JP2009043083A - Database connection system, database connection program, and database connection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a database connection system capable of registering to a database metadata along with data to be written to the database by a processing module without altering the processing module. <P>SOLUTION: A setting information storage means 51 stores setting information which defines items of the metadate to be written to the database along with data in writing the data to the database. A metadata collecting means 52 requests metadata of the item defined by the setting information of a metadata holding means 54 for holding the metadata and collects the metadata from the metadata holding means 54. A write query conversion means 53 converts a query for instructing writing the data to the database into a query for instructing writing the data and the metadata of the item defined by the data and the setting information. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a database connection system for performing database operations, a database connection program, and a database connection method.

  A system that performs data analysis using data such as WWW (World-Wide Web), existing information systems, or voice / images due to the spread of the Internet or an increase in the amount of data stored in the company, and stores and manages the results. Has become important.

  FIG. 15 is an explanatory diagram showing an example of a database connection system related to the present invention. The database connection system illustrated in FIG. 15 includes a database system 85, and an application program 81 and a database access module 82 are installed. The application program 81 and the database access module 82 are software, and the database connection system actually operates according to these software. However, in the description of FIG.

  The application program 81 that performs data analysis records data, audio / image data, and the like held by the WWW or an existing information system in the database system 85. Further, if necessary, intermediate data in the middle of processing is written and read into the database system 85 while analyzing and extracting data, and finally obtaining the result is written into the database system 85.

  The database access module 82 provides a standard database connection interface 83 to the application program 81. The database access module 82 includes a library 84 that converts commands such as data writing and reading in a database language (for example, SQL) into commands of a protocol unique to the database system 85.

  When reading / writing the database, the application program 81 passes a database read / write command in a database language such as SQL to the database access module 82. The database access module 82 converts the command into a command of a protocol specific to the database by the library 84 and causes the database system 85 to read or write data.

  On the other hand, a data analysis system called BI (Business Intelligence) is known. This data analysis system collects data from various information sources, such as the WWW and corporate databases, and performs analysis by various processing modules for performing various processes such as morphological analysis, voice recognition, and data mining. The data obtained by the module analysis is analyzed further across the board, and the analysis results are provided to the user. In such a data analysis system, a processing module for performing all desired information analysis cannot be prepared in advance, and a process for performing an analysis desired by the analyst on the collected data. Modules may be added to extend the functionality to perform the desired analysis.

  When performing such a cross-sectional analysis, it is necessary to retain the metadata of the data together with the data of the analysis result and selectively use appropriate data based on the metadata. The metadata is data related to the data of the analysis result (analysis result by the processing module before performing the cross-sectional analysis).

  Examples of metadata include data creation date and time, data source information that is the source of analysis result data, and parameter setting statuses of processing modules. In addition, when the analysis results are derived by linking multiple processing modules, the information on how the analysis results were obtained is also the metadata necessary to confirm the reliability of the analysis result data. May be. In addition, in the case of a processing module that collects rainfall, surrounding sound volume, and the like at the position of the apparatus on which the processing module is mounted, the information on the position may also be metadata. In addition, when a plurality of similar processing modules are used by a plurality of users or are operated by a plurality of devices (for example, personal computers), the name of the device executing each processing module (for example, PC name) or the name of the user who uses the device is also metadata.

  Non-Patent Document 1 describes a method for replacing a database system with a different database system without changing an application program. In the method described in Non-Patent Document 1, the SQL issued by the application program is converted into the SQL for a new database system without changing the application program. This conversion is performed in the database connection layer that connects the application program to the database system. Non-Patent Document 1 describes reserved word conversion for converting a unique reserved word, logic conversion for converting a logic including a parameter as well as a reserved word, and prepared conversion as SQL conversion modes. In the method described in Non-Patent Document 1, conversion to a new database system can be realized without changing the application program itself by converting the SQL issued by the application program.

Kenichiro Fujiyama, Koji Kida, Satoshi Nakamura, “DBMS replacement support by SQL transformation in database connection layer”, IPSJ National Conference Proceedings, VOL. 69th, NO. 1, PAGE. 1.357-1.358, 2007

  As described above, in order to perform a further cross-sectional analysis on the analysis result of the processing module in the data analysis system such as BI, the metadata of the data is held together with the analysis result data, and is based on the metadata. It is necessary to selectively use appropriate data. Here, the processing modules are generally processing modules developed individually for various application programs, and only necessary data corresponding to the purpose of the application program is processed. Further, the schema of the database read and written by the processing module is determined in advance. For these reasons, the metadata cannot be added to the analysis result by the processing module unless the processing module is modified so that the metadata along with the data of the analysis result is registered in the database. However, since only an execution module (execution file) is provided as a processing module, the analyst cannot view the processing module source code and modify the processing module.

  The method described in Non-Patent Document 1 is a method for converting a query without modifying a processing module. This conversion is performed by a database language in a query such as reserved word conversion or logic conversion. Formal conversion. Therefore, metadata that is not included in the original query cannot be added to the analysis result record by the processing module and recorded together with the record.

  An object of the present invention is to provide a database connection system, a database connection program, and a database connection method that enable metadata to be registered in a database together with data written to the database by the processing module without modifying the processing module. To do.

  The database connection system of the present invention has a setting information storage unit that stores setting information that defines metadata items to be written to the database together with the data when the data is written to the database, and a metadata holding unit that holds the metadata. Requesting metadata for the item defined in the setting information, collecting metadata from the metadata holding means, and a query for instructing data writing to the database, the data and the setting Write query conversion means for converting into a query for instructing writing of metadata of an item defined by information.

  Further, the database connection program of the present invention is a database connection program installed in a computer provided with setting information storage means for storing setting information defining metadata items to be written to the database together with the data when the data is written to the database. Metadata that collects metadata from the metadata holding unit by requesting metadata of an item defined by setting information from the metadata holding unit that holds metadata to the computer A collection query and a write query conversion process for converting a query instructing writing of data into a database into a query instructing writing of metadata of an item defined by the data and setting information are executed. .

  The database connection method of the present invention is a database applied to a database connection system provided with setting information storage means for storing setting information that defines metadata items to be written to the database together with the data when the data is written to the database. A connection method, wherein the metadata collecting unit requests metadata of an item defined in the setting information from the metadata holding unit holding the metadata, and collects the metadata from the metadata holding unit The write query conversion means converts the query for instructing data writing to the database into a query for instructing to write metadata of items defined by the data and setting information.

  According to the present invention, not only the data written to the database by the processing module but also the metadata can be written to the database without modifying the processing module.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing an example of a database connection system of the present invention. The database connection system 10 of this embodiment includes a storage device 11, a DB access virtualization unit 100, a DB access unit 200, a process execution unit 300, and an execution control unit 400. The database connection system 10 is connected to a database system 500 including the database 501 via a communication network, and operates the database 501. The database system 500 also includes a database management system (RDBMS, not shown). DB means a database.

  Each of the database connection system 10 and the database system 500 is realized by an information processing apparatus such as a personal computer, for example.

  The storage device 11 is a storage device that stores a DB access virtualization module 101, a database access module (hereinafter referred to as a DB access module) 201, a processing module 301, and an execution control module 401. Each of these modules 101, 201, 301, 401 is software. The DB access virtualization unit 100 operates according to the DB access virtualization module 101. The DB access means 200 operates according to the DB access module 201. The process execution unit 300 operates according to the process module 301. The execution control unit 400 operates according to the execution control module 401.

