Virtual table index mechanism and method for multi-attribute compound condition query

• US 20150095342A1
• Filed: 12/27/2012
• Published: 04/02/2015
• Est. Priority Date: 12/19/2012
• Status: Abandoned Application

• Current Assignee: Guangdong Electronics Industry Institute Company Limited
• Original Assignee: Guangdong Electronics Industry Institute Company Limited
• Inventors: Li, Xiaolin, Xie, Yi, Xu, Zhiwei, Yue, Qiang
• Application Number:
  US14/381,214
• Publication Number:
  US 20150095342A1
• Time in Patent Office: Days
• US Class Current: 707/741
• CPC Class Codes: G06F 16/2264 Multidimensional index structures
• CPC Class Codes: G06F 16/31 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06F 16/3341 using boolean model

Abstract

A virtual table index mechanism and a method for a multi-attribute compound condition query relate to a field of computer application technology. The virtual table index mechanism for a multi-attribute compound condition query includes: an index manager; a condition analyzer; and a pre-execution engine. The present invention forms index key values corresponding to a plurality of attributes in a virtual table. When querying, a query condition for the virtual table is calculated. Referring to a key value index, whether a mapping physical table of the virtual table should be executed is independently pre judged according to the attributes, so as to ensure that query is applied only to a virtual sub-table satisfying a corresponding key value condition. The present invention is effective for multi-attribute compound condition query, and is suitable for indexing virtual tables.

Claims

1-8. -8. (canceled)

9. A virtual table index mechanism for a multi-attribute compound condition query, comprising:

an index manager;

a condition analyzer; and

a pre-execution engine;

wherein the index manager manages index key values of attributes of a virtual table, and supports a management of a single key value and an interval key value of the multiple attributes of the virtual table;

the condition analyzer disassembles a query condition applied to the virtual table and analyzes a predicate;

the condition analyzer sequentially analyzes the whole query condition for every indexed attribute according to an SQL (Structure Query Language) grammar, and, after masking a predicate expression of other attributes with a true value, only maintains the predicate condition of the indexed attribute for judging whether an index of the indexed attribute satisfies the query condition;

meanwhile, if the index key value of the attribute is an interval value, the analyzer further calculates whether the predicate of the attribute in the query condition masked with the true value is true or false according to the interval value, and replaces the predicate with a Boolean result (true/false);

if the predicate is not able to be calculated, an inferred result needing querying the virtual table is directly returned and displayed;

the pre-execution engine judges whether the query condition partially replaced by a true value expression and the Boolean result is true or false according to the index key value of the corresponding attribute, and determines whether the index of the corresponding attribute satisfies the query condition;

if the query condition is not satisfied, an inferred result not querying the virtual table is directly returned;

if the query condition is satisfied, an index of other the attributes is judged;

in some cases, an execution condition is satisfied by default;

if the index key value is the interval value, only a beginning value and an ending value are utilized.

10. A virtual table index method for a multi-attribute compound condition query, comprising steps of:

forming index key values corresponding to a plurality of attributes in a virtual table;

when querying, calculating a query condition applied to the virtual table;

referring to a key value index, independently pre judging whether a mapping physical table of the virtual table should be executed for each of the attributes, so as to ensure that query is applied only to a virtual sub-table satisfying a corresponding key value condition.

11. The virtual table index method, as recited in claim 10, wherein for pre judging which sub-table comprises data records satisfying the query condition, the virtual table index method further comprises:

in a virtual layer, according to an index based on the sub-table on each sub-table, pre judging whether a data collection of the sub-table satisfies the query condition, only positioning the sub-tables satisfying the query condition for querying;

for exceptions of no query filter condition, no index or difficulty in pre judging based on the sub-table, directly querying the sub-table.

12. The virtual table index method, as recited in claim 10, further comprising:

forming a sub-table index in a B+ tree structure according to a certain attribute key value, wherein each leaf node of a B+ tree is one of the sub-tables;

forming an index for one or more of the (other) attributes at each leaf node, wherein an index value is a single value or an interval value.

13. The virtual table index method, as recited in claim 11, further comprising:

forming a sub-table index in a B+ tree structure according to a certain attribute key value, wherein each leaf node of a B+ tree is one of the sub-tables;

forming an index for one or more of the (other) attributes at each leaf node, wherein an index value is a single value or an interval value.

14. The virtual table index method, as recited in claim 10, wherein two replacements of the query condition is independently analyzing and judging each index attribute according to a language tree formed by analyzing an SQL query condition.

15. The virtual table index method, as recited in claim 11, wherein two replacements of the query condition is independently analyzing and judging each index attribute according to a language tree formed by analyzing an SQL query condition.

16. The virtual table index method, as recited in claim 12, wherein two replacements of the query condition is independently analyzing and judging each index attribute according to a language tree formed by analyzing an SQL query condition.

17. The virtual table index method, as recited in claim 13, wherein two replacements of the query condition is independently analyzing and judging each index attribute according to a language tree formed by analyzing an SQL query condition.

18. The virtual table index method, as recited in claim 10, wherein for a value in an interval, a predicate expression is transformed to comparison logic for calculating Boolean results by substituting with the interval value.

19. The virtual table index method, as recited in claim 11, wherein for a value in an interval, a predicate expression is transformed to comparison logic for calculating Boolean results by substituting with the interval value.

20. The virtual table index method, as recited in claim 12, wherein for a value in an interval, a predicate expression is transformed to comparison logic for calculating Boolean results by substituting with the interval value.

21. The virtual table index method, as recited in claim 13, wherein for a value in an interval, a predicate expression is transformed to comparison logic for calculating Boolean results by substituting with the interval value.

22. The virtual table index method, as recited in claim 14, wherein for a value in an interval, a predicate expression is transformed to comparison logic for calculating Boolean results by substituting with the interval value.

23. The virtual table index method, as recited in claim 15, wherein for a value in an interval, a predicate expression is transformed to comparison logic for calculating Boolean results by substituting with the interval value.

24. The virtual table index method, as recited in claim 16, wherein for a value in an interval, a predicate expression is transformed to comparison logic for calculating Boolean results by substituting with the interval value.

25. The virtual table index method, as recited in claim 17, wherein for a value in an interval, a predicate expression is transformed to comparison logic for calculating Boolean results by substituting with the interval value.