JP2005149249A - Instruction generator, database system, instruction generation method, and program thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an instruction generator for automatically generates an Xpath. <P>SOLUTION: When definition data are given, a hierarchical structure acquisition means 6 traces lower nodes of a root node up to a terminal node to store their arrangement relation and determines whether an undetected younger brother node linked to a node in a hierarchy one higher than that of the terminal node exists or not and traces lower-hierarchy nodes of the younger brother node up to a terminal node to store their arrangement relation in the case of the existence of the undetected younger brother node and traces lower-hierarchy nodes of a node in a hierarchy one higher than that of this terminal node up to a terminal node to store their arrangement relation in the case of the existence of a younger brother node. This processing is repeated. A transmission means 7 extracts a tag having no child nodes and generates an input area for the tag as well as a control button for inputting an instruction to input data. They are transmitted to a terminal 1. An Xpath is generated from reply data. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an apparatus for generating an instruction to a database of a hierarchical self-describing document, and more particularly, to generation of an input screen that facilitates input according to the degree of automation of the hierarchical self-descripting document.

  Today, an XML database system (hereinafter referred to as XMLDB) that can handle structured documents is attracting attention.

JP 2001-34619 A

  In such an XML DB, a system has been proposed in which an input screen as shown in FIG. 9 is prepared in advance to simplify data registration from a terminal.

  On the other hand, XML data can be in a flexible input format by freely attaching tags. For example, the number of e-mail addresses is set to three, or the address of the parents' home is designated as a contact address.

  In other words, the flexibility and simple data registration of such XML data conflict with each other.

  In order to solve such a problem, it may be possible to display all items of the greatest common divisor, but this is unrealistic.

  An object of the present invention is to provide an instruction generation device that can easily register data in a database without restricting the degree of freedom of XML data capable of a flexible input format. It is another object of the present invention to provide an instruction generation device or method for easily registering data in a database without limiting the degree of freedom of a hierarchical self-describing document.

  1) In the instruction generation apparatus according to the present invention, a tag representing an attribute of the element data is added to each element data, and the element data to which each tag is added is arranged according to the arrangement positional relationship with respect to other tags in the document. An instruction generation device for giving a database processing instruction to an XML database that manages XML data defined by a structure, A) definition data storage means for storing definition data including the tag arrangement positional relationship, and B) in the definition data Hierarchical structure acquisition means for acquiring the hierarchical structure of each tag according to the arrangement position relationship of the tags, C) A tag located at the end of the hierarchical structure is extracted, together with an input area for inputting element data for the tag, Input screen data that can display a command input section for inputting an instruction for the input data by an operator of the terminal computer is displayed. A transmission means for transmitting to the computer; D) a processing instruction means for giving a database processing instruction inputted by the terminal computer to the XML database upon receipt of reply data from the terminal computer; E) the hierarchical structure acquisition means; 1) Acquiring the hierarchical structure of the tag by the following processing: 1) Tracing the tag of the highest node in the hierarchical structure as the node of interest, and the node of the tag in the lower hierarchy to the lower hierarchy sequentially 2) determine whether there is an undetected younger brother node of the terminal node that is connected to this node as a node of interest in the hierarchy one level above the terminal node, and 3) If an undetected younger brother node exists, the younger brother node is regarded as the node of interest, and the nodes in the lower hierarchy are sequentially moved down to the lower hierarchy. 4) When there is no sibling node of the end node, 4) If there is no sibling node of the end node, the node in the hierarchy one level higher than the end node is set as the target node, and the node in the lower hierarchy is set as the end node. Store the arrangement relationship of each node while sequentially tracing down to the lower level of the node. 5) Repeat the above 2) to 4).

  In this way, the entire hierarchical structure is determined from the definition data, a tag located at the end of the hierarchical structure is extracted from the hierarchical structure, and an input area for inputting element data for the tag is included, By generating input screen data that can display a command input unit for inputting commands for input data, data can be easily registered in the database without limiting the degree of freedom of XML data that allows a flexible input format. An instruction generation device can be provided.

  2) In the instruction generation device according to the present invention, an attribute label indicating an attribute of the element data is added to each element data, and each attribute label is added according to an arrangement positional relationship with respect to other attribute labels in the document. A command generation device that gives a database processing command to a database that manages a hierarchical self-describing document in which data is defined in a hierarchical structure, and A) definition data storage means for storing definition data including the tag arrangement positional relationship, B) Hierarchical structure acquisition means for acquiring the hierarchical structure of each attribute label according to the arrangement position relation of the attribute label in the definition data, C) Extracting the attribute label located at the end of the hierarchical structure, In addition to the input area for entering data, the operator of the terminal computer enters instructions for the input data. A transmission means for transmitting input screen data capable of displaying an instruction input unit to the terminal computer; D) a processing instruction means for giving a database processing instruction inputted by the terminal computer to the database when receiving reply data from the terminal computer It has.

  In this way, the entire hierarchical structure is determined from the definition data, and an attribute label located at the end of the hierarchical structure is extracted from the hierarchical structure, together with an input area for inputting element data for the attribute label. By generating input screen data that can display an instruction input unit for inputting instructions for the input data, it is possible to store in a database without restricting the degree of freedom of a hierarchical self-describing document capable of flexible input formats. It is possible to provide an instruction generation device capable of easily executing database processing instructions.

  3) In the instruction generation device according to the present invention, the hierarchical structure acquisition means acquires the hierarchical structure of the attribute label by the following processing. 1) Using the attribute label of the highest node of the hierarchical structure as a target node, sequentially storing the nodes of the attribute labels of the lower hierarchy up to the lower hierarchy, storing the arrangement relationship of each node, and 2) the terminal Determine whether there is an undetected younger brother node of the end node connected to this node as a node of the hierarchy one level higher than the node, and 3) if there is an undetected younger brother node, the younger brother node , With the nodes in the lower hierarchy being sequentially traced to the lower hierarchy, storing the positional relationship of each node, and 4) if there is no brother node of the end node, Using the node one level higher as the node of interest, the nodes in the lower layer are sequentially traced to the lower layer until the end node, and the arrangement relationship of each node is stored. 5) Repeat the above 2) to 4)Therefore, the hierarchical structure of the attribute label can be acquired with certainty.

