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]  Kuang & Gezelter,  J. Chem. Phys. 133, 164101 (2010).
 * [5]  Vardeman, Stocker & Gezelter, J. Chem. Theory Comput. 7, 834 (2011).
 */

#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 */
    unsigned 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 (unsigned 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:	1782
Committed:	Wed Aug 22 02:28:28 2012 UTC (12 years, 8 months ago) by gezelter
File size:	18456 byte(s)
Log Message:	MERGE OpenMD development branch 1465:1781 into trunk
#	User	Rev	Content
1	tim	279	/*
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	279	* 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	279	* 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	1782	* [4] Kuang & Gezelter, J. Chem. Phys. 133, 164101 (2010).
40			* [5] Vardeman, Stocker & Gezelter, J. Chem. Theory Comput. 7, 834 (2011).
41	tim	279	*/
42
43			#include "selection/SelectionCompiler.hpp"
44	tim	281	#include "utils/StringUtils.hpp"
45	gezelter	1390	namespace OpenMD {
46	tim	279
47	cli2	1364	bool SelectionCompiler::compile(const std::string& filename,
48			const std::string& script) {
49	tim	279
50			this->filename = filename;
51			this->script = script;
52			lineNumbers.clear();
53			lineIndices.clear();
54			aatokenCompiled.clear();
55
56	tim	281	if (internalCompile()) {
57	gezelter	507	return true;
58	tim	279	}
59
60			int icharEnd;
61	tim	281	if ((icharEnd = script.find('\r', ichCurrentCommand)) == std::string::npos &&
62			(icharEnd = script.find('\n', ichCurrentCommand)) == std::string::npos) {
63	gezelter	507	icharEnd = script.size();
64	tim	279	}
65			errorLine = script.substr(ichCurrentCommand, icharEnd);
66			return false;
67	gezelter	507	}
68	tim	279
69	gezelter	507	bool SelectionCompiler::internalCompile(){
70	tim	279
71			cchScript = script.size();
72			ichToken = 0;
73			lineCurrent = 1;
74
75			error = false;
76
77	tim	281	//std::vector<Token> lltoken;
78			aatokenCompiled.clear();
79	tim	279	std::vector<Token> ltoken;
80
81	tim	281	Token tokenCommand;
82			int tokCommand = Token::nada;
83	tim	279
84			for ( ; true; ichToken += cchToken) {
85	gezelter	507	if (lookingAtLeadingWhitespace())
86			continue;
87			//if (lookingAtComment())
88			// continue;
89			bool endOfLine = lookingAtEndOfLine();
90			if (endOfLine \|\| lookingAtEndOfStatement()) {
91			if (tokCommand != Token::nada) {
92			if (! compileCommand(ltoken)) {
93			return false;
94			}
95			aatokenCompiled.push_back(atokenCommand);
96			lineNumbers.push_back(lineCurrent);
97			lineIndices.push_back(ichCurrentCommand);
98			ltoken.clear();
99			tokCommand = Token::nada;
100			}
101	tim	279
102	gezelter	507	if (ichToken < cchScript) {
103			if (endOfLine)
104			++lineCurrent;
105			continue;
106			}
107			break;
108			}
109	tim	279
110	gezelter	507	if (tokCommand != Token::nada) {
111			if (lookingAtString()) {
112			std::string str = getUnescapedStringLiteral();
113			ltoken.push_back(Token(Token::string, str));
114			continue;
115			}
116			//if ((tokCommand & Token::specialstring) != 0 &&
117			// lookingAtSpecialString()) {
118			// std::string str = script.substr(ichToken, ichToken + cchToken);
119			// ltoken.push_back(Token(Token::string, str));
120			// continue;
121			//}
122			if (lookingAtDecimal((tokCommand & Token::negnums) != 0)) {
123			float value = lexi_cast<float>(script.substr(ichToken, cchToken));
124			ltoken.push_back(Token(Token::decimal, boost::any(value)));
