From: Michael Widenius
Date: March 19 1999 6:10pm
Subject: bug in CURDATE()?
List-Archive: http://lists.mysql.com/mysql/628
Message-Id: <14066.37533.920717.714251@monty.pp.sci.fi>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

>>>>> "h" == h radners <h.radners@stripped> writes:

>> Description:
h> 	CURDATE() in string-context gives wrong year.
>> How-To-Repeat:
h>         SELECT SYSDATE();     => 1999-03-19 14:37:58   => OK
h> 	SELECT CURDATE()+0;   => 19990319              => OK
h> 	SELECT CURDATE();     => 6703-03-19            => WRONG
>> Originator:	Horst-W. Radners
>> Organization:
h>   IWiS GmbH, Markgrafenstr. 11, D-10969 Berlin
h>   Tel.: +49 30 253769-12, Fax.: -77, e-mail: h.radners@stripped
>> MySQL support: none
>> Synopsis:	CURDATE() in string-context gives wrong year.
>> Severity:	serious
>> Priority:	medium
>> Category:	mysql
>> Class:		sw-bug
>> Release:	mysql-3.22.20a (Source distribution)

>> Environment:

h> System: IRIX  6.2  IP22

<cut>

Hi!

Testing...

mysql> SELECT CURDATE();
+------------+
| CURDATE()  |
+------------+
| 1999-03-19 |
+------------+
1 row in set (0.00 sec)

mysql> SELECT CURDATE()+0;
+-------------+
| CURDATE()+0 |
+-------------+
|    19990319 |
+-------------+

Any change you know gdb good enough to debug this ?
(This may be a gcc bug ...)

If you can put a breakpoint at:

Item_date::str

and single step through this function, you should be able to find out
what's wrong ?

Regards,
Monty