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]  Kuang & Gezelter,  J. Chem. Phys. 133, 164101 (2010).
 * [5]  Vardeman, Stocker & Gezelter, J. Chem. Theory Comput. 7, 834 (2011).
 */
#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()) > 0 ? true : false;
  }


  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(unsigned int i = 0; i < str.size(); ++i){
      if (!std::isdigit(str[i])) {
        return false;
      }
    }
    return true;
  }
}
Revision:	1803
Committed:	Wed Oct 3 14:20:07 2012 UTC (12 years, 6 months ago) by gezelter
File size:	9125 byte(s)
Log Message:	Merging trunk changes back to development branch
#	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	gezelter	1665	* [4] Kuang & Gezelter, J. Chem. Phys. 133, 164101 (2010).
40			* [5] Vardeman, Stocker & Gezelter, J. Chem. Theory Comput. 7, 834 (2011).
41	tim	283	*/
42			#include "selection/NameFinder.hpp"
43	tim	284	#include "utils/wildcards.hpp"
44	tim	287	#include "utils/StringTokenizer.hpp"
45			#include "primitives/Molecule.hpp"
46	tim	303	#include "utils/StringUtils.hpp"
47	gezelter	1390	namespace OpenMD {
48	tim	283
49	gezelter	507	TreeNode::~TreeNode(){
50	tim	284	std::map<std::string, TreeNode*>::iterator i;
51			for ( i = children.begin(); i != children.end(); ++i) {
52	gezelter	507	i->second->~TreeNode();
53	tim	284	}
54			children.clear();
55	gezelter	507	}
56	tim	284
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	gezelter	507	NameFinder::~NameFinder(){
63	tim	284	delete root_;
64	gezelter	507	}
65	tim	283
66	gezelter	507	void NameFinder::loadNames() {
67	tim	283
68	tim	284	std::map<std::string, TreeNode*>::iterator foundIter;
69	tim	283	SimInfo::MoleculeIterator mi;
70			Molecule* mol;
71			Molecule::AtomIterator ai;
72			Atom* atom;
73			Molecule::RigidBodyIterator rbIter;
74			RigidBody* rb;
75	tim	284
76			root_ = new TreeNode;
77	tim	287	root_->bs.resize(nStuntDouble_);
78	tim	284	root_->bs.setAll(); //
79
80	gezelter	1803	for (mol = info_->beginMolecule(mi); mol != NULL;
81			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	1803	for (rb = mol->beginRigidBody(rbIter); rb != NULL;
95			rb = mol->nextRigidBody(rbIter)) {
96	gezelter	507	std::string rbName = rb->getType();
97			TreeNode* currentRbNode = createNode(currentMolNode, rbName);
98	tim	284
99	gezelter	507	currentMolNode->bs.setBitOn(rb->getGlobalIndex());
100			currentRbNode->bs.setBitOn(rb->getGlobalIndex());
101	tim	284
102	gezelter	507	//create nodes for atoms belong to this rigidbody
103			for(atom = rb->beginAtom(ai); atom != NULL; atom = rb->nextAtom(ai)) {
104			std::string rbAtomName = atom->getType();
105			TreeNode* currentRbAtomNode = createNode(currentRbNode, rbName);;
106	tim	288
107	gezelter	507	currentRbAtomNode->bs.setBitOn(atom->getGlobalIndex());
108			}
109			}
110	gezelter	1803	}
111	gezelter	507	}
112	tim	283
113	gezelter	507	TreeNode* NameFinder::createNode(TreeNode* parent, const std::string& name) {
114	tim	288	TreeNode* node;
115			std::map<std::string, TreeNode*>::iterator foundIter;
116			foundIter = parent->children.find(name);
117			if ( foundIter == parent->children.end()) {
118	gezelter	507	node = new TreeNode;
119			node->name = name;
120			node->bs.resize(nStuntDouble_);
121			parent->children.insert(std::make_pair(name, node));
122	tim	288	}else {
123	gezelter	507	node = foundIter->second;
124	tim	288	}
125			return node;
126	gezelter	507	}
127	tim	283
128	gezelter	1390	OpenMDBitSet NameFinder::match(const std::string& name){
129			OpenMDBitSet bs(nStuntDouble_);
130	tim	288
131	tim	283	StringTokenizer tokenizer(name, ".");
132
133			std::vector<std::string> names;
134			while(tokenizer.hasMoreTokens()) {
135	gezelter	507	names.push_back(tokenizer.nextToken());
136	tim	283	}
137
138			int size = names.size();
139			switch(size) {
140	gezelter	507	case 1 :
141			//could be molecule name, atom name and rigidbody name
142			matchMolecule(names[0], bs);
143			matchStuntDouble("*", names[0], bs);
144	tim	284
145	gezelter	507	break;
146			case 2:
147			//could be molecule.(include atoms and rigidbodies) or rigidbody.(atoms belong to rigidbody)
148	tim	303
149	gezelter	507	if (!isInteger(names[1])){
150			matchRigidAtoms("*", names[0], names[1], bs);
