JP2001117934A - Electronic document management method and system, and recording medium - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To provide an electronic document management method capable of easily providing information for full-text search of a document without the user being aware of the content, type, and classification of the content. A full-text search index file creation function 105 collectively specifies documents, extracts text from the specified documents, and registers them as a full-text search index file. At the time of text extraction, extraction processing is performed using a text extraction library or function defined for each document type. If the application that created the document could not be specified or if a predetermined condition was satisfied, there is provided a means for acquiring bibliographic items of the document and creating an index file for full-text search from this information. This process can be executed as a batch job at a time convenient for the user.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a computer-based electronic document management system, and more particularly to a method and system for creating a full-text search index for a document created by various applications such as various document creation programs, and a recording medium. It is.

[0002]

2. Description of the Related Art In a conventional document management method, a user defines information such as a class, a property, a content, an index file for full-text search for each document, and manages the document. In addition, the function of extracting only a text portion from a document in order to perform a full-text search can only extract text from a document created by a predetermined type of application. In the case of a search index, it is necessary to separately register this file as a full-text search index file.

As described in Japanese Patent Application Laid-Open No. 9-204442, in the conventional document management method, when a document is updated, a data update trigger is generated, and when the trigger is received, the updated document data is updated. The index is updated based on.

[0004]

In the prior art, in order to create a full-text search index from documents created by various applications, a user must prepare a separate text extraction program and individually create and register it. However, at this time, the applications from which text can be extracted are limited, and processing cannot be performed for non-target applications. In addition, if the data to be extracted is text or images, or if the data is complicated, such as the security of the contents, processing cannot be performed because the text is not to be extracted. In some cases, normal text extraction processing cannot be performed, extracted characters are garbled, or meaningful information cannot be obtained. For this reason,
In the prior art, the types of documents that can be indexed are limited.

[0005] Also, in order to extract text and create an index from a document created by an application, a great deal of time and computer processing power are generally required. In the prior art, an index is created when a document is updated. There was only.

[0006] In the present invention, the user is required
It is an object of the present invention to provide a computerized document management method that provides information for easily performing full-text search of a document without being aware of the type and classification. It is another object of the present invention to provide an electronic document management method capable of setting a time for performing an index creation process.

[0007]

The object of the present invention is to specify a document in a lump and to extract text from a specified document file, and to specify the application that created the document file or to perform the full-text search. Full-text search index file creation processing comprising: means for creating a full-text search index from bibliographic items corresponding to the document file when the processing for creating a search index satisfies a predetermined condition; and means for registering the index This is achieved by providing means. Here, the means for performing text extraction has means for performing extraction processing using a text extraction library or function defined for each document type when extracting text.
Further, the means for creating a full-text search index from bibliographic information includes a means for acquiring bibliographic items of a predetermined document from a database when the type of the document cannot be identified, and registering the bibliographic information as a full-text search index. Have. The condition for creating a full-text search index from bibliographic information is that text cannot be extracted from the contents of the document, or that text extraction is difficult and processing takes longer than the time defined by the user, or is subject to processing. The document may be a rendition not defined in the rendition definition file in advance, or the user may want to save resources required for the index file area. Therefore, the predetermined condition may be at least one of, for example, the time for full-text search index creation processing, the length of the full-text search index, and the occurrence of an error in the full-text search index creation processing.

Further, according to the present invention, the index creation processing can be executed as a batch job at a time convenient for the user (for example, at night or on a weekend when the burden is relatively small). These store the document update state, and the full-text search index creation processing means:
This is achieved by providing means for receiving at least one of the start time of the creation processing and the elapsed time from the current time, and starting the full-text search processing after a time satisfying the received specification.

Incidentally, in order to achieve the above object, a recording medium storing a program for realizing the functions realized by the above-described method and system may be used.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows an example of a system configuration of an electronic document management method having a full-text search index creation function according to the present invention. The document management software 101 is a CD-ROM,
It is installed in a client through a medium such as an FD, is connected to a document database 102 on a server 103 via a network, and can perform an input operation from a terminal 104 for online processing. The full-text search index creation function 105 is a part of the document management software 101, but may be located on another machine.

FIG. 2 shows an input / output outline of a full-text search index creation function according to the present invention. The index creation function 105 is a part of the document management software 101, reads a rendition definition file 202 and an environment setting file 203 as input files, and extracts an index creation target document from the document database 102 as an input document 204. After creating the index, the full-text search index file 205 is output and stored in the document database 102, and the error log file 206 and the trace log file 207 are output. The content of the document and the bibliographic items of the document are stored in the document database 102 in advance.

