src/selection/NameFinder.cpp

/*
 * Copyright (c) 2005 The University of Notre Dame. All Rights Reserved.
 *
 * The University of Notre Dame grants you ("Licensee") a
 * non-exclusive, royalty free, license to use, modify and
 * redistribute this software in source and binary code form, provided
 * that the following conditions are met:
 *
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 *
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the
 *    distribution.
 *
 * This software is provided "AS IS," without a warranty of any
 * kind. All express or implied conditions, representations and
 * warranties, including any implied warranty of merchantability,
 * fitness for a particular purpose or non-infringement, are hereby
 * excluded.  The University of Notre Dame and its licensors shall not
 * be liable for any damages suffered by licensee as a result of
 * using, modifying or distributing the software or its
 * derivatives. In no event will the University of Notre Dame or its
 * licensors be liable for any lost revenue, profit or data, or for
 * direct, indirect, special, consequential, incidental or punitive
 * damages, however caused and regardless of the theory of liability,
 * arising out of the use of or inability to use software, even if the
 * University of Notre Dame has been advised of the possibility of
 * such damages.
 *
 * SUPPORT OPEN SCIENCE!  If you use OpenMD or its source code in your
 * research, please cite the appropriate papers when you publish your
 * work.  Good starting points are:
 *                                                                      
 * [1]  Meineke, et al., J. Comp. Chem. 26, 252-271 (2005).             
 * [2]  Fennell & Gezelter, J. Chem. Phys. 124, 234104 (2006).          
 * [3]  Sun, Lin & Gezelter, J. Chem. Phys. 128, 24107 (2008).          
 * [4]  Vardeman & Gezelter, in progress (2009).                        
 */
#include "selection/NameFinder.hpp"
#include "utils/wildcards.hpp"
#include "utils/StringTokenizer.hpp"
#include "primitives/Molecule.hpp"
#include "utils/StringUtils.hpp"
namespace OpenMD {

  TreeNode::~TreeNode(){
    std::map<std::string, TreeNode*>::iterator i;
    for ( i = children.begin(); i != children.end(); ++i) {
      i->second->~TreeNode();
    }
    children.clear();
  }


  NameFinder::NameFinder(SimInfo* info) : info_(info), root_(NULL){
    nStuntDouble_ = info_->getNGlobalAtoms() + info_->getNGlobalRigidBodies();
    loadNames();
  }


  NameFinder::~NameFinder(){
    delete root_;
  }

  void NameFinder::loadNames() {

    std::map<std::string, TreeNode*>::iterator foundIter;
    SimInfo::MoleculeIterator mi;
    Molecule* mol;
    Molecule::AtomIterator ai;
    Atom* atom;
    Molecule::RigidBodyIterator rbIter;
    RigidBody* rb;

    root_ = new TreeNode;
    root_->bs.resize(nStuntDouble_);
    root_->bs.setAll(); //
    
    for (mol = info_->beginMolecule(mi); mol != NULL; mol = info_->nextMolecule(mi)) {
           
      std::string molName = mol->getMoleculeName();
      TreeNode* currentMolNode = createNode(root_, molName);
        
      for(atom = mol->beginAtom(ai); atom != NULL; atom = mol->nextAtom(ai)) {
        std::string atomName = atom->getType();
        TreeNode* currentAtomNode = createNode(currentMolNode, atomName);
            
        currentMolNode->bs.setBitOn(atom->getGlobalIndex());
        currentAtomNode->bs.setBitOn(atom->getGlobalIndex());
      }

      for (rb = mol->beginRigidBody(rbIter); rb != NULL; rb = mol->nextRigidBody(rbIter)) {
        std::string rbName = rb->getType();
        TreeNode* currentRbNode = createNode(currentMolNode, rbName);
            
        currentMolNode->bs.setBitOn(rb->getGlobalIndex());
        currentRbNode->bs.setBitOn(rb->getGlobalIndex());

        //create nodes for atoms belong to this rigidbody
        for(atom = rb->beginAtom(ai); atom != NULL; atom = rb->nextAtom(ai)) {
          std::string rbAtomName = atom->getType();
          TreeNode* currentRbAtomNode = createNode(currentRbNode, rbName);;

          currentRbAtomNode->bs.setBitOn(atom->getGlobalIndex());
        }

      }
        
    }    

  }

  TreeNode* NameFinder::createNode(TreeNode* parent, const std::string& name) {
    TreeNode* node;    
    std::map<std::string, TreeNode*>::iterator foundIter;
    foundIter = parent->children.find(name);
    if ( foundIter  == parent->children.end()) {
      node = new TreeNode;
      node->name = name;
      node->bs.resize(nStuntDouble_);
      parent->children.insert(std::make_pair(name, node));
    }else {
      node = foundIter->second;
    }
    return node;
  }

