Database access using partitioned data areas

A database system including a database in which a structured document having a hierarchical structure is registered, the database system comprising:

• a data management unit which generates and manages a value partition in which a unit subtree including target data to be partitioned in the structured document is separated and registered according to a value of the target data, and a base partition which includes a subtree other than the unit subtree and an index for uniquely identifying the unit subtree;
• an SQL processing unit which processes a search request for searching the structured document registered in the database, the search request designating the hierarchical structure of the structured document;
• and
• a search result generating unit which generates a search result with respect to the search request by using a result of individual search performed on the value partition by the SQL processing unit;
• the data management unit including:
• a partition generating unit which generates the partitions;
• a parser unit which identifies positions of a start tag and an end tag defining an attribute in the structured document to generate a cutting position list corresponding to a hierarchical structure of an attribute value and register the cutting position list in a storage area;
• a cutting unit which separates the unit subtree and the index for referring to the unit subtree from the structured document by referring to the cutting position list to identify the start tag position and the end tag position of the unit subtree to be cut, the unit subtree including the target data;
• and
• a registration unit which registers the unit subtree and the index for identifying the unit subtree, separated by the cutting unit, in the separate partitions.

An method having a database, the database including a structured document, the method comprising:

• separating target data into partitioned data based on a value associated with the target data;
• generating a value partition having a unit subtree, the unit subtree having the partitioned data therein;
• generating a base partition, the base partition having a first subtree distinct from the unit subtree and an index associated with the unit subtree;
• identifying positions of a start tag and an end tag associated with an attribute in the structured document, the attribute having a hierarchical structure;
• generating a cutting position list corresponding to the hierarchical structure based on the start tag and the end tag;
• storing the cutting position list in a storage area;
• separating the unit subtree and the index from the structured document based on the cutting position list to identify the start tag position and the end tag position of the unit subtree to be cut;
• and
• registering the unit subtree and the index to identify the unit subtree, separated in the separating step, in the base and value partitions;
• searching the database based on at least one of the value partition and the base partition;
• and
• displaying the results of the search.

A computer program product, the computer program product comprising:

• a computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising:
• computer readable program code configured to:
• generate a value partition in which a unit subtree including target data to be partitioned in the structured document is separated and registered according to a value of the target data, and a base partition which includes a subtree other than the unit subtree and an index for uniquely identifying the unit subtree;
• identify positions of a start tag and an end tag defining an attribute in the structured document to generate a cutting position list corresponding to a hierarchical structure of an attribute value and register the cutting position list in a storage area;
• separate the unit subtree and the index for referring to the unit subtree from the structured document by referring to the cutting position list to identify the start tag position and the end tag position of the unit subtree to be cut, the unit subtree including the target data;
• and
• register the unit subtree and the index for identifying the unit subtree, separated in the separating step, in the separate partitions.

An information processing apparatus including a database in which a structured document having a hierarchical structure is managed, the information processing apparatus comprising:

• a partition generating unit which generates a value partition in which a unit subtree including target data to be partitioned in the structured document is separated and registered according to a value of the target data, and a base partition which includes a subtree other than the unit subtree and an index for uniquely identifying the unit subtree;
• a parser unit which identifies positions of a start tag and an end tag defining an attribute in the structured document to generate a cutting position list corresponding to a hierarchical structure of an attribute value and register the cutting position list in a storage area;
• a cutting unit which separates the unit subtree and the index for referring to the unit subtree from the structured document by referring to the cutting position list to identify the start tag position and the end tag position of the unit subtree to be cut, the unit subtree including the target data;
• and
• a registration unit which registers the unit subtree and the index for identifying the unit subtree, separated by the cutting unit, in the separate partitions;
• wherein the unit subtree includes a plurality of start tags, and the index is generated from a first one of the start tags in the unit subtree,
• wherein the registration unit adds pointers to the unit subtree and the index for uniquely identifying the unit subtree to allow cross-reference between the unit subtree and the index, before registering the unit subtree and the index in the separate partitions,
• wherein in order to separate the unit subtree for registration in the separate partition by referring to the cutting position list, the cutting unit generates a subtree data list and registers the generated list in a storage area, the subtree data list including a range-defining pair in which a start position and an end position of the unit subtree included in the structured document are associated with each other and an occurrence counter in which the order of description of the unit subtree in the structured document is registered, and
• wherein the structured document is described in XML, and the database is an XMLDB which enables storage and search of XML documents.