125			continue;
126			}
127			if (lookingAtInteger((tokCommand & Token::negnums) != 0)) {
128	tim	295
129	gezelter	507	int val = lexi_cast<int>(script.substr(ichToken, cchToken));
130			ltoken.push_back(Token(Token::integer, boost::any(val)));
131			continue;
132			}
133			}
134	tim	279
135	gezelter	507	if (lookingAtLookupToken()) {
136			std::string ident = script.substr(ichToken, cchToken);
137			Token token;
138			Token* pToken = TokenMap::getInstance()->getToken(ident);
139			if (pToken != NULL) {
140			token = *pToken;
141			} else {
142			token = Token(Token::identifier, ident);
143			}
144	tim	279
145	gezelter	507	int tok = token.tok;
146	tim	279
147	gezelter	507	switch (tokCommand) {
148			case Token::nada:
149			ichCurrentCommand = ichToken;
150			//tokenCommand = token;
151			tokCommand = tok;
152			if ((tokCommand & Token::command) == 0)
153			return commandExpected();
154			break;
155	tim	279
156	gezelter	507	case Token::define:
157			if (ltoken.size() == 1) {
158			// we are looking at the variable name
159			if (tok != Token::identifier &&
160			(tok & Token::predefinedset) != Token::predefinedset)
161			return invalidExpressionToken(ident);
162			} else {
163			// we are looking at the expression
164			if (tok != Token::identifier &&
165			(tok & (Token::expression \| Token::predefinedset)) == 0)
166			return invalidExpressionToken(ident);
167			}
168	tim	279
169	gezelter	507	break;
170	tim	279
171	gezelter	507	case Token::select:
172			if (tok != Token::identifier && (tok & Token::expression) == 0)
173			return invalidExpressionToken(ident);
174			break;
175			}
176			ltoken.push_back(token);
177			continue;
178			}
179	tim	279
180	gezelter	507	if (ltoken.size() == 0) {
181			return commandExpected();
182			}
183	tim	279
184	gezelter	507	return unrecognizedToken();
185	tim	279	}
186
187			return true;
188			}
189
190
191			bool SelectionCompiler::lookingAtLeadingWhitespace() {
192
193			int ichT = ichToken;
194			while (ichT < cchScript && std::isspace(script[ichT])) {
195			++ichT;
196			}
197			cchToken = ichT - ichToken;
198			return cchToken > 0;
199			}
200
201			bool SelectionCompiler::lookingAtEndOfLine() {
202			if (ichToken == cchScript)
203			return true;
204			int ichT = ichToken;
205			char ch = script[ichT];
206			if (ch == '\r') {
207			++ichT;
208			if (ichT < cchScript && script[ichT] == '\n')
209	gezelter	507	++ichT;
210	tim	279	} else if (ch == '\n') {
211			++ichT;
212			} else {
213			return false;
214			}
215			cchToken = ichT - ichToken;
216			return true;
217			}
218
219			bool SelectionCompiler::lookingAtEndOfStatement() {
220			if (ichToken == cchScript \|\| script[ichToken] != ';')
221			return false;
222			cchToken = 1;
223			return true;
224			}
225
226			bool SelectionCompiler::lookingAtString() {
227			if (ichToken == cchScript)
228			return false;
229			if (script[ichToken] != '"')
230			return false;
231			// remove support for single quote
232			// in order to use it in atom expressions
233			// char chFirst = script.charAt(ichToken);
234			// if (chFirst != '"' && chFirst != '\'')
235			// return false;
236			int ichT = ichToken + 1;
237			// while (ichT < cchScript && script.charAt(ichT++) != chFirst)
238			char ch;
239	tim	281	bool previousCharBackslash = false;
240	tim	279	while (ichT < cchScript) {
241	tim	281	ch = script[ichT++];
242	tim	279	if (ch == '"' && !previousCharBackslash)
243			break;
244			previousCharBackslash = ch == '\\' ? !previousCharBackslash : false;
