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. Acknowledgement of the program authors must be made in any
 *    publication of scientific results based in part on use of the
 *    program.  An acceptable form of acknowledgement is citation of
 *    the article in which the program was described (Matthew
 *    A. Meineke, Charles F. Vardeman II, Teng Lin, Christopher
 *    J. Fennell and J. Daniel Gezelter, "OOPSE: An Object-Oriented
 *    Parallel Simulation Engine for Molecular Dynamics,"
 *    J. Comput. Chem. 26, pp. 252-271 (2005))
 *
 * 2. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 *
 * 3. 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.
 */

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

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));        
                std::cout << "encount an decimal: " << value << std::endl;
                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));
                std::cout << "encount an integer: " << val << std::endl;
                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;


    std::cout << "lookingAtString: encount " << script.substr(ichToken, cchToken) << std::endl; 
    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 '-':
        case '[':
        case ']':
        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 '?': // include question marks in identifier for atom expressions
            while (ichT < cchScript && !std::isspace(ch = script[ichT]) && (std::isalpha(ch) ||std::isdigit(ch) ||
                ch == '_' || ch == '?') ) {

                ++ichT;
            }
        break;
    }

    cchToken = ichT - ichToken;

    std::cout << "lookingAtLookupToken: encount " << script.substr(ichToken, cchToken) << std::endl; 
    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:
            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() {
    std::string chemObjName;
    int tok = tokPeek();
    if (!clauseName(chemObjName)){
        return false;
    }


    tok = tokPeek();
    //allow two dot at most
    if (tok == Token::dot) {
        tokenNext();
        chemObjName += ".";
        if (!clauseName(chemObjName)) {
            return false;
        }
        tok = tokPeek();
        if (tok == Token::dot) {
            tokenNext();
            chemObjName += ".";

            if (!clauseName(chemObjName)) {
                return false;
            }
        }        
    }

    return addTokenToPostfix(Token(Token::name, chemObjName));
}

bool SelectionCompiler:: clauseName(std::string& name) {

    int tok = tokPeek();

    if (tok == Token::asterisk || tok == Token::identifier || tok == Token::integer) {

        Token token = tokenNext();
        if (token.value.type() == typeid(std::string)) {
            name += boost::any_cast<std::string>(token.value);
        } else if (token.value.type() == typeid(int)){
            int intVal = boost::any_cast<int>(token.value);
            char buffer[255];
            sprintf(buffer,"%d", intVal);
            name += buffer; /** @todo */
            //name += toString<int>(intVal);
        }
        while(true){
            tok = tokPeek();
            switch (tok) {
                case Token::asterisk :
                    name += "*";
                    tokenNext();
                    break;
                case Token::identifier :
                    name += boost::any_cast<std::string>(tokenNext().value);
                    break;
                case Token::integer :
                    name += toString(boost::any_cast<int>(tokenNext().value));
                    break;
                case Token::dot :
                    return true;
                default :
                    return true;
            }
        }
        