FIG. 6 shows an example of the rendition definition file 202. The rendition definition file 202 defines what kind of document creation program, that is, an application, the document is created for each document type, that is, for each rendition. Referring to the relationship between the rendition and the application, This is for creating an index by selecting an appropriate text extraction library or function suitable for the application. This file can be substituted with a file that defines the association between the extension of the document file and the application.

FIG. 7 shows an example of the environment setting file 203. The environment setting file 203 includes various prerequisites required for performing the index creation processing, for example, a storage location of a text extraction library or a command,
Defines the upper limit of the entire text extraction processing time, the upper limit of the processing time per document, the upper limit of the text length extracted per document, whether or not bibliographic information is acquired, and the name of the bibliographic information to be acquired File for

FIG. 3 is an example of a processing flow for creating a full-text search index according to the present invention. When this program is started, the rendition definition file 20
2 and the environment setting file 203 are read and stored in the internal memory as internal tables, respectively, and these values are referred to during execution of the program.

Next, in process 302, the document database 102
Connect to

In step 303, the property of the document to be processed, which is given to the program as an argument when the command is executed, is acquired from the document database 102.

In step 304, the registration status, which is one of the properties of the acquired document, is confirmed. If the document can be processed, the processing is continued. Possible processing targets are (1) an unprocessed document, (2) a document that has been processed but updated afterwards, and (3) if overwriting is specified for all documents when the command is executed, All documents are covered.

If it is determined in step 304 that the document is not a processing target, for example, if the process has already been executed to create an index document for full-text search, and there is no change in the document, the process returns to step 303. To the next document to be processed.

In step 305, the contents of the document to be processed are obtained from the document database 102.

In the process 306, a conditional branch is performed depending on the content.

First, if the document type is defined in the rendition definition file, a text extraction process is performed as a default process.

The text extraction process is performed using a library for extracting text from contents. This library performs a text extraction process when the number of characters is within the time limit and the number of characters given by the command or the environment setting file 203, and creates a text file for the index. FIG. 5 shows an example of this text file. When an error occurs during the text extraction process, or when the text extraction process exceeds the time limit, the process proceeds to the process of creating an index from the bibliographic information of the process 307.

If the document type is not defined in the rendition definition file at the time of acquiring the content in the process 306, the process of creating an index from the bibliographic information in the process 307 is also performed.

If the user wants to save resources for the index, the process 306 is skipped and the process 3
07 can always be set, for example, the text extraction processing time and the extraction length can be defined to be extremely small.

Process 307 is a process for creating an index from bibliographic information. The details of this process are shown in FIG.

In step 401, it is determined whether the document has no application defined in the rendition definition file, that is, whether the application cannot be specified, or whether the index creation time satisfies a predetermined condition. The predetermined condition is, for example, whether the upper limit of the text extraction processing time has been exceeded, the text length has exceeded the upper limit, or whether an error has occurred during text extraction.

In process 402, the bibliographic item name (column name) specified in the environment setting file 203 is obtained.

In step 403, the value of the bibliographic item specified in step 402 is obtained from the document database.

In step 404, the bibliographic items obtained in step 403 are converted into plain text format.

In step 405, a text file for creating a full-text search index is created from the text format of the bibliographic items obtained in step 404. FIG. 5 shows an example of a text file to be created. Here, a case is shown where a document title, a comment, and an author name are specified as bibliographic items.

When the process 405 is completed, the process returns to the process 308.

In step 308, step 306 or step 3
The text file obtained in step 07 is converted into a format suitable for a full-text search index file, and registered in a database as a full-text search index file. Or
In the present embodiment, the registration in the database can be performed at a time convenient for the user as a batch job in addition to the sequential execution according to the definition of the user. Process 30 for delayed execution as a batch job
In step 8, information necessary for the delay processing is registered. FIG. 8 shows an example of information necessary for the delay processing. The information necessary for the delay processing includes at least the processing start time, information for acquiring the column of the target table, and information on the operation performed on the database. When this information is defined,
The database can acquire the information necessary for the delay processing and start each processing for the database after a predetermined start time. If the document has been updated by this time, the update result cannot be reflected. In general, since full-text search index creation and registration processing takes time, the user operates and manages the document database so as to perform delayed updates at predetermined intervals so that the document update contents do not need to be sequentially reflected.

In step 309, the registration status of the document registered in step 308 is rewritten.

In the process 310, the document database 102 is updated, and the process returns to the process 303 for the next document.

After all the documents have been processed, the connection with the document database 102 is disconnected in a process 311, and an end process is performed.