245			}
246			cchToken = ichT - ichToken;
247	tim	295
248	tim	279	return true;
249			}
250
251
252	gezelter	507	std::string SelectionCompiler::getUnescapedStringLiteral() {
253	tim	281	/** @todo */
254			std::string sb(cchToken - 2, ' ');
255
256	tim	279	int ichMax = ichToken + cchToken - 1;
257			int ich = ichToken + 1;
258
259			while (ich < ichMax) {
260	gezelter	507	char ch = script[ich++];
261			if (ch == '\\' && ich < ichMax) {
262			ch = script[ich++];
263			switch (ch) {
264			case 'b':
265			ch = '\b';
266			break;
267			case 'n':
268			ch = '\n';
269			break;
270			case 't':
271			ch = '\t';
272			break;
273			case 'r':
274			ch = '\r';
275			// fall into
276			case '"':
277			case '\\':
278			case '\'':
279			break;
280			case 'x':
281			case 'u':
282			int digitCount = ch == 'x' ? 2 : 4;
283			if (ich < ichMax) {
284			int unicode = 0;
285			for (int k = digitCount; --k >= 0 && ich < ichMax; ) {
286			char chT = script[ich];
287			int hexit = getHexitValue(chT);
288			if (hexit < 0)
289			break;
290			unicode <<= 4;
291			unicode += hexit;
292			++ich;
293			}
294			ch = (char)unicode;
295			}
296			}
297			}
298			sb.append(1, ch);
299	tim	279	}
300
301	tim	281	return sb;
302	gezelter	507	}
303	tim	279
304	gezelter	507	int SelectionCompiler::getHexitValue(char ch) {
305	tim	279	if (ch >= '0' && ch <= '9')
306	gezelter	507	return ch - '0';
307	tim	279	else if (ch >= 'a' && ch <= 'f')
308	gezelter	507	return 10 + ch - 'a';
309	tim	279	else if (ch >= 'A' && ch <= 'F')
310	gezelter	507	return 10 + ch - 'A';
311	tim	279	else
312	gezelter	507	return -1;
313			}
314	tim	279
315	gezelter	507	bool SelectionCompiler::lookingAtSpecialString() {
316	tim	279	int ichT = ichToken;
317			char ch = script[ichT];
318			while (ichT < cchScript && ch != ';' && ch != '\r' && ch != '\n') {
319	gezelter	507	++ichT;
320	tim	279	}
321			cchToken = ichT - ichToken;
322			return cchToken > 0;
323	gezelter	507	}
324	tim	279
325	gezelter	507	bool SelectionCompiler::lookingAtDecimal(bool allowNegative) {
326	tim	279	if (ichToken == cchScript) {
327	gezelter	507	return false;
328	tim	279	}
329
330			int ichT = ichToken;
331			if (script[ichT] == '-') {
332	gezelter	507	++ichT;
333	tim	279	}
334	tim	281	bool digitSeen = false;
335	tim	279	char ch = 'X';
336			while (ichT < cchScript && std::isdigit(ch = script[ichT])) {
337	gezelter	507	++ichT;
338			digitSeen = true;
339	tim	279	}
340
341			if (ichT == cchScript \|\| ch != '.') {
342	gezelter	507	return false;
343	tim	279	}
344
345	tim	303	// to support DMPC.1, let's check the character before the dot
346			if (ch == '.' && (ichT > 0) && std::isalpha(script[ichT - 1])) {
347	gezelter	507	return false;
348	tim	279	}
349
350			++ichT;
351			while (ichT < cchScript && std::isdigit(script[ichT])) {
352	gezelter	507	++ichT;
353			digitSeen = true;
354	tim	279	}
355			cchToken = ichT - ichToken;
356			return digitSeen;
357	gezelter	507	}
358	tim	279
359	gezelter	507	bool SelectionCompiler::lookingAtInteger(bool allowNegative) {
360	tim	279	if (ichToken == cchScript) {
361	gezelter	507	return false;
362	tim	279	}
363			int ichT = ichToken;
364			if (allowNegative && script[ichToken] == '-') {
365	gezelter	507	++ichT;
366	tim	279	}
367			int ichBeginDigits = ichT;
368			while (ichT < cchScript && std::isdigit(script[ichT])) {
369	gezelter	507	++ichT;
370	tim	279	}
371			if (ichBeginDigits == ichT) {
372	gezelter	507	return false;
373	tim	279	}
