Apparatus, program product and method for structured document management

1. A structured document management apparatus comprising:

a document data accepting unit that accepts input of structured document data having a hierarchical logic structure;

a structure guide data storage unit that stores structure guide data which is a summary of hierarchical structure information of the structured document, wherein the structure guide data contains a guide node where unique identification information is allocated;

a structure stream converting unit that executes a syntax analysis of the structured document data, and converts the structured document data into structure stream data as one-dimensional sequence data by arranging the unique identification information of guide nodes corresponding to nodes that pass when depth-prioritized tracing is performed from a route node of the structured document data using the structured guide data;

a structure stream data storage unit that stores the structure stream data;

a query data accepting unit that accepts input of query data;

a path pattern compile unit that creates a path pattern processing table which specifies a processing procedure specialized for the query data by

executing a syntax analysis of the accepted query data and creating a primary structure graph in which a relationship between tags is expressed by a tree format, and

collating the primary structure graph and the structured guide data, converting the primary structure graph and the structured guide data into the unique identification information, and creating a secondary structure graph where unnecessary nodes are removed, the path pattern processing table including:

an Entry Table which has a sequence element corresponding to the unique identification information and executes a process corresponding to the unique identification information;

a Place which is a storage area for holding a queue of Tokens as intermediate data, the Place representing a position of the unique identification information; and

a Trans which connects one Place and another Place and flows a Token held at a high-order Place to a low-order Place,

the path pattern processing table being created by

pasting the process corresponding to the unique identification information of the Entry Table and linking the Place and the Trans recursively from the route node to a terminal node of the secondary structure graph; and

a structure stream scanning unit that acquires the structure stream data from the structure stream data storage unit, and gives the structure stream to the path pattern processing table so as to execute the processing procedure.

2. The apparatus according to claim 1 , wherein the path pattern compile unit synthesizes the path pattern processing tables relating to the respective query data so as to create a path pattern processing table of the plural query data when the plural query data are processed.

3. The apparatus according to claim 1 , wherein the path pattern compile unit incorporates a procedure for skipping a portion of the structure stream data in the path pattern processing table when structure information of the structured document data is defined.

4. The apparatus according to claim 1 , wherein the path pattern compile unit incorporates a procedure for skipping a portion of the structure stream data in the path pattern processing table when the structure information appears due to statistics information of the structured document data.

5. The apparatus according to claim 1 , wherein the structure guide data hold the following conditions (1) to (3):

(1) all paths that appear in the structured document data appear in the structure guide data;

(2) all paths that appears in the structure guide data appear in the structured document data; and

(3) all paths which appear in the structure guide data are unique.

6. A computer program product having a computer readable storage medium including programmed instructions for managing a structured document, wherein the instructions, when executed by a computer, cause the computer to perform:

accepting input of structured document data having a hierarchical logic structure;

executing a syntax analysis of the structured document data, and converting the structured document data into structure stream data as one-dimensional sequence data by arranging unique identification information which is a summary of hierarchical structure information of the structured document data,

wherein the unique identification information of the guide nodes corresponds to nodes that pass when depth-prioritized tracing is performed from a route node of the structured document data using the structured guide data, and

the structured guide data contains a guide node where the unique identification information is allocated;

storing the structure stream data in a structure stream data storage unit;

accepting input of query data;

creating a path pattern processing table which specifies a processing procedure specialized for the query data by

executing a syntax analysis of the accepted query data and creating a primary structure graph in which a relationship between tags is expressed by a tree format, and

collating the primary structure graph and the structured guide data, converting the primary structure graph and the structured guide data into the unique identification information, and creating a secondary structure graph where unnecessary nodes are removed, the path pattern processing table including:

an Entry Table which has a sequence element corresponding to the unique identification information and executes a process corresponding to the unique identification information;

a Place which is a storage area for holding a queue of Tokens as intermediate data, the Place representing a position of the unique identification information; and

a Trans which connects one Place and another Place and flows a Token held at a high-order Place to a low-order Place,

the path pattern processing table being created by

pasting the process corresponding to the unique identification information of the Entry Table and linking the Place and the Trans recursively from the route node to a terminal node of the secondary structure graph; and

acquiring the structure stream data from the structure stream data storage unit and giving the structure stream to the path pattern processing table so as to execute the processing procedure.

7. The computer program product according to claim 6 , wherein the path pattern processing tables relating to the respective query data are synthesized so as to create a path pattern processing table of the plural query data when the plural query data are processed.

8. The computer program product according to claim 6 , wherein a procedure for skipping a portion of the structure stream data is incorporated in the path pattern processing table when structure information of the structured document data is defined.

9. The computer program product according to claim 6 , wherein a procedure for skipping a portion of the structure stream data is incorporated in the path pattern processing table when the structure information appears due to statistics information of the structured document data.

10. A method of managing structured document comprising:

accepting input of structured document data having a hierarchical logic structure;

executing a syntax analysis of the structured document data, and converting structure information in the structured document data into structure stream data as one-dimensional sequence data by arranging unique identification information which is a summary of hierarchical structure information of the structured document data, wherein the unique identification information of guide nodes corresponds to nodes that pass when depth-prioritized tracing is performed from a route node of the structured document data using the structured guide data, the structured guide data contains a guide node where the unique identification information is allocated;

storing the structure stream data in a structure stream data storage unit;

accepting input of query data;

creating a path pattern processing table which specifies a processing procedure specialized for the query data by

executing a syntax analysis of the accepted query data and creating a primary structure graph in which a relationship between tags is expressed by a tree format, and

collating the primary structure graph and the structured guide data, converting the primary structure graph and the structured guide data into the unique identification information, and creating a secondary structure graph where unnecessary nodes are removed, the path pattern processing table including:

an Entry Table which has a sequence element corresponding to the unique identification information and executes a process corresponding to the unique identification information;

a Place which is a storage area for holding a queue of Tokens as intermediate data, the Place representing a position of the unique identification information; and

a Trans which connects one Place and another Place and flows a Token held at a high-order Place to a low-order Place,

the path pattern processing table being created by

pasting the process corresponding to the unique identification information of the Entry Table and linking the Place and the Trans recursively from the route node to a terminal node of the secondary structure graph; and

acquiring the structure stream data from the structure stream data storage unit and giving the structure stream to the path pattern processing table so as to execute the processing procedure.

11. The method according to claim 10 , wherein the path pattern processing tables relating to the respective query data are synthesized so as to create a path pattern processing table of the plural query data when the plural query data are processed.

12. The method according to claim 10 , wherein a procedure for skipping a portion of the structure stream data is incorporated in the path pattern processing table when structure information of the structured document data is defined.

13. The method according to claim 10 , wherein a procedure for skipping a portion of the structure stream data is incorporated in the path pattern processing table when the structure information appears due to statistics information of the structured document data.