  4) In the database system according to the present invention, an attribute label representing an attribute of the element data is added to each element data, and each attribute label is added according to the arrangement positional relationship with respect to other attribute labels in the document. A database system for managing a hierarchical self-describing document in which data is defined in a hierarchical structure, wherein 1) definition data storage means for storing definition data including the tag arrangement position relationship, 2) attribute labels in the definition data 3) Hierarchical structure acquisition means for acquiring the hierarchical structure of each attribute label according to the arrangement positional relationship of 3), and 3) an input area for extracting the attribute label located at the end of the hierarchical structure and inputting element data for the attribute label In addition, an instruction input unit for inputting an instruction for the input data by an operator of the terminal computer can be displayed. Transmitting means for transmitting input screen data to the terminal computer, 4) When receiving reply data from the terminal computer, it comprises execution means for executing database processing based on the database processing command input by the terminal computer. . In this way, the entire hierarchical structure is determined from the definition data, and an attribute label located at the end of the hierarchical structure is extracted from the hierarchical structure, together with an input area for inputting element data for the attribute label. By generating input screen data that can display a command input unit for inputting commands for the input data, it is easy to limit the flexibility of hierarchical self-describing documents that allow flexible input formats. A database system capable of database processing instructions can be provided.

  5) In the instruction generating device according to the present invention, an attribute label representing an attribute of the element data is added to each element data, and each attribute label is added according to the arrangement positional relationship with respect to other attribute labels in the document. An instruction generation device that gives a database processing instruction to a database that manages a hierarchical self-describing document in which data is defined in a hierarchical structure, and 1) definition data storage means for storing definition data including the tag arrangement positional relationship, 2) A hierarchical structure acquisition unit is provided for acquiring the hierarchical structure of each attribute label according to the positional relationship of the attribute labels in the definition data. In this way, the entire hierarchical structure is determined from the definition data, and an attribute label located at the end of the hierarchical structure is extracted from the hierarchical structure, together with an input area for inputting element data for the attribute label. Database processing without restricting the degree of freedom of hierarchical self-describing documents that allow flexible input formats by generating input screen data that can display an instruction input unit for inputting instructions for the input data An instruction generation device capable of generating an instruction can be provided.

  6) In the instruction generation method according to the present invention, an attribute label indicating an attribute of the element data is added to each element data, and each attribute label is added according to the arrangement positional relationship with respect to other attribute labels in the document. A method of using a computer that gives a database processing instruction to a database that manages a hierarchical self-describing document in which data is defined in a hierarchical structure, wherein the computer executes the following processing: 1) the tag arrangement Definition data including a positional relationship is stored in the definition data storage unit, 2) the hierarchical structure of each attribute label is acquired according to the positional relationship of the attribute label in the definition data, and 3) positioned at the end of the hierarchical structure. The attribute label to be extracted is extracted, and the input data is input together with an input area for inputting element data for the attribute label. Input screen data that can display a command input unit for an operator of the terminal computer to input a command to the terminal computer. 4) When reply data is received from the terminal computer, the database is input to the database. Give database processing instructions.

  In this way, the entire hierarchical structure is determined from the definition data, an attribute label located at the end of the hierarchical structure is extracted from the hierarchical structure, and an input area for inputting element data for the attribute label, A database is generated without restricting the degree of freedom of a hierarchical self-describing document capable of a flexible input format by causing a computer to generate input screen data that can display an instruction input unit for inputting an instruction for the input data. In addition, it is possible to provide an instruction generation method capable of easily executing a database processing instruction.

  7) A program according to the present invention is a program for causing a computer to function as an instruction generation device that gives a database processing instruction to a database, and for causing the computer to execute the following processing. 1) store definition data including the tag placement position relationship, 2) obtain the hierarchical structure of each attribute label according to the placement position relation of the attribute label in the definition data, and 3) attributes located at the end of the hierarchical structure The input screen data which can display the command input part which the operator of a terminal computer inputs the instruction | indication with respect to the input data with the input area for extracting a label and inputting element data about the said attribute label is transmitted to a terminal computer 4) When the reply data is received from the terminal computer, the database processing instruction inputted by the terminal computer is given to the database.

  In this way, the entire hierarchical structure is determined from the definition data, an attribute label located at the end of the hierarchical structure is extracted from the hierarchical structure, and an input area for inputting element data for the attribute label, A database can be generated without restricting the degree of freedom of a hierarchical self-describing document capable of a flexible input format by causing a computer to generate input screen data that can display an instruction input unit for inputting an instruction for the input data. In addition, it is possible to provide an instruction generation method capable of easily executing a database processing instruction.

  8) In the database system according to the present invention, A) the execution means is: a1) hierarchical self-describing document storage means for storing a hierarchical self-describing document; a2) names of a plurality of attribute labels having a hierarchical structure; Given a specified search expression, search means for searching for data that matches the hierarchical self-describing document, a3) Correspondence table between each attribute label and its address in the hierarchical self-describing document Correspondence table storage means for storing a4) Hierarchical structure notation data storage for storing hierarchical structure notation data in which the hierarchical structure of each attribute label in the hierarchical self-describing document is represented by an address at which each attribute label is stored B) The search means, when given the search expression, b1) a first step with the highest node in the search expression as the node of interest, b2) before The address of the node of interest is obtained by referring to the correspondence table, the address of the node of the next layer of the node of interest is obtained by referring to the hierarchical structure notation data, and the next node of the node of interest in the search formula For the attribute label of the hierarchy, the address is obtained by referring to the correspondence table, and an address in which the address of the next hierarchy node, the address of the next hierarchy node, and the address of the attribute label of the next hierarchy match is detected. B3) It is determined whether the second step has been executed up to the lowest hierarchy for the search expression. If it is executed, the current detection address is used as the search result. If not, the third step is repeated to repeat the second step. In this way, for each node, a search can be performed at high speed by separately searching the storage position and the attribute label of each node.