An outline of the operation of the database connection system 10 will be described.
The processing module 301 is a software module that causes the processing execution unit 300 to execute processing involving data writing to the database 501 or data reading from the database 501. The storage device 11 may store a plurality of types of processing modules 301. The execution control module 401 is software that causes the execution control unit 400 to perform process module call control (for example, call order control). The process execution unit 300 calls the processing module 301 under the control of the execution control unit 400, and generates a query (command) for writing data into the database or reading data from the database. The processing module 301 generates a query described in a database language (for example, SQL).

  The DB access virtualization unit 100 is preliminarily input with setting information indicating the correspondence between the data written in the database 501 and the metadata items to be added to the data. The DB access virtualization unit 100 stores the setting information. Remember. When the process execution unit 300 creates a query for writing data, the DB access virtualization unit 100 refers to the setting information stored in advance, and not only the data designated for writing by the query, The query is converted so that metadata to be added to the data is also written in the database 501. Data instructed to be written by the processing module 301 and its metadata are written to the database 501 based on the converted query.

  The processing module 301 is a processing module for proceeding with processing without using metadata, and the processing execution means 300 that operates according to the processing module 301 does not generate a query that instructs writing of metadata to the database.

  When the DB access virtualization unit 100 reads data from the database 501, the DB access virtualization unit 100 deletes metadata from the read data. The process execution unit 300 performs a target process of the processing module 301 based on the data from which the metadata has been deleted. The DB access virtualization unit 100 prevents the metadata unrelated to the process from being sent to the process execution unit 300 by deleting the metadata.

  Also, metadata is written in the database 501 together with data instructed to be written by the query. Therefore, a data analysis system such as BI (not shown in FIG. 1) can perform data analysis with reference to data and metadata stored in the database 501.

Next, each component of the database connection system 10 will be described.
The DB access virtualization unit 100 has an interface through which setting information is input, and stores setting information input through the interface in advance. When the process execution unit 300 generates a query for writing data, the DB access virtualization unit 100 refers to the setting information, and writes the query to the database 501 in addition to the data specified by the query. Convert. Further, the execution control unit 400 has metadata, and the DB access virtualization unit 100 collects metadata to be written in the database 501 from the execution control unit 400. When the DB access virtualization unit 100 converts a query for writing data, the DB access virtualization unit 100 outputs the query to the DB access unit 200.

  When the process execution unit 300 generates a query for reading data from the database 501, the DB access virtualization unit 100 outputs the query to the DB access unit 200 without performing conversion. As a result, data is read from the database 501 and the result is returned from the DB access module 200. At this time, if the read data read from the database 501 includes metadata added by query conversion, the DB access virtualization unit 100 deletes the metadata and stores the metadata. The removed data is returned to the process execution means 300. If the read data does not include metadata, the DB access virtualization unit 100 returns the data to the process execution unit 300.

  The DB access virtualization unit 100 executes these processes according to the DB access virtualization module 101. The DB access virtualization module 101 is a database connection program of the present invention.

  When a query written in a database language such as SQL is input from the DB access virtualization module 100, the DB access means 200 rewrites the query into a command uniquely defined in the database system 500, To the database system 500. The database system 500 reads and writes data according to a unique command corresponding to the type of the database system 500.

  When the DB access unit 200 transmits a command to the database system 500, the database system 500 reads and writes data, and transmits a result value (response result to the command) to the DB access unit 200. This response result is also described in a unique expression of the database system 500. When receiving the response result from the database system 500, the DB access unit 200 rewrites the response result into a description in the database language, and sends the rewritten result to the DB access virtualization module 100.

  The DB access unit 200 executes these rewrite processes according to the DB access module 201. The DB access module 201 is, for example, a JDBC (Java DataBase Connectivity) or an ODBC (Open DataBase Connectivity) corresponding to the database system 500. Java is a registered trademark.

  The process execution unit 300 generates a query for instructing data writing to the database 501 or a query for instructing data reading from the database 501 in accordance with the processing module 301. Further, the process execution unit 300 executes a process according to the purpose of the processing module 301 using read data read from the database 501 and the like. For example, information extraction from data read from the database 501 and data analysis are performed. The content of this process is not limited. When the process execution unit 300 receives the activation notification of the process module 301 from the execution control module 400, the process execution unit 300 executes the process according to the notified process module 301. When the process is completed, the process execution unit 300 notifies the execution control unit 400 of the completion of the process.

  An example of the processing module 301 is shown below. The processing module 301 is, for example, a software module that causes the processing execution unit 300 to sequentially read data (sentences) from the database 501, execute morpheme analysis on the sentences, and write the morpheme analysis result to the database 501 as new data. It may be. The processing module 301 may be a software module that causes the processing execution unit 300 to collect Web pages via the Internet and write the Web pages as new data in the database 501. The processing module 301 is a software module that causes the processing execution unit 300 to collect information about a file from an information processing apparatus such as a personal computer that stores the file, and to write the information in the database 501 as new data. May be. The processing module 301 may be a software module that causes the processing execution unit 300 to measure the outside air temperature and writes the measurement result in the database 501.

  The execution control module 401 is software that describes or sets the cooperation of processing modules. That is, the execution control module 401 is software that defines the order of processing of one or a plurality of processing modules. The execution control module 401 is created by a user of the database connection system 10 and stored in the storage device 11. The execution control unit 400 controls the order of processing modules to be executed according to the execution control module 401. For example, the execution control unit 400 notifies the processing execution unit 300 of the processing modules to be executed in the order determined by the execution control module 401, thereby causing the processing execution unit 300 to sequentially execute the processing modules.

  A user of this system defines an application (combination of processing modules) that performs the entire intended processing by one processing module or by linking a plurality of processing modules. The user creates an execution control module 401 that defines the execution order of the processing modules in this application. The database connection system 10 includes an interface for inputting the execution control module 401, and stores the execution control module 401 in the storage device 11 when the user inputs the execution control module 401 via the interface.

  The execution control unit 400 causes the process execution unit 300 to sequentially execute the processing modules according to the execution control module 401. For example, the execution control unit 400 sends an activation notification of a processing module 301 to the process execution unit 300 according to the execution control module 401, and causes the process execution unit 300 to execute the process. When the processing execution unit 300 completes the processing of the notified processing module 301, the processing execution unit 300 sends a completion notification to the execution control unit 400. When the execution control unit 400 receives the completion notification, the processing execution unit 300 sends a startup notification of the next processing module 301. Send to 300. The execution control unit 400 repeats the above processing until there is no processing module to be executed.