  OpenMDBitSet NameFinder::match(const std::string& name){
    OpenMDBitSet bs(nStuntDouble_);
  
    StringTokenizer tokenizer(name, ".");

    std::vector<std::string> names;
    while(tokenizer.hasMoreTokens()) {
      names.push_back(tokenizer.nextToken());
    }

    int size = names.size();
    switch(size) {
    case 1 :
      //could be molecule name, atom name and rigidbody name
      matchMolecule(names[0], bs);
      matchStuntDouble("*", names[0], bs);
            
      break;
    case 2:
      //could be molecule.*(include atoms and rigidbodies) or rigidbody.*(atoms belong to rigidbody)

      if (!isInteger(names[1])){
        matchRigidAtoms("*", names[0], names[1], bs);
        matchStuntDouble(names[0], names[1], bs);
      } else {
        int internalIndex = lexi_cast<int>(names[1]);
        if (internalIndex < 0) {
          std::cerr << names[0] << ". " << names[1] << " is an invalid name" << std::endl;           
        } else {
          matchInternalIndex(names[0], internalIndex, bs);
        }
      }
            
      break;
    case 3:
      //must be molecule.rigidbody.*
      matchRigidAtoms(names[0], names[1], names[2], bs);
      break;
    default:      
      std::cerr << "invalid name: " << name << std::endl;
      break;           
    }

    return bs; 
  }

  void NameFinder::matchMolecule(const std::string& molName, OpenMDBitSet& bs) {
    std::vector<TreeNode*> molNodes = getMatchedChildren(root_, molName);            
    std::vector<TreeNode*>::iterator i;
    for( i = molNodes.begin(); i != molNodes.end(); ++i ) {
      bs |= (*i)->bs;
    }    
  }

  void NameFinder::matchStuntDouble(const std::string& molName, const std::string& sdName, OpenMDBitSet& bs){
    std::vector<TreeNode*> molNodes = getMatchedChildren(root_, molName);            
    std::vector<TreeNode*>::iterator i;
    for( i = molNodes.begin(); i != molNodes.end(); ++i ) {
      std::vector<TreeNode*> sdNodes = getMatchedChildren(*i, sdName);   
      std::vector<TreeNode*>::iterator j;
      for (j = sdNodes.begin(); j != sdNodes.end(); ++j) {
        bs |= (*j)->bs;
      }
    }

  }

  void NameFinder::matchRigidAtoms(const std::string& molName, const std::string& rbName, const std::string& rbAtomName, OpenMDBitSet& bs){
    std::vector<TreeNode*> molNodes = getMatchedChildren(root_, molName);            
    std::vector<TreeNode*>::iterator i;
    for( i = molNodes.begin(); i != molNodes.end(); ++i ) {
      std::vector<TreeNode*> rbNodes = getMatchedChildren(*i, rbName);   
      std::vector<TreeNode*>::iterator j;
      for (j = rbNodes.begin(); j != rbNodes.end(); ++j) {
        std::vector<TreeNode*> rbAtomNodes = getMatchedChildren(*j, rbAtomName);
        std::vector<TreeNode*>::iterator k;
        for(k = rbAtomNodes.begin(); k != rbAtomNodes.end(); ++k){
          bs |= (*k)->bs;
        }
      }
    }

  }


  std::vector<TreeNode*> NameFinder::getMatchedChildren(TreeNode* node, const std::string& name) {
    std::vector<TreeNode*> matchedNodes;
    std::map<std::string, TreeNode*>::iterator i;
    for (i = node->children.begin(); i != node->children.end(); ++i) {
      if (isMatched( i->first, name)) {
        matchedNodes.push_back(i->second);
      }
    }

    return matchedNodes;
  }

  bool NameFinder::isMatched(const std::string& str, const std::string& wildcard) {
    return Wildcard::wildcardfit (wildcard.c_str(), str.c_str());
  }


  void NameFinder::matchInternalIndex(const std::string& name, int internalIndex, OpenMDBitSet& bs){

    std::map<std::string, TreeNode*>::iterator foundIter;
    SimInfo::MoleculeIterator mi;
    Molecule* mol;

    for (mol = info_->beginMolecule(mi); mol != NULL; mol = info_->nextMolecule(mi)) {
           
      if (isMatched(mol->getMoleculeName(), name) ) {
        int natoms = mol->getNAtoms();
        int nrigidbodies = mol->getNRigidBodies();
        if (internalIndex >= natoms + nrigidbodies) {
          continue;
        } else if (internalIndex < natoms) {
          bs.setBitOn(mol->getAtomAt(internalIndex)->getGlobalIndex());
          continue;
        } else if ( internalIndex < natoms + nrigidbodies) {
          bs.setBitOn(mol->getRigidBodyAt(internalIndex - natoms)->getGlobalIndex());
        }
      }
        
