JP2012073706A - Structured document storage device, structured document storage method and structured document storage program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To quickly determine an ancestor-descendant relationship between nodes forming a structured document.SOLUTION: An XML analysis part 101 analyzes an XML document and outputs information about nodes of the XML document. A storage part 102 determines at least an order number showing an appearance order of a processing object node within the information about the nodes of the XML document and a second order number which is an order number of a first node among the nodes in the information appearing after the processing object node except descendant nodes of the processing object node, and stores them in a node table 2 together with a node type etc. of the processing object node. A retrieval part 105 has a descendant node retrieval means 105g determining that a comparison source node is a descendant node of a comparison destination node when the order number of the comparison source node in the stored information is larger than the order number of the comparison destination node and the order number of the comparison source node is smaller than the second order number of the comparison destination node.

Description

  The present invention relates to a structured document storage device, a structured document storage method, and a structured document storage program, and more particularly to a structured document that stores a structured document represented by XML (Extensible Markup Language). The present invention relates to a storage device, a structured document storage method, and a structured document storage program.

  As a method of storing a structured document represented by XML and searching for elements constituting the stored structured document at high speed, there is a method of storing the structured document in a relational database composed of tabular tables.

  For example, in Patent Document 1, the contents of each element of XML data and the attribute value items and the link relationship between the data are stored in a table, and a sequential number is assigned in the order of appearance of the data. A method for storing XML data for storing the sequence numbers in a table is disclosed. In the XML data described in Patent Document 1, that is, the structured document storage method, the parent-child relationship between elements can be determined at high speed based on the order number of the element stored in the table and the order number of the parent element of the element. It is also possible to determine the hierarchical relationship between data.

Japanese Patent No. 3560043

  However, the structured document storage method described in Patent Document 1 cannot immediately determine the ancestor-descendant relationship between elements. Alternatively, a method of determining the ancestor-descendant relationship by tracing the elements while sequentially determining the parent-child relationship between the elements is conceivable, but it is a problem that the method cannot be determined at high speed.

  The present invention has been made in view of the above points, and provides a structured document storage device, a structured document storage method, and a structured document that can quickly determine an ancestor-descendant relationship between nodes constituting a structured document. An object is to provide a document storage program.

  In order to achieve the above object, a structured document storage device according to a first aspect of the invention comprises a structured document analyzing means for analyzing a structured document and outputting information about nodes included in the structured document, and a structured document. A storage means having a node table for storing information relating to the nodes of the node, a sequence number representing the order of appearance of the target nodes in the information relating to the nodes of the structured document output from the structured document analyzing means, and a target node The order number of the node that appears first in the information excluding the descendant node of the target node is determined as the next number to be assigned to the target node, and the order number and the next number together with the information about the target node are determined. Storage means for storing in the node table and search means for outputting the search result by referring to the storage information of the node table according to the supplied search procedure list When the order number of the comparison source node in the stored information is larger than the order number of the comparison destination node and the order number of the comparison source node is smaller than the next number of the comparison destination node, the search means It is characterized by having descendant node search means for judging that it is a descendant node of the comparison destination node.

  In order to achieve the above object, the structured document storage device according to the second aspect of the invention is characterized in that the structured document in the first aspect is not defined with the first node in which the order of appearance is defined and the order of appearance. And the storage means assigns an integer greater than or equal to 0 that increases by 2 in the order of appearance of the first node to the order number of the first node, and the order number of the second node A value obtained by adding 1 to the order number of the parent node that is one node above the second node is assigned, and the descendant node search means adds the order number of the comparison source node to the order number of the comparison destination node. The comparison source node is determined to be a descendant node of the comparison destination node when the value is larger than the value and the order number of the comparison source node is smaller than the next number of the comparison destination node.

  In order to achieve the above object, a structured document storage method according to a third aspect of the invention comprises a structured document analysis step of analyzing a structured document and outputting information on nodes included in the structured document; A sequence number representing the appearance order of the target node in the information related to the node of the structured document obtained by the structured document analysis step is determined, and the target node among the nodes appearing in the information after the target node is determined. A node in which the order number of the node that appears first except for the descendant node is determined as the next number to be assigned to the target node, and the order number and the next number are stored together with the information about the target node, and the information about the node of the structured document is stored. A storage step for storing in the table, and a search step for outputting the search result by referring to the storage information of the node table according to the supplied search procedure list. In the search step, when the sequence number of the comparison source node in the stored information is larger than the sequence number of the comparison destination node and the sequence number of the comparison source node is smaller than the next number of the comparison destination node, the comparison source node The method includes a descendant node search step for determining that the node is a descendant node of the comparison destination node.

  In order to achieve the above object, the structured document storage method according to the fourth aspect of the present invention is the structured document according to the third aspect, in which the first node in which the order of appearance is defined and the order in which the order of appearance is not defined. And the storing step assigns an integer greater than or equal to 0 increasing in order of appearance of the first node to the sequence number of the first node, and assigns the sequence number of the second node to the sequence number of the first node. A value obtained by adding 1 to the order number of the parent node that is one node higher than the node 2 and the descendant node search step is obtained by adding 1 to the order number of the comparison destination node. The comparison source node is determined to be a descendant node of the comparison destination node when the comparison source node is larger and the order number of the comparison source node is smaller than the next number of the comparison destination node.

  In order to achieve the above object, a structured document storage program according to a fifth aspect of the present invention provides a computer with a structured document analysis step, a storage step, and a descendant node search step in the structured document storage method of the third aspect of the invention. A search step including: is executed.

  Furthermore, to achieve the above object, the structured document storage program according to the sixth aspect of the invention is the structured document according to the fifth aspect, in which the appearance order is not defined in the first node in which the order of appearance is defined. And the storing step assigns an integer greater than or equal to 0 increasing in order of appearance of the first node to the sequence number of the first node, and assigns the sequence number of the second node to the sequence number of the first node. A value obtained by adding 1 to the order number of the parent node that is one node higher than the node 2 is assigned, and the descendant node search means adds 1 to the order number of the comparison destination node. In this configuration, the comparison source node is determined to be a descendant node of the comparison destination node when the comparison number is larger than the value and the order number of the comparison source node is smaller than the next number of the comparison destination node.

