src/selection/SelectionCompiler.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/SelectionCompiler.hpp"
#include "utils/StringUtils.hpp"
namespace OpenMD {

  bool SelectionCompiler::compile(const std::string& filename, 
                                  const std::string& script) {

    this->filename = filename;
    this->script = script;
    lineNumbers.clear();
    lineIndices.clear();
    aatokenCompiled.clear();
        
    if (internalCompile()) {
      return true;
    }
    
    int icharEnd;
    if ((icharEnd = script.find('\r', ichCurrentCommand)) == std::string::npos &&
        (icharEnd = script.find('\n', ichCurrentCommand)) == std::string::npos) {
      icharEnd = script.size();
    }
    errorLine = script.substr(ichCurrentCommand, icharEnd);
    return false;
  }

  bool SelectionCompiler::internalCompile(){

    cchScript = script.size();
    ichToken = 0;
    lineCurrent = 1;

    error = false;

    //std::vector<Token> lltoken;
    aatokenCompiled.clear();
    std::vector<Token> ltoken;

    Token tokenCommand;
    int tokCommand = Token::nada;

    for ( ; true; ichToken += cchToken) {
      if (lookingAtLeadingWhitespace())
        continue;
      //if (lookingAtComment())
      //    continue;
      bool endOfLine = lookingAtEndOfLine();
      if (endOfLine || lookingAtEndOfStatement()) {
        if (tokCommand != Token::nada) {
          if (! compileCommand(ltoken)) {
            return false;
          }
          aatokenCompiled.push_back(atokenCommand);
          lineNumbers.push_back(lineCurrent);
          lineIndices.push_back(ichCurrentCommand);
          ltoken.clear();
          tokCommand = Token::nada;
        }
            
        if (ichToken < cchScript) {
          if (endOfLine)
            ++lineCurrent;
          continue;
        }
        break;
      }

      if (tokCommand != Token::nada) {
        if (lookingAtString()) {
          std::string str = getUnescapedStringLiteral();
          ltoken.push_back(Token(Token::string, str));
          continue;
        }
        //if ((tokCommand & Token::specialstring) != 0 &&
        //    lookingAtSpecialString()) {
        //    std::string str = script.substr(ichToken, ichToken + cchToken);
        //    ltoken.push_back(Token(Token::string, str));
        //    continue;
        //}
        if (lookingAtDecimal((tokCommand & Token::negnums) != 0)) {
          float value = lexi_cast<float>(script.substr(ichToken, cchToken));        
          ltoken.push_back(Token(Token::decimal, boost::any(value)));
          continue;
        }
        if (lookingAtInteger((tokCommand & Token::negnums) != 0)) {

          int val = lexi_cast<int>(script.substr(ichToken, cchToken));
          ltoken.push_back(Token(Token::integer,   boost::any(val)));
          continue;
        }
      }
      
      if (lookingAtLookupToken()) {
        std::string ident = script.substr(ichToken, cchToken);
        Token token;            
        Token* pToken = TokenMap::getInstance()->getToken(ident);
        if (pToken != NULL) {
          token = *pToken;
        } else {
          token = Token(Token::identifier, ident);
        }
            
        int tok = token.tok;
            
        switch (tokCommand) {
        case Token::nada:
          ichCurrentCommand = ichToken;
          //tokenCommand = token;
          tokCommand = tok;
          if ((tokCommand & Token::command) == 0)
            return commandExpected();
          break;

        case Token::define:
          if (ltoken.size() == 1) {
            // we are looking at the variable name
            if (tok != Token::identifier &&
                (tok & Token::predefinedset) != Token::predefinedset)
              return invalidExpressionToken(ident);
          } else {
            // we are looking at the expression
            if (tok != Token::identifier && 
                (tok & (Token::expression | Token::predefinedset)) == 0)
              return invalidExpressionToken(ident);
          }
                    
          break;

        case Token::select:
          if (tok != Token::identifier && (tok & Token::expression) == 0)
            return invalidExpressionToken(ident);
          break;
        }
        ltoken.push_back(token);
        continue;
      }

      if (ltoken.size() == 0) {
        return commandExpected();
      }
        
      return unrecognizedToken();
    }

    return true;
  }


  bool SelectionCompiler::lookingAtLeadingWhitespace() {

    int ichT = ichToken;
    while (ichT < cchScript && std::isspace(script[ichT])) {
      ++ichT;
    }
    cchToken = ichT - ichToken;
    return cchToken > 0;
  }

  bool SelectionCompiler::lookingAtEndOfLine() {
    if (ichToken == cchScript)
      return true;
    int ichT = ichToken;
    char ch = script[ichT];
    if (ch == '\r') {
      ++ichT;
      if (ichT < cchScript && script[ichT] == '\n')
        ++ichT;
    } else if (ch == '\n') {
      ++ichT;
    } else {
      return false;
    }
    cchToken = ichT - ichToken;
    return true;
  }

