[Bug 1007] nightly testing + valgrind results

Wednesday, 26 May 2010

http://bugs.meego.com/show_bug.cgi?id=1007

--- Comment #11 from yongsheng <yongsheng.zhu(a)intel.com&gt; 2010-05-26 20:14:37 PDT
---
...
 I agree with your root cause analysis, but not with this solution. I
believe it
 leads to accessing the cost value at index #-2, which falls of the lower end of
 the array. I don't think so. for i and j will never equal 0 when calculating
this
expression.
please see lcs.h:145.
145             if (i == 0 || j == 0) {
146                 sub[i][j].choice = NONE;
147                 sub[i][j].length = 0;
148                 sub[i][j].cost = 0;
149             } else if (access.entry_at(a, i - 1) == access.entry_at(b, j -
1)) {

...
 IMHO the cost function was called with correct indices. Basically
both one
 before array and one after the end are valid and should give the same result as
 the indices just inside the array. I don't quite understand what the
'cost' is. but here, based on your
description, the first cost should be equal to the cost of second element
substract to that of the first element.
...
 The implementation only handled index -1, but not index == size(). It
also did
 not work for empty arrays.
 Here's a more complete solution, committed to branch mbc1007_pohly. Yongsheng,
 do you agree? ok, please do.

-- 
Configure bugmail: http://bugs.meego.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching someone on the CC list of the bug.

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009