  According to the present invention, the procedure for searching the ancestor-descendant relationship between the nodes of the stored structured document can be performed in a short time compared to the conventional method. As a result, the structured document can be searched compared with the conventional method. It can be done at high speed.

It is a block diagram of one embodiment of a structured document storage device of the present invention. It is a figure which shows an example of a structure of the node table in the relational database in FIG. FIG. 3 is a diagram illustrating an example of a node type list in the node table of FIG. 2. 3 is a flowchart for explaining document storage processing in a storage unit of the structured document storage device in FIG. 1. It is a flowchart explaining the detail of the node storing process in FIG. It is a flowchart explaining the detail of the attribute storage process in FIG. It is a figure which shows an example of an XML document. It is a figure which shows the example of the node table which stored the XML document of FIG. It is a figure which shows the list of the classification of the search procedure which the XQuery analysis part in FIG. 1 produces | generates. It is a flowchart explaining the search process in the search part in FIG. It is a flowchart explaining the detail of the search procedure execution process in FIG. 12 is a flowchart illustrating details of a descendant node search process in FIG. 11. 12 is a flowchart illustrating details of an ancestor node search process in FIG. 11. It is a flowchart explaining the detail of the subtree data extraction process in FIG. It is a figure which shows an example of the search query input into the XQuery analysis part in FIG. It is a figure which shows an example of the search procedure list | wrist produced | generated by the XQuery analysis part in FIG. FIG. 17 is a diagram showing transition of a context node list by a search process performed by the search unit in FIG. 1 using the search procedure list in FIG. 16. It is a figure which shows an example of the search result output from the XML shaping | molding part in FIG.

  DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 shows a block diagram of an embodiment of a structured document storage apparatus according to the present invention.

  As shown in FIG. 1, the structured document storage device 1 of the present embodiment includes an XML analysis unit 101, a storage unit 102, a relational database 103, a query (XQuery) analysis unit 104, a search unit 105, An XML molding unit 106 is included. The structured document storage device 1 stores an XML document that is an input structured document in the relational database 103 and searches the XML document in the search unit 105 according to the input search query, and matches the search query. A structured document storage device that outputs a list of nodes as a search result.

  The XML analysis unit 101 analyzes an XML document input as text data, and generates DOM (Document Object Model) data in which objects corresponding to each node constituting the XML document are arranged in a tree structure. The data is supplied to the storage unit 102.

  Here, an outline of the XML document will be described. As is well known, in an XML document, a set of text and elements is described between a start tag represented by “<element name>” and an end tag represented by “</ element name>”. An element completed with one tag is called an empty element. In addition, an attribute can be described as “<element name attribute name =“ attribute value ”>” in the start tag or empty element tag. An XML document that is a structured document can be converted into a tree structure based on the XML logical structure. Each element, attribute, and text of the XML document is called a node in the tree structure.

  In this node, the expressions “parent node”, “child node”, “descendant node”, and “ancestor node” may be used. The “parent node” is a node one level higher than the processing target node, the “child node” is a node one level lower than the processing target node, and the “descendant node” is all the lower level nodes of the processing target node. “Ancestor node” indicates all upper nodes of the processing target node. Since the element is a kind of node, the start tag can be expressed as “<node name>” and the end tag can be expressed as “</ node name>”.

  When the DOM data is supplied from the XML analysis unit 101, the storage unit 102 stores an XML document represented by the supplied DOM data in the relational database 103. In terms of its function, the storage unit 102 includes a node ID determination unit 102a, A sequence number determining unit 102b, a node data inserting unit 102c, an attribute data inserting unit 102d, an attribute storing unit 102e, and a child node storing unit 102f are provided.

  The node ID determination unit 102a determines a node ID that uniquely identifies the processing target node. Specifically, the node ID determination unit 102a selects a positive integer other than 0 that does not match any of the node IDs stored in the node table 2 in the relational database 103, and selects it as the node of the processing target node. ID.

  The order number determination unit 102b determines the order number representing the order of appearance of the nodes for the processing target node. Further, the sequence number determination unit 102b determines the next number, which is the sequence number of the node that appears first, excluding the descendant nodes of the node to be processed among the nodes that appear after the node to be processed. . However, if all nodes that appear after the processing target node are descendant nodes of the processing target node, the next number is a virtual node when it is assumed that there is a virtual node that is not a descendant node of the processing target node after the processing target node It becomes order number.

  The node data insertion unit 102 c inserts the node data of the processing target node into the node table 2 in the relational database 103. The attribute data insertion unit 102 d inserts the attribute data of the processing target attribute into the node table 2 in the relational database 103 when the processing target node is an attribute that is a kind of node.

  The attribute data insertion unit 102e performs processing for storing an attribute held by the processing target node, that is, an attribute that is a child node of the processing target node. The child node storage unit 102f performs processing for storing child nodes other than the attribute of the processing target node.

  The relational database 103 is configured by a storage device such as a hard disk and a processing unit that reads and writes the storage device and provides a function as a relational database. The relational database 103 has a tabular node table 2 and is stored in the node table 2. It has a function to search data.

  When a search query described in the XQuery (XML Query Language) format is input, the XQuery analysis unit 104 analyzes the input search query and generates a search procedure list for executing search processing. Generated and supplied to the search unit 105. XQuery is formulated as an inquiry language for searching XML data.

  When the search procedure list is supplied from the XQuery analysis unit 104, the search unit 105 executes a search on the XML document stored in the relational database 103 in accordance with the supplied search procedure list, and a list of subtree data as search results To the XML forming unit 106. In terms of its function, the search control unit 105a, the node type search unit 105b, the node name search unit 105c, the node value search unit 105d, the child node search unit 105e, A parent node search unit 105f, a descendant node search unit 105g, an ancestor node search unit 105h, and a subtree data extraction unit 105i are provided.

  The search control unit 105 a controls the entire search process in the search unit 105. The node type search means 105b executes a node type search when the search procedure is a node type search. The node name search means 105c executes a search by node name when the search procedure is a node name search. The node value search unit 105d executes a search based on the node value when the search procedure is a node value search.

  The child node search means 105e executes a child node search when the search procedure is a child node search. When the search procedure is a parent node search, the parent node search unit 105f executes a parent node search. The descendant node search means 105g executes a search for a descendant node when the search procedure is a descendant node search. The ancestor node search means 105h executes ancestor node search when the search procedure is ancestor node search.