    }else {
        return 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:	1987
Committed:	Mon Feb 7 22:36:32 2005 UTC (19 years, 4 months ago) by tim
File size:	22135 byte(s)
Log Message:	support select internal index selection, for example, select DMPC.3
#	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. Acknowledgement of the program authors must be made in any
10	* publication of scientific results based in part on use of the
11	* program. An acceptable form of acknowledgement is citation of
12	* the article in which the program was described (Matthew
13	* A. Meineke, Charles F. Vardeman II, Teng Lin, Christopher
14	* J. Fennell and J. Daniel Gezelter, "OOPSE: An Object-Oriented
15	* Parallel Simulation Engine for Molecular Dynamics,"
16	* J. Comput. Chem. 26, pp. 252-271 (2005))
17	*
18	* 2. Redistributions of source code must retain the above copyright
19	* notice, this list of conditions and the following disclaimer.
20	*
21	* 3. Redistributions in binary form must reproduce the above copyright
22	* notice, this list of conditions and the following disclaimer in the
23	* documentation and/or other materials provided with the
24	* distribution.
25	*
26	* This software is provided "AS IS," without a warranty of any
27	* kind. All express or implied conditions, representations and
28	* warranties, including any implied warranty of merchantability,
29	* fitness for a particular purpose or non-infringement, are hereby
30	* excluded. The University of Notre Dame and its licensors shall not
31	* be liable for any damages suffered by licensee as a result of
32	* using, modifying or distributing the software or its
33	* derivatives. In no event will the University of Notre Dame or its
34	* licensors be liable for any lost revenue, profit or data, or for
35	* direct, indirect, special, consequential, incidental or punitive
36	* damages, however caused and regardless of the theory of liability,
37	* arising out of the use of or inability to use software, even if the
38	* University of Notre Dame has been advised of the possibility of
39	* such damages.
40	*/
41
42	#include "selection/SelectionCompiler.hpp"
43	#include "utils/StringUtils.hpp"
44	namespace oopse {
45
46	bool SelectionCompiler::compile(const std::string& filename, const std::string& script) {
47
48	this->filename = filename;
49	this->script = script;
50	lineNumbers.clear();
51	lineIndices.clear();
52	aatokenCompiled.clear();
53
54	if (internalCompile()) {
55	return true;
56	}
57
58	int icharEnd;
59	if ((icharEnd = script.find('\r', ichCurrentCommand)) == std::string::npos &&
60	(icharEnd = script.find('\n', ichCurrentCommand)) == std::string::npos) {
61	icharEnd = script.size();
62	}
63	errorLine = script.substr(ichCurrentCommand, icharEnd);
64	return false;
65	}
66
67	bool SelectionCompiler::internalCompile(){
68
69	cchScript = script.size();
70	ichToken = 0;
71	lineCurrent = 1;
72
73	error = false;
74
75	//std::vector<Token> lltoken;
76	aatokenCompiled.clear();
77	std::vector<Token> ltoken;
78
79	Token tokenCommand;
80	int tokCommand = Token::nada;
81
82	for ( ; true; ichToken += cchToken) {
83	if (lookingAtLeadingWhitespace())
84	continue;
85	//if (lookingAtComment())
86	// continue;
87	bool endOfLine = lookingAtEndOfLine();
88	if (endOfLine \|\| lookingAtEndOfStatement()) {
89	if (tokCommand != Token::nada) {
90	if (! compileCommand(ltoken)) {
91	return false;
92	}
93	aatokenCompiled.push_back(atokenCommand);
94	lineNumbers.push_back(lineCurrent);
95	lineIndices.push_back(ichCurrentCommand);
96	ltoken.clear();
97	tokCommand = Token::nada;
98	}
99
100	if (ichToken < cchScript) {
101	if (endOfLine)
102	++lineCurrent;
103	continue;
104	}
105	break;
106	}
107
108	if (tokCommand != Token::nada) {
109	if (lookingAtString()) {
110	std::string str = getUnescapedStringLiteral();
111	ltoken.push_back(Token(Token::string, str));
112	continue;
113	}
114	//if ((tokCommand & Token::specialstring) != 0 &&
115	// lookingAtSpecialString()) {
116	// std::string str = script.substr(ichToken, ichToken + cchToken);
117	// ltoken.push_back(Token(Token::string, str));
118	// continue;
119	//}
120	if (lookingAtDecimal((tokCommand & Token::negnums) != 0)) {
121	float value = lexi_cast<float>(script.substr(ichToken, cchToken));
122	std::cout << "encount an decimal: " << value << std::endl;