151			matchStuntDouble(names[0], names[1], bs);
152			} else {
153			int internalIndex = lexi_cast<int>(names[1]);
154			if (internalIndex < 0) {
155			std::cerr << names[0] << ". " << names[1] << " is an invalid name" << std::endl;
156			} else {
157			matchInternalIndex(names[0], internalIndex, bs);
158			}
159			}
160	tim	284
161	gezelter	507	break;
162			case 3:
163			//must be molecule.rigidbody.*
164			matchRigidAtoms(names[0], names[1], names[2], bs);
165			break;
166			default:
167			std::cerr << "invalid name: " << name << std::endl;
168			break;
169	tim	283	}
170
171	tim	288	return bs;
172	gezelter	507	}
173	tim	283
174	gezelter	1390	void NameFinder::matchMolecule(const std::string& molName, OpenMDBitSet& bs) {
175	tim	284	std::vector<TreeNode*> molNodes = getMatchedChildren(root_, molName);
176			std::vector<TreeNode*>::iterator i;
177			for( i = molNodes.begin(); i != molNodes.end(); ++i ) {
178	gezelter	507	bs \|= (*i)->bs;
179	tim	284	}
180	gezelter	507	}
181	tim	284
182	gezelter	1390	void NameFinder::matchStuntDouble(const std::string& molName, const std::string& sdName, OpenMDBitSet& bs){
183	tim	284	std::vector<TreeNode*> molNodes = getMatchedChildren(root_, molName);
184			std::vector<TreeNode*>::iterator i;
185			for( i = molNodes.begin(); i != molNodes.end(); ++i ) {
186	gezelter	507	std::vector<TreeNode> sdNodes = getMatchedChildren(i, sdName);
187			std::vector<TreeNode*>::iterator j;
188			for (j = sdNodes.begin(); j != sdNodes.end(); ++j) {
189			bs \|= (*j)->bs;
190			}
191	tim	284	}
192
193	gezelter	507	}
194	tim	284
195	gezelter	1390	void NameFinder::matchRigidAtoms(const std::string& molName, const std::string& rbName, const std::string& rbAtomName, OpenMDBitSet& bs){
196	tim	284	std::vector<TreeNode*> molNodes = getMatchedChildren(root_, molName);
197			std::vector<TreeNode*>::iterator i;
198			for( i = molNodes.begin(); i != molNodes.end(); ++i ) {
199	gezelter	507	std::vector<TreeNode> rbNodes = getMatchedChildren(i, rbName);
200			std::vector<TreeNode*>::iterator j;
201			for (j = rbNodes.begin(); j != rbNodes.end(); ++j) {
202			std::vector<TreeNode> rbAtomNodes = getMatchedChildren(j, rbAtomName);
203			std::vector<TreeNode*>::iterator k;
204			for(k = rbAtomNodes.begin(); k != rbAtomNodes.end(); ++k){
205			bs \|= (*k)->bs;
206			}
207			}
208	tim	284	}
209
210	gezelter	507	}
211	tim	284
212
213	gezelter	507	std::vector<TreeNode> NameFinder::getMatchedChildren(TreeNode node, const std::string& name) {
214	tim	284	std::vector<TreeNode*> matchedNodes;
215			std::map<std::string, TreeNode*>::iterator i;
216			for (i = node->children.begin(); i != node->children.end(); ++i) {
217	gezelter	507	if (isMatched( i->first, name)) {
218			matchedNodes.push_back(i->second);
219			}
220	tim	284	}
221
222			return matchedNodes;
223	gezelter	507	}
224	tim	284
225	gezelter	507	bool NameFinder::isMatched(const std::string& str, const std::string& wildcard) {
226	gezelter	1767	return Wildcard::wildcardfit(wildcard.c_str(), str.c_str()) > 0 ? true : false;
227	gezelter	507	}
228	tim	284
229	tim	303
230	gezelter	1390	void NameFinder::matchInternalIndex(const std::string& name, int internalIndex, OpenMDBitSet& bs){
231	tim	303
232			std::map<std::string, TreeNode*>::iterator foundIter;
233			SimInfo::MoleculeIterator mi;
234			Molecule* mol;
235
236	gezelter	1803	for (mol = info_->beginMolecule(mi); mol != NULL;
237			mol = info_->nextMolecule(mi)) {
238	tim	303
239	gezelter	507	if (isMatched(mol->getMoleculeName(), name) ) {
240			int natoms = mol->getNAtoms();
241			int nrigidbodies = mol->getNRigidBodies();
242			if (internalIndex >= natoms + nrigidbodies) {
243			continue;
244			} else if (internalIndex < natoms) {
245			bs.setBitOn(mol->getAtomAt(internalIndex)->getGlobalIndex());
246			continue;
247			} else if ( internalIndex < natoms + nrigidbodies) {
248			bs.setBitOn(mol->getRigidBodyAt(internalIndex - natoms)->getGlobalIndex());
249			}
250			}
251	gezelter	1803	}
252	gezelter	507	}
253	tim	303
254	gezelter	507	bool NameFinder::isInteger(const std::string str) {
255	gezelter	1767	for(unsigned int i = 0; i < str.size(); ++i){
256	gezelter	507	if (!std::isdigit(str[i])) {
257			return false;
258			}
259	tim	303	}
260			return true;
261	gezelter	507	}
262	tim	303	}