  Further, the execution control unit 400 holds metadata written in the database 501 together with data generated by the process execution unit 300. Examples of metadata include, for example, the user name of a user who created data to be processed by the processing module 301, and identification information of a device that creates and holds data to be processed by the processing module 301 (for example, Device name, IP address, etc.). For example, when the processing module 301 is a file crawler for collecting files from a PC (personal computer) on a communication network, the user name of the user who created the file, the device name of the PC holding the file, etc. Meta information can be used. Further, the user name of the user who is performing an operation for causing the execution control module 401 to execute the execution control module 401 may be used as metadata. Examples of metadata include the module name of the processing module 301 to be executed by the processing execution unit 300 and the version name of the processing module 301. Further, the module name of the processing module that is currently executed immediately before the processing module 301 that is being executed by the processing execution unit 300 and the module name of the processing module that is executed next can also be used as metadata. The execution time of the processing module 301 is also metadata. The name of the application (combination of modules) (for example, the name of the execution control module 401) may be used as metadata. Further, when executing the processing module 301, information (referred to as data source information) input to the processing execution means 300 by the execution control module 401 and parameters set in the processing module 301 are held as metadata. Also good. When the processing module 301 is a program that causes the process execution unit 300 to output data to the execution control unit 400 when the process is completed or during the process, the execution control unit 400 performs the process when the process is completed or during the process. The output data may be metadata. Each piece of information such as the user name and time given here is an example of metadata, and the execution control unit 400 may hold information other than the above as metadata.

  For example, the execution control unit 400 may hold the time when the execution of the processing module 301 is instructed (that is, the time when the activation notification is sent to the processing execution unit 300) as the execution time of the processing module 301. Alternatively, the execution control unit 400 measures the time, sends an activation notification of the processing module 301 to the process execution unit 300, and then sets the time from the DB access virtualization unit 100 (specifically, the metadata collection unit 140 described later). The current time may be provided as the execution time of the processing module 301 when requested.

  The user name of the user who created the data to be processed by the processing module 301 may be clearly specified in the execution control module 401, for example. For example, the creator of the execution control module 401 instructing execution of the file crawler describes the user name of the creator of the file collected by the file crawler as metadata in advance in the execution control module 401, and executes the execution control means. 400 may read the user name and hold it as metadata. When the processing module 301 is a program that causes the process execution unit 300 to output a user name to the execution control unit 400 when the process is completed or during the process, the execution control module 401 outputs from the process execution unit 300. It is also possible to hold the user name that has been set.

  Further, in order to use the user name of the user who is performing the operation of causing the execution control module 401 to execute the execution control module 401 as metadata, the execution control unit 400 may hold the user name as follows. That is, the execution control unit 400 converts the user name into metadata when a login operation for inputting the user name is performed in an apparatus (the database connection system 10 in the example shown in FIG. 1) including the execution control unit 400 itself. As long as it holds.

  Upon receiving the metadata acquisition request from the DB access virtualization unit 100, the execution control unit 400 sends the requested metadata to the DB access virtualization unit 100.

  Examples of the processing module 301 and the execution control module 401 will be described. For example, each processing module 301 may be a program, and the execution control module 401 may be a batch file (for example, a shell script in UNIX) that defines the execution order of each program. UNIX is a registered trademark. Further, the processing module 301 may be a module called by a DLL (Dynamic Link Library), and the execution control module 401 may be a file for calling the module. Further, the processing module 301 may be a class library in Java, and the execution control module 401 may be information for calling the class library.

  When the storage device 11 stores a plurality of types of processing modules 301, the DB access virtualization module 101 and the DB access module 201 are also prepared for each processing module 301, and the storage device 11 corresponds to each processing module 301. Storing module 101 and DB access module 201 are stored. For example, the storage device 11 stores the processing module 301, the DB access virtualization module 101, and the DB access module 201 in this way. Alternatively, even if a plurality of types of processing modules 301 are prepared, the DB access virtualization module 101 and the DB access module 201 may each be stored in the storage device 11 as a single module.

  The database system 500 reads or writes data from the database 501 in accordance with a command for instructing reading or writing of data received from the DB access unit 200.

  Further, an integrated data analysis system (not shown in FIG. 1) that can refer to the database 501 is connected to the database system 500. The integrated data analysis system is, for example, a BI, and comprehensively analyzes the execution results of the various processing modules 301 by referring to the data registered in the database 501 by the execution of the various processing modules 301 and the metadata registered together with the data. To do. The processing content of analysis performed by the integrated data analysis system is not particularly limited.

  FIG. 2 is a block diagram showing the DB access virtualization unit 100. The DB access virtualization unit 100 includes a query rewrite unit 110, a rewrite setting unit 120, a rewrite setting storage unit 130, a metadata collection unit 140, and a metadata storage unit 150.

  The rewrite setting unit 120 has an interface (for example, an input device such as a keyboard) to which setting information indicating a correspondence relationship between data written in the database 501 and metadata items to be added to the data is input. Is stored in the rewrite setting storage means 130. The rewrite setting storage unit 130 is a storage device that stores setting information.

  In the setting information, for each table of the database 501, metadata items to be written together with data to be written to the table are determined. Therefore, the data to be written to the table is associated with the metadata item.

  3 and 4 are explanatory diagrams showing examples of setting information. The setting information illustrated in FIG. 3 indicates that when data is written in the table A, “user name” and “execution time” are written as metadata together with the data. “User name” is, for example, the user name of the user who created the data to be processed by the processing module 301, but may be the user name of another user. The “execution time” is the execution time of the processing module 301 that causes the process execution unit 300 to generate a query for writing data to the table A.

  In the setting information, the metadata item is described so that the metadata field in the table to which the metadata is written can be specified. For example, metadata field names in a table to which metadata is written are described in the setting information as metadata items. In the example shown in FIG. 3, “user name” and “execution time” that are metadata field names in the table A are described in the setting information.

  Since the process execution unit 300 does not handle metadata, the table in the database 501 is not initially provided with a metadata field. For this reason, for example, the user manually checks the schema of the database 501 and changes the table to add metadata fields.

  The metadata may be stored in a table different from the table specified by the query generated by the process execution unit 300. In the setting information illustrated in FIG. 4, when data is written to the table B, a specific field (“data 1” in this example) in the table B is used as an external key, and the external key (“data 1”) is stored in another table C. ) Key value, “user name”, and “execution time”. In this example, when “data 1” and “data 2” are written in the table B, the “user name” and “execution time” as metadata are written in another table C. The setting information describes that “user name” and “execution time” are metadata items.

  3 and 4 show examples of setting information that determines metadata items for each table. However, when there is no need to specify a table (for example, one processing target for each processing module 301 is one table). In some cases, only metadata items may be defined.

  The metadata collection unit 140 requests the execution control unit 400 for various metadata defined in the setting information, and collects metadata from the execution control unit 400.

  The metadata storage unit 150 causes the metadata collection unit 140 to collect metadata. The metadata accumulating unit 150 includes a storage device that stores metadata, and stores the metadata collected by the metadata collection unit 140 in the storage device. Further, the metadata accumulating unit 150 provides the stored metadata to the query rewriting unit 110 in response to a request from the query rewriting unit 110.

  When a data write query is generated by the process execution unit 300, the query rewrite unit 110 refers to setting information stored in the rewrite setting storage unit 130, and adds metadata to the data instructed to write by the query Identify items. For example, it is assumed that the table in the database 501 and the metadata item are associated with each other as in the setting information illustrated in FIGS. 3 and 4. The query rewriting means 110 identifies the metadata item corresponding to the table specified by the query by referring to the setting information. The query rewriting unit 110 requests the metadata corresponding to the item from the metadata storage unit 150. When the query rewriting unit 110 receives the metadata, the query rewriting unit 110 converts the query so that the metadata is also written in the database 501, and converts the query after conversion. It is sent to the DB access means 200.

