FIXED: [V4RB] Two V4RB 1.11 crashing problems -- db attached

[Ruslan Zasukhin]

On 12/1/04 10:02 PM, "Keith DeLong" <delong at REDCORT.COM> wrote:

FIXED BUG.

    in case sub-query produce NULL value then IN () operator crashes.


> Hi Ruslan,
> I'm encouraged by the continuing victories being reported in the Ukraine!
> :-)
> 
> Attached is a db that demonstrates the crashes I reported.
> Here's some example selects that crash ViSQL:
> 
> This works
> 
> SELECT * FROM Employees WHERE Salary IN
> (select max(Salary) from Employees WHERE RecID > 40)
> 
> This crashes:
> 
> SELECT * FROM Employees WHERE Salary IN
> (select max(Salary) from Employees WHERE RecID > 999)
> 
> 
> This works:
> 
> Select * FROM timecards
> WHERE insecs IN (select max(insecs) from timecards,tasks WHERE Insecs <=
> 3190000000 AND EmpPtr = 5 AND Manual = false)
> AND EMPPTR = 5
> 
> Change Manual to False (no records in subselect) and it crashes:
> 
> Select * FROM timecards
> WHERE insecs IN (select max(insecs) from timecards,tasks WHERE Insecs <=
> 3190000000 AND EmpPtr = 5 AND Manual = true)
> AND EMPPTR = 5
> 
> 
> It appears any sub-select that results in no records selected crashes.
> 
> Also, in REALbasic, this crashes if there is no recid = 9999:
> 
> VCursor(InDB,"Select * from Employees WHERE RecID = 9999",kV_Server,
> kV_NoLocks kV_Random)
> 
> 
> Thanks Ruslan,
> 
> Keith
> 
> 
>> On 11/29/04 2:45 AM, "Keith DeLong" <delong at redcort.com> wrote:
>> 
>>> Hi Ruslan,
>>> 
>>> Problem #1
>>> V4RB crashes if an invalid RecID is passed. For example:
>>> 
>>> VCursor(InDB,"Select * from Names WHERE RecID = 9999",kV_Server, kV_NoLocks,
>>> kV_Random)
>>> 
>>> This crashes the application if 9999 is an invalid recID
>>> 
>>> 
>>> Problem #2
>>> V4RB crashes if a sub select results in no records selected. For example:
>>> 
>>> SELECT * FROM Employees WHERE Salary IN
>>> (select max(Salary) from Employees WHERE Active = false)
>>> 
>>> This crashes the application if all employees are active.
>>> 
>>> I tested and verified both of these bugs using ViSQL.
>>> 
>>> I know you are focused on 2.0 but these are show stopping, deal breaking
>>> critical problems. They crash my application without warning or error.
>>> 
>>> How quickly can we get a fix for these?
>> 
>> Hi Keith,
>> 
>> I need to check.
>> 
>> Well, it looks I have 4-5 reports on 1.11 for check.
>> I will try to check this in 1-2 days.
>> 
>> If you have db or else to send to me please send/.
>> 
> 
> 
> 
> Scanned by evaluation version of Dr.Web antivirus Daemon
> http://drweb.ru/unix/
> 

-- 
Best regards,
Ruslan Zasukhin      [ I feel the need...the need for speed ]
-------------------------------------------------------------
e-mail: ruslan at paradigmasoft.com
web: http://www.paradigmasoft.com

To subscribe to the Valentina mail list go to:
http://lists.macserve.net/mailman/listinfo/valentina
-------------------------------------------------------------