Difference between revisions of "AK-index segfault"

From ScarletDME
Jump to navigation Jump to search
(Steps to reproduce)
 
 
(3 intermediate revisions by 2 users not shown)
Line 1: Line 1:
 +
==Scope of problem==
 +
I've tried to reproduce this bug, and it appears to be fixed in the 2.6-6 release of the GPL OpenQM GPL.  Hence it should also be fixed in the CVS, but nobody has checked yet.
 +
 +
''' As of 22Jun12, this bug has been verified as fixed in the current source tree (Revision 98). '''
 +
 
==Steps to reproduce==
 
==Steps to reproduce==
  
Line 10: Line 15:
  
 
Now run this program to get some data in:
 
Now run this program to get some data in:
 +
<nowiki>
 
  open "TESTFILE" to file else stop
 
  open "TESTFILE" to file else stop
 
   
 
   
Line 23: Line 29:
 
   
 
   
 
  end
 
  end
 +
</nowiki>
  
 
Now
 
Now

Latest revision as of 14:06, 22 June 2012

Scope of problem

I've tried to reproduce this bug, and it appears to be fixed in the 2.6-6 release of the GPL OpenQM GPL. Hence it should also be fixed in the CVS, but nobody has checked yet.

As of 22Jun12, this bug has been verified as fixed in the current source tree (Revision 98).

Steps to reproduce

Create a file called TESTFILE with a dictionary entry F1 of:

1: D
2: 1
3: 
4: F1
5: 5R
6: S

Now run this program to get some data in:

 open "TESTFILE" to file else stop
 
 for i = 1 to 2000
   data = i
   write data to file, "rec":i
 next i
 
 data = ''
 for i = 2001 to 2100
   write data to file, "rec":i
 next i
 
 end
 

Now

SELECT TESTFILE WITH F1 > 1500

works fine;

CREATE-INDEX TESTFILE F1
BUILD-INDEX TESTFILE F1
SELECT TESTFILE WITH F1 > 1500

Segfault.