  Assume that the setting information illustrated in FIG. 3 is defined and a query that specifies data writing to the table A is generated. Then, the query rewriting unit 110 identifies items such as “user name” and “execution time” associated with the table A from the setting information, and sets metadata (for example, “User1”) of the items as metadata. The query is converted so as to be received from the storage means 150 and to write their metadata into the table A as well. For example, metadata field names and metadata in the database 501 are added to the query. By converting the query in this way, metadata such as the user name and execution time is also written in the table A in the database 501 as shown in the lower part of FIG.

  Further, for example, it is assumed that the setting information illustrated in FIG. 4 is defined and a query that specifies data writing to the table B is generated. Then, the query rewriting unit 110 identifies items such as “user name” and “execution time” associated with the table B from the setting information, and sets metadata (for example, “User1”) of the items as metadata. Receive from storage means 150. Then, the query rewriting means 110 newly creates a query for writing the key value of the foreign key included in the query for designating data writing to the table B and the metadata of “user name” and “execution time” to the table C. To create. By the query obtained as a result, each data and metadata are written in the tables B and C in the database 501 as shown in the lower part of FIG.

  The query rewriting unit 110 may convert the query generated by the process execution unit 300 into a query including each field name of data and metadata to be written in the database 501 and the data and metadata. The conversion may be a conversion for adding a new query as described with reference to FIG.

  Further, when the processing execution unit 300 generates a query for reading data from the database 501, the query rewriting unit 110 sends the query to the DB access unit 200 without converting the query. When read data read from the database 501 is returned from the DB access means 200, the query rewrite means 110 refers to the setting information, and the metadata specified by the setting information is included in the read data. It is determined whether or not. In the setting information, metadata items are described so that metadata fields in the table can be specified. For example, field names are described as metadata items. The query rewriting unit 110 may refer to the setting information to identify the field name of the metadata and determine whether or not the data of the field is included in the read data. If the read data includes metadata defined by the setting information, the query rewriting unit 110 deletes the metadata from the read data and returns the data from which the metadata has been deleted to the process execution unit 300. If the read data does not include metadata, the query rewrite unit 110 sends the data to the process execution unit 300.

  FIG. 1 shows a case where the storage device 11 stores the DB access virtualization module 101 corresponding to each processing module 301. When the DB access virtualization module 101 is created for each processing module 301, a portion corresponding to a command for storing the input setting information in the rewrite setting storage unit 130 may be created as a common module. Then, a command portion for the query rewriting unit 110, the metadata collection unit 140, and the metadata storage unit 150 may be created for each processing module 301 and stored in the storage device 11.

  The query rewriting unit 110, the metadata collection unit 140, the DB access unit 200, the process execution unit 300, and the execution control unit 400 are realized by a CPU that operates according to a program, for example. The rewrite setting unit 120 is realized by, for example, a CPU that operates according to a program and an input device such as a keyboard. The metadata storage unit 150 is realized by a CPU that operates according to a program and a storage device. The same CPU may be sufficient as CPU for implement | achieving said each means.

  In the above description, as an example of adding a metadata field to a table in the database 501, a case where a user manually adds a metadata field is illustrated. Processing for adding a metadata field to a table in the database 501 may be automatically performed. For example, a module that causes the DB access virtualization unit 100 (more specifically, the query rewriting unit 110) to execute a process of adding a metadata field defined by setting information to a query instructing generation of a new table Remember me. When the query rewriting unit 110 generates a query (for example, an SQL create table statement) that instructs generation of a new table, the setting information is referred to, and the metadata item associated with the new table is displayed. If there is an item of metadata by searching, the query may be converted to add the item to the table as a new field. When there is no table information in the setting information and only an item of metadata is specified, a query for instructing generation of a new table may be rewritten so that the item is added to the table as a new field. Here, the database connection system 10 may be provided with means (not shown) for creating a query that instructs generation of a new table. Alternatively, a query instructing generation of a new table may be input by a device other than the database connection system 10 or a user of the database connection system 10.

  Further, the module for converting the query so that the metadata field is added to the new table is not created as a module separate from the DB access virtualization module 101, but the DB access virtualization module 101 and They may be created as the same module. That is, the DB access virtualization module 101 may include an instruction that causes the DB access virtualization unit 100 to convert a query so that a metadata field is added to the new table.

  Next, another configuration example of the database connection system of the present invention is shown. In FIG. 5 and FIG. 6 described below, the same components as those shown in FIG. 1 are denoted by the same reference numerals as those in FIG. FIG. 5 is a block diagram illustrating a configuration example when the database system and the database connection system 10 are the same device. As shown in FIG. 5, the database connection system 10 may hold the database 501.

  FIG. 6 is a block diagram illustrating a configuration example when the processing modules 301 are distributed among a plurality of database connection systems 10. In this case, each database connection system 10 includes DB access means 200, DB access virtualization means 100, and process execution means 300. Each database connection system 10 includes a storage device 11 that stores a DB access module 201, a DB access virtualization module 101, and a processing module 301. In addition, an information processing apparatus 70 including a storage device 71 that stores the execution control module 401 and an execution control unit 400 is provided. The information processing apparatus 70 and each database connection system 10 are connected via a communication network. The execution control unit 400 sequentially transmits a processing module activation notification to be executed to the process execution unit 300 of each database connection system 10 according to the execution control module. When the processing of the processing module 301 notified to the execution control unit 400 is completed, the process execution unit 300 transmits a completion notification to the execution control unit 400. Further, the DB access virtualization unit 100 transmits the metadata to be collected to the execution control unit 400 via the communication network. The execution control unit 400 transmits the requested metadata to the DB access virtualization unit 100. 6 shows a case where the information processing apparatus 70 including the execution control unit 400 is provided separately from each database connection system 10, but any one of the plurality of database connection systems 10 stores the execution control module 401. In addition, the execution control means 400 may be provided.

Next, the operation will be described.
FIG. 7 is a flowchart showing an example of operations of the database connection system 10 and the integrated data analysis system (not shown in FIG. 1). First, the user of the database connection system 10 determines an application for performing a desired process by one processing module or by linking a plurality of processing modules. The application is a combination of processing modules, and the user creates an execution control module that defines the execution order of the processing modules, and inputs the execution control module to the database integrated system 10. When the execution control module 401 is input, the database integration system 10 (see FIG. 1) stores the execution control module 401 in the storage device 11 (step S1).

  Next, the user creates setting information. Data that is not used by the processing execution unit 300 that operates according to the processing module 301 at the time of processing execution and that does not instruct writing to the database 501 and that the integrated data analysis system uses at the time of analysis is used as metadata by the user. Determine. Then, the user creates setting information indicating the correspondence between the data that the process execution unit 300 instructs to write to the database 301 and the metadata items. When the setting information is input, the database connection system 10 stores the setting information (step S2).

Subsequently, the process execution unit 300 executes each process module 301 defined as an application by the user (step S3). At this time, the execution control unit 400 controls the execution order of the processing modules 301. In step S <b> 3, the process execution unit 300 performs the process defined by each process module 301 while reading data stored in the database 501 and writing data to the database 501. When the process execution unit 300 generates a query that instructs to write data to the database 501, the DB access virtualization unit 100 converts the query so that metadata corresponding to the data is also written to the database 501. The DB access means 200 converts the converted query into a command unique to the database system 500, and the database system 500 stores data and metadata according to the command. As a result, metadata is also written in the database 501 in addition to the data that the process execution means 300 is trying to write.