  bool SelectionCompiler::lookingAtEndOfStatement() {
    if (ichToken == cchScript || script[ichToken] != ';')
      return false;
    cchToken = 1;
    return true;
  }

  bool SelectionCompiler::lookingAtString() {
    if (ichToken == cchScript)
      return false;
    if (script[ichToken] != '"')
      return false;
    // remove support for single quote
    // in order to use it in atom expressions
    //    char chFirst = script.charAt(ichToken);
    //    if (chFirst != '"' && chFirst != '\'')
    //      return false;
    int ichT = ichToken + 1;
    //    while (ichT < cchScript && script.charAt(ichT++) != chFirst)
    char ch;
    bool previousCharBackslash = false;
    while (ichT < cchScript) {
      ch = script[ichT++];
      if (ch == '"' && !previousCharBackslash)
        break;
      previousCharBackslash = ch == '\\' ? !previousCharBackslash : false;
    }
    cchToken = ichT - ichToken;

    return true;
  }

  
  std::string SelectionCompiler::getUnescapedStringLiteral() {
    /** @todo */
    std::string sb(cchToken - 2, ' ');
    
    int ichMax = ichToken + cchToken - 1;
    int ich = ichToken + 1;

    while (ich < ichMax) {
      char ch = script[ich++];
      if (ch == '\\' && ich < ichMax) {
        ch = script[ich++];
        switch (ch) {
        case 'b':
          ch = '\b';
          break;
        case 'n':
          ch = '\n';
          break;
        case 't':
          ch = '\t';
          break;
        case 'r':
          ch = '\r';
          // fall into
        case '"':
        case '\\':
        case '\'':
          break;
        case 'x':
        case 'u':
          int digitCount = ch == 'x' ? 2 : 4;
          if (ich < ichMax) {
            int unicode = 0;
            for (int k = digitCount; --k >= 0 && ich < ichMax; ) {
              char chT = script[ich];
              int hexit = getHexitValue(chT);
              if (hexit < 0)
                break;
              unicode <<= 4;
              unicode += hexit;
              ++ich;
            }
            ch = (char)unicode;
          }
        }
      }
      sb.append(1, ch);
    }

    return sb;
  }

  int SelectionCompiler::getHexitValue(char ch) {
    if (ch >= '0' && ch <= '9')
      return ch - '0';
    else if (ch >= 'a' && ch <= 'f')
      return 10 + ch - 'a';
    else if (ch >= 'A' && ch <= 'F')
      return 10 + ch - 'A';
    else
      return -1;
  }

  bool SelectionCompiler::lookingAtSpecialString() {
    int ichT = ichToken;
    char ch = script[ichT];
    while (ichT < cchScript && ch != ';' && ch != '\r' && ch != '\n') {
      ++ichT;
    }
    cchToken = ichT - ichToken;
    return cchToken > 0;
  }

  bool SelectionCompiler::lookingAtDecimal(bool allowNegative) {
    if (ichToken == cchScript) {
      return false;
    }
    
    int ichT = ichToken;
    if (script[ichT] == '-') {
      ++ichT;
    }
    bool digitSeen = false;
    char ch = 'X';
    while (ichT < cchScript && std::isdigit(ch = script[ichT])) {
      ++ichT;
      digitSeen = true;
    }

    if (ichT == cchScript || ch != '.') {
      return false;
    }

    // to support DMPC.1, let's check the character before the dot
    if (ch == '.' && (ichT > 0) && std::isalpha(script[ichT - 1])) {
      return false;
    }

    ++ichT;
    while (ichT < cchScript && std::isdigit(script[ichT])) {
      ++ichT;
      digitSeen = true;
    }
    cchToken = ichT - ichToken;
    return digitSeen;
  }

  bool SelectionCompiler::lookingAtInteger(bool allowNegative) {
    if (ichToken == cchScript) {
      return false;
    }
    int ichT = ichToken;
    if (allowNegative && script[ichToken] == '-') {
      ++ichT;
    }
    int ichBeginDigits = ichT;
    while (ichT < cchScript && std::isdigit(script[ichT])) {
      ++ichT;
    }
    if (ichBeginDigits == ichT) {
      return false;
    }
    cchToken = ichT - ichToken;
    return true;
  }

  bool SelectionCompiler::lookingAtLookupToken() {
    if (ichToken == cchScript) {
      return false;
    }