  9) In the database system according to the present invention, A) the execution means includes: a1) hierarchical self-describing document storage means for storing a hierarchical self-describing document; a2) names of a plurality of attribute labels having a hierarchical structure; Given a specified search expression, search means for searching for data matching the hierarchical self-describing document, a3) Correspondence between each attribute label in the hierarchical self-describing document and its hash value Corresponding table storage means for storing a table, a4) Hierarchical structure notation data storage means for storing hierarchical structure notation data in which the hierarchical structure of each attribute label in the hierarchical structure self-describing document is represented by a hash value of each attribute label B) The search means, when given the search formula, b1) a first step in which the highest node in the search formula is the node of interest, b2) the pair A hash value of the node of interest is obtained by referring to the table, a hash value of a node of the next layer of the node of interest is obtained by referring to the hierarchical structure notation data, and the next node of the node of interest for the search expression The second step of obtaining the hash value of the attribute label of the hierarchy by referring to the correspondence table and detecting the matching of the two hash values, b3) The second of the search expression up to the lowest hierarchy It is determined whether or not the step has been executed. If it is executed, the current detected hash value is used as the address of the search result, and if it is not executed, the second step is repeated. Perform steps. As described above, for each node, a high-speed search can be performed by separately searching the storage position and the attribute label of each node using the hash method.

  10) In the database system according to the present invention, A) the execution means includes: a1) hierarchical self-describing document storage means for storing a hierarchical self-describing document; a2) names of a plurality of attribute labels having a hierarchical structure; Given a specified search expression, search means for searching for data that matches the hierarchical self-describing document, a3) Correspondence table between each attribute label and its address in the hierarchical self-describing document Correspondence table storage means for storing, a4) hierarchical structure notation data storage for storing hierarchical structure notation data in which the hierarchical structure of each attribute label in the hierarchical self-describing document is represented by an address at which each attribute label is stored B) The search means refers to the correspondence table, obtains the address of the node of interest, refers to the hierarchical structure notation data, and Obtaining the address of the next layer node of the node of interest; b2) obtaining the address of the attribute label of the next layer of the node of interest with reference to the search formula; b3) obtaining the address of the next layer node; The address of the matching data is searched by executing the step of setting the address and the address of the attribute label of the next layer as the detection address. In this way, for each node, a search can be performed at high speed by separately searching the storage position and the attribute label of each node.

  In this specification, “Xpath” is a language for designating a specific element of an XML document, and in this embodiment, designates an element position of XML data stored in an XML database. Further, the “attribute label” corresponds to a tag in the embodiment. “Database processing” refers to “addition”, “update”, “deletion”, and “search” processing of data in the database.

1. Description of Functional Block Diagram One embodiment of the present invention will be described with reference to the drawings.
A database system 1 shown in FIG. 1 includes a Web server 2 and a database 3 which are instruction generation devices. The database 3 manages hierarchical self-describing documents. In a hierarchical self-describing type document, an attribute label representing an attribute of the element data is added to each element data, and the element data to which each attribute label is added according to the arrangement positional relationship with respect to other attribute labels in the document. Data defined in a hierarchical structure.

  The Web server 2 includes a receiving unit 3, a definition data storage unit 4, a hierarchical structure acquisition unit 6, a transmission unit 7, and a processing instruction unit 8.

  When the definition data including the tag arrangement positional relationship is given from the user terminal 1a, the receiving means 3 stores this in the definition data storage means. The hierarchical structure acquisition unit 6 acquires the hierarchical structure of each attribute label according to the arrangement position relationship of the attribute labels in the definition data. In the present embodiment, the hierarchical structure of each attribute label is acquired by the following processing. 1) In the definition data, the attribute label of the highest node in the hierarchical structure is used as a node of interest, the node of the attribute label of the lower hierarchy is sequentially traced to the lower hierarchy, and the arrangement relationship of each node is stored. 2) Determine whether there is an undetected younger brother node that is connected to this node as a node of interest that is one node above the last node. 3) If there is an undetected younger brother node , With the younger brother node as the node of interest, the nodes in the lower hierarchy are sequentially traced to the lower hierarchy, storing the arrangement relationship of each node, and 4) if the younger brother node of the end node does not exist, An arrangement relationship of each node is stored while a node in a hierarchy one level higher than the terminal node is set as a target node, and a node in a lower hierarchy is sequentially traced to a lower hierarchy up to the terminal node. 5) Repeat 2) to 4) above.

  The transmission means 7 extracts an attribute label located at the end of the hierarchical structure, and inputs an instruction for the input data together with an input area for inputting element data for the attribute label. Input screen data capable of displaying the input unit is transmitted to the terminal computer. When receiving the reply data from the terminal computer, the processing command means 8 gives the database processing command input by the terminal computer to the database.

2. Hardware Configuration The hardware configuration of the Web server 2 shown in FIG. 1 will be described with reference to FIG. FIG. 2 is an example of a hardware configuration configured using a CPU.
The Web server 2 includes a CPU 23, a memory 27, a hard disk 26, a monitor 30, a CDD (CDROM drive) 25, an input device 28, a communication board 32, and a bus line 29. The CPU 23 controls each unit via the bus line 29 according to each program stored in the hard disk 26.

  The hard disk 26 includes an operating system program (hereinafter abbreviated as OS) storage unit 26o and an automatically generated program storage unit 26a.

  The operating system program (OS) 26o employs Linux (trademark) in this embodiment, but is not limited to this.

  As will be described later, the automatic generation program 26a generates input screen data that can display a command input unit input by an operator of the terminal computer based on definition data received from the user terminal or another computer, and stores it in the terminal computer. Send. When a reply is received from the terminal computer, it is passed to the database 3. Details will be described later.

  The program is read from the CDROM 25a storing the program via the CDD 25 and installed in the hard disk 26. In addition to the CDROM, a program such as a flexible disk (FD) or an IC card may be installed on a hard disk from a computer-readable recording medium. Furthermore, it may be downloaded using a communication line.

  In the present embodiment, the program stored in the CDROM is indirectly executed by the computer by installing the program from the CDROM into the hard disk 26. However, the present invention is not limited to this, and the program stored in the CDROM may be directly executed from the CDD 25. Note that programs that can be executed by a computer are not only programs that can be directly executed by being installed as they are, but also programs that need to be converted into other forms (for example, those that have been compressed) In addition, those that can be executed in combination with other module parts are also included.