  When the database connection system 10 executes a plurality of applications, the processes in steps S1 to S3 may be repeated.

  The integrated data analysis system uses the metadata accumulated in the database 501 to analyze the data (step S4). For example, after a plurality of applications are executed, the integrated data analysis system performs analysis using the data and the entire metadata stored in the database 501 by executing each application.

  Next, the processing in steps S2 and S3 will be described in more detail. 8 to 10 are flowcharts showing the processes of steps S2 and S3 in more detail. Specifically, FIG. 8 shows the operation from when the setting information is input to when the application execution is started. FIG. 9 is a flowchart illustrating an example of an operation when the process execution unit 300 generates a query for instructing writing of data into the database. FIG. 10 is a flowchart illustrating an example of an operation when the process execution unit 300 generates a query for instructing reading of data from the database.

  The metadata provided by the execution control unit 400 to the DB access virtualization unit 100 may or may not change as the process execution unit 300 sequentially executes each processing module 301. An example of the former is the execution time of the processing module 301. As an example of the latter, there is a name of an application (a combination of modules) (for example, a name of the execution control module 401). In the following description, an example will be described in which the DB access virtualization unit 100 first collects the latter metadata, and the process execution unit 300 collects the latter metadata when the processing module 301 is executed. .

  First, the operation up to the start of application execution will be described with reference to FIG. The user creates setting information indicating a correspondence relationship between data written in the database 501 and metadata items. In the rewrite setting unit 120, setting information is input by the user (step S11). Then, the rewrite setting unit 120 stores the input setting information in the rewrite setting storage unit 130 (step S12).

  Thereafter, the execution control unit 400 activates the application (step S13). That is, the process of sequentially sending the activation notifications of the processing modules 301 to the processing execution unit 300 in the order determined by the execution control module 401 is started.

  Further, when the processing execution unit 300 first receives the activation notification of the processing module 301 from the execution control unit 400, for example, the processing execution unit 300 notifies the metadata collection unit 140 via the query rewriting unit 110. When notified that the processing module 301 has received the first activation notification, the metadata collection unit 140 refers to the setting information stored in the rewrite setting information storage unit 130 and identifies the item of the metadata. The metadata collection unit 140 sends metadata items that do not change with the progress of processing of the processing execution unit 300 among the metadata items to the execution control unit 400 and requests the metadata. In the setting information, the user may determine whether the metadata changes along with the processing progress of the processing execution means 300 together with the metadata item. The metadata collection unit 140 identifies metadata items that are determined not to change, and requests the metadata from the execution control unit 400. The execution control unit 400 sends the requested metadata to the metadata collection unit 140, and the metadata collection unit 140 stores the metadata received from the execution control unit 400 in the metadata storage unit 150 (step S14).

  When the processing execution unit 300 receives the activation notification of the processing module 301 from the execution control unit 400, the processing execution unit 300 starts the processing of the designated processing module 301. The process execution means 300 generates a query for writing data into the database 501 or a query for reading data from the database 501 according to the processing module 301. The operation when these queries are generated will be described below.

  First, with reference to FIG. 9, the operation when the process execution unit 300 generates a query for writing data will be described. The processing execution means 300 performs processing according to the designated processing module 301, and generates a query that instructs writing of data to the database 501 in the processing (step S15). The process execution unit 300 generates a query using a database language such as SQL, for example.

  When a query for instructing data writing is generated, the query rewriting unit 110 (see FIG. 2) included in the DB access virtualization unit 100 analyzes the query (step S16). Specifically, the query rewriting unit 110 reads the table name of the table to which data is written from the query. Since the query format is defined in the database language, the query rewriting means 110 may read the table name from a predetermined location of the query.

  The query rewrite unit 110 refers to the setting information stored in the rewrite setting storage unit 130, and identifies the metadata item to be written to the database together with the data instructed to be written in the query. That is, the query rewriting unit 110 reads the metadata item associated with the table read in step S16 from the setting information. Further, the query rewriting unit 110 selects an item of metadata that changes with the progress of processing of the processing execution unit 300 from among the items. The query rewriting unit 110 specifies the selected item and requests the metadata storage unit 150 to store the metadata of the item (step S17). As already described, it is only necessary to determine whether the metadata changes with the progress of the process of the process execution means 300 together with the metadata item in the setting information. Then, the query rewriting unit 110 may select an item of metadata that is determined to change.

  The execution time or the like of the processing module 301 changes with the progress of the processing of the processing execution unit 300, and these metadata are not fixed at the time of step S14, and are stored in the metadata storage unit 150 in step S17.

  When a metadata item is specified by the query rewriting unit 110, the metadata storage unit 150 notifies the metadata collection unit 140 of the item and causes the metadata collection unit 140 to collect metadata. The metadata collection unit 140 requests metadata of the notified item from the execution control unit 400, and the execution control unit 400 sends the requested metadata to the metadata collection unit 140. The metadata collection unit 140 stores the metadata received from the execution control unit 400 in the metadata storage unit 150.

  After step S17, the query rewriting unit 110 requests metadata to be written to the database together with the data instructed to be written in the query to the metadata storage unit 150. The metadata item to be written in the database is an item associated with the table in the setting information, and this item is specified in step S17. The query rewriting unit 110 requests the metadata storage unit 150 for the metadata of the item. The metadata storage unit 150 sends the requested metadata to the query rewrite unit 110. Further, the query rewriting unit 110 converts the query generated in step S15 into a query including each field name of data and metadata to be written in the database 501 and the data and metadata (step S18).

  After step S18, the query rewriting unit 110 passes the converted query to the DB access unit 200 (step S19). The DB access unit 200 rewrites the query into a command uniquely defined in the database system 500 and transmits the command to the database system 500. As a result, the database system 500 writes the specified data and metadata into the database 501 in accordance with the command received from the DB access means 200. Further, the database system returns the processing result to the DB access means 200. The DB access unit 200 rewrites the processing result described in a format unique to the database system 500 into a format determined by the database language. The DB access virtualization unit 100 sends the processing result to the process execution unit 300 (step S20).

  Here, the case of collecting metadata in steps S14 and S17 has been described as an example. However, regardless of whether the metadata changes with the progress of the processing of the processing execution unit 300, each step in step S17 Metadata may be collected. However, it is preferable to collect metadata in steps S14 and S17 because the metadata can be collected efficiently.

  Next, with reference to FIG. 10, an operation when the process execution unit 300 generates a query for reading data will be described. The processing execution means 300 performs processing according to the designated processing module 301, and generates a query that instructs reading of data from the database 501 in the processing (step S21). The process execution unit 300 generates a query using a database language such as SQL, for example.

  When a query for instructing data reading is generated, the query rewriting means 110 (see FIG. 2) of the DB access virtualization means 100 sends the query to the DB access means 200 without conversion (step S22). The DB access unit 200 rewrites the query into a command uniquely defined in the database system 500 and transmits the command to the database system 500. As a result, the database system 500 reads data from the database 501 according to the command received from the DB access means 200 and transmits it to the DB access means 200. At this time, the read data read from the database 501 may include metadata added at the time of data writing.

  Data transmitted from the database system 500 to the DB access means 200 is described in a format unique to the database system 500. The DB access means 200 rewrites the data into a format determined by the database language and sends it to the query rewrite means 110.

