Test generator for database management systems providing tight joins

• US 7,007,007 B2
• Filed: 11/03/2004
• Issued: 02/28/2006
• Est. Priority Date: 05/14/1998
• Status: Expired due to Term

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Slutz, Donald R.
• Primary Examiner(s): Mofiz, Apu M.
• Application Number:
  US10/980,734
• Publication Number:
  US 20050097118A1
• Time in Patent Office: 482 Days
• Field of Search: 707/2, 707/7, 707/200, 707/100, 707/101, 707/4, 707/102, 707/3, 707/10, 707/104.1, 707/5, 707/103.R, 712/200, 717/169, 714/20
• US Class Current: 707/809
• CPC Class Codes: G06F 11/3672 Test management
• CPC Class Codes: G06F 16/21 Design, administration or maintenance of databases
• CPC Class Codes: G06F 16/242 Query formulation
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

A test generator produces a set of database query-language statements comprised of randomly chosen elements for testing one or more database management systems on arbitrary databases. The statements are syntactically correct according to the query language, and are semantically correct according to the query language and according to the schema of the target database. A configuration file further specifies parameters of the test statements, in terms of maximum elements, weights of different elements, etc. The generated statements include predicates in which tables in a from clause are tightly joined. In addition, a dictionary of words randomly selected from text columns in a test database is maintained and used to create predicates having words that actually appear in the row data.

Claims

1. A computer-implemented system for creating a SQL statement, the system comprising:

means for generating a SQL statement having a from clause, the from clause include a plurality of table names;

means for creating a plurality of table sets from the plurality of table names;

means for iterating of the plurality of table sets until one table set remains and for each iteration performing the tasks of;

means for selecting a first table set and a second table set from the plurality of table sets;

means for selecting a first table from the first table set and selecting a second table from the second table set;

means for selecting a first colunm in the first table and selecting a second colunm from the second table;

means for inserting into a where clause for the SQL statement a predicate including the first column and the second column; and

means for merging the first table set and the second table set.

2. The system of claim 1, further comprising means for deleting the first table set and the second table set after merging.

3. The system of claim 1, wherein the first table set and the second table set are uniformly selected.

4. The system of claim 1, wherein the first table and the second table are uniformly selected.

5. The system of claim 1, wherein the first column and the second column are uniformly selected.

6. The system of claim 1, wherein the first column and the second column are selected based on a compatibility of a data type of the first column and a data type of the second column.

7. The system of claim 1, wherein the first column and the second column are selected based on a substantial similarity between a column name of the first column and a column name of the second column.

8. The system of claim 7, wherein the substantial similarity is between a prefix of the column name of the first column and a prefix of the column name of the second column.

9. The methed system of claim 7, wherein the substantial similarity is between a suffix of the column name of the first column and a suffix of the column name of the second column.

10. A system for performing a method for creating a SQL statement, comprising:

a plurality of table names;

a SQL statement having a from clause, the from clause including the plurality of table names;

a plurality of table sets created from the plurality of table names, wherein the plurality of table sets are iterated until one table set remains and for each iteration;

a first table set and a second table set selected from the plurality of table sets;

a first table selected from the first table set;

a second table selected from the second table set;

a first column selected from the first table; and

a second column selected from the second table, wherein a predicate is inserted into a where clause for the SQL statement, and wherein the predicate includes the first column and the second column, and wherein the first table set and the second table set are merged.

11. The system of claim 10, wherein the first table set and the second table set are deleted after merging.

12. The system of claim 10, wherein the first table set and the second table set are uniformly selected.

13. The system of claim 10, wherein the first table and the second table are uniformly selected.

14. The system of claim 10, wherein the first column and the second column are uniformly selected.

15. The system of claim 10, wherein the first column and the second column are selected based on a compatibility of a data type of the first column and a data type of the second column.

16. The system of claim 10, wherein the first column and the second column are selected based on a substantial similarity between a column name of the first column and a column name of the second column.

17. The system of claim 10, wherein the substantial similarity is between a prefix of the column name of the first column and a prefix of the column name of the second column.

18. The system of claim 10, wherein the substantial similarity is between a suffix of the column name of the first column and a suffix of the column name of the second column.