  As shown in FIG. 3, the user terminal 2 has substantially the same hardware configuration except that the hard disk 126 has a browser program storage unit 126 b that stores the browser program, and a description thereof will be omitted. As shown in FIG. 4, the database 3 is a normal database system for XML data, and stores XML data related to the data storage unit 226d. The program stored in the data management program storage unit 226a is also a program that performs normal database processing.

3. Flowchart Next, processing in this system will be described with reference to FIG. Hereinafter, in the present embodiment, an example will be described in which definition data is given from a user terminal, and XML data having a hierarchical structure desired by the user is additionally registered in the database 3 using the definition data.
The operator of the user terminal 1 transmits definition data defining the data structure of XML data desired to be input to the Web server 2 (step S1). An example is shown in FIG. In this data, data surrounded by <ns: record></ ns: record> is added to the database 3 as one record in the database 3. Note that <ns: origin></ ns: origin> is information for specifying an input / output position as each record.

  When the CPU 23 of the Web server 2 receives this definition data, it performs an XHTML automatic generation process. Details of XHTML automatic generation processing by the automatic generation program will be described with reference to FIG. First, CPU23 acquires the tag used as a root node <step S31 of FIG. 7]. Since the root node is the node located at the top of one record, in this case, the tag serving as the root node is <Contact>. The CPU 23 stores the tag in the memory 27 as a root node.

  Next, the CPU 23 outputs the header part of the XHTML data (step S32). Such a header part may be stored in advance. In this embodiment, data as shown in FIG. 8A is stored, and the tag of the root node (in this case, “contact”) is input to the variable% 1 and output.

  Next, the CPU 23 acquires a child node (the eldest brother if there are a plurality of nodes) of the root node (step S33), and determines whether or not a child node has been acquired for the node <contact address> (step S34). In this case, the node <name> as the elder brother node among the child nodes exists in the root node <contact address> Therefore, the CPU 23 determines in step S34 that the child node has been acquired and stores it in advance. In this embodiment, the XHTML text data as shown in FIG.8C is stored, and the node name is directly stored in the variable% 3 (in this case). Name "), and the variable% 3 is inserted with the node name including the upper node (as a path name) (in this case," contact ") / Name ").

  Next, the CPU 23 acquires the younger brother node of the node <name> (step S37), and determines whether the younger brother node has been acquired (step S39). The younger brother node is a node that is located behind the node hanging from the same parent. In this case, since there is a younger brother node <telephone> of the node <name>, the CPU 23 obtains a child node (elder brother) of the node <telephone> (step S33), assuming that the younger brother node can be obtained. It is determined whether or not a child node has been acquired for (telephone) (step S34). In this case, since there is no child node, the CPU 23 determines that the child node could not be acquired, and outputs an XHTML text box (step S36). That is, the node name (in this case, “telephone”) is inserted into the variable% 3 of the XHTML text data as shown in FIG. 8C, and the node name is inserted into the variable% 3 including the upper node (as a path name). (In this case “Contact / Phone”).

Next, the CPU 23 acquires the younger brother node of the node <telephone> (step S37), and determines whether or not the younger brother node has been acquired (step S39). In this case, since there is a younger brother node <address> of the node <phone>, the CPU 23 proceeds to step S33 to acquire a child node of the node <address> (step S33), and for the node <address>, the child node It is determined whether or not can be acquired (step S34). In this case, since the node <address> includes the elder brother node <address 1> among the child nodes, the CPU 23 determines in step S34 that the child node has been acquired, and outputs the XHTML frame header data. (Step S35). In the present embodiment, the node <address>, which is the node from which the child node can be acquired, is inserted into the variable% 2 of the XHTML frame header data stored in FIG. 8B and output.
Next, the CPU 23 acquires a child node of the node <address 1> (step S33), and determines whether or not a child node has been acquired for the node <address 1> (step S34). In this case, since there is no child node, the CPU 23 outputs an XHTML text box (step S36). In this case, XHTML text data in which “address 1” is inserted in variable% 3 and “contact / address / address 1” is inserted in variable% 4 is output.

  Next, the CPU 23 acquires a younger brother node (step S37), and determines whether or not a younger brother node has been acquired (step S39). In this case, since there is a younger brother node <address 2> of the node <address 1>, the CPU 23 proceeds to step S33 and acquires a child node for the acquired node <address 2> (step S33). It is determined whether or not a child node has been acquired for <address 2> (step S34). In this case, since there is no child node of the node <address 2>, the CPU 23 outputs XHTML text data (step S36). In this case, XHTML text data in which “address 2” is inserted into variable% 3 and “contact / address / address 2” is inserted into variable% 4 is output.

  Next, the CPU 23 acquires the younger brother node of the node <address 2> (step S37), and determines whether the younger brother node has been acquired (step S39). In this case, the CPU 23 proceeds to step S41, acquires a parent node (step S41), and determines whether the acquired node is a root node (step S43). In this case, since the node <address> is not the root node, the frame footer part is output (step S45). In the present embodiment, data as shown in FIG. 8D is stored and output.

  CPU23 acquires a younger brother node (step S37 of FIG. 7), and judges whether the younger brother node was able to be acquired (step S39). In this case, since the younger brother node cannot be acquired, the CPU 23 proceeds to step S41, acquires the parent node (step S41), and determines whether the acquired node is the root node (step S43).

  In this case, since the node <contact address> is the parent node, the CPU 23 outputs the footer part (step S47). In the present embodiment, XHTML footer data as shown in FIG. 8E is stored and output. As will be described later, the footer portion is data for displaying a control button (command input portion) for allowing the operator of the user terminal to specify a processing command for the database 3. In this example, four control buttons “add”, “update”, “delete”, and “search” are displayed.

  In this way, XHTML data as shown in FIGS. 9a and 9b is automatically generated.

  The generated XHTML data is transmitted to the user terminal 1a (step S5 in FIG. 5). The user terminal 1a reads and displays it by the browser program (step S7). On the user terminal 1a, an input box having a hierarchical structure as shown in FIG. 10 and control buttons 55 to 58 are displayed based on the XHTML data shown in FIGS. 9a and 9b.