  When the query rewrite means 110 receives the read data from the DB access means 200, the query rewrite means 110 determines whether or not the read data contains metadata. For example, the field name of the metadata is specified with reference to the setting information, and it is determined whether or not the data of the field is included in the read data. If there is metadata field data in the read data, the query rewriting means 110 deletes the metadata (step S23). Then, the query rewriting unit 110 sends the data after deleting the metadata to the process execution unit 300 (step S24). If there is no metadata in the read data, the query rewrite unit 110 sends the data to the process execution unit 300 as it is.

Next, the effect of this embodiment will be described.
The database connection system according to the present invention stores in advance setting information that defines items of metadata to be written to the database 501 together with data to be written by a query. The metadata collection unit 140 collects metadata from the execution control unit 400. When a query instructing data writing to the database 501 is generated, the query rewriting unit 110 writes not only the data instructed by the query but also the metadata defined by the setting information to the database 501. Convert. Therefore, metadata can be written in the database 501 without modifying the processing module 301. Further, the integrated data analysis system (not shown) can analyze the execution result of each processing module 301 by using the data and metadata written in the database 501 by the execution of each processing module 301.

  Further, when the processing execution unit 300 executes the processing module 301 and generates a query instructing to read data from the database 501, and the read data read based on the query includes metadata, the DB The access virtualization unit 100 deletes metadata from the read data. Therefore, it is possible to prevent the metadata from being sent to the process execution unit 300 that does not perform the process on the metadata.

  In addition, setting information is input to the rewrite setting unit 120, and the rewrite setting unit 120 stores the input setting information in the rewrite setting storage unit 130. Therefore, the user can input desired setting information and add metadata based on the setting information.

  In the setting information, the execution time of the processing module is preferably defined as a metadata item. In this case, the metadata collecting unit 140 collects the execution time of the processing module from the execution control unit 400, and the query rewriting unit 110 rewrites the query so that the execution time is also written in the database. As a result, the integrated data analysis system (not shown) analyzes how the data amount and quality of the data written in the database 501 are converted along the time series of the execution time of the processing module 301. be able to.

  In the setting information, it is preferable that the user name of the user who created the data to be processed by the processing module 301 is defined as a metadata item. In this case, data analysis can be performed for each user. For example, it is assumed that the processing module 301 is a file crawler and the process execution unit 300 collects files. In this case, if the user name is defined as metadata in the setting information, the integrated data analysis system (not shown) can perform file analysis for each file creator.

  In the above-described embodiment, the case where the metadata is deleted from the read data read from the database 501 has been described. When the processing execution unit 300 generates a query that includes symbols indicating data of all fields and instructs the reading of the data of all fields, the query rewriting unit 110 rewrites the query, Data may not be read. Also in this case, it is possible to prevent the metadata from being sent to the process execution unit 300.

  In this case, the query rewriting unit 110 stores all the field names specified by the query generated by the process execution unit 300 and instructing to write data to the database 501. When the processing execution unit 300 generates a query including a symbol (for example, “*”) indicating data of all fields and instructing reading of the data of all fields, it means the data of all fields. May be replaced with a stored field name (each field name specified in a query instructing to write data). In this case, the data of all fields is not read from the database 501, but the data of the field specified by the query for writing data (that is, data other than the metadata) is read. Therefore, the query rewriting unit 110 does not have to delete the metadata from the read data. Further, it is possible to prevent the metadata from being sent to the process execution unit 300.

  Next, a specific example of the operation of the present invention will be described. FIG. 11 is a block diagram illustrating an example of a database connection system. Components similar to those shown in FIG. 1 are denoted by the same reference numerals as those in FIG. Each of the PC file crawler 301a and the analysis module 301b shown in FIG. 11 corresponds to the processing module 301 shown in FIG. Hereinafter, the case where the process execution unit 300 operates according to the PC file crawler 301a and the analysis module 301b will be described as an example. FIG. 11 illustrates a case where the DB access module 201 is a JDBC.

  The PC file crawler 301a is a processing module that causes the process execution unit 300 to collect information on files (Web pages) stored in a PC (personal computer) on a communication network. The process execution unit 300 collects information about files stored in the PC on the intranet according to the PC file crawler 301a, and generates a query for writing the information to the database 501. Here, it is assumed that the information collected by the process execution means 300 includes file position information in the PC and the text of the file (text). The file position information includes the folder name and file name of the folder in which the file is stored.

  The analysis module 301b is a processing module that causes the process execution unit 300 to read information stored in the database 501 (information related to a file) and analyze the information. The process execution means 300 performs processes such as obtaining the file size from the position information of the file in the PC and the text (text) of the file and creating a search index according to the analysis module 301b.

  A file search (PC desktop search) of each PC can be realized by the combination of the PC file crawler 301a and the analysis module 301b, and the combination of the PC file crawler 301a and the analysis module 301b is prepared for each PC.

  The integrated data analysis system 91 is an analysis system that refers to the database 501 and analyzes information on files collected during a certain period to analyze changes in time series such as file delivery status between PCs.

  The PC file crawler 301a causes the process execution unit 300 to execute a process of collecting file position information and a set of text from the PC, adding the file ID, and writing the data to the database 501. FIG. 12 is an explanatory diagram showing an example of a table used in such processing. The PC file crawler 301a does not cause the processing execution unit 300 to perform processing for writing metadata or processing for reading metadata. Therefore, the table in which the process execution unit 300 operating according to the PC file crawler 301a reads and writes data includes file ID, position information, and text fields as illustrated in FIG. 12, and includes metadata fields. Is not included. The database connection system 10 may include means (not shown) for generating a query that instructs to create a new table illustrated in FIG. Alternatively, a query instructing to create a new table may be input by a device other than the database connection system 10 or a user of the database connection system 10.

  The DB access virtualization unit 100 may send a query for instructing the creation of a new table to the DB access unit 200 as it is. In this case, the database system 500 creates a table having fields illustrated in FIG. In this example, it is assumed that the integrated data analysis system 91 uses the date and time when the file is collected and the PC name that is the collection target of the file as metadata. For this reason, it is necessary to add a file collection date and a PC name field as a file collection target to the table illustrated in FIG. FIG. 13 is an explanatory diagram showing an example of a table in which fields of date and time as metadata are added to the table illustrated in FIG. In order for the query rewriting unit 110 to convert a query so that metadata is also written, and to enable writing of metadata to the database by the query, a table illustrated in FIG. 13 must be prepared. After creating the table illustrated in FIG. 12, for example, the user of the database connection system 10 may add the date / time and PC name fields manually. Further, the user of the database connection system 10 inputs the setting information determined so that the date and the PC name are added to the data to be written in the database 501 to the rewrite setting unit 120, and the rewrite setting unit 120 is input. The set information is stored in the rewrite setting storage means 130.

  Further, the DB access virtualization unit 100 may automatically perform the process of adding the metadata field to the table instead of the manual operation by the user. In this case, it is assumed that metadata items are determined in advance in the setting information. The query rewriting unit 110 of the DB access virtualization unit 100 converts the query so that a metadata item defined in the setting information is added to the new table as a new field in response to the query instructing generation of the new table. To do. For example, it is assumed that an SQL statement “create table t1 (id int, position information text, text text)” is input as a query for newly creating the table shown in FIG. In the setting information, it is assumed that the date and time and the PC name are defined as metadata. The query rewriting means 110 adds the metadata field name to the above SQL statement so that the metadata item is added to the new table as a new field. In other words, the field name of the metadata is added to the part describing the field name in the SQL statement (create table statement) instructing generation of a new table (that is, in the parentheses of the create table statement). Since the create table statement describes the field name and data type, the query rewrite means 110 may add the metadata field name and data type to the create table statement. For example, the query rewriting unit 110 converts the input SQL sentence into an SQL sentence of create talbe t1 (id int, position information text, text text, PC name text, time time). The data type may be determined in the setting information, for example. The query rewriting unit 110 sends the converted query to the DB access unit 200. As a result, the table illustrated in FIG. 13 is automatically created.