    int ichT = ichToken;
    char ch;
    switch (ch = script[ichT++]) {
    case '(':
    case ')':
    case ',':
    case '[':
    case ']':
      break;
    case '&':
    case '|':
      if (ichT < cchScript && script[ichT] == ch) {
        ++ichT;
      }
      break;
    case '<':
    case '=':
    case '>':
      if (ichT < cchScript && ((ch = script[ichT]) == '<' || ch == '=' || ch == '>')) {
        ++ichT;
      }
      break;
    case '/':
    case '!':
      if (ichT < cchScript && script[ichT] == '=') {
        ++ichT;
      }
      break;
    default:
      if ((ch < 'a' || ch > 'z') && (ch < 'A' && ch > 'Z') && ch != '_') {
        return false;
      }
    case '*':
    case '?': // include question marks in identifier for atom expressions
      while (ichT < cchScript && !std::isspace(ch = script[ichT]) && 
             (std::isalpha(ch) ||std::isdigit(ch) || ch == '_' || ch == '.' || ch == '*' || ch == '?' || ch == '+' || ch == '-' || ch == '[' || ch == ']') ){

        ++ichT;
      }
      break;
    }

    cchToken = ichT - ichToken;

    return true;
  }

  bool SelectionCompiler::compileCommand(const std::vector<Token>& ltoken) {
    const Token& tokenCommand = ltoken[0];
    int tokCommand = tokenCommand.tok;

    atokenCommand = ltoken;
    if ((tokCommand & Token::expressionCommand) != 0 && !compileExpression()) {
      return false;
    }
    
    return true;
  }

  bool SelectionCompiler::compileExpression() {
    /** todo */
    int i = 1;
    int tokCommand = atokenCommand[0].tok;
    if (tokCommand == Token::define) {
      i = 2;
    } else if ((tokCommand & Token::embeddedExpression) != 0) {
      // look for the open parenthesis
      while (i < atokenCommand.size() &&
             atokenCommand[i].tok != Token::leftparen)
        ++i;
    }

    if (i >= atokenCommand.size()) {
      return true;
    }
    return compileExpression(i);
  }

                  
  bool SelectionCompiler::addTokenToPostfix(const Token& token) {
    ltokenPostfix.push_back(token);
    return true;
  }

  bool SelectionCompiler::compileExpression(int itoken) {
    ltokenPostfix.clear();
    for (int i = 0; i < itoken; ++i) {
      addTokenToPostfix(atokenCommand[i]);
    }
    
    atokenInfix = atokenCommand;
    itokenInfix = itoken;

    addTokenToPostfix(Token::tokenExpressionBegin);
    if (!clauseOr()) {
      return false;
    }
    
    addTokenToPostfix(Token::tokenExpressionEnd);
    if (itokenInfix != atokenInfix.size()) {
      return endOfExpressionExpected();
    }

    atokenCommand = ltokenPostfix;
    return true;
  }

  Token SelectionCompiler::tokenNext() {
    if (itokenInfix == atokenInfix.size()) {
      return Token();
    }
    return atokenInfix[itokenInfix++];
  }

  boost::any SelectionCompiler::valuePeek() {
    if (itokenInfix == atokenInfix.size()) {
      return boost::any();
    } else {
      return atokenInfix[itokenInfix].value;
    }
  }

  int SelectionCompiler::tokPeek() {
    if (itokenInfix == atokenInfix.size()) {
      return 0;
    }else {
      return atokenInfix[itokenInfix].tok;
    }
  }

  bool SelectionCompiler::clauseOr() {
    if (!clauseAnd()) {
      return false;
    }
    
    while (tokPeek() == Token::opOr) {
      Token tokenOr = tokenNext();
      if (!clauseAnd()) {
        return false;
      }
      addTokenToPostfix(tokenOr);
    }
    return true;
  }

  bool SelectionCompiler::clauseAnd() {
    if (!clauseNot()) {
      return false;
    }

    while (tokPeek() == Token::opAnd) {
      Token tokenAnd = tokenNext();
      if (!clauseNot()) {
        return false;
      }
      addTokenToPostfix(tokenAnd);
    }
    return true;
  }

  bool SelectionCompiler::clauseNot() {
    if (tokPeek() == Token::opNot) {
      Token tokenNot = tokenNext();
      if (!clauseNot()) {
        return false;
      }
      return addTokenToPostfix(tokenNot);
    }
    return clausePrimitive();
  }

  bool SelectionCompiler::clausePrimitive() {
    int tok = tokPeek();
    switch (tok) {
    case Token::within:
      return clauseWithin();

    case Token::asterisk:
    case Token::identifier:
      return clauseChemObjName();

    case Token::integer :
      return clauseIndex();
    default:
      if ((tok & Token::atomproperty) == Token::atomproperty) {
        return clauseComparator();
      }
      if ((tok & Token::predefinedset) != Token::predefinedset) {
        break;
      }
      // fall into the code and below and just add the token
    case Token::all:
    case Token::none:
    case Token::hull:
      return addTokenToPostfix(tokenNext());
    case Token::leftparen:
      tokenNext();
      if (!clauseOr()) {
        return false;
      }
      if (tokenNext().tok != Token::rightparen) {
        return rightParenthesisExpected();
      }
      return true;
    }
    return unrecognizedExpressionToken();
  }

  bool SelectionCompiler::clauseComparator() {
    Token tokenAtomProperty = tokenNext();
    Token tokenComparator = tokenNext();
    if ((tokenComparator.tok & Token::comparator) == 0) {
      return comparisonOperatorExpected();
    }