  The operator of the user terminal 1a inputs data in the input boxes 50, 51, 53, and 54, and selects one of the control buttons with the input device (step S9 in FIG. 5). In this case, since it is an additional process, the operator of the user terminal 1a selects the control button 55. When any one of the control buttons is selected, the user terminal 1a transmits the ID of the input box and the input value to the input box to the Web server 2 by the browser program together with the command specifying information for performing additional processing ( Step S11). In the present embodiment, the instruction specifying information “insert”. For example, “Taro Yamada”, “090-xxxx-xxxx”, “Osaka City Osaka xxx-xxx”, “Kobe City Hyogo xxx-xxx” are entered in the input boxes 50 to 54 shown in FIG. , When the button “insert” is selected, the user terminal 1a inserts “contact / name” = Taro Yamada, “contact / phone” = 090-xxxx-xxxx, “contact / address / address 1” = Send Osaka city xxx-xxx, “contact / address / address 2” = Kobe city xxx-xxx, Hyogo prefecture.

  The Web server 2 generates a command statement (including Xpath) specifying the processing content from the received data, and transmits it to the database 3 (step S13). In this case, the instruction specifying information generates an instruction “insertflagment” corresponding to “insert”. Since the processing content of the command generated from the command specifying information is determined every time it is determined in the output destination database, it may be registered in the Web server 2. Further, as the Xpath, an input box ID and its value (for example, “contact / name” = Taro Yamada,...) Are generated and transmitted to the database 3.

  The database 3 determines whether or not a processing instruction is given from the Web server 2 (step S15). When the instruction is given, the designated instruction is executed (step S17). Such processing is the same as in the prior art, but will be briefly described with reference to FIG.

  The CPU 223 of the database 3 determines whether the instruction type is “add”, “update”, “delete”, or “search” (step S81). In this case, since the instruction “insertflagment” is given, an additional position determination process is performed (step S83). Although the addition position may be arbitrary (for example, always at the very end), in the present embodiment, if there is a record having a common node, it is added next to improve the search speed. Therefore, in this case, a record having a hierarchical structure of <name>, <telephone>, and <address> is searched as a lower node of the node <contact address> specified by Xpath, and the next record is set as an additional position. .

  The CPU 223 adds a record at the determined additional position (step S85). This terminates the process using Xpath and transmits the result (step S19). When the Web server 2 receives this, the process ends.

  Note that the “update” and “deletion” processes differ from the determination of the additional position in that the target record is specified (steps S93 and S103 in FIG. 11). For the “search” process, a search is performed according to the given search condition (step S115), and the result is transmitted (step S117).

  As described above, when the instance definition data is given from the user terminal, the definition data is automatically analyzed, the hierarchical structure of the definition data is determined, and a screen in which the input box is associated with the terminal node is automatically generated. On this screen, an instruction box for inputting the type of processing instruction to the database that manages the XML data is displayed. Such screen data is transmitted to the user terminal, and when input data and instruction data are given from the terminal, a command statement is given to the database based on the input data. Thereby, the XML data of the tree structure desired by the user can be database processed in the database.

  Unlike this case, it is theoretically possible to prepare all the tree structures of input boxes that all users would like in advance. However, in reality, it is necessary to prepare a huge number of combinations, and there is a problem that a large number of candidates are required. On the other hand, like this system, by analyzing the definition data given by the user and dynamically generating the input screen, there is no need to register such candidates and flexible handling is possible. It becomes possible.

4). Other Embodiments In this embodiment, the definition data is given from the user terminal. However, one or more of the definition data may be stored in the hard disk in advance, and the user may select one of them. . Alternatively, only the hierarchical structure of the nodes may be stored, and the definition data may be generated by the web server when a node name is input thereto.
In the above embodiment, the definition data is given from the user terminal. However, the definition data registered in the Web server 2 in advance may be automatically determined for each user attribute to generate the XHTML data. Is possible. The generated data is not limited to XHTML, and any data format having regularity such as HTML or XML may be used.

  In the above embodiment, the XHTML data in which the input box is arranged is generated for each node when there is no child node. However, the hierarchical structure of the acquired nodes is determined, and the XHTML is collectively collected. Data may be generated. In this case, for example, when the acquired node is a root node in step S43, the hierarchical structure of each node is determined among the acquired nodes, and the frame header data is output when a child node exists, XML text data may be generated for a node having no child node.

  In addition, it can be grasped not as an instruction generation device but as a hierarchical structure determination device.

5). Regarding the search method in the database 3 5.1 Problems of the search method in the database 3 In the above embodiment, the same search method as the conventional one is used for the record extraction in the database 3. However, in such a method, it is necessary to determine whether or not all nodes of all records match, and if the number of records increases or the hierarchical structure of one record becomes complicated, the search speed decreases. is there. For example, assume that XML data having a hierarchical structure as shown in FIG. 14 is stored in the database. In this case, when searching for the node <name list / contact / company / address>, the first node <name list> in the search formula is usually searched. In this case, the nodes 305 and 405 match. Next, it is determined whether or not the attribute name of the node immediately below the node 305 is <name>. In this case, since the attribute name of the node 307 is <name>, it does not match the search expression. Since the attribute name of the next node 309 is <company>, it matches the search formula. Next, it is determined whether or not the attribute name of the node immediately below the node 309 is <address>. In this case, since the attribute name of the node 311 is <zip code>, it does not match the search expression. Since the attribute name of the next node 313 is <address>, it matches the search expression. Since the attribute name of the node 313 is <telephone number>, it does not match the search expression. The same search is performed for the node 405. In this way, it is necessary to determine whether or not all the nodes match one by one.

  However, this requires more search time as the amount of accumulated data increases. A hash method is known as a technique for speeding up the data location search. The hash method is one of the methods for performing high-speed data retrieval. The key value is calculated by a certain function (hash function), and the hash value is referenced from the obtained value (hash value). The method of searching by specifying the storage location.

  For example, the node <name list / contact / company / address> is 33, 45, 67..., <Name list / contact / home / address> is 55, 86. If a correspondence table is stored and a search expression is given, it is possible to perform a search in one shot by referring to this hash table and reading the data of the corresponding address. However, in a data group defined in a self-describing language having a hierarchical structure, not all nodes are designated by a search command, but a node <name list / contact /? ? ? / Address> may be partially specified by a wild card, or may be specified at an intermediate level such as <name list / contact address>. Therefore, it is unrealistic to store all combinations as hash values in advance. In particular, in a database, such a problem becomes more serious considering that some data is updated, added, or deleted.