374			cchToken = ichT - ichToken;
375			return true;
376	gezelter	507	}
377	tim	279
378	gezelter	507	bool SelectionCompiler::lookingAtLookupToken() {
379	tim	279	if (ichToken == cchScript) {
380	gezelter	507	return false;
381	tim	279	}
382
383			int ichT = ichToken;
384			char ch;
385			switch (ch = script[ichT++]) {
386	gezelter	507	case '(':
387			case ')':
388			case ',':
389			case '[':
390			case ']':
391			break;
392			case '&':
393			case '\|':
394			if (ichT < cchScript && script[ichT] == ch) {
395			++ichT;
396			}
397			break;
398			case '<':
399			case '=':
400			case '>':
401			if (ichT < cchScript && ((ch = script[ichT]) == '<' \|\| ch == '=' \|\| ch == '>')) {
402			++ichT;
403			}
404			break;
405			case '/':
406			case '!':
407			if (ichT < cchScript && script[ichT] == '=') {
408			++ichT;
409			}
410			break;
411			default:
412			if ((ch < 'a' \|\| ch > 'z') && (ch < 'A' && ch > 'Z') && ch != '_') {
413			return false;
414			}
415			case '*':
416			case '?': // include question marks in identifier for atom expressions
417			while (ichT < cchScript && !std::isspace(ch = script[ichT]) &&
418			(std::isalpha(ch) \|\|std::isdigit(ch) \|\| ch == '_' \|\| ch == '.' \|\| ch == '*' \|\| ch == '?' \|\| ch == '+' \|\| ch == '-' \|\| ch == '[' \|\| ch == ']') ){
419	tim	288
420	gezelter	507	++ichT;
421			}
422			break;
423	tim	279	}
424	tim	295
425	tim	279	cchToken = ichT - ichToken;
426	tim	295
427	tim	279	return true;
428	gezelter	507	}
429	tim	279
430	gezelter	507	bool SelectionCompiler::compileCommand(const std::vector<Token>& ltoken) {
431	tim	281	const Token& tokenCommand = ltoken[0];
432	tim	279	int tokCommand = tokenCommand.tok;
433	tim	281
434			atokenCommand = ltoken;
435			if ((tokCommand & Token::expressionCommand) != 0 && !compileExpression()) {
436	gezelter	507	return false;
437	tim	279	}
438	tim	281
439	tim	279	return true;
440	gezelter	507	}
441	tim	279
442	gezelter	507	bool SelectionCompiler::compileExpression() {
443	tim	279	/** todo */
444	gezelter	1782	unsigned int i = 1;
445	tim	279	int tokCommand = atokenCommand[0].tok;
446	tim	281	if (tokCommand == Token::define) {
447	gezelter	507	i = 2;
448	tim	281	} else if ((tokCommand & Token::embeddedExpression) != 0) {
449	gezelter	507	// look for the open parenthesis
450			while (i < atokenCommand.size() &&
451			atokenCommand[i].tok != Token::leftparen)
452	tim	279	++i;
453			}
454	tim	281
455			if (i >= atokenCommand.size()) {
456	gezelter	507	return true;
457	tim	281	}
458	tim	279	return compileExpression(i);
459			}
460
461
462	gezelter	507	bool SelectionCompiler::addTokenToPostfix(const Token& token) {
463	tim	279	ltokenPostfix.push_back(token);
464			return true;
465	gezelter	507	}
466	tim	279
467	gezelter	507	bool SelectionCompiler::compileExpression(int itoken) {
468	tim	281	ltokenPostfix.clear();
469			for (int i = 0; i < itoken; ++i) {
470	gezelter	507	addTokenToPostfix(atokenCommand[i]);
471	tim	281	}
472
473	tim	279	atokenInfix = atokenCommand;
474			itokenInfix = itoken;
475
476	tim	281	addTokenToPostfix(Token::tokenExpressionBegin);
477	tim	279	if (!clauseOr()) {
478	gezelter	507	return false;
479	tim	279	}
480
481	tim	281	addTokenToPostfix(Token::tokenExpressionEnd);
482			if (itokenInfix != atokenInfix.size()) {
483	gezelter	507	return endOfExpressionExpected();
484	tim	279	}
485
486			atokenCommand = ltokenPostfix;
487			return true;
488	gezelter	507	}
489	tim	279
490	gezelter	507	Token SelectionCompiler::tokenNext() {