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	std::cout << "encount an integer: " << val << std::endl;
130	ltoken.push_back(Token(Token::integer, boost::any(val)));
131	continue;
132	}
133	}
134
135	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
145	int tok = token.tok;
146
147	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
156	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
169	break;
170
171	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
180	if (ltoken.size() == 0) {
181	return commandExpected();
182	}
183
184	return unrecognizedToken();
185	}
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	++ichT;
210	} 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	bool previousCharBackslash = false;
240	while (ichT < cchScript) {
241	ch = script[ichT++];
242	if (ch == '"' && !previousCharBackslash)
243	break;
244	previousCharBackslash = ch == '\\' ? !previousCharBackslash : false;
245	}
246	cchToken = ichT - ichToken;
247
248
249	std::cout << "lookingAtString: encount " << script.substr(ichToken, cchToken) << std::endl;
250	return true;
251	}
252
253
254	std::string SelectionCompiler::getUnescapedStringLiteral() {
255	/** @todo */
256	std::string sb(cchToken - 2, ' ');
257
258	int ichMax = ichToken + cchToken - 1;
259	int ich = ichToken + 1;
260
261	while (ich < ichMax) {
262	char ch = script[ich++];
263	if (ch == '\\' && ich < ichMax) {
264	ch = script[ich++];
265	switch (ch) {
266	case 'b':
267	ch = '\b';
268	break;
269	case 'n':
270	ch = '\n';
271	break;
272	case 't':
273	ch = '\t';
274	break;
275	case 'r':
276	ch = '\r';
277	// fall into
278	case '"':
279	case '\\':
280	case '\'':
281	break;
282	case 'x':
283	case 'u':
284	int digitCount = ch == 'x' ? 2 : 4;
285	if (ich < ichMax) {
286	int unicode = 0;
287	for (int k = digitCount; --k >= 0 && ich < ichMax; ) {
288	char chT = script[ich];
289	int hexit = getHexitValue(chT);
290	if (hexit < 0)
291	break;
292	unicode <<= 4;
293	unicode += hexit;
294	++ich;
295	}
296	ch = (char)unicode;
297	}
298	}
299	}
300	sb.append(1, ch);
301	}
302
303	return sb;
304	}
305
306	int SelectionCompiler::getHexitValue(char ch) {
307	if (ch >= '0' && ch <= '9')
308	return ch - '0';
309	else if (ch >= 'a' && ch <= 'f')
310	return 10 + ch - 'a';
311	else if (ch >= 'A' && ch <= 'F')
312	return 10 + ch - 'A';
313	else
314	return -1;
315	}
316
317	bool SelectionCompiler::lookingAtSpecialString() {
318	int ichT = ichToken;
319	char ch = script[ichT];
320	while (ichT < cchScript && ch != ';' && ch != '\r' && ch != '\n') {
321	++ichT;
322	}
323	cchToken = ichT - ichToken;
324	return cchToken > 0;
325	}
326
327	bool SelectionCompiler::lookingAtDecimal(bool allowNegative) {
328	if (ichToken == cchScript) {
329	return false;
330	}
331
332	int ichT = ichToken;
333	if (script[ichT] == '-') {
334	++ichT;
335	}
336	bool digitSeen = false;
337	char ch = 'X';
338	while (ichT < cchScript && std::isdigit(ch = script[ichT])) {
339	++ichT;
340	digitSeen = true;
341	}
342
343	if (ichT == cchScript \|\| ch != '.') {
344	return false;
345	}
346
347	// to support DMPC.1, let's check the character before the dot
348	if (ch == '.' && (ichT > 0) && std::isalpha(script[ichT - 1])) {
349	return false;
350	}
351
352	++ichT;
353	while (ichT < cchScript && std::isdigit(script[ichT])) {
354	++ichT;
355	digitSeen = true;
356	}
357	cchToken = ichT - ichToken;
358	return digitSeen;
359	}
360
361	bool SelectionCompiler::lookingAtInteger(bool allowNegative) {
362	if (ichToken == cchScript) {
363	return false;
364	}
365	int ichT = ichToken;
366	if (allowNegative && script[ichToken] == '-') {
367	++ichT;
368	}
369	int ichBeginDigits = ichT;
370	while (ichT < cchScript && std::isdigit(script[ichT])) {
371	++ichT;
372	}
373	if (ichBeginDigits == ichT) {
374	return false;
375	}
376	cchToken = ichT - ichToken;
377	return true;
378	}
379
380	bool SelectionCompiler::lookingAtLookupToken() {
381	if (ichToken == cchScript) {
382	return false;
383	}
384
385	int ichT = ichToken;
386	char ch;
387	switch (ch = script[ichT++]) {
388	case '(':
389	case ')':
390	case ',':
391	case '*':
392	case '-':
393	case '[':
394	case ']':
395	case '+':
396	case ':':
397	case '@':
398	case '.':
399	case '%':
400	break;
401	case '&':
402	case '\|':
403	if (ichT < cchScript && script[ichT] == ch) {
404	++ichT;
405	}
406	break;
407	case '<':
408	case '=':
409	case '>':