  Therefore, the inventor has invented a new search method in order to solve the above problem. An object of the following invention is to provide a search method that enables high-speed search in a data group defined by a self-describing language having a hierarchical structure, and an apparatus for realizing the method.

5.2 Functional Blocks of Means for Solving FIG. 12 shows a functional block diagram of such a search device. The data search apparatus 500 includes a hierarchical self-describing document storage unit 501, a search unit 503, a change unit 504, a correspondence table storage unit 505, and a hierarchical structure notation data storage unit 507. The hierarchical self-describing document storage unit 501 stores a hierarchical self-describing document. Retrieval means 503, when given a retrieval formula in which the names of a plurality of attribute labels are specified in a hierarchical structure, retrieves data matching the hierarchical structure self-describing document. The correspondence table storage unit 505 stores a correspondence table of each attribute label and its address in the hierarchical self-describing document. The hierarchical structure notation data storage means 507 stores hierarchical structure notation data in which the hierarchical structure of each attribute label in the hierarchical structure self-describing document is represented by an address at which each attribute label is stored.
The search means 503 receives 1) the first step when the search expression is given, and 2) obtains the address of the target node by referring to the correspondence table. , Referring to the hierarchical structure notation data, obtaining the address of the node next to the node of interest, and referring to the correspondence table for the attribute label of the next layer of the node of interest for the search expression And a second step in which an address where the address of the next hierarchy node, the address of the next hierarchy node and the address of the attribute label of the next hierarchy match is used as a detection address, and 3) the lowest hierarchy for the search expression Whether or not the second step has been executed, and if so, the current detection address is used as the search result address with the matching data as the search result address. When to not to perform the third step of repeating the second step.

  The change unit 504 executes an add, change, or delete instruction when given to the hierarchical self-describing document stored in the hierarchical self-describing document storage unit 501, and also executes the corresponding table storage unit 505 and the hierarchical table. Corresponding data in the structure notation data storage means 507 is changed. For example, when there is an addition, it is added to the correspondence table storage unit 505 and the hierarchical structure notation data storage unit 507. The same applies to changes and deletions.

5.3 Detailed Description of the Invention Since the hardware configuration for realizing the invention is the same as described above, the description thereof is omitted.
Differences from the database 3 shown in FIG. 4 will be described. In the present embodiment, in order to solve the above problem, a tree structure storage unit 226t and a hash table storage unit 226h are provided, and further, a data management program for performing data search using these is provided.

5.4 Outline Next, an outline of processing in this apparatus will be described. In this embodiment, high-speed search is possible as follows. A tree structure table in FIG. 15 and a hash table in FIG. 16 are prepared in advance on the hard disk 226. Such a tree structure table and a hash table are generated as follows. First, the hash value is calculated for each node of data to be added. For example, the hash value may be determined using a predetermined hash function in the same manner as a general hash value. If the same node already exists for each node, it is added to the hash table shown in FIG. 16, and if there is no same node, a new node name and its hash value are additionally stored in the hash table. That's fine. Further, each node in the tree structure of the added XML data is replaced with the hash value and added to the tree structure table shown in FIG.
Then, if a node specified by a relative path of A / B / C is specified as a node as a search expression, the hash value of the first node A is first acquired from the hash table. For the acquired hash value, the hash value of the child node of node A is acquired from the tree structure table. Further, with reference to the search expression, it is understood that the child node of node A is node B. The hash value of node B is acquired from the hash table. The hash value of the node B is compared with the hash value of the child node of the node A obtained from the tree structure table to obtain a matching hash value. For these (these), the child nodes are similarly acquired from the tree structure. Then, referring to the search formula, it is understood that the child node of node B is node C, and the hash value of node C is acquired from the hash table. Get the hash value that matches the two. By repeating this, it is possible to retrieve the node name and the arrangement position separately, and finally obtain the address of the node that matches the retrieval formula.

5.5 Flowchart Next, processing in this apparatus will be described with reference to FIG. Hereinafter, the processing of the CPU 423 by the data management program storage unit 426s will be described by taking as an example the case where the node “name list / contact / name / company / address” is given as the search target node.
First, the CPU 423 initializes the node of interest Ni (step S61 in FIG. 17). As a result, the 0th node in the Xpath becomes the target node. In this case, the “name list” becomes the attention node.

  The CPU 423 obtains the hash value of the node of interest Ni from the hash table shown in FIG. 15 (step S65). In this case, the hash value “2” is obtained. The CPU 423 extracts all the hash values of the child nodes for the hash value of the node of interest using the tree structure table for the hash values obtained from the hash table (step S67). In this case, hash values “3”, “13”, “23”, and “43” are obtained. The CPU 423 obtains the hash value from the hash table for the child node of the node of interest in the search target node “name list / contact address / name / company / address” (step S69). In this case, since the child node of the node of interest is “contact”, the hash values “3”, “13”, “23”, and “43” are obtained by referring to the hash table. The CPU 423 stores a node where the hash value of the extracted child node matches the hash value obtained from the hash table (step S71). In this case, all four nodes match.

  Next, the CPU 423 determines whether or not processing has been completed for all nodes in the Xpath (step S73). In this case, since processing has not been completed for all nodes, the process proceeds to step S75, and the node of interest Ni is incremented. As a result, Ni = 1, and the node of interest becomes “contact”.

  Next, the CPU 423 extracts all the hash values of the child nodes for the hash value of the node of interest using the tree structure table for the hash values obtained from the hash table (step S67). In this case, hash values “4, 5, 14, 15, 24, 25, 29, 44, 45” are obtained. The CPU 423 obtains the hash value from the hash table for the child node of the node of interest (step S69). In this case, since the child node of the node of interest is “name”, the hash value 4, 14, 24, 44 ”is obtained by referring to the hash table. The CPU 423 obtains the hash value of the extracted child node and the hash value. The node that matches the hash value obtained from the table is stored (step S71). In this case, the nodes 4, 14, 24, and 44 ″ match.