  Next, an operation in which the process execution unit 300 collects files (Web pages) according to the PC file crawler 301a will be described. The process execution unit 300 sequentially collects Web pages from the Internet, and generates a query (for example, an SQL insert into statement) that writes the ID, position information, and text (text) of the collected Web pages to a table. In the insert into statement, first specify the table name, specify the field name with the next parenthesis, and specify the data to be written in the parentheses following the string value that follows. As the ID, a numerical value indicating the order of the collected Web pages may be used. The process execution means 300 collects Web pages and inserts, for example, insert into t1 (id, position information, text) values (100, 'c: \ foo \ index.html', '<html> ... </ html > ') Etc. is generated. In this way, the process execution unit 300 generates a query including the field name of the data to be written and the data.

  The DB access virtualization unit 100 adds metadata field names (PC name, date and time) and metadata to the query. That is, the DB access virtualization unit 100 converts the above insert into statement into insert into t1 (id, location information, text, PC name, time) values (100, 'c: \ foo \ index.html', '< html> ... </ html> ',' PC1 ',' 2007/1/1 10:00:01 '). In the converted query, in addition to the field name and data of the data defined from the beginning, the metadata field name (PC name, time) and its metadata ('PC1', '2007/1/1 10: 00:01 ') is also included.

  The processing of the DB access virtualization unit 100 will be described more specifically. It is assumed that setting information “add date as DATE; add pcname as PC name” is input to the rewrite setting unit 120 in advance by the user, and the rewrite setting unit 120 stores the setting information in the rewrite setting storage unit 130. The above “add date as DATE; add pcname as PC name” is an example of setting information, and the setting information may be described in another format. The metadata collection means 140 requests the execution control means 400 for metadata items (pcname and date) defined by the setting information. In response to this request, the execution control unit 400 provides the PC name and the write time (current time) to the database to the metadata collection unit 140, and the metadata collection unit 140 supplies the PC name and time to the metadata storage unit 150. To accumulate. Further, the query rewriting unit 110 requests the metadata storage unit 150 for the PC name and date determined by the setting information, and acquires the PC name and date. Then, by adding the metadata field name and metadata to the query generated by the process execution means 300, insert into t1 (id, location information, text, PC name, time) values (100, 'c : \ foo \ index.html ',' <html> ... </ html> ',' PC1 ',' 2007/1/1 10:00:01 ') The data is sent to the DB access means 200.

  The DB access means 200 converts the converted query into a command unique to the database system 500, and transmits the converted command to the database system 500. In accordance with the command, the database system 500 writes a record in which metadata (PC name and date / time) is added to the ID, position information, and text specified by the process execution unit 300 in the database 501.

  Further, when the process execution unit 300 operates in accordance with the analysis module 301b, the process execution unit 300 generates SQL for reading data in the database 501 and performs analysis processing. At this time, metadata that is not processed by the analysis module 301b may also be read from the database 501 by SQL for reading data. For example, when all the data in the database written by the PC file crawler 301a is read and sequentially analyzed, a query using symbols representing the data of all fields is generated. For example, an SQL statement such as “select * from t1” is generated using a “*” symbol indicating data of all fields. This SQL statement means reading all data in the table t1. Since the analysis module 301b does not handle metadata, even if the above SQL statement is generated for the purpose of reading all data from the table illustrated in FIG. 12, the PC name and date / time are actually determined from the table illustrated in FIG. Will also be loaded. The DB access means 200 converts “select * from t1” into a command unique to the database system 500, and the database system 500 uses the ID, position information, text, PC name, and ID from the table illustrated in FIG. 13 based on the command. Read all date and time data. The read data read by the database connection system 500 is sent to the DB access virtualization means 100 via the DB access means 200. The DB access virtualization unit 100 refers to the setting information to determine that the PC name and date / time are metadata, and from the read data, data whose field name is the PC name and data whose field name is the date / time. And delete. Then, the deleted data is sent to the process execution means 300. As a result, ID, position information, and text that are not metadata are sent to the process execution unit 300, and the process execution unit 300 performs an analysis process using these data.

  Note that the query rewriting unit 110 performs conversion by replacing the symbol “*” indicating the data of all fields in “select * from t1” generated by the processing execution unit 300 with the field names of the data of each field other than the metadata. You may go. For example, a query “select * from t1” may be converted into a query such as “select id, url, body from t1 where PC name like PC1”. Then, data may be read based on the converted query.

  When the query for data reading generated by the process execution unit 300 is rewritten as described above, the query rewrite unit 110 uses the field name (id, position) specified in the query for data writing generated by the process execution unit 300. Information, PC name). And if the query for reading data contains a symbol (for example, “*”) that means the data of all fields, the symbol is used for all field names specified in the query for writing data. Replace with (id, location information, PC name). In this case, since data other than metadata is read out, it is not necessary to delete the metadata.

  By such an operation, even if the PC file crawler 301a and the analysis module 301b are not modified, metadata (PC name and date / time in this example) used when the integrated data analysis system performs analysis is also written in the database 501. Can be made. Further, when reading data from the database 501, it is possible to select only data not to be processed by the processing module and prevent unnecessary metadata from being sent to the processing execution means 300.

  Note that by adding the PC name and date and time as metadata as in the above example, the integrated data analysis system 91 analyzes from which PC the file has been moved to which PC over time, etc. Can do.

  For example, it is assumed that the file “file1.txt” is collected from the PC1 and the file “file2.txt” is collected from the PC2 at the time t1. It is assumed that information regarding “file1.txt” is written in the database 501 together with the metadata t1 and PC1, and information regarding “file2.txt” is written in the database 501 along with the metadata t1 and PC2. Similarly, it is assumed that the file “file1.txt” is collected from the PC1 and PC3 and the file “file2.txt” is collected from the PC3 at time t2. Information about “file1.txt” is written into the database 501 together with the metadata t2 and PC1, information about “file1.txt” is written into the database 501 together with the metadata t2 and PC3, and “file2. It is assumed that information regarding “txt” is written in the database 501. Based on these data, the integrated data analysis system 91 can analyze that “file1.txt” has been copied from PC1 to PC3 and “file2.txt” has been moved from PC2 to PC3.

  Next, the outline of the present invention will be described. FIG. 14 is an explanatory diagram showing an outline of the present invention. The database connection system 10 of the present invention includes setting information storage means 51, metadata collection means 52, and write query conversion means 53.

  The metadata holding unit 54 holds the metadata and provides the metadata to the metadata collecting unit 52 in response to a request from the metadata collecting unit 52.

  The setting information storage unit 51 stores setting information. The metadata collection unit 52 requests the metadata holding unit 54 for metadata of items determined by the setting information, and collects metadata from the metadata holding unit 54.