    Token tokenValue = tokenNext();
    if (tokenValue.tok != Token::integer && tokenValue.tok != Token::decimal) {
      return numberExpected();
    }
    
    float val;
    if (tokenValue.value.type() == typeid(int)) {
      val = boost::any_cast<int>(tokenValue.value);
    } else if (tokenValue.value.type() == typeid(float)) {
      val = boost::any_cast<float>(tokenValue.value);
    } else {
      return false;
    }

    boost::any floatVal;
    floatVal = val;
    return addTokenToPostfix(Token(tokenComparator.tok,
                                   tokenAtomProperty.tok, floatVal));
  }

  bool SelectionCompiler::clauseWithin() {
    tokenNext();                             // WITHIN
    if (tokenNext().tok != Token::leftparen) {  // (
      return leftParenthesisExpected();
    }
    
    boost::any distance;
    Token tokenDistance = tokenNext();       // distance
    switch(tokenDistance.tok) {
    case Token::integer:
    case Token::decimal:
      distance = tokenDistance.value;
      break;
    default:
      return numberOrKeywordExpected();
    }

    if (tokenNext().tok != Token::opOr) {       // ,
      return commaExpected();
    }
    
    if (! clauseOr()) {                        // *expression*
      return false;
    }
    
    if (tokenNext().tok != Token::rightparen) { // )T
      return rightParenthesisExpected();
    }
    
    return addTokenToPostfix(Token(Token::within, distance));
  }

  bool SelectionCompiler::clauseChemObjName() {
    Token token = tokenNext();
    if (token.tok == Token::identifier && token.value.type() == typeid(std::string)) {

      std::string name = boost::any_cast<std::string>(token.value);
      if (isNameValid(name)) {
        return addTokenToPostfix(Token(Token::name, name));
      } else {
        return compileError("invalid name: " + name);
      }
    } 

    return false;
        
  }

  bool SelectionCompiler::isNameValid(const std::string& name) {
    int nbracket = 0;
    int ndot = 0;
    for (int i =0 ; i < name.size(); ++i) {
      switch(name[i]) {

      case '[' :
        ++nbracket;
        break;
      case ']' :
        --nbracket;
        break;
      case '.' :
        ++ndot;
        break;       
      }
    }

    //only allow 3 dots at most
    return (ndot <=3 && nbracket == 0) ? true : false;
  }

  bool SelectionCompiler::clauseIndex(){
    Token token = tokenNext();
    if (token.tok == Token::integer) {
      int index = boost::any_cast<int>(token.value);
      int tok = tokPeek();
      std::cout << "Token::to is " << Token::to << ", tok = " << tok << std::endl;
      if (tok == Token::to) {
        tokenNext();
        tok = tokPeek();
        if (tok != Token::integer) {
          return numberExpected();
        }
            
        boost::any intVal = tokenNext().value;
        int first = index;
        if (intVal.type() != typeid(int)){
          return false;
        }
        int second = boost::any_cast<int>(intVal);

        return addTokenToPostfix(Token(Token::index, boost::any(std::make_pair(first, second))));
            
      }else {
        return addTokenToPostfix(Token(Token::index, boost::any(index)));
      }
    } else {
      return numberExpected();
    }
  }