    }    
    
  }

  bool NameFinder::isInteger(const std::string str) {
    for(int i =0; i < str.size(); ++i){
      if (!std::isdigit(str[i])) {
        return false;
      }
    }

    return true;
  }

}
Revision:	1390
Committed:	Wed Nov 25 20:02:06 2009 UTC (15 years, 11 months ago) by gezelter
File size:	9036 byte(s)
Log Message:	Almost all of the changes necessary to create OpenMD out of our old project (OOPSE-4)
#	User	Rev	Content
1	tim	283	/*
2			* Copyright (c) 2005 The University of Notre Dame. All Rights Reserved.
3			*
4			* The University of Notre Dame grants you ("Licensee") a
5			* non-exclusive, royalty free, license to use, modify and
6			* redistribute this software in source and binary code form, provided
7			* that the following conditions are met:
8			*
9	gezelter	1390	* 1. Redistributions of source code must retain the above copyright
10	tim	283	* notice, this list of conditions and the following disclaimer.
11			*
12	gezelter	1390	* 2. Redistributions in binary form must reproduce the above copyright
13	tim	283	* notice, this list of conditions and the following disclaimer in the
14			* documentation and/or other materials provided with the
15			* distribution.
16			*
17			* This software is provided "AS IS," without a warranty of any
18			* kind. All express or implied conditions, representations and
19			* warranties, including any implied warranty of merchantability,
20			* fitness for a particular purpose or non-infringement, are hereby
21			* excluded. The University of Notre Dame and its licensors shall not
22			* be liable for any damages suffered by licensee as a result of
23			* using, modifying or distributing the software or its
24			* derivatives. In no event will the University of Notre Dame or its
25			* licensors be liable for any lost revenue, profit or data, or for
26			* direct, indirect, special, consequential, incidental or punitive
27			* damages, however caused and regardless of the theory of liability,
28			* arising out of the use of or inability to use software, even if the
29			* University of Notre Dame has been advised of the possibility of
30			* such damages.
31	gezelter	1390	*
32			* SUPPORT OPEN SCIENCE! If you use OpenMD or its source code in your
33			* research, please cite the appropriate papers when you publish your
34			* work. Good starting points are:
35			*
36			* [1] Meineke, et al., J. Comp. Chem. 26, 252-271 (2005).
37			* [2] Fennell & Gezelter, J. Chem. Phys. 124, 234104 (2006).
38			* [3] Sun, Lin & Gezelter, J. Chem. Phys. 128, 24107 (2008).
39			* [4] Vardeman & Gezelter, in progress (2009).
40	tim	283	*/
41			#include "selection/NameFinder.hpp"
42	tim	284	#include "utils/wildcards.hpp"
43	tim	287	#include "utils/StringTokenizer.hpp"
44			#include "primitives/Molecule.hpp"
45	tim	303	#include "utils/StringUtils.hpp"
46	gezelter	1390	namespace OpenMD {
47	tim	283
48	gezelter	507	TreeNode::~TreeNode(){
49	tim	284	std::map<std::string, TreeNode*>::iterator i;
50			for ( i = children.begin(); i != children.end(); ++i) {
51	gezelter	507	i->second->~TreeNode();
52	tim	284	}
53			children.clear();
54	gezelter	507	}
55	tim	284
56
57	gezelter	507	NameFinder::NameFinder(SimInfo* info) : info_(info), root_(NULL){
58	tim	287	nStuntDouble_ = info_->getNGlobalAtoms() + info_->getNGlobalRigidBodies();
59	tim	283	loadNames();
60	gezelter	507	}
61	tim	283
62
63	gezelter	507	NameFinder::~NameFinder(){
64	tim	284	delete root_;
65	gezelter	507	}
66	tim	283
67	gezelter	507	void NameFinder::loadNames() {
68	tim	283
69	tim	284	std::map<std::string, TreeNode*>::iterator foundIter;
70	tim	283	SimInfo::MoleculeIterator mi;