In the index creation process of this embodiment, bibliographic information can be registered as an index in case text cannot be extracted from a document. Thus, for example, when extracting text from a document created by an application, if some or all of the text is garbled, or if the text extraction process cannot be completed within the time limit, or if it cannot be performed at all, the image When the data volume is large and complicated such as data, voice, security, etc., the bibliographic information of the document to be processed is converted to text and used as an index file for full-text search, thereby shortening the processing time and interrupting the program with an error. By obtaining bibliographic information even if meaningful information could not be obtained due to garbled content of the content, it is possible to give the user some clue about the document when performing full-text search, Also, full-text search function for bibliographic information as well as content Some, similar words search, full-width-half-width search, kana-katakana search, such as a variety of search functions will be able to use. Therefore, by using the full-text search index creation method of the present embodiment, the user acquires meaningful information for easily searching for a document without being conscious of the type, classification, image, sound, and the like of the content. The effect that it becomes possible is obtained.

Further, according to the present embodiment, the text information of the document content can be acquired from the document database to create and register the full-text search index file. Thus, an effect of executing the program at a time convenient for the user is obtained.

As described above, according to the present invention, a full-text search index can be created from bibliographic items even when the application that created the document cannot be specified. There is an effect that information for easily performing a full-text search of a document can be provided without being conscious of the content, type, and classification of the content. Also, during the process of identifying an application and extracting text and creating an index, if a predetermined condition is satisfied, a full-text search index is created from bibliographic items in the same way, so that the program is not interrupted. Also, there is an effect that the index can be created while saving the resources of the database depending on the specification of the user.

According to the present invention, in creating the full-text search index, at least one of the start time of the creation process and the elapsed time from the current time can be accepted, and after the time satisfying the accepted designation, Since the full-text search index creation process can be started, there is an effect that the full-text search index creation process can be performed at a time convenient for the user (for example, at night or on a weekend).

[0041]

According to the present invention, a user can easily acquire information for searching a document without being conscious of the type, classification, image, sound, or the like of the content.

[Brief description of the drawings]

FIG. 1 is a diagram showing a system configuration example of an electronic document management method having a full-text search index creation function according to the present invention.

FIG. 2 is a schematic diagram of an input / output function of a full-text search index creation function according to the present invention.

FIG. 3 is an example of a processing flow for creating a full-text search index according to the present invention.

FIG. 4 is an example of a process flow for creating an index from bibliographic information in a full-text search index creating process according to the present invention.

FIG. 5 is an example of a text file for creating a full-text search index created by the present invention.

FIG. 6 is an example of a rendition definition file.

FIG. 7 is an example of an environment setting file.

FIG. 8 is an example of information necessary for a delay creation function.

[Explanation of symbols]

101 Document Management Software 102 Document Database 105 Index Creation Function 202 Rendition Definition File 203 Configuration File 204 Input Document 205 Full Text Search Index File 501 Text File for Creating Full Text Search Index When Text is Extracted from Document Content Example 502 Example of text file for creating full-text search index when created from bibliographic items 801 Example of information required for delayed creation function

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Hiroshi Majima 5030 Totsuka-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture F-term in the Software Division, Hitachi, Ltd. (Reference) 5B009 SA12 5B075 ND03 NK02 NK04 NK24 NR05

Claims (5)

[Claims] 1. A method for managing a digitized document, comprising: storing a bibliographic item of the document file in a predetermined format when registering the document file; and creating a full-text search index of the document file. In the case where the process of creating the full-text search index satisfies a predetermined condition,
A computerized document management method, wherein a full-text search index is created from the bibliographic items corresponding to the document file. 2. The electronic document management method according to claim 1, wherein the predetermined conditions are: a full-text search index creation processing time, a full-text search index length, and a full-text search index creation process. An electronic document management method, characterized in that at least one error occurs. 3. A method for managing an electronic document, comprising: receiving a full-text search index creation processing command from a user and specifying a start time of the creation processing; extracting a document file to be indexed; An electronic document management method, wherein the full-text search index creation process is started after a start time. 4. A computerized document management system, comprising: receiving a designation of a document file from a user; extracting text from the document file by a line text extracting unit; Computerized bibliographic item extraction means for extracting a bibliographic item corresponding to a document file; and means for creating an index for full-text search in response to information from the text extracting unit or the bibliographic item extracting unit. Document management system. 5. A recording medium storing a method for managing a digitized document, wherein the method for managing a digitized document includes the following: when registering a document file, a bibliographic item of the document file is defined in a predetermined format If the document creation program that created the document file cannot be specified when creating a full-text search index for the document file, a full-text search index is created from the bibliographic items corresponding to the document file.