}
Revision:	1629
Committed:	Wed Sep 14 21:15:17 2011 UTC (14 years, 1 month ago) by gezelter
File size:	18371 byte(s)
Log Message:	Merging changes from old branch into development branch
#	Content
1	/*
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	* 1. Redistributions of source code must retain the above copyright
10	* notice, this list of conditions and the following disclaimer.
11	*
12	* 2. Redistributions in binary form must reproduce the above copyright
13	* 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	*
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	*/
41
42	#include "selection/SelectionCompiler.hpp"
43	#include "utils/StringUtils.hpp"
44	namespace OpenMD {
45
46	bool SelectionCompiler::compile(const std::string& filename,
47	const std::string& script) {
48
49	this->filename = filename;
50	this->script = script;
51	lineNumbers.clear();
52	lineIndices.clear();
53	aatokenCompiled.clear();
54
55	if (internalCompile()) {
56	return true;
57	}
58
59	int icharEnd;
60	if ((icharEnd = script.find('\r', ichCurrentCommand)) == std::string::npos &&
61	(icharEnd = script.find('\n', ichCurrentCommand)) == std::string::npos) {
62	icharEnd = script.size();
63	}
64	errorLine = script.substr(ichCurrentCommand, icharEnd);
65	return false;
66	}
67
68	bool SelectionCompiler::internalCompile(){
69
70	cchScript = script.size();
71	ichToken = 0;
72	lineCurrent = 1;
73
74	error = false;
75
76	//std::vector<Token> lltoken;
77	aatokenCompiled.clear();
78	std::vector<Token> ltoken;
79
80	Token tokenCommand;
81	int tokCommand = Token::nada;
82
83	for ( ; true; ichToken += cchToken) {
84	if (lookingAtLeadingWhitespace())
85	continue;
86	//if (lookingAtComment())
87	// continue;
88	bool endOfLine = lookingAtEndOfLine();
89	if (endOfLine \|\| lookingAtEndOfStatement()) {
90	if (tokCommand != Token::nada) {
91	if (! compileCommand(ltoken)) {
92	return false;
93	}
94	aatokenCompiled.push_back(atokenCommand);
95	lineNumbers.push_back(lineCurrent);
96	lineIndices.push_back(ichCurrentCommand);
97	ltoken.clear();
98	tokCommand = Token::nada;
99	}
100
101	if (ichToken < cchScript) {
102	if (endOfLine)
103	++lineCurrent;
104	continue;
105	}
106	break;
107	}
108
109	if (tokCommand != Token::nada) {
110	if (lookingAtString()) {
111	std::string str = getUnescapedStringLiteral();
112	ltoken.push_back(Token(Token::string, str));
113	continue;
114	}
115	//if ((tokCommand & Token::specialstring) != 0 &&
116	// lookingAtSpecialString()) {
117	// std::string str = script.substr(ichToken, ichToken + cchToken);
118	// ltoken.push_back(Token(Token::string, str));
119	// continue;
120	//}
121	if (lookingAtDecimal((tokCommand & Token::negnums) != 0)) {
122	float value = lexi_cast<float>(script.substr(ichToken, cchToken));
123	ltoken.push_back(Token(Token::decimal, boost::any(value)));
124	continue;
125	}
126	if (lookingAtInteger((tokCommand & Token::negnums) != 0)) {
127
128	int val = lexi_cast<int>(script.substr(ichToken, cchToken));
129	ltoken.push_back(Token(Token::integer, boost::any(val)));
130	continue;
131	}
132	}
133
134	if (lookingAtLookupToken()) {
135	std::string ident = script.substr(ichToken, cchToken);
136	Token token;
137	Token* pToken = TokenMap::getInstance()->getToken(ident);
138	if (pToken != NULL) {
139	token = *pToken;
140	} else {
141	token = Token(Token::identifier, ident);
142	}
143
144	int tok = token.tok;
145
146	switch (tokCommand) {
147	case Token::nada:
148	ichCurrentCommand = ichToken;
149	//tokenCommand = token;
150	tokCommand = tok;
151	if ((tokCommand & Token::command) == 0)
152	return commandExpected();
153	break;
154
155	case Token::define:
156	if (ltoken.size() == 1) {
157	// we are looking at the variable name
158	if (tok != Token::identifier &&
159	(tok & Token::predefinedset) != Token::predefinedset)
160	return invalidExpressionToken(ident);
161	} else {
162	// we are looking at the expression
163	if (tok != Token::identifier &&
164	(tok & (Token::expression \| Token::predefinedset)) == 0)
165	return invalidExpressionToken(ident);
166	}
167
168	break;
169
170	case Token::select:
171	if (tok != Token::identifier && (tok & Token::expression) == 0)
172	return invalidExpressionToken(ident);
173	break;
174	}
175	ltoken.push_back(token);