  Similarly, when the name / company / address is specified, the process proceeds from step S73 to step S77, and the candidate is specified as a processing target node. In this case, finally, the nodes 7, 17, and 27 are specified.

  As described above, when the hash table in which the node name is represented by the hash value and the hierarchical structure relation of the hash value between the nodes are separately stored and given the search target hierarchical structure data, By sequentially searching child nodes for nodes, it is possible to search for hierarchical structure data using hash values.

  Note that when the record change instruction is given to the XML data stored in the database, when the structure itself is changed, the tree structure shown in FIG. 15 and the correspondence table shown in FIG. 16 are changed. The For example, in FIG. 14, when an instruction to change as shown in FIG. 18A to FIG. 18B is newly given, the corresponding record is searched, and the record stored in the database is as shown in FIG. 18B. Be changed. Since the node <company> is changed to <home>, the tree structure table is changed as shown in FIG. 19, and the hash table is changed as shown in FIG. The same applies to other modifications. When the data structure of the record is not changed, the tree structure table and the hash table are not changed.

5.6 Others Although the above embodiment has been described as a search method performed when the Xpath is automatically generated, the search method can naturally be adopted even when such automatic generation is not performed.
In the above embodiment, the hash value is stored. However, the hash value is not necessarily stored, and a correspondence table between each node address and each node name in the XML data may be used.

  In the above embodiment, the CPU is used to realize the function shown in FIG. 1, and this is realized by software. However, some or all of them may be realized by hardware such as a logic circuit.

  In addition, you may make it make an operating system (OS) process a part of said program.

It is a functional block diagram of database system 10 concerning the present invention. It is a figure which shows an example of the hardware constitution which implement | achieved Web server 2 shown in FIG. 1 using CPU. It is a figure which shows an example of the hardware constitution which implement | achieved the user terminal 1 shown in FIG. 1 using CPU. It is a figure which shows an example of the hardware constitution which implement | achieved the database 3 shown in FIG. 1 using CPU. It is the whole flowchart. It is a figure which shows an example of definition data. It is a detailed flowchart of screen data creation by XHTML. It is a figure which shows the XHTML basic data memorize | stored beforehand. It is a figure which shows the produced | generated XHTML data. It is a figure which shows the produced | generated XHTML data. It is a figure which shows the screen which displayed the XHTML data of FIG. 9a, FIG. 9b. It is a flowchart which shows the database process in a database. It is a functional block diagram of database 400 which improved search speed. It is a figure which shows an example of the hardware constitutions which implement | achieved the database 400 shown in FIG. 12 using CPU. It is a figure which shows an example of the XML data memorize | stored in the database. It is a figure which shows an example of the data structure of a tree structure memory | storage part. It is a figure which shows an example of the data structure of a hash table. It is a flowchart of the search process of the database. The record data before and after the change. It is a figure which shows the data structure of the tree structure memory | storage part after a change. It is a figure which shows the data structure of the hash table after a change.

Explanation of symbols

23 ... CPU
27 ... Memory

Claims (10)