  The subtree data extraction unit 105i reads the subtree data of each node in the node list of the search result from the relational database 103, rearranges the read subtree data in the order of the order numbers included in the node data, and supplies the rearranged data to the XML forming unit 106. The “node subtree data” is node data of each node of the subtree including the target node and all nodes (descendant nodes) that are subordinate to the target node in the tree structure of the XML document.

  The search unit 105 includes a search procedure directly related to the present invention among search procedures that can be described by XQuery, and search units 105b to 105h for executing a representative search procedure. It is desirable to have a configuration including a search means for executing other search procedures that can be described by XQuery.

  When the subtree data list is supplied from the search unit 105, the XML forming unit 106 forms the supplied subtree data into an XML document format described as text data, and outputs the result as a search result.

  Next, the configuration of a tabular node table for storing node data in the relational database 103 of the structured document storage device 1 will be described.

  FIG. 2 shows the configuration of the node table in the relational database 103. In FIG. 2, the “column name” column of the node table 2 indicates the name of each column of the table, and the “content” column indicates the content of data stored in each column.

  The node table 2 includes an id column 201, a type column 202, a parent column 203, an order column 204, a next column 205, a name column 206, and a value column 207.

  The id column 201 is a column that stores a positive integer other than 0, which is a node ID that uniquely identifies the target node. The type column 202 is a column for storing numerical values representing the node type (type) of the target node. The parent column 203 is a column that stores an integer other than “0” that is the node ID of the parent node of the target node, or a numerical value “0” indicating that the parent node does not exist.

  The order column 204 is a column that stores the order numbers of the target nodes. In XML, there are nodes that are not defined as nodes in which the appearance order is defined depending on the type of node, but the order number stored in the order column 204 is “0” in the order of appearance for the nodes in which the order of appearance is defined. The number is an even number obtained by allocating every second integer starting from “,” and for a node whose appearance order is not defined, an odd value obtained by adding “1” in the order of appearance of the parent node.

  The next column 205 is a column for storing the next number of the target node. However, if the target node is a node whose appearance order is not defined, the same value as the order number is stored. The name column 206 is a column for storing the node name (name) of the target node. However, if the node type of the target node does not have a node name, the name column 206 is empty (null). The value column 207 is a column for storing the node value (value) of the target node. However, when the node type of the target node does not have a node value, the value column 207 is empty (null).

  Next, numerical values representing the node types stored in the type column 202 in the node table 2 of the structured document storage device 1 according to the embodiment of the present invention shown in FIG. 2 will be described.

  FIG. 3 shows a list of node types stored in the type column 202 in the node table 2. The column “type value” in FIG. 3 indicates a numerical value indicating the node type stored in the type column 202 in the node table 2, and the column “node type” indicates a node type corresponding to the “type value”. .

  The node type includes an element 301, an attribute 302, a text node 303, a CDATA section 304, an entity reference 305, and an entity. 306, Processing Instruction 307, Comment 308, Document 309, Document Type 310, Document Fragment 311, and Notation 312 There is. A numerical value shown in the “type value” column is assigned to each of these node types.

  In XML, a node type other than the attribute 302 defines a node appearance order, but the attribute 302 does not define an appearance order. That is, in two XML documents composed of the same node, if the appearance order of nodes other than the attribute 302 is different, the XML documents are different. However, if the apparent appearance order of the attribute 302 is different, the same XML document is used. It becomes.

  Also, in XML, some node types including the attribute 302 have no child nodes. In XML, a descendant node of a processing target node does not include an attribute held by the processing target node, that is, an attribute that becomes a child node of the processing target node.

  Next, details of document storage processing in the storage unit 102 of the structured document storage device 1 according to the embodiment of the present invention will be described.

  FIG. 4 is a flowchart for explaining document storage processing in the storage unit 102 of the structured document storage device 1 of this embodiment.

  When the DOM data is supplied from the XML analysis unit 101, the node ID determination unit 102a of the storage unit 102 starts document storage processing and substitutes “0” for the variable p (step S401). The variable p is a variable for storing the node ID of the parent node. Next, the sequence number determination unit 102b of the storage unit 102 substitutes “0” for the variable o (step S402). The variable o is a variable for storing a sequential number.

  Next, the child node storage unit 102f of the storage unit 102 sets the root node as a processing target node (step S403). The root node is a node that does not have a parent node, and the value of the parent column 203 in the node table 2 is “0”. Then, the storage unit 102 executes a node storage process described later (step S404). Note that the variable p assigned the value in step S401, the variable o assigned the value in step S402, and the processing target node determined in step S403 are taken over by the node storing process.

  Next, details of the node storage process in step S404 of the document storage process in the storage unit 102 of the structured document storage apparatus 1 in FIG. 4 will be described.

  FIG. 5 is a flowchart for explaining node storage processing in the storage unit 102 of the structured document storage device 1 of this embodiment.

  When the node ID determination unit 102a of the storage unit 102 starts the node storage process, the node ID determination unit 102a determines the node ID of the processing target node determined in step S403 or step S515, and substitutes the determined node ID into the variable i (step S501). . Specifically, a positive integer other than “0” that does not match any node ID stored in the id column 201 of the node table 2 is selected as a node ID.

  Next, the node ID determination unit 102a of the storage unit 102 sets the value of the variable p as a parent node ID of the processing target node, that is, a value to be inserted into the parent column 203 of the node table 2 (step S502). Next, the sequence number determination unit 102b of the storage unit 102 sets the value of the variable o as the sequence number of the processing target node, that is, a value to be inserted into the order column 204 of the node table 2 (step S503).

  Next, the node data insertion unit 102c of the storage unit 102 inserts node data other than the next number of the processing target node into the node table 2 (step S504). Specifically, the node ID assigned to the variable i in step S501 is used as the value of the node ID in the id column 201, the node type is acquired from the DOM object of the processing target node, and the node type in the node type list 3 is supported. The type value to be used is the value in the type column 202. Further, the parent node ID determined in step S502 is set as the value in the parent column 203, and the order number determined in step S503 is set as the value in the order column 204. Further, the node name is acquired from the DOM object of the processing target node, the node name is set as the value of the name column 206, the node value is acquired from the DOM object of the processing target node, and the value is set as the value of the value column 207. , A new row is added to the node table 2. The next column 205 of the row to be added is left empty (null).