176	continue;
177	}
178
179	if (ltoken.size() == 0) {
180	return commandExpected();
181	}
182
183	return unrecognizedToken();
184	}
185
186	return true;
187	}
188
189
190	bool SelectionCompiler::lookingAtLeadingWhitespace() {
191
192	int ichT = ichToken;
193	while (ichT < cchScript && std::isspace(script[ichT])) {
194	++ichT;
195	}
196	cchToken = ichT - ichToken;
197	return cchToken > 0;
198	}
199
200	bool SelectionCompiler::lookingAtEndOfLine() {
201	if (ichToken == cchScript)
202	return true;
203	int ichT = ichToken;
204	char ch = script[ichT];
205	if (ch == '\r') {
206	++ichT;
207	if (ichT < cchScript && script[ichT] == '\n')
208	++ichT;
209	} else if (ch == '\n') {
210	++ichT;
211	} else {
212	return false;
213	}
214	cchToken = ichT - ichToken;
215	return true;
216	}
217
218	bool SelectionCompiler::lookingAtEndOfStatement() {
219	if (ichToken == cchScript \|\| script[ichToken] != ';')
220	return false;
221	cchToken = 1;
222	return true;
223	}
224
225	bool SelectionCompiler::lookingAtString() {
226	if (ichToken == cchScript)
227	return false;
228	if (script[ichToken] != '"')
229	return false;
230	// remove support for single quote
231	// in order to use it in atom expressions
232	// char chFirst = script.charAt(ichToken);
233	// if (chFirst != '"' && chFirst != '\'')
234	// return false;
235	int ichT = ichToken + 1;
236	// while (ichT < cchScript && script.charAt(ichT++) != chFirst)
237	char ch;
238	bool previousCharBackslash = false;
239	while (ichT < cchScript) {
240	ch = script[ichT++];
241	if (ch == '"' && !previousCharBackslash)
242	break;
243	previousCharBackslash = ch == '\\' ? !previousCharBackslash : false;
244	}
245	cchToken = ichT - ichToken;
246
247	return true;
248	}
249
250
251	std::string SelectionCompiler::getUnescapedStringLiteral() {
252	/** @todo */
253	std::string sb(cchToken - 2, ' ');
254
255	int ichMax = ichToken + cchToken - 1;
256	int ich = ichToken + 1;
257
258	while (ich < ichMax) {
259	char ch = script[ich++];
260	if (ch == '\\' && ich < ichMax) {
261	ch = script[ich++];
262	switch (ch) {
263	case 'b':
264	ch = '\b';
265	break;
266	case 'n':
267	ch = '\n';
268	break;
269	case 't':
270	ch = '\t';
271	break;
272	case 'r':
273	ch = '\r';
274	// fall into
275	case '"':
276	case '\\':
277	case '\'':
278	break;
279	case 'x':
280	case 'u':
281	int digitCount = ch == 'x' ? 2 : 4;
282	if (ich < ichMax) {
283	int unicode = 0;
284	for (int k = digitCount; --k >= 0 && ich < ichMax; ) {
285	char chT = script[ich];
286	int hexit = getHexitValue(chT);
287	if (hexit < 0)
288	break;
289	unicode <<= 4;
290	unicode += hexit;
291	++ich;
292	}
293	ch = (char)unicode;
294	}
295	}
296	}
297	sb.append(1, ch);
298	}
299
300	return sb;
301	}
302
303	int SelectionCompiler::getHexitValue(char ch) {
304	if (ch >= '0' && ch <= '9')
305	return ch - '0';
306	else if (ch >= 'a' && ch <= 'f')
307	return 10 + ch - 'a';
308	else if (ch >= 'A' && ch <= 'F')
309	return 10 + ch - 'A';
310	else
311	return -1;
312	}
313
314	bool SelectionCompiler::lookingAtSpecialString() {
315	int ichT = ichToken;
316	char ch = script[ichT];
317	while (ichT < cchScript && ch != ';' && ch != '\r' && ch != '\n') {
318	++ichT;
319	}
320	cchToken = ichT - ichToken;
321	return cchToken > 0;
322	}
323
324	bool SelectionCompiler::lookingAtDecimal(bool allowNegative) {
325	if (ichToken == cchScript) {
326	return false;
327	}
328
329	int ichT = ichToken;
330	if (script[ichT] == '-') {
331	++ichT;
332	}
333	bool digitSeen = false;
334	char ch = 'X';
335	while (ichT < cchScript && std::isdigit(ch = script[ichT])) {
336	++ichT;
337	digitSeen = true;
338	}
339
340	if (ichT == cchScript \|\| ch != '.') {
341	return false;
342	}
343
344	// to support DMPC.1, let's check the character before the dot
345	if (ch == '.' && (ichT > 0) && std::isalpha(script[ichT - 1])) {
346	return false;
347	}
348
349	++ichT;
350	while (ichT < cchScript && std::isdigit(script[ichT])) {
351	++ichT;
352	digitSeen = true;
353	}
354	cchToken = ichT - ichToken;
355	return digitSeen;
356	}
357
358	bool SelectionCompiler::lookingAtInteger(bool allowNegative) {
359	if (ichToken == cchScript) {
360	return false;
361	}
362	int ichT = ichToken;
363	if (allowNegative && script[ichToken] == '-') {