71			Molecule* mol;
72			Molecule::AtomIterator ai;
73			Atom* atom;
74			Molecule::RigidBodyIterator rbIter;
75			RigidBody* rb;
76	tim	284
77			root_ = new TreeNode;
78	tim	287	root_->bs.resize(nStuntDouble_);
79	tim	284	root_->bs.setAll(); //
80
81	tim	283	for (mol = info_->beginMolecule(mi); mol != NULL; mol = info_->nextMolecule(mi)) {
82	tim	288
83	gezelter	507	std::string molName = mol->getMoleculeName();
84			TreeNode* currentMolNode = createNode(root_, molName);
85	tim	283
86	gezelter	507	for(atom = mol->beginAtom(ai); atom != NULL; atom = mol->nextAtom(ai)) {
87			std::string atomName = atom->getType();
88			TreeNode* currentAtomNode = createNode(currentMolNode, atomName);
89	tim	288
90	gezelter	507	currentMolNode->bs.setBitOn(atom->getGlobalIndex());
91			currentAtomNode->bs.setBitOn(atom->getGlobalIndex());
92			}
93	tim	284
94	gezelter	507	for (rb = mol->beginRigidBody(rbIter); rb != NULL; rb = mol->nextRigidBody(rbIter)) {
95			std::string rbName = rb->getType();
96			TreeNode* currentRbNode = createNode(currentMolNode, rbName);
97	tim	284
98	gezelter	507	currentMolNode->bs.setBitOn(rb->getGlobalIndex());
99			currentRbNode->bs.setBitOn(rb->getGlobalIndex());
100	tim	284
101	gezelter	507	//create nodes for atoms belong to this rigidbody
102			for(atom = rb->beginAtom(ai); atom != NULL; atom = rb->nextAtom(ai)) {
103			std::string rbAtomName = atom->getType();
104			TreeNode* currentRbAtomNode = createNode(currentRbNode, rbName);;
105	tim	288
106	gezelter	507	currentRbAtomNode->bs.setBitOn(atom->getGlobalIndex());
107			}
108	tim	284
109	gezelter	507	}
110	tim	283
111			}
112	tim	284
113	gezelter	507	}
114	tim	283
115	gezelter	507	TreeNode* NameFinder::createNode(TreeNode* parent, const std::string& name) {
116	tim	288	TreeNode* node;
117			std::map<std::string, TreeNode*>::iterator foundIter;
118			foundIter = parent->children.find(name);
119			if ( foundIter == parent->children.end()) {
120	gezelter	507	node = new TreeNode;
121			node->name = name;
122			node->bs.resize(nStuntDouble_);
123			parent->children.insert(std::make_pair(name, node));
124	tim	288	}else {
125	gezelter	507	node = foundIter->second;
126	tim	288	}
127			return node;
128	gezelter	507	}
129	tim	283
130	gezelter	1390	OpenMDBitSet NameFinder::match(const std::string& name){
131			OpenMDBitSet bs(nStuntDouble_);
132	tim	288
133	tim	283	StringTokenizer tokenizer(name, ".");
134
135			std::vector<std::string> names;
136			while(tokenizer.hasMoreTokens()) {
137	gezelter	507	names.push_back(tokenizer.nextToken());
138	tim	283	}
139
140			int size = names.size();
141			switch(size) {
142	gezelter	507	case 1 :
143			//could be molecule name, atom name and rigidbody name
144			matchMolecule(names[0], bs);
145			matchStuntDouble("*", names[0], bs);
146	tim	284
147	gezelter	507	break;
148			case 2:
149			//could be molecule.(include atoms and rigidbodies) or rigidbody.(atoms belong to rigidbody)
150	tim	303
151	gezelter	507	if (!isInteger(names[1])){
152			matchRigidAtoms("*", names[0], names[1], bs);
153			matchStuntDouble(names[0], names[1], bs);
154			} else {
155			int internalIndex = lexi_cast<int>(names[1]);
156			if (internalIndex < 0) {
157			std::cerr << names[0] << ". " << names[1] << " is an invalid name" << std::endl;
158			} else {
159			matchInternalIndex(names[0], internalIndex, bs);
160			}
161			}
162	tim	284
163	gezelter	507	break;
164			case 3:
165			//must be molecule.rigidbody.*
166			matchRigidAtoms(names[0], names[1], names[2], bs);
167			break;
168			default:
169			std::cerr << "invalid name: " << name << std::endl;
170			break;
171	tim	283	}
172
