Bug #50231 Select shows different outputs for 'bigint' column containing 'null' values
Submitted: 11 Jan 2010 13:10 Modified: 15 Jan 2010 7:51
Reporter: Nirbhay Choubey Email Updates:
Status: Can't repeat Impact on me:
None 
Category:MySQL Server: Data Types Severity:S3 (Non-critical)
Version:5.1-bugteam OS:Any
Assigned to: Ramil Kalimullin CPU Architecture:Any
Tags: regression

[11 Jan 2010 13:10] Nirbhay Choubey 
Description:
Following set of statements shows invalid output when the 'BIGINT' column contains 'NULL' :

DROP TABLE IF EXISTS t1;
CREATE TABLE t1 (c1 BIGINT);
INSERT INTO t1 VALUES (1),(null),(2);

5.1-bugteam build produces :

SELECT * FROM t1 WHERE c1 <> 2;
+------+
| c1   |
+------+
|    1 |
| NULL |
+------+

While 5.1-main tree build gives :

+------+
| c1   |
+------+
|    1 |
+------+

How to repeat:
See the description above.
[11 Jan 2010 15:02] Nirbhay Choubey 
Changed the synopsis from 'invalid' to 'different outputs' as its unclear which of the 2 outputs should be considered as correct/expected output.

TIME, YEAR data types show the same difference,
while <, <= operators also behave similar to <>, as described in the bug.
[11 Jan 2010 23:45] Sveta Smirnova 
Thank you for the report.

Verified as described.
[15 Jan 2010 5:21] Ramil Kalimullin 
couldn't repeat using the latest mysql-5.1-bugteam.