364	++ichT;
365	}
366	int ichBeginDigits = ichT;
367	while (ichT < cchScript && std::isdigit(script[ichT])) {
368	++ichT;
369	}
370	if (ichBeginDigits == ichT) {
371	return false;
372	}
373	cchToken = ichT - ichToken;
374	return true;
375	}
376
377	bool SelectionCompiler::lookingAtLookupToken() {
378	if (ichToken == cchScript) {
379	return false;
380	}
381
382	int ichT = ichToken;
383	char ch;
384	switch (ch = script[ichT++]) {
385	case '(':
386	case ')':
387	case ',':
388	case '[':
389	case ']':
390	break;
391	case '&':
392	case '\|':
393	if (ichT < cchScript && script[ichT] == ch) {
394	++ichT;
395	}
396	break;
397	case '<':
398	case '=':
399	case '>':
400	if (ichT < cchScript && ((ch = script[ichT]) == '<' \|\| ch == '=' \|\| ch == '>')) {
401	++ichT;
402	}
403	break;
404	case '/':
405	case '!':
406	if (ichT < cchScript && script[ichT] == '=') {
407	++ichT;
408	}
409	break;
410	default:
411	if ((ch < 'a' \|\| ch > 'z') && (ch < 'A' && ch > 'Z') && ch != '_') {
412	return false;
413	}
414	case '*':
415	case '?': // include question marks in identifier for atom expressions
416	while (ichT < cchScript && !std::isspace(ch = script[ichT]) &&
417	(std::isalpha(ch) \|\|std::isdigit(ch) \|\| ch == '_' \|\| ch == '.' \|\| ch == '*' \|\| ch == '?' \|\| ch == '+' \|\| ch == '-' \|\| ch == '[' \|\| ch == ']') ){
418
419	++ichT;
420	}
421	break;
422	}
423
424	cchToken = ichT - ichToken;
425
426	return true;
427	}
428
429	bool SelectionCompiler::compileCommand(const std::vector<Token>& ltoken) {
430	const Token& tokenCommand = ltoken[0];
431	int tokCommand = tokenCommand.tok;
432
433	atokenCommand = ltoken;
434	if ((tokCommand & Token::expressionCommand) != 0 && !compileExpression()) {
435	return false;
436	}
437
438	return true;
439	}
440
441	bool SelectionCompiler::compileExpression() {
442	/** todo */
443	int i = 1;
444	int tokCommand = atokenCommand[0].tok;
445	if (tokCommand == Token::define) {
446	i = 2;
447	} else if ((tokCommand & Token::embeddedExpression) != 0) {
448	// look for the open parenthesis
449	while (i < atokenCommand.size() &&
450	atokenCommand[i].tok != Token::leftparen)
451	++i;
452	}
453
454	if (i >= atokenCommand.size()) {
455	return true;
456	}
457	return compileExpression(i);
458	}
459
460
461	bool SelectionCompiler::addTokenToPostfix(const Token& token) {
462	ltokenPostfix.push_back(token);
463	return true;
464	}
465
466	bool SelectionCompiler::compileExpression(int itoken) {
467	ltokenPostfix.clear();
468	for (int i = 0; i < itoken; ++i) {
469	addTokenToPostfix(atokenCommand[i]);
470	}
471
472	atokenInfix = atokenCommand;
473	itokenInfix = itoken;
474
475	addTokenToPostfix(Token::tokenExpressionBegin);
476	if (!clauseOr()) {
477	return false;
478	}
479
480	addTokenToPostfix(Token::tokenExpressionEnd);
481	if (itokenInfix != atokenInfix.size()) {
482	return endOfExpressionExpected();
483	}
484
485	atokenCommand = ltokenPostfix;
486	return true;
487	}
488
489	Token SelectionCompiler::tokenNext() {
490	if (itokenInfix == atokenInfix.size()) {
491	return Token();
492	}
493	return atokenInfix[itokenInfix++];
494	}
495
496	boost::any SelectionCompiler::valuePeek() {
497	if (itokenInfix == atokenInfix.size()) {
498	return boost::any();
499	} else {
500	return atokenInfix[itokenInfix].value;
501	}
502	}
503
504	int SelectionCompiler::tokPeek() {
505	if (itokenInfix == atokenInfix.size()) {
506	return 0;
507	}else {
508	return atokenInfix[itokenInfix].tok;
509	}
510	}
511
512	bool SelectionCompiler::clauseOr() {
513	if (!clauseAnd()) {
514	return false;
515	}
516
517	while (tokPeek() == Token::opOr) {
518	Token tokenOr = tokenNext();
519	if (!clauseAnd()) {
520	return false;
521	}
522	addTokenToPostfix(tokenOr);
523	}
524	return true;
525	}
526
527	bool SelectionCompiler::clauseAnd() {
528	if (!clauseNot()) {
529	return false;
530	}
531
532	while (tokPeek() == Token::opAnd) {
533	Token tokenAnd = tokenNext();
534	if (!clauseNot()) {
535	return false;
536	}
537	addTokenToPostfix(tokenAnd);
538	}
539	return true;
540	}
541
542	bool SelectionCompiler::clauseNot() {
543	if (tokPeek() == Token::opNot) {
544	Token tokenNot = tokenNext();
545	if (!clauseNot()) {
546	return false;
547	}
548	return addTokenToPostfix(tokenNot);
549	}
