windows-nt/Source/XPSP1/NT/enduser/troubleshoot/tshoot/nodestate.cpp

110 lines
2.8 KiB
C++
Raw Permalink Normal View History

2020-09-26 03:20:57 -05:00
//
// MODULE: NODESTATE.CPP
//
// PURPOSE: Implement some functions relevant to CNodeState
//
// PROJECT: Generic Troubleshooter DLL for Microsoft AnswerPoint
//
// COMPANY: Saltmine Creative, Inc. (206)-284-7511 support@saltmine.com
//
// AUTHOR: Joe Mabel
//
// ORIGINAL DATE: 10/99
//
// NOTES:
// 1.
//
// Version Date By Comments
//--------------------------------------------------------------------
// V3.0 10/15/99 JM original
//
#include "nodestate.h"
#include <algorithm>
// operator-= removes only identical node/state pairs.
// This is the appropriate behavior if lhs represents all current node states and rhs
// represents sniffed values: if the node already deviates from a sniffed value, and
// we are removing sniffed values, this sniffed value is irrelevant & shouldn't be
// removed from lhs.
// this is an N-squared algorithm.
// >>> $MAINT There might be a case for working with sorted lists and STL generic
// algorithms, which could reduce this to N log N
CBasisForInference& operator-=(CBasisForInference& lhs, const CBasisForInference& rhs)
{
CBasisForInference::iterator i = lhs.begin();
while ( i != lhs.end() )
{
NID inid = i->nid();
IST istate = i->state();
bool bMatch = false;
for (CBasisForInference::const_iterator j = rhs.begin(); j != rhs.end(); ++j)
{
if (j->nid() == inid && j->state() == istate)
{
bMatch = true;
break;
}
}
if (bMatch)
i = lhs.erase(i);
else
++i;
}
return lhs;
}
// operator+= adds only pairs for which there is no match to any node already in lhs.
// This is the appropriate behavior if lhs represents node states obtained by means other
// than sniffing and rhs represents re-sniffed values: if the node already has a value
// assigned by other means, the sniffed values are irrelevant & shouldn't be
// added to lhs.
// this is an N-squared algorithm.
// >>> $MAINT There might be a case for working with sorted lists and STL generic
// algorithms, which could reduce this to N log N
CBasisForInference& operator+=(CBasisForInference& lhs, const CBasisForInference& rhs)
{
for (CBasisForInference::const_iterator j = rhs.begin(); j != rhs.end(); ++j)
{
NID jnid = j->nid();
bool bMatch = false;
for (CBasisForInference::const_iterator i = lhs.begin(); i != lhs.end(); ++i)
{
if (i->nid() == jnid)
{
bMatch = true;
break;
}
}
if (!bMatch)
lhs.push_back(*j);
}
return lhs;
}
vector<NID>& operator-=(vector<NID>& lhs, const CBasisForInference& rhs)
{
for (long i = 0; i < rhs.size(); i++)
{
vector<NID>::iterator found = find(lhs.begin(), lhs.end(), rhs[i].nid());
if (found < lhs.end())
lhs.erase(found);
}
return lhs;
}
vector<NID>& operator+=(vector<NID>& lhs, const CBasisForInference& rhs)
{
for (long i = 0; i < rhs.size(); i++)
lhs.push_back(rhs[i].nid());
return lhs;
}