Indexing Strategy With Improved DML Performance and Space Usage for Node-Aware Full-Text Search Over XML

• US 20100185683A1
• Filed: 12/30/2008
• Published: 07/22/2010
• Est. Priority Date: 12/30/2008
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Baby, Thomas, Liu, Zhen Hua, Lin, Wesley
• Granted Patent: US 8,126,932 B2
• Time in Patent Office: Days
• US Class Current: 707/803
• CPC Class Codes: G06F 16/81 Indexing, e.g. XML tags; Data structures therefor; Storage structures

Abstract

Techniques are provided for searching within a collection of XML documents. A relational table stores an entry for each node of a set of nodes in a collection of XML documents. Each entry of the relational table stores an order key and a path identifier along with the atomized value of the node. Instead of storing the atomized value in a full-text index, a virtual column can be created to represent, for each node, the atomized value of the node. Alternately, each entry of the relational table stores an order key and a path identifier along with, for simple nodes, the atomized value, and for complex nodes, a null value. For a complex node with a descendant text node, a separate entry is stored for the descendant text node in the relational table.

Claims

1. A computer-implemented method comprising:

storing a table that stores data for a plurality of nodes in a collection of XML documents, the table comprising an entry for each node of the plurality of nodes, the entry of each node comprising;

an order key that specifies a hierarchical position of the node within the XML document;

path data that specifies a path, through the structure of the XML document, to the node;

for a simple node, an atomized value of the node;

for a complex node, a null value;

for a mixed content node, storing a separate entry for a descendant text node of the mixed content node, the entry comprising the text value of the descendant text node; and

storing a full-text index of the values stored in the entries of the table.

2. The computer-implemented method of claim 1, wherein said atomized value is formed by concatenating the text values of said node.

3. The computer-implemented method of claim 1, wherein said atomized values are stored in a column of said table, said column being of a character large object type.

4. The computer-implemented method of claim 1, further comprising:

receiving a query that requires a text search of the collection of XML documents; and

rewriting said query to produce a rewritten query, wherein said rewritten query causes using said full-text index to compute said rewritten query.

5. The computer-implemented method of claim 1, further comprising:

receiving a query that specifies certain criteria, said criteria comprising;

one or more criteria for text contained in a node; and

one or more criteria for a path of a node;

rewriting said query to produce a rewritten query, wherein said rewritten query causes using said full-text index and said table to compute which one or more nodes of said plurality of nodes satisfy said criteria.

6. The computer-implemented method of claim 5, further comprising computing said rewritten query, wherein computing said rewritten query comprises computing whether the one or more criteria for text exist in the atomized values of adjacent nodes.

7. The computer-implemented method of claim 6, wherein said computing said rewritten query is based on the order key and node values.

8. The computer-implemented method of claim 6, further comprising using a first text value for a first adjacent node with a second text value for a second adjacent node to compute whether the one or more criteria for text exist in the atomized values of adjacent text in the adjacent nodes.

17. A volatile or non-volatile computer-readable storage medium storing one or more sequences of instruction, wherein execution of the one or more sequences of instruction by one or more processors causes the one or more processors to perform the steps of claim 1.

9. A computer-implemented method comprising:

receiving a query that requests data from a collection of XML documents;

wherein said collection of XML documents comprises a plurality of nodes;

wherein a certain index comprises a table that comprises a plurality of rows, each row of said plurality of rows corresponding to a node of said plurality of nodes;

wherein each row of said plurality of rows includes;

an order key that indicates a hierarchical position of the node within the XML document, said order key being stored in a order key column of said table; and

path data that specifies a path, through the structure of the XML document, to the node, said path identifier being stored in a path identifier column of said table;

for a simple node, an atomized value of the node, said atomized value being contained in a value column of said table;

for a complex node, a null value, said null value being contained in the value column of said table;

for a mixed content node, storing a separate entry for a descendant text node of the mixed content node, the entry storing the text value of the descendant text node in the value column of said table; and

wherein said certain index of said XML collection comprises a full-text index, said value column being indexed by said full-text index;

rewriting said query to produce a rewritten query, wherein said rewritten query causes using said full-text index to compute said rewritten query.

10. The computer-implemented method of claim 9, further comprising computing whether a part of the requested data exists in the atomized values of adjacent nodes.

11. The computer-implemented method of claim 10, wherein said computing is based on the order key and node values.

12. The computer-implemented method of claim 10, further comprising using a first text value for a first adjacent node with a second text value for a second adjacent node to compute whether the requested data exists in the atomized values of adjacent text in the adjacent nodes.

13. A computer-implemented method comprising:

receiving a query that requests data from a collection of XML documents;

wherein said collection of XML documents comprises a plurality of nodes;

wherein a certain index comprises a table that stores data for a plurality of nodes in a collection of XML documents, the table comprising a value column, the value column comprising;

for a simple node, an atomized value of the node;

for a complex node, a null value;

for a mixed content node, the table storing a separate entry for a descendant text node of the mixed content node, the separate entry storing a text value of the descendant text node in the value column; and

wherein said value column is indexed by a full-text index;

rewriting said query to produce a rewritten query, wherein said rewritten query causes using said full-text index to compute said rewritten query.

14. The computer-implemented method of claim 13, further comprising computing whether a part of the requested data exists in the atomized values of adjacent nodes.

15. The computer-implemented method of claim 14, wherein said computing is based on the order key and node values.

16. The computer-implemented method of claim 14, further comprising using a first text value for a first adjacent node with a second text value for a second adjacent node to compute whether the requested data exists in the atomized values of adjacent text in the adjacent nodes.