550	return clausePrimitive();
551	}
552
553	bool SelectionCompiler::clausePrimitive() {
554	int tok = tokPeek();
555	switch (tok) {
556	case Token::within:
557	return clauseWithin();
558
559	case Token::asterisk:
560	case Token::identifier:
561	return clauseChemObjName();
562
563	case Token::integer :
564	return clauseIndex();
565	default:
566	if ((tok & Token::atomproperty) == Token::atomproperty) {
567	return clauseComparator();
568	}
569	if ((tok & Token::predefinedset) != Token::predefinedset) {
570	break;
571	}
572	// fall into the code and below and just add the token
573	case Token::all:
574	case Token::none:
575	case Token::hull:
576	return addTokenToPostfix(tokenNext());
577	case Token::leftparen:
578	tokenNext();
579	if (!clauseOr()) {
580	return false;
581	}
582	if (tokenNext().tok != Token::rightparen) {
583	return rightParenthesisExpected();
584	}
585	return true;
586	}
587	return unrecognizedExpressionToken();
588	}
589
590	bool SelectionCompiler::clauseComparator() {
591	Token tokenAtomProperty = tokenNext();
592	Token tokenComparator = tokenNext();
593	if ((tokenComparator.tok & Token::comparator) == 0) {
594	return comparisonOperatorExpected();
595	}
596
597	Token tokenValue = tokenNext();
598	if (tokenValue.tok != Token::integer && tokenValue.tok != Token::decimal) {
599	return numberExpected();
600	}
601
602	float val;
603	if (tokenValue.value.type() == typeid(int)) {
604	val = boost::any_cast<int>(tokenValue.value);
605	} else if (tokenValue.value.type() == typeid(float)) {
606	val = boost::any_cast<float>(tokenValue.value);
607	} else {
608	return false;
609	}
610
611	boost::any floatVal;
612	floatVal = val;
613	return addTokenToPostfix(Token(tokenComparator.tok,
614	tokenAtomProperty.tok, floatVal));
615	}
616
617	bool SelectionCompiler::clauseWithin() {
618	tokenNext(); // WITHIN
619	if (tokenNext().tok != Token::leftparen) { // (
620	return leftParenthesisExpected();
621	}
622
623	boost::any distance;
624	Token tokenDistance = tokenNext(); // distance
625	switch(tokenDistance.tok) {
626	case Token::integer:
627	case Token::decimal:
628	distance = tokenDistance.value;
629	break;
630	default:
631	return numberOrKeywordExpected();
632	}
633
634	if (tokenNext().tok != Token::opOr) { // ,
635	return commaExpected();
636	}
637
638	if (! clauseOr()) { // expression
639	return false;
640	}
641
642	if (tokenNext().tok != Token::rightparen) { // )T
643	return rightParenthesisExpected();
644	}
645
646	return addTokenToPostfix(Token(Token::within, distance));
647	}
648
649	bool SelectionCompiler::clauseChemObjName() {
650	Token token = tokenNext();
651	if (token.tok == Token::identifier && token.value.type() == typeid(std::string)) {
652
653	std::string name = boost::any_cast<std::string>(token.value);
654	if (isNameValid(name)) {
655	return addTokenToPostfix(Token(Token::name, name));
656	} else {
657	return compileError("invalid name: " + name);
658	}
659	}
660
661	return false;
662
663	}
664
665	bool SelectionCompiler::isNameValid(const std::string& name) {
666	int nbracket = 0;
667	int ndot = 0;
668	for (int i =0 ; i < name.size(); ++i) {
669	switch(name[i]) {
670
671	case '[' :
672	++nbracket;
673	break;
674	case ']' :
675	--nbracket;
676	break;
677	case '.' :
678	++ndot;
679	break;
680	}
681	}
682
683	//only allow 3 dots at most
684	return (ndot <=3 && nbracket == 0) ? true : false;
685	}
686
687	bool SelectionCompiler::clauseIndex(){
688	Token token = tokenNext();
689	if (token.tok == Token::integer) {
690	int index = boost::any_cast<int>(token.value);
691	int tok = tokPeek();
692	std::cout << "Token::to is " << Token::to << ", tok = " << tok << std::endl;
693	if (tok == Token::to) {
694	tokenNext();
695	tok = tokPeek();
696	if (tok != Token::integer) {
697	return numberExpected();
698	}
699
700	boost::any intVal = tokenNext().value;
701	int first = index;
702	if (intVal.type() != typeid(int)){
703	return false;
704	}
705	int second = boost::any_cast<int>(intVal);
706
707	return addTokenToPostfix(Token(Token::index, boost::any(std::make_pair(first, second))));
708
709	}else {
710	return addTokenToPostfix(Token(Token::index, boost::any(index)));
711	}
712	} else {
713	return numberExpected();
714	}
715	}
716
717	}