  Then, the write query conversion means 53 converts the query for instructing writing of data into the database into a query for instructing writing of metadata of items defined by the data and setting information.

  Thus, in the present invention, the generated query is converted instead of the processing module that generates the query. Further, in the present invention, the query for instructing the writing of data to the database is converted into a query for writing the data and the metadata of the item defined in the setting information. Both metadata can be written to the database. Therefore, in the present invention, it is possible to write both the data instructed by the query generated by the processing module and the metadata to the database without modifying the processing module.

  Further, the above embodiment discloses a configuration including metadata deleting means for deleting metadata of items defined by setting information from read data read from a database.

  In the above embodiment, the field name storage means for storing all the field names specified by the query for instructing the data writing to the database, and all the fields included in the query for instructing the reading of the data for all fields. There is also disclosed a configuration including reading query conversion means for converting the query by replacing symbols representing data with respective field names stored in the field name storage means.

  Further, in the above embodiment, table generation for converting a query by adding a metadata item defined in the setting information to the field name of the new table specified by the query instructing generation of the new table. A configuration including query conversion means is disclosed.

  Further, in the above-described embodiment, when the first software among the one or a plurality of software that generates a query instructing writing of data to the database is started by the metadata collecting unit, each software is sequentially executed. A configuration is disclosed that collects metadata that does not change with the progress of execution and collects metadata that changes with the progress of each software when each software is started up. Yes.

  In the above embodiment, a configuration is disclosed that includes setting information input means for inputting setting information, and setting information registration means for storing the setting information input to the setting information input means in the setting information storage means. Yes.

  The present invention is suitably applied to a system for registering data other than data specified by a query in a database without modifying a processing module that generates the query.

It is a block diagram which shows the example of the database connection system of this invention. It is a block diagram which shows DB access virtualization means. It is explanatory drawing which shows the example of setting information. It is explanatory drawing which shows the example of setting information. It is a block diagram which shows the structural example in case a database system and a database connection system are the same apparatuses. It is a block diagram which shows the example of a structure in case a processing module is distributed by the several database connection system. It is a flowchart which shows the example of operation | movement of a database connection system and an integrated data analysis system. It is a flowchart which shows the example of operation | movement from the setting information input to the time of application execution start. It is a flowchart which shows the example of operation | movement when a process execution means produces | generates the query which instruct | indicates writing of the data to a database. It is a flowchart which shows the example of operation | movement when a process execution means produces | generates the query which instruct | indicates reading of the data from a database. It is a block diagram which shows the example of a database connection system. It is explanatory drawing which shows the example of the table used by the process of PC file crawler. It is explanatory drawing which shows the example of the table which added the date and time field used as metadata to the table illustrated in FIG. 12, and a file name. It is explanatory drawing which shows the outline | summary of this invention. It is explanatory drawing which shows the example of the database connection system relevant to this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Database connection system 11 Storage apparatus 100 DB access virtualization means 101 DB access virtualization module 200 DB access means 201 DB access module 300 Process execution means 301 Processing module 400 Execution control means 401 Execution control module 500 Database system 501 Database

Claims (18)

Setting information storage means for storing setting information that defines metadata items to be written to the database together with the data when the data is written to the database;
A metadata collection unit that requests metadata of an item defined in setting information to a metadata holding unit that holds metadata, and collects metadata from the metadata holding unit;
A database connection system comprising: a write query conversion unit that converts a query instructing writing of data into a database into a query instructing writing of metadata of an item defined by the data and setting information. The database connection system according to claim 1, further comprising metadata deleting means for deleting metadata of an item determined by setting information from read data read from a database. Field name storage means for storing all field names specified in a query for instructing writing of data to the database;
Read query conversion means for converting the query by replacing symbols representing the data of all fields included in the query instructing reading of data of all fields with field names stored in the field name storage means. The database connection system according to claim 1. The table generation query conversion means for converting the query by adding an item of metadata defined by setting information to the field name of the new table specified by the query instructing generation of the new table. The database connection system according to any one of claims 3 to 4. Metadata collection means
Collects metadata that does not change over time as each software is executed sequentially when the first of the software or software that generates a query that directs the writing of data to the database is launched And
The database connection system according to any one of claims 1 to 4, wherein when the individual software is activated, metadata that changes as the software is sequentially executed is collected. A setting information input means for inputting the setting information;
The database connection system according to any one of claims 1 to 5, further comprising setting information registration means for storing the setting information input to the setting information input means in the setting information storage means. A database connection program installed in a computer comprising setting information storage means for storing setting information defining metadata items to be written to the database together with the data when writing data to the database,
In the computer,
A metadata collection process for collecting metadata from the metadata holding means by requesting metadata of the items defined in the setting information from the metadata holding means for holding the metadata, and data of the data to the database A database connection program for executing a write query conversion process for converting a query instructing writing into a query instructing to write metadata of an item defined by the data and setting information. On the computer,
The database connection program according to claim 7, wherein metadata deletion processing is executed to delete metadata of an item determined by setting information from read data read from a database. On the computer,
A field name storage process that stores all the field names specified in the query that instructs to write data to the database, and a symbol that means the data of all fields included in the query that instructs to read the data of all fields. The database connection program according to claim 7, wherein a read query conversion process for converting the query is executed by replacing each field name stored in the name storage process. On the computer,
The table generation query conversion process for converting the query is executed by adding a metadata item defined in the setting information to the field name of the new table specified in the query instructing the generation of the new table. The database connection program according to any one of claims 7 to 9. On the computer,
Collects metadata that does not change over time as each software is executed sequentially when the first of the software or software that generates a query that directs the writing of data to the database is launched Let
The database connection program according to any one of claims 7 to 10, wherein when the individual software is started, metadata that changes as the software is sequentially executed is collected. . It is mounted on a computer having setting information input means for inputting setting information,
In the computer,
The program for database connection according to any one of claims 7 to 11, wherein a setting information registration process for storing the setting information input to the setting information input unit in the setting information storage unit is executed. A database connection method that is applied to a database connection system that includes setting information storage means for storing setting information that defines metadata items to be written to the database together with the data when the data is written to the database,
The metadata collection means requests metadata of the items defined in the setting information to the metadata holding means that holds the metadata, collects metadata from the metadata holding means,
A database connection method, wherein the write query conversion means converts a query instructing writing of data into the database into a query instructing writing of metadata of an item defined by the data and setting information. The database connection method according to claim 13, wherein the metadata deletion means deletes metadata of items determined by setting information from read data read from the database. The field name storage means stores all the field names specified by the query instructing writing of data to the database,
The read query conversion means converts the query by replacing symbols representing the data of all fields included in the query instructing the reading of data of all fields with each field name stored in the field name storage means. Item 14. The database connection method according to Item 13. 14. The table generation query conversion means converts the query by adding an item of metadata defined by setting information to the field name of the new table specified by the query instructing generation of the new table. The database connection method according to claim 1. Metadata collection means
Collects metadata that does not change over time as each software is executed sequentially when the first of the software or software that generates a query that directs the writing of data to the database is launched And
The database connection method according to any one of claims 13 to 16, wherein when the individual software is activated, metadata that changes as the software is sequentially executed is collected. Applied to a database connection system comprising setting information input means for inputting setting information;
The database connection method according to any one of claims 13 to 17, wherein the setting information registration unit stores the setting information input to the setting information input unit in the setting information storage unit.

Images