173	tim	288	return bs;
174	gezelter	507	}
175	tim	283
176	gezelter	1390	void NameFinder::matchMolecule(const std::string& molName, OpenMDBitSet& bs) {
177	tim	284	std::vector<TreeNode*> molNodes = getMatchedChildren(root_, molName);
178			std::vector<TreeNode*>::iterator i;
179			for( i = molNodes.begin(); i != molNodes.end(); ++i ) {
180	gezelter	507	bs \|= (*i)->bs;
181	tim	284	}
182	gezelter	507	}
183	tim	284
184	gezelter	1390	void NameFinder::matchStuntDouble(const std::string& molName, const std::string& sdName, OpenMDBitSet& bs){
185	tim	284	std::vector<TreeNode*> molNodes = getMatchedChildren(root_, molName);
186			std::vector<TreeNode*>::iterator i;
187			for( i = molNodes.begin(); i != molNodes.end(); ++i ) {
188	gezelter	507	std::vector<TreeNode> sdNodes = getMatchedChildren(i, sdName);
189			std::vector<TreeNode*>::iterator j;
190			for (j = sdNodes.begin(); j != sdNodes.end(); ++j) {
191			bs \|= (*j)->bs;
192			}
193	tim	284	}
194
195	gezelter	507	}
196	tim	284
197	gezelter	1390	void NameFinder::matchRigidAtoms(const std::string& molName, const std::string& rbName, const std::string& rbAtomName, OpenMDBitSet& bs){
198	tim	284	std::vector<TreeNode*> molNodes = getMatchedChildren(root_, molName);
199			std::vector<TreeNode*>::iterator i;
200			for( i = molNodes.begin(); i != molNodes.end(); ++i ) {
201	gezelter	507	std::vector<TreeNode> rbNodes = getMatchedChildren(i, rbName);
202			std::vector<TreeNode*>::iterator j;
203			for (j = rbNodes.begin(); j != rbNodes.end(); ++j) {
204			std::vector<TreeNode> rbAtomNodes = getMatchedChildren(j, rbAtomName);
205			std::vector<TreeNode*>::iterator k;
206			for(k = rbAtomNodes.begin(); k != rbAtomNodes.end(); ++k){
207			bs \|= (*k)->bs;
208			}
209			}
210	tim	284	}
211
212	gezelter	507	}
213	tim	284
214
215	gezelter	507	std::vector<TreeNode> NameFinder::getMatchedChildren(TreeNode node, const std::string& name) {
216	tim	284	std::vector<TreeNode*> matchedNodes;
217			std::map<std::string, TreeNode*>::iterator i;
218			for (i = node->children.begin(); i != node->children.end(); ++i) {
219	gezelter	507	if (isMatched( i->first, name)) {
220			matchedNodes.push_back(i->second);
221			}
222	tim	284	}
223
224			return matchedNodes;
225	gezelter	507	}
226	tim	284
227	gezelter	507	bool NameFinder::isMatched(const std::string& str, const std::string& wildcard) {
228	tim	284	return Wildcard::wildcardfit (wildcard.c_str(), str.c_str());
229	gezelter	507	}
230	tim	284
231	tim	303
232	gezelter	1390	void NameFinder::matchInternalIndex(const std::string& name, int internalIndex, OpenMDBitSet& bs){
233	tim	303
234			std::map<std::string, TreeNode*>::iterator foundIter;
235			SimInfo::MoleculeIterator mi;
236			Molecule* mol;
237
238			for (mol = info_->beginMolecule(mi); mol != NULL; mol = info_->nextMolecule(mi)) {
239
240	gezelter	507	if (isMatched(mol->getMoleculeName(), name) ) {
241			int natoms = mol->getNAtoms();
242			int nrigidbodies = mol->getNRigidBodies();
243			if (internalIndex >= natoms + nrigidbodies) {
244			continue;
245			} else if (internalIndex < natoms) {
246			bs.setBitOn(mol->getAtomAt(internalIndex)->getGlobalIndex());
247			continue;
248			} else if ( internalIndex < natoms + nrigidbodies) {
249			bs.setBitOn(mol->getRigidBodyAt(internalIndex - natoms)->getGlobalIndex());
250			}
251			}
252	tim	303
253			}
254
255	gezelter	507	}
256	tim	303
257	gezelter	507	bool NameFinder::isInteger(const std::string str) {
258	tim	303	for(int i =0; i < str.size(); ++i){
259	gezelter	507	if (!std::isdigit(str[i])) {
260			return false;
261			}
262	tim	303	}
263
264			return true;
265	gezelter	507	}
266	tim	303
267			}