  Next, the node ID determination unit 102a of the storage unit 102 substitutes the value of the variable i for the variable p (step S505). Next, the sequence number determination unit 102b of the storage unit 102 adds “1” to the value of the variable o and substitutes it into the variable o (step S506). Next, the attribute storage unit 102e of the storage unit 102 substitutes the attribute of the processing target node, that is, the number of attributes that are child nodes of the processing target node, into the variable Na, and substitutes “0” into the variable a (step) S507). Subsequently, the attribute storage unit 102e of the storage unit 102 determines whether or not the value of the variable a is smaller than the value of the variable Na (step S508). If it is determined that the value of the variable a is smaller than the value of the variable Na, the process proceeds to step S509. If it is determined that the value of the variable a is not smaller than the value of the variable Na, that is, greater than or equal to the value of the variable Na, the process proceeds to step S511. Migrate processing.

  If it is determined in step S508 that the value of the variable a is smaller than the value of the variable Na, the attribute storage unit 102e of the storage unit 102 adds “1” to the value of the variable a and substitutes it for the variable a (step S508). S509). Subsequently, the attribute storage unit 102e of the storage unit 102 sets a variable a-th attribute after the addition as a processing target and executes an attribute storage process to be described later (step S510). In addition, since the order is not defined between the attributes of a certain node, for example, the attributes may be arranged in the order of the dictionary of the node names and set as the order of the attributes. In addition, the variable p and the variable o and the processing target attribute determined in this step are taken over by the attribute storing process.

  On the other hand, if it is determined in step S508 that the value of the variable a is not smaller than the value of the variable Na, that is, greater than or equal to the variable Na, the sequence number determination unit 102b of the storage unit 102 sets “1” to the value of the variable o. Are added to variable o (step S511).

  Next, the child node storage unit 102f of the storage unit 102 substitutes the number of child nodes other than the attribute of the processing target node into the variable Nc, and substitutes “0” into the variable c (step S512). Subsequently, the child node storage unit 102f determines whether or not the value of the variable c is smaller than the value of the variable Na (step S513). If it is determined that the value of the variable c is smaller than the value of the variable Nc, the process proceeds to step S514. If it is determined that the value of the variable c is not smaller than the value of the variable Nc, that is, greater than or equal to the value of the variable Nc, the process proceeds to step S516. Migrate processing.

  When it is determined in step S513 that the value of the variable c is smaller than the value of the variable Nc, the child node storage unit 102f of the storage unit 102 adds “1” to the value of the variable c and substitutes it into the variable c ( Step S514). Subsequently, the child node storage unit 102f sets the variable c-th child node after the addition as a processing target, and executes node storage processing recursively (step S515). At this time, the child nodes of the processing target node are arranged in the order of appearance to obtain the child node order. In addition, the variable p and the variable o and the node to be processed determined in step S515 are succeeded to the node storing process that is recursively executed.

  On the other hand, if it is determined in step S513 that the value of the variable c is not smaller than the value of the variable Nc, that is, greater than or equal to the value of the variable Nc, the sequential number determination unit 102b of the storage unit 102 processes the value of the variable o. The next number of the target node is set (step S516). Note that the processing target node in step S516 is the processing target node at the start of the node storing process, and is not the processing target node determined in step S515.

  Next, the node data insertion unit 102c of the storage unit 102 sets the next number of the processing target node in the node table 2 (step S517). Specifically, the id column 201 of the node table 2 is searched using the node ID of the processing target node stored in the variable i, and the value of the node ID in the id column 201 matches the value of the variable i. The value of the next column 205 of the row to be set is set to the value of the next number determined in step S516.

  Next, details of the attribute storage processing in step S510 of the node storage processing in the storage unit 102 of the structured document storage device 1 of the present embodiment in FIG. 5 will be described with reference to the flowchart in FIG.

The node ID determination unit 102a of the storage unit 102 determines the node ID of the processing target attribute when the attribute storage process of step S510 is started (step S601). A specific determination method is the same as that in step S501. Subsequently, the node ID determination unit 102a sets the value of the variable p as a parent node ID of the processing target attribute, that is, a value to be inserted into the parent column 203 of the node table 2 (step S602).
Next, the sequence number determination unit 102b of the storage unit 102 sets the value of the variable o as the sequence number of the processing target attribute, that is, the value to be inserted into the order column 204 of the node table 2, and sets the same value as the processing target attribute. The next number, that is, a value to be inserted into the next column 205 of the node table 2 is set (step S603).

  Next, the attribute data insertion unit 102d of the storage unit 102 inserts the node data of the processing target attribute into the node table 2 (step S604). Specifically, the node ID determined in step S601 is the value of the node ID in the id column 201, the type value corresponding to the attribute 302 in the node type list 3 is the value of the type column 202, and the parent node determined in step S602 The ID is the value of the parent column 203, the order number determined in step S603 is the value of the order column 204, the next number determined in step S603 is the value of the next column 205, and the node name is acquired from the DOM object of the processing target attribute Then, the node name is set as the value of the name column 206, the node value is acquired from the DOM object of the processing target attribute, and the node value is set as the value of the value column 207 to add a new row to the node table 2.

  Next, a specific example of the XML document stored in the node table 2 in the structured document storage device 1 according to the embodiment of the present invention will be described.

  FIG. 7 shows a specific example of an XML document stored in the node table 2 in the structured document storage device 1. A document showing the entire XML document 7 shown in FIG. As a child node of the root node, there is an element whose element name is “library”. A part surrounded by angle brackets such as “book” or “title” is also an element. An element whose element name is “book” has an attribute whose node name is “id”. Also, “Title 1” sandwiched between the start tag “<title>” and the end tag “</ title>”, and “Author” sandwiched between the start tag “<author>” and the end tag “</ author>”. The part of the character string such as “1” is a text node whose character string is a node value.

  FIG. 8 shows a specific example of the node table 2 in the structured document storage device 1 of the present embodiment. A node table example 8 shown in FIG. 8 shows a state in which the XML document 7 shown in FIG. 7 is stored. That is, in FIG. 8, the id column stores a node ID value that increases by “1” with “1” as an initial value. In FIG. 8, as shown in FIG. 2, the type column stores type values in the node type list 3 (that is, values for identifying node types). The parent column ID value is stored in the parent column, the order number is stored in the order column, the next number is stored in the next column, the node name (element name) is stored in the name column, and the node value is stored in the value column. Has been.

