format converted by the document format homogenization unit, and a partial structure A document management system, comprising: a document data holding unit that holds a partial structure extracted for each application by an extraction unit in association with each other. 2. The document management system according to claim 1, wherein the document registration unit receives the format name or the document type together with the document, and the partial structure extraction unit applies the document from the format name or the document type received by the document registration unit. System-independent conversion rules to be specified, the condition description of the application-dependent conversion rules is expanded based on the condition descriptions of the system-incorporated conversion rules, and a partial structure is extracted from the document converted into the internal format. And a document management system. 3. The document management system according to claim 1, wherein the conversion rule holding unit holds a definition of a partial structure that can be extracted by a system built-in type conversion rule, and holds the definition in the conversion rule holding unit. Substructure definition reference means for presenting the definition of the substructure to the user, and a conversion rule holding means for referring to the substructure definition presented by the substructure definition reference means to create an application-dependent conversion rule created by the user. And a conversion rule registering means for registering the document in the document management system.