491	tim	281	if (itokenInfix == atokenInfix.size()) {
492	gezelter	507	return Token();
493	tim	281	}
494			return atokenInfix[itokenInfix++];
495	gezelter	507	}
496	tim	279
497	gezelter	507	boost::any SelectionCompiler::valuePeek() {
498	tim	281	if (itokenInfix == atokenInfix.size()) {
499	gezelter	507	return boost::any();
500	tim	279	} else {
501	gezelter	507	return atokenInfix[itokenInfix].value;
502	tim	279	}
503	gezelter	507	}
504	tim	279
505	gezelter	507	int SelectionCompiler::tokPeek() {
506	tim	281	if (itokenInfix == atokenInfix.size()) {
507	gezelter	507	return 0;
508	tim	279	}else {
509	gezelter	507	return atokenInfix[itokenInfix].tok;
510	tim	279	}
511	gezelter	507	}
512	tim	279
513	gezelter	507	bool SelectionCompiler::clauseOr() {
514	tim	279	if (!clauseAnd()) {
515	gezelter	507	return false;
516	tim	279	}
517
518	tim	281	while (tokPeek() == Token::opOr) {
519	gezelter	507	Token tokenOr = tokenNext();
520			if (!clauseAnd()) {
521			return false;
522			}
523			addTokenToPostfix(tokenOr);
524	tim	279	}
525			return true;
526	gezelter	507	}
527	tim	279
528	gezelter	507	bool SelectionCompiler::clauseAnd() {
529	tim	279	if (!clauseNot()) {
530	gezelter	507	return false;
531	tim	279	}
532
533	tim	281	while (tokPeek() == Token::opAnd) {
534	gezelter	507	Token tokenAnd = tokenNext();
535			if (!clauseNot()) {
536			return false;
537			}
538			addTokenToPostfix(tokenAnd);
539	tim	279	}
540			return true;
541	gezelter	507	}
542	tim	279
543	gezelter	507	bool SelectionCompiler::clauseNot() {
544	tim	281	if (tokPeek() == Token::opNot) {
545	gezelter	507	Token tokenNot = tokenNext();
546			if (!clauseNot()) {
547			return false;
548			}
549			return addTokenToPostfix(tokenNot);
550	tim	279	}
551			return clausePrimitive();
552	gezelter	507	}
553	tim	279
554	gezelter	507	bool SelectionCompiler::clausePrimitive() {
555	tim	279	int tok = tokPeek();
556			switch (tok) {
557	gezelter	507	case Token::within:
558			return clauseWithin();
559	tim	283
560	gezelter	507	case Token::asterisk:
561			case Token::identifier:
562			return clauseChemObjName();
563	tim	295
564	gezelter	507	case Token::integer :
565			return clauseIndex();
566			default:
567			if ((tok & Token::atomproperty) == Token::atomproperty) {
568			return clauseComparator();
569			}
570			if ((tok & Token::predefinedset) != Token::predefinedset) {
571			break;
572			}
573			// fall into the code and below and just add the token
574			case Token::all:
575			case Token::none:
576	kstocke1	1523	case Token::hull:
577	gezelter	507	return addTokenToPostfix(tokenNext());
578			case Token::leftparen:
579			tokenNext();
580			if (!clauseOr()) {
581			return false;
582			}
583			if (tokenNext().tok != Token::rightparen) {
584			return rightParenthesisExpected();
585			}
586			return true;
587	tim	279	}
588			return unrecognizedExpressionToken();
589	gezelter	507	}
590	tim	279
591	gezelter	507	bool SelectionCompiler::clauseComparator() {
592	tim	279	Token tokenAtomProperty = tokenNext();
593			Token tokenComparator = tokenNext();
594	tim	281	if ((tokenComparator.tok & Token::comparator) == 0) {
595	gezelter	507	return comparisonOperatorExpected();
596	tim	279	}
597
598			Token tokenValue = tokenNext();
599	tim	288	if (tokenValue.tok != Token::integer && tokenValue.tok != Token::decimal) {
600	gezelter	507	return numberExpected();
601	tim	279	}
602	tim	288
603			float val;
604			if (tokenValue.value.type() == typeid(int)) {
605	gezelter	507	val = boost::any_cast<int>(tokenValue.value);