  For example, for the start tag “<library>” on the first line of the XML document 7, the value of the node ID in the id column is “2”, and the type value in the type column is “1” indicating that the node type is an element. , “1” is stored as the value of the parent node ID in the parent column. Further, “<library>” of the node ID value “2” is a node in which the order of appearance is defined, and the order number of the even value “2” is stored in the parent column in the order of appearance, and the node ID value as the next number The same “34” as “1” is stored, and the node name “library” is stored in the name column.

  Since “<book id =“ 1 ”>” in the second line of the XML document 7 includes a node “book” in which the occurrence is defined and a node “id =“ 1 ”” in which the occurrence is not defined, the value of the node ID In the column “3”, the type value is “1”, the parent node ID value is “2”, the next number “4” in the order of appearance, the next number “14”, the node name “book” "Is stored correspondingly. The subsequent node ID value is “4”, the type value is “2” indicating the attribute, the parent node ID value is “3”, the node name “id”, and “1” is stored as the node value in the value column. Is done. Further, since the occurrence of “id =“ 1 ”” is not defined, the sequential number is “5” obtained by adding “1” to the sequential number of the parent node “book”, and the next number is “5”. It becomes.

  In the node table example 8 in FIG. 8, for convenience, the rows are arranged in the order in which the corresponding nodes appear, but the order of the rows of the node table 2 in the relational database 103 is indefinite.

  Next, details of a search procedure generated by the XQuery analysis unit 104 of the structured document storage device 1 according to the embodiment of the present invention will be described. Note that details of the process of generating a search procedure by analyzing a query (XQuery) are not directly related to the present invention, and thus the description thereof is omitted.

  FIG. 9 shows a list of types of search procedures generated by the XQuery analysis unit 104 of the structured document storage device 1 of this embodiment. In FIG. 9, the “search type” column indicates the type of search procedure, the “argument” column indicates an argument associated with the search procedure, and the “processing content” column indicates the content of the search process.

  The types of search procedures include a node type search 901, a node name search 902, a node value search 903, a child node search 904, a parent node search 905, a descendant node search 906, and an ancestor node search 907. .

  The node type search 901 is a procedure for taking a type value of the node type list 3 as an argument and searching for a node whose numerical value representing the node type matches the type value of the argument. Node name search 902 is a procedure for taking a character string as an argument and searching for a node whose node name matches the argument character string. The node value search 903 is a procedure for taking a character string as an argument and searching for a node whose node value matches the argument character string.

  The child node search 904 is a procedure for searching a child node of a context node without taking an argument. The “context node” is a node searched by executing a search procedure immediately before a search procedure to be executed in the process of executing a list of search procedures.

  The parent node search 905 is a procedure for searching for a parent node of a context node without taking an argument. The descendant node search 906 is a procedure for searching for a descendant node of a context node without taking an argument. An ancestor node search 907 is a procedure for searching an ancestor node of a context node without taking an argument.

  The search procedure list 9 includes search procedures 901 to 907 for executing a search procedure directly related to the present invention and a typical search procedure among search procedures that can be described by a query (XQuery). However, it is desirable to include a search means for executing other search procedures that can be described by a query (XQuery).

  Next, details of the search processing in the search unit 105 of the structured document storage device 1 of the present embodiment will be described.

  FIG. 10 is a flowchart for explaining search processing in the search unit 105. When the search procedure list is supplied from the XQuery analysis unit 104, the search control unit 105a in the search unit 105 shown in FIG. 1 starts the search process, acquires the node ID of the root node, and adds it to the context node list (Step S1001). Specifically, a row in which the value of the parent column 203 in the node table 2 in the relational database 103 is “0” is retrieved, and the value of the node ID in the id column 201 of that row is acquired as the node ID of the root node. Then, a context node list including only the node ID of the acquired root node is generated.

  Next, the search control means 105a in the search unit 105 substitutes the number of search procedures included in the search procedure list supplied from the XQuery analysis unit 104 into a variable N, and substitutes “0” into a variable i (step) S1002).

  Next, the search control means 105a in the search unit 105 determines whether or not the value of the variable i is smaller than the value of the variable N (that is, the number of search procedures included in the search procedure list) (step S1003). When it is determined that the value of the variable i is smaller than the value of the variable N, the search control unit 105a adds “1” to the value of the variable i and substitutes it into the variable i (step S1004). Subsequently, the search control unit 105a executes a variable i-th search procedure included in the search procedure list supplied from the XQuery analysis unit 104 (step S1005). After the process of step S1005, the process returns to step S1003, and it is determined again whether or not the value of the variable i is smaller than the value of the variable N.

  On the other hand, when it is determined in step S1002 that the value of the variable i is not smaller than the value of the variable N, that is, the value of the variable N is greater than or equal to the value of the variable N, the search control unit 105a It is determined that the search procedure has been executed. Subsequently, the number of nodes included in the context node list is substituted into the variable M, and “0” is substituted into the variable j (step S1006).

  Next, the search control unit 105a determines whether or not the value of the variable j is smaller than the value of the variable M (that is, the number of nodes included in the context node list) (step S1007). When it is determined that the value is smaller than the value of the variable M, the search control unit 105a adds “1” to the value of the variable j and substitutes it for the variable j (step S1008). Subsequently, the search control unit 105a executes a subtree data extraction process of the variable j-th node included in the context node list as described later (step S1009). After the process of step S1009, the process returns to step S1007 to determine whether or not the value of the variable j is smaller than the value of the variable M.

  On the other hand, if it is determined in step S1007 that the value of the variable j is not smaller than the value of the variable M, that is, the search control unit 105a determines that the value of the variable M is greater than or equal to the value of the variable M, It is determined that the subtree data extraction processing has been executed, and the subtree data extracted in step S1009 is rearranged by node order number (step S1010). Specifically, the subtree data is rearranged in ascending order of the order numbers using the order numbers of the context nodes included in the subtree data corresponding to the context nodes extracted in step S1009. As a result, the subtree data of the search results are arranged in the order of appearance in the XML document.