A tag that represents an attribute of the element data is added to each element data, and XML that manages the XML data in which the element data to which each tag is added is defined in a hierarchical structure according to the arrangement positional relationship with respect to other tags in the document. An instruction generation device for giving a database processing instruction to a database,
Definition data storage means for storing definition data including the tag arrangement positional relationship;
Hierarchical structure acquisition means for acquiring the hierarchical structure of each tag according to the tag placement position relationship in the definition data,
An input screen for extracting a tag located at the end of the hierarchical structure and displaying an instruction input unit for an operator of the terminal computer to input an instruction for the input data together with an input area for inputting element data for the tag Transmitting means for transmitting data to the terminal computer;
Processing instruction means for giving a database processing instruction inputted at the terminal computer to the XML database upon receiving reply data from the terminal computer;
With
The hierarchical structure acquisition means acquires the hierarchical structure of the tag by the following processing;
1) Using the tag of the highest node in the hierarchical structure as a target node, sequentially storing the tags of the lower hierarchy tags to the lower hierarchy, storing the arrangement relationship of each node,
2) Determine whether there is an undetected younger brother node of the terminal node that is connected to the node in the hierarchy one level higher than the terminal node,
3) If there is an undetected younger brother node, the younger brother node is the node of interest, the lower layer nodes are sequentially traced to the lower layer, and the arrangement relationship of each node is stored.
4) If there is no younger brother node of the end node, each node is traced down to the end node sequentially from the node in the lower hierarchy to the end node with the node in the hierarchy one level higher than the end node as the target node. Memorize the arrangement of
5) Repeat 2) to 4) above,
An instruction generation device characterized by the above. An attribute label representing the attribute of the element data is added to each element data, and the element data to which each attribute label is added is defined in a hierarchical structure according to the arrangement positional relationship with respect to other attribute labels in the document. An instruction generation device for giving a database processing instruction to a database for managing a descriptive document,
Definition data storage means for storing definition data including the tag arrangement positional relationship;
Hierarchical structure acquisition means for acquiring the hierarchical structure of each attribute label according to the positional relationship between the attribute labels in the definition data,
An attribute label located at the end of the hierarchical structure is extracted, and an instruction input unit for inputting an instruction for the input data together with an input area for inputting element data for the attribute label can be displayed. Transmitting means for transmitting input screen data to a terminal computer;
Processing instruction means for receiving a reply processing data from the terminal computer and giving a database processing instruction input by the terminal computer to the database;
An instruction generation device comprising: The instruction generation device according to claim 2,
The hierarchical structure acquisition means acquires the hierarchical structure of the attribute label by the following processing;
1) Using the attribute label of the highest node of the hierarchical structure as a target node, sequentially storing the node of the attribute label of the lower hierarchy to the lower hierarchy, storing the arrangement relationship of each node,
2) Determine whether there is an undetected younger brother node of the terminal node that is connected to the node in the hierarchy one level higher than the terminal node,
3) If there is an undetected younger brother node, the younger brother node is the node of interest, the lower layer nodes are sequentially traced to the lower layer, and the arrangement relationship of each node is stored.
4) If there is no younger brother node of the end node, each node is traced down to the end node sequentially from the node in the lower hierarchy to the end node with the node in the hierarchy one level higher than the end node as the target node. Memorize the arrangement of
5) Repeat 2) to 4) above,
It is characterized by. An attribute label representing the attribute of the element data is added to each element data, and the element data to which each attribute label is added is defined in a hierarchical structure according to the arrangement positional relationship with respect to other attribute labels in the document. A database system for managing descriptive documents,
Definition data storage means for storing definition data including the tag arrangement positional relationship;
Hierarchical structure acquisition means for acquiring the hierarchical structure of each attribute label according to the positional relationship between the attribute labels in the definition data,
An attribute label located at the end of the hierarchical structure is extracted, and an instruction input unit for inputting an instruction for the input data together with an input area for inputting element data for the attribute label can be displayed. Transmitting means for transmitting input screen data to a terminal computer;
Execution means for executing database processing based on a database processing command input by the terminal computer upon receiving reply data from the terminal computer;
A database system characterized by comprising: An attribute label representing the attribute of the element data is added to each element data, and the element data to which each attribute label is added is defined in a hierarchical structure according to the arrangement positional relationship with respect to other attribute labels in the document. An instruction generation device for giving a database processing instruction to a database for managing a descriptive document,
Definition data storage means for storing definition data including the tag arrangement positional relationship;
Hierarchical structure acquisition means for acquiring the hierarchical structure of each attribute label according to the positional relationship of the attribute labels in the definition data,
An instruction generation device comprising: An attribute label representing the attribute of the element data is added to each element data, and the element data to which each attribute label is added is defined in a hierarchical structure according to the arrangement positional relationship with respect to other attribute labels in the document. A method of using a computer that gives a database processing instruction to a database managing a descriptive document, causing the computer to execute the following processing:
Definition data including the tag arrangement positional relationship is stored in a definition data storage unit,
The hierarchical structure of each attribute label is acquired according to the arrangement position relationship of the attribute label in the definition data,
An attribute label located at the end of the hierarchical structure can be extracted, and an instruction input unit for inputting an instruction for the input data together with an input area for inputting element data for the attribute label can be displayed. Send input screen data to the terminal computer,
When receiving reply data from the terminal computer, the database processing instruction input in the terminal computer is given to the database.
An instruction generation method characterized by the above. A program for causing a computer to function as an instruction generation device that gives a database processing instruction to a database, and causing the computer to execute the following processing.
Storing definition data including the tag placement positional relationship;
The hierarchical structure of each attribute label is acquired according to the arrangement position relationship of the attribute label in the definition data,
An attribute label located at the end of the hierarchical structure can be extracted, and an instruction input unit for inputting an instruction for the input data together with an input area for inputting element data for the attribute label can be displayed. Send input screen data to the terminal computer,
When the reply data is received from the terminal computer, the database processing instruction inputted by the terminal computer is given to the database. The database system of claim 4,
The execution means includes
Hierarchical self-describing document storage means for storing a hierarchical self-describing document;
Search means for searching for data matching the hierarchical structure self-describing document when given a search expression in which names of a plurality of attribute labels are specified in a hierarchical structure;
Correspondence table storage means for storing a correspondence table between each attribute label and its address in the hierarchical self-describing document;
Hierarchical structure notation data storage means for storing hierarchical structure notation data in which the hierarchical structure of each attribute label in the hierarchical structure self-describing document is represented by an address storing each attribute label;
With
The search means includes
Given the search expression, a first step in which the highest node in the search expression is the node of interest;
The address of the node of interest is obtained by referring to the correspondence table, the address of the node of the next layer of the node of interest is obtained by referring to the hierarchical structure notation data, and the next of the node of interest for the search expression For the attribute label of the hierarchy, the address is obtained by referring to the correspondence table, and an address in which the address of the next hierarchy node, the address of the next hierarchy node, and the address of the attribute label of the next hierarchy match is detected. The second step of addressing,
It is determined whether or not the second step has been executed up to the lowest hierarchy for the search formula. If it is executed, the current detection address is used as the search result address and the search result is not executed. Is a third step that repeats the second step,
It is characterized by executing. The database system of claim 4,
The execution means includes
Hierarchical self-describing document storage means for storing a hierarchical self-describing document;
Search means for searching for data matching the hierarchical structure self-describing document when given a search expression in which names of a plurality of attribute labels are specified in a hierarchical structure;
Correspondence table storage means for storing a correspondence table of each attribute label and its hash value in the hierarchical self-describing document;
Hierarchical structure notation data storage means for storing hierarchical structure notation data in which the hierarchical structure of each attribute label in the hierarchical structure self-describing document is represented by a hash value of each attribute label;
With
The search means includes
Given the search expression, a first step in which the highest node in the search expression is the node of interest;
The hash value of the node of interest is obtained by referring to the correspondence table, the hash value of the node of the next layer of the node of interest is obtained by referring to the hierarchical structure notation data, and the node of interest for the search expression A second step of obtaining the hash value of the attribute label of the next layer by referring to the correspondence table and detecting a match between the two hash values;
It is determined whether or not the second step has been executed up to the lowest hierarchy for the search expression, and if it is executed, the currently detected hash value is used as the address of the search result and is not executed. In this case, a third step that repeats the second step,
It is characterized by executing. The database system of claim 4,
The execution means includes
Hierarchical self-describing document storage means for storing a hierarchical self-describing document;
Search means for searching for data matching the hierarchical structure self-describing document when given a search expression in which names of a plurality of attribute labels are specified in a hierarchical structure;
Correspondence table storage means for storing a correspondence table between each attribute label and its address in the hierarchical self-describing document;
Hierarchical structure notation data storage means for storing hierarchical structure notation data in which the hierarchical structure of each attribute label in the hierarchical structure self-describing document is represented by an address storing each attribute label;
With
The search means includes
Obtaining an address of a node of interest by referring to the correspondence table, obtaining an address of a node of the next layer of the node of interest by referring to the hierarchical structure notation data;
For the attribute label of the next layer of the node of interest with respect to the search expression, referring to the correspondence table, obtaining the address;
The detection address is an address where the address of the next layer node and the address of the attribute label of the next layer match,
To retrieve the address of the matching data by executing
It is characterized by.

Images