606	tim	288	} else if (tokenValue.value.type() == typeid(float)) {
607	gezelter	507	val = boost::any_cast<float>(tokenValue.value);
608	tim	288	} else {
609	gezelter	507	return false;
610	tim	288	}
611
612	tim	295	boost::any floatVal;
613			floatVal = val;
614	tim	281	return addTokenToPostfix(Token(tokenComparator.tok,
615	gezelter	507	tokenAtomProperty.tok, floatVal));
616			}
617	tim	279
618	gezelter	507	bool SelectionCompiler::clauseWithin() {
619	tim	279	tokenNext(); // WITHIN
620	tim	281	if (tokenNext().tok != Token::leftparen) { // (
621	gezelter	507	return leftParenthesisExpected();
622	tim	279	}
623
624	tim	281	boost::any distance;
625	tim	279	Token tokenDistance = tokenNext(); // distance
626			switch(tokenDistance.tok) {
627	gezelter	507	case Token::integer:
628			case Token::decimal:
629			distance = tokenDistance.value;
630			break;
631			default:
632			return numberOrKeywordExpected();
633	tim	279	}
634
635	tim	281	if (tokenNext().tok != Token::opOr) { // ,
636	gezelter	507	return commaExpected();
637	tim	279	}
638
639			if (! clauseOr()) { // expression
640	gezelter	507	return false;
641	tim	279	}
642
643	tim	281	if (tokenNext().tok != Token::rightparen) { // )T
644	gezelter	507	return rightParenthesisExpected();
645	tim	279	}
646
647	tim	281	return addTokenToPostfix(Token(Token::within, distance));
648	gezelter	507	}
649	tim	279
650	gezelter	507	bool SelectionCompiler::clauseChemObjName() {
651	tim	452	Token token = tokenNext();
652			if (token.tok == Token::identifier && token.value.type() == typeid(std::string)) {
653	tim	279
654	gezelter	507	std::string name = boost::any_cast<std::string>(token.value);
655			if (isNameValid(name)) {
656			return addTokenToPostfix(Token(Token::name, name));
657			} else {
658			return compileError("invalid name: " + name);
659			}
660	tim	452	}
661	tim	288
662	tim	452	return false;
663
664	gezelter	507	}
665	tim	279
666	gezelter	507	bool SelectionCompiler::isNameValid(const std::string& name) {
667			int nbracket = 0;
668	tim	452	int ndot = 0;
669	gezelter	1782	for (unsigned int i = 0 ; i < name.size(); ++i) {
670	gezelter	507	switch(name[i]) {
671	tim	283
672	gezelter	507	case '[' :
673			++nbracket;
674			break;
675			case ']' :
676			--nbracket;
677			break;
678			case '.' :
679			++ndot;
680			break;
681			}
682	tim	283	}
683
684	tim	452	//only allow 3 dots at most
685			return (ndot <=3 && nbracket == 0) ? true : false;
686	gezelter	507	}
687	tim	279
688	gezelter	507	bool SelectionCompiler::clauseIndex(){
689	tim	295	Token token = tokenNext();
690			if (token.tok == Token::integer) {
691	gezelter	507	int index = boost::any_cast<int>(token.value);
692			int tok = tokPeek();
693			std::cout << "Token::to is " << Token::to << ", tok = " << tok << std::endl;
694			if (tok == Token::to) {
695			tokenNext();
696			tok = tokPeek();
697			if (tok != Token::integer) {
698			return numberExpected();
699			}
700	tim	295
701	gezelter	507	boost::any intVal = tokenNext().value;
702			int first = index;
703			if (intVal.type() != typeid(int)){
704			return false;
705			}
706			int second = boost::any_cast<int>(intVal);
707	tim	283
708	gezelter	507	return addTokenToPostfix(Token(Token::index, boost::any(std::make_pair(first, second))));
709	tim	295
710	gezelter	507	}else {
711			return addTokenToPostfix(Token(Token::index, boost::any(index)));
712			}
713	tim	295	} else {
714	gezelter	507	return numberExpected();
715	tim	295	}
716	gezelter	507	}
717	tim	295
718			}