  Next, details of the search procedure execution processing in step S1005 by the search unit 105 shown in FIG. 10 will be described with reference to the flowchart of FIG.

  When starting the search procedure execution process, the search control unit 105a of the search unit 105 determines whether or not the search procedure to be executed is a node type search 901 (step S1101). When it is determined that the node type search 901 is performed, the node type search unit 105b of the search unit 105 illustrated in FIG. 1 selects a node whose node type matches the condition from the context node list (step S1102). Specifically, the node type search means 105b matches the node ID of the id column 201 with each node ID included in the context node list and the node type of the type column 202 with respect to the node table 2. A row whose value matches the type value that is an argument of the search procedure is searched, the value of the node ID in the id column 201 of each row that matches the search is read as the node ID, and the context node list is obtained from the list of read node IDs. replace.

  On the other hand, if it is determined in step S1101 that the search procedure to be executed is not the node type search 901, the search control unit 105a determines whether or not the search procedure to be executed is the node name search 902 (step S1103).

  If it is determined in step S1103 that the search procedure to be executed is the node name search 902, the node name search unit 105c of the search unit 105 illustrated in FIG. 1 selects a node whose node name matches the condition from the context node list. Select (step S1104). Specifically, the node name search unit 105c matches the node ID in the id column 201 with each node ID included in the context node list and the node name in the name column 206 is the same as the node table 2. A line that matches the character string that is an argument of the search procedure is searched, the value of the node ID in the id column 201 of each line that matches the search is read as the node ID, and the context node list is replaced with the read list of node IDs.

  On the other hand, if it is determined in step S1103 that the search procedure to be executed is not the node name search 902, the search control unit 105a of the search unit 105 determines whether or not the search procedure to be executed is the node value search 903 (step S1103). S1105).

  If it is determined in step S1105 that the search procedure to be executed is the node value search 903, the node value search unit 105d of the search unit 105 illustrated in FIG. 1 selects a node whose node value matches the condition from the context node list. Select (step S1106). Specifically, the node value search unit 105d determines that the node ID value in the id column 201 matches each node ID included in the context node list and the node value in the value column 207 matches the node table 2. A line that matches the character string that is an argument of the search procedure is searched, the value of the node ID in the id column 201 of each line that matches the search is read as the node ID, and the context node list is replaced with the read list of node IDs.

  On the other hand, if it is determined in step S1105 that the search procedure to be executed is not the node value search 903, the search control unit 105a of the search unit 105 determines whether or not the search procedure to be executed is a child node search 904 (step S1105). S1107).

  If it is determined in step S1107 that the search procedure to be executed is the child node search 904, the child node search unit 105e of the search unit 105 illustrated in FIG. 1 searches for a child node of each node in the context node list ( Step S1108). Specifically, the child node search unit 105e searches the node table 2 for a row in which the value of the parent node ID in the parent column 203 matches each node ID included in the context node list, and matches the search. The node ID value in the id column 201 of each row is read as the node ID, and the context node list is replaced with the read node ID list.

  On the other hand, if it is determined in step S1107 that the search procedure to be executed is not the child node search 904, the search control unit 105a of the search unit 105 determines whether or not the search procedure to be executed is the parent node search 905 (step S1107). S1109).

  If it is determined in step S1109 that the search procedure to be executed is the parent node search 905, the parent node search means 105f of the search unit 105 shown in FIG. 1 searches for the parent node of each node in the context node list ( Step S1110). Specifically, the parent node search unit 105f searches the node table 2 for a row in which the value of the node ID in the id column 201 matches each node ID included in the context node list, and each row that matches the search. The parent node ID value in the parent column 203 is read as a node ID, and the context node list is replaced with the read node ID list.

  On the other hand, if it is determined in step S1109 that the search procedure to be executed is not the parent node search 905, the search control means 105a of the search unit 105 determines whether or not the search means to be executed is a descendant node search 906 (step S1109). S1111).

  If it is determined in step S1111 that the search procedure to be executed is the descendant node search 906, the descendant node search unit 105g of the search unit 105 illustrated in FIG. 1 executes a later-described descendant node search process (step S1112). . On the other hand, if it is determined in step S1111 that the search procedure to be executed is not the descendant node search 906, the search control unit 105a of the search unit 105 determines whether the search procedure to be executed is the ancestor node search 907 (step S1111). S1113).

  If it is determined in step S1113 that the search procedure to be executed is the ancestor node search 907, the ancestor node search means 105h of the search unit 105 shown in FIG. 1 executes an ancestor node search process to be described later (step S1114). . On the other hand, if it is determined in step S1113 that the search procedure to be executed is not the ancestor node search 907, the search control unit 105a of the search unit 105 executes other search procedures not listed in the search procedure list 9 (step S1115). ). The details of the other search procedure execution processing will be omitted.

  Next, details of the descendant node search processing in step S1112 by the search unit 105 shown in FIG. 11 will be described with reference to the flowchart of FIG.

  When the descendant node search unit 105g of the search unit 105 starts the descendant node search process, the node table 2 searches the node table 2 for a row where the value of the node ID in the id column 201 matches the node ID included in the context node list. (Step S1201).

  Next, the descendant node search unit 105g of the search unit 105 sets the order number value of the order column 204 to “1” for the order number of the order column 204 of each row searched in step S1201 for the node table 2. A row that is larger than the value obtained by adding and less than the next number value in the next column 205 of each row retrieved in step S1201 is retrieved as a descendant node (step S1202).

  Next, the descendant node search unit 105g of the search unit 105 reads the descendant node searched in step S1202 and the node ID value of the id column 201 of each searched row, and displays the context node list in the read node ID list. Replace (step S1203).

  In step S1202, by adding “1” to the order number value in the order column 204 and performing the search, there is an effect that the attribute of the context node can be excluded from the search for the descendant nodes.

  Next, details of the ancestor node search processing in step S1114 by the search unit 105 shown in FIG. 11 will be described with reference to the flowchart of FIG.

  When the ancestor node search unit 105h of the search unit 105 starts the ancestor node search process, the node table 2 searches the node table 2 for a row whose node ID value in the id column 201 matches the node ID included in the context node list. (Step S1301).

  Next, the ancestor node search unit 105h of the search unit 105 sets the order number value in the order column 204 to the node table 2 to be smaller than the order number value in the order column 204 of each row searched in step S1301. A row whose value of the next number in the column 205 is larger than the value of the order number in the order column 204 of each row searched in step S1301 is searched (step S1302).

  Next, the ancestor node search unit 105h of the search unit 105 reads the value of the node ID in the id column 201 of each row searched in step S1302, and replaces the context node list with the list of read node IDs (S1303).

  Next, details of the subtree data extraction processing in step S1009 by the search unit 105 shown in FIG. 10 will be described with reference to the flowchart of FIG.

  When the subtree data extraction unit 105i of the search unit 105 illustrated in FIG. 1 starts the subtree data extraction process, a row in which the value of the node ID in the id column 201 matches the node ID of the processing target node with respect to the node table 2. Is searched (step S1401).

  Next, the subtree data extraction unit 105i of the search unit 105 determines that the order number value in the order column 204 is greater than or equal to the order number value in the order column 204 of the row searched in step S1401 for the node table 2. A line smaller than the value of the next number in the next column 205 of the line searched in step S1401 is searched (step S1402). Then, the subtree data extraction unit 105i of the search unit 105 reads the value of the id column of each row searched in step S1402 (step S1403).

  Next, a specific example of search in the structured document storage device 1 according to the embodiment of the present invention will be described.

  FIG. 15 shows an example of a search query input to the XQuery analysis unit 104 in the structured document storage device 1 according to the embodiment of this invention. The search query 15 shown in FIG. 15 is based on the assumption that the XML document example 7 is stored, and is used when searching for a “book” element that includes a text node whose node value matches “Author 1” as a descendant. It is.

  FIG. 16 shows an example of a search procedure list generated by the XQuery analysis unit 104. This search procedure list 16 is generated by analyzing the search query 15 and includes six descending node search, node type search, node value search, ancestor node search, node type search, and node name search. The search procedure is shown.

  FIG. 17 shows an example of transition of the context node list in the search process of the search unit 105 of the structured document storage device 1 according to the embodiment of the present invention. This example shows the transition of the context node list when the text node search process is performed using the search procedure list 16 shown in FIG.

  That is, as indicated by 170 in FIG. 17, the context node list initially includes only the node ID = 1 of the root node shown in the first row of the column of the id column of the node table example 8 shown in FIG.

  Next, the search unit 105 executes a descendant node search of the search procedure 1 shown in FIG. Accordingly, as indicated by 170 in FIG. 17, the context node list includes “2”, “3”, “4”, “5” which are node IDs of descendant nodes in the node table example 8 shown in FIG. , “6”, “7”, “8”, “9”, “10”, “11”, “12”, “13”, “14”, “15”, “16”, “17”, “ “18”, “19”, and “20” are described.

  Subsequently, the search unit 105 executes a node type search of the argument 3 in the search procedure 2 shown in FIG. Accordingly, as indicated by reference numeral 171 in FIG. 17, the context node list includes a node ID indicating the node type of the text node whose type value is “3” from the argument 3 in the node table example 8 illustrated in FIG. “6”, “8”, “12”, “14”, “18”, “20” are described.

  Subsequently, the search unit 105 executes a node value search for the argument “Author1” in the search procedure 3 shown in FIG. Accordingly, as indicated by 173 in FIG. 17, the context node list includes “8” in which “value” indicating the node value from the argument “Author1” is “Author1” in the node table example 8 illustrated in FIG. 8. , “20” is described.

  Subsequently, the search unit 105 executes an ancestor node search of the search procedure 4 shown in FIG. Accordingly, as indicated by 174 in FIG. 17, the context node list includes “1”, which is the ancestor node of the node ID “8” among the node IDs “8” and “20” as the search results of the search procedure 3. , “2”, “3”, and “7”, and “15” and “19” that are ancestor nodes of the node D “20” are described.

  Subsequently, the search unit 105 executes a node type search of argument 1 in the search procedure 5 shown in FIG. Accordingly, as indicated by 175 in FIG. 17, the context node list includes a node indicating the node type of the text node of the argument 1 (type value is “1”) shown in FIG. IDs “2”, “3”, “7”, “15”, and “19” are described.

  Then, the search unit 105 executes a node name search for the argument “book” in the last search procedure 6 shown in FIG. Accordingly, as indicated by 176 in FIG. 17, the node ID of the search result whose node name shown in the column of the name column in FIG. 8 is “book” in the search result of the search procedure 5 is displayed in the context node list. “3” and “15” are described. Thus, each time the search unit 105 executes the search procedure one by one, the node ID included in the context node list changes, and finally a desired search result based on the search query is obtained.

  FIG. 18 shows an example of a search result output from the XML forming unit 106 of the structured document storage device 1 according to the embodiment of this invention. This search result shows the search result shown in FIG. 17 obtained based on the search query 15 shown in FIG.

  That is, the first to fourth lines in FIG. 18 use the node name “book” indicated by the node ID “3” in the search result of FIG. 17 in the XML document example 7 shown in FIG. Indicates a text node whose node value whose end tag is “/ book” matches “Author 1”. Also, the 5th to 8th lines in FIG. 18 use the node name “book” indicated by the node ID “15” in the search result of FIG. 17 in the XML document example 7 shown in FIG. Indicates a text node whose node value whose end tag is “/ book” matches “Author 1”.

  As described above, according to the structured document storage device 1 of the present embodiment, the sequence number indicating the appearance order of the target node of the structured document and the descendant node of the target node among the nodes appearing after the target node are displayed. The next number indicating the sequence number of the node that appears first among the excluded nodes is stored in the node table 2, and the search unit 105 refers to the node table 2 to determine the sequence number of the comparison source node of the structured document. Since the comparison source node is compared with the order number and next number of the comparison destination node to determine whether or not the comparison source node is a descendant node of the comparison destination node, the number of search procedures is the same as before, but compared with the conventional case Thus, the procedure for searching for the ancestor-descendant relationship between the nodes of the structured document can be performed in a short time, and as a result, the retrieval of the structured document can be performed at a higher speed than in the past.

  The present invention is not limited to the above embodiment. For example, in the structured document storage device 1 of the present embodiment, an XML document is input and stored as text data. The method is not limited to text data. An XML document input method may be based on DOM data or other methods. In the structured document storage device 1 of the present embodiment, the search result is output as text data. However, the output method of the search result is not limited to text data. The search result output method may be based on DOM data or other methods.

  The structured document storage apparatus 1 according to the present embodiment stores an XML document including an attribute whose appearance order is not defined, but stores a structured document in which the appearance order of all nodes is defined. It is good. In this case, the storage unit 102 does not need the attribute data insertion unit 102d and the attribute storage unit 102e.

  The order number stored in the order column 204 in the node table 2 may be a value assigned with an integer starting from “0” and increasing by “1” in the order of appearance of the nodes. If the node does not have an attribute, steps S506 to S510 in the node storing process in FIG. 5 are not necessary. Also, in step S1202 of the descendant node search process of FIG. 12, the process of adding “1” to the value of the order column 204 is not necessary.

  Furthermore, the present invention also includes a structured document storage method for realizing the operation of the structured document storage device 1 described above, and a structured document storage program for causing the computer to execute the operation of the structured document storage device 1. For example, the structured document storage program may be read from a recording medium in which the structured document storage program is stored, loaded into a computer, and executed to configure the structured document storage device. The structured document storage device may be configured by being distributed via a communication network, installed in a computer, and executed.

DESCRIPTION OF SYMBOLS 1 Structured document storage apparatus 2 Node table 3 Node type list 7 XML document example 8 Node table example 9 Search procedure list 101 XML analysis part 102 Storage part 102a Node ID determination means 102b Sequence number determination means 102c Node data insertion means 102d Attribute data Insertion unit 102e Attribute storage unit 102f Child node storage unit 103 Relational database 104 XQuery analysis unit 105 Search unit 105a Search control unit 105b Node type search unit 105c Node name search unit 105d Node value search unit 105e Child node search unit 105f Parent node search unit 105g Descendant node search means 105h Ancestor node search means 105i Subtree data extraction means 106 XML forming unit

Claims (6)

A structured document analyzing means for analyzing the structured document and outputting information about nodes included in the structured document;
Storage means having a node table in which information about nodes of the structured document is stored;
Determining a sequence number representing the order of appearance of the target node in the information related to the node of the structured document output from the structured document analyzing means, and the target among the nodes appearing in the information after the target node The order number of the node that appears first except for the descendant nodes of the node is determined as a next number to be assigned to the target node, and the order number and the next number are stored in the node table together with information on the target node. Storage means;
Search means for referring to the stored information in the node table according to the supplied search procedure list and outputting a search result, wherein the search means is configured such that the sequence number of the comparison source node in the storage information is a comparison destination node. A descendant node that determines that the comparison source node is a descendant node of the comparison destination node when the order number of the comparison source node is less than the next number of the comparison destination node. A structured document storage device comprising a search means. The structured document includes a first node whose appearance order is defined and a second node whose appearance order is not defined,
The storage means assigns an integer of 0 or more that increases by 2 in the order of appearance of the first node to the order number of the first node, and assigns the second number to the order number of the second node. Assign a value obtained by adding 1 to the order number of the parent node that is one node higher than the node,
The descendant node search means is configured such that the order number of the comparison source node is larger than a value obtained by adding 1 to the order number of the comparison destination node, and the order number of the comparison source node is the value of the comparison destination node. 2. The structured document storage device according to claim 1, wherein when the number is smaller than a next number, the comparison source node is determined to be a descendant node of the comparison destination node. A structured document analysis step of analyzing the structured document and outputting information about nodes included in the structured document;
Determining a sequence number representing the order of appearance of the target node in the information related to the node of the structured document obtained by analysis in the structured document analyzing step, and determining a node number that appears in the information after the target node; The order number of the node that appears first except for the descendant nodes of the target node is determined as the next number to be assigned to the target node, and the order number and the next number together with information on the target node are determined as the structure A storage step of storing in a node table in which information on nodes of the document is stored;
A search step of outputting a search result by referring to the stored information in the node table according to the supplied search procedure list, wherein the search step includes the sequence number of the comparison source node in the storage information as a comparison destination node A descendant node that determines that the comparison source node is a descendant node of the comparison destination node when the order number of the comparison source node is less than the next number of the comparison destination node. A structured document storage method comprising a retrieval step. The structured document includes a first node whose appearance order is defined and a second node whose appearance order is not defined,
The storing step assigns an integer of 0 or more that increases by 2 in the order of appearance of the first node to the order number of the first node, and assigns the second number to the order number of the second node. Assign a value obtained by adding 1 to the order number of the parent node that is one node higher than the node,
In the descendant node search step, the order number of the comparison source node is larger than a value obtained by adding 1 to the order number of the comparison destination node, and the order number of the comparison source node is the value of the comparison destination node. 4. The structured document storage method according to claim 3, wherein when the number is smaller than a next number, it is determined that the comparison source node is a descendant node of the comparison destination node. On the computer,
A structured document analysis step of analyzing the structured document and outputting information about nodes included in the structured document;
Determining a sequence number representing the order of appearance of the target node in the information related to the node of the structured document obtained by analysis in the structured document analyzing step, and determining a node number that appears in the information after the target node; The order number of the node that appears first except for the descendant nodes of the target node is determined as the next number to be assigned to the target node, and the order number and the next number together with information on the target node are determined as the structure A storage step for storing in a node table in which information on nodes of the document is stored;
A search step of referring to the stored information of the node table and outputting a search result according to the supplied search procedure list, wherein the search step includes a comparison destination node in which the order number of the comparison source node is compared A descendant that determines that the comparison source node is a descendant node of the comparison destination node when the order number of the comparison source node is greater than the sequence number of the node and the comparison source node is less than the next number of the comparison destination node A structured document storage program comprising a node search step. The structured document includes a first node whose appearance order is defined and a second node whose appearance order is not defined,
The storing step assigns an integer of 0 or more that increases by 2 in the order of appearance of the first node to the order number of the first node, and assigns the second number to the order number of the second node. Assign a value obtained by adding 1 to the order number of the parent node that is one node higher than the node,
In the descendant node search step, the order number of the comparison source node is larger than a value obtained by adding 1 to the order number of the comparison destination node, and the order number of the comparison source node is the value of the comparison destination node. 6. The structured document storage program according to claim 5, wherein when it is smaller than a next number, it is determined that the comparison source node is a descendant node of the comparison destination node.

Images