410	if (ichT < cchScript && ((ch = script[ichT]) == '<' \|\| ch == '=' \|\| ch == '>')) {
411	++ichT;
412	}
413	break;
414	case '/':
415	case '!':
416	if (ichT < cchScript && script[ichT] == '=') {
417	++ichT;
418	}
419	break;
420	default:
421	if ((ch < 'a' \|\| ch > 'z') && (ch < 'A' && ch > 'Z') && ch != '_') {
422	return false;
423	}
424	case '?': // include question marks in identifier for atom expressions
425	while (ichT < cchScript && !std::isspace(ch = script[ichT]) && (std::isalpha(ch) \|\|std::isdigit(ch) \|\|
426	ch == '_' \|\| ch == '?') ) {
427
428	++ichT;
429	}
430	break;
431	}
432
433	cchToken = ichT - ichToken;
434
435	std::cout << "lookingAtLookupToken: encount " << script.substr(ichToken, cchToken) << std::endl;
436	return true;
437	}
438
439	bool SelectionCompiler::compileCommand(const std::vector<Token>& ltoken) {
440	const Token& tokenCommand = ltoken[0];
441	int tokCommand = tokenCommand.tok;
442
443	atokenCommand = ltoken;
444	if ((tokCommand & Token::expressionCommand) != 0 && !compileExpression()) {
445	return false;
446	}
447
448	return true;
449	}
450
451	bool SelectionCompiler::compileExpression() {
452	/** todo */
453	int i = 1;
454	int tokCommand = atokenCommand[0].tok;
455	if (tokCommand == Token::define) {
456	i = 2;
457	} else if ((tokCommand & Token::embeddedExpression) != 0) {
458	// look for the open parenthesis
459	while (i < atokenCommand.size() &&
460	atokenCommand[i].tok != Token::leftparen)
461	++i;
462	}
463
464	if (i >= atokenCommand.size()) {
465	return true;
466	}
467	return compileExpression(i);
468	}
469
470
471	bool SelectionCompiler::addTokenToPostfix(const Token& token) {
472	ltokenPostfix.push_back(token);
473	return true;
474	}
475
476	bool SelectionCompiler::compileExpression(int itoken) {
477	ltokenPostfix.clear();
478	for (int i = 0; i < itoken; ++i) {
479	addTokenToPostfix(atokenCommand[i]);
480	}
481
482	atokenInfix = atokenCommand;
483	itokenInfix = itoken;
484
485	addTokenToPostfix(Token::tokenExpressionBegin);
486	if (!clauseOr()) {
487	return false;
488	}
489
490	addTokenToPostfix(Token::tokenExpressionEnd);
491	if (itokenInfix != atokenInfix.size()) {
492	return endOfExpressionExpected();
493	}
494
495	atokenCommand = ltokenPostfix;
496	return true;
497	}
498
499	Token SelectionCompiler::tokenNext() {
500	if (itokenInfix == atokenInfix.size()) {
501	return Token();
502	}
503	return atokenInfix[itokenInfix++];
504	}
505
506	boost::any SelectionCompiler::valuePeek() {
507	if (itokenInfix == atokenInfix.size()) {
508	return boost::any();
509	} else {
510	return atokenInfix[itokenInfix].value;
511	}
512	}
513
514	int SelectionCompiler::tokPeek() {
515	if (itokenInfix == atokenInfix.size()) {
516	return 0;
517	}else {
518	return atokenInfix[itokenInfix].tok;
519	}
520	}
521
522	bool SelectionCompiler::clauseOr() {
523	if (!clauseAnd()) {
524	return false;
525	}
526
527	while (tokPeek() == Token::opOr) {
528	Token tokenOr = tokenNext();
529	if (!clauseAnd()) {
530	return false;
531	}
532	addTokenToPostfix(tokenOr);
533	}
534	return true;
535	}
536
537	bool SelectionCompiler::clauseAnd() {
538	if (!clauseNot()) {
539	return false;
540	}
541
542	while (tokPeek() == Token::opAnd) {
543	Token tokenAnd = tokenNext();
544	if (!clauseNot()) {
545	return false;
546	}
547	addTokenToPostfix(tokenAnd);
548	}
549	return true;
550	}
551
552	bool SelectionCompiler::clauseNot() {
553	if (tokPeek() == Token::opNot) {
554	Token tokenNot = tokenNext();
555	if (!clauseNot()) {
556	return false;
557	}
558	return addTokenToPostfix(tokenNot);
559	}
560	return clausePrimitive();
561	}
562
563	bool SelectionCompiler::clausePrimitive() {
564	int tok = tokPeek();
565	switch (tok) {
566	case Token::within:
567	return clauseWithin();
568
569	case Token::asterisk:
570	case Token::identifier:
571	return clauseChemObjName();
572
573	case Token::integer :
574	return clauseIndex();
575	default:
576	if ((tok & Token::atomproperty) == Token::atomproperty) {
577	return clauseComparator();
578	}
579	if ((tok & Token::predefinedset) != Token::predefinedset) {
580	break;
581	}
582	// fall into the code and below and just add the token
583	case Token::all:
584	case Token::none:
585	return addTokenToPostfix(tokenNext());
586	case Token::leftparen:
587	tokenNext();
588	if (!clauseOr()) {
589	return false;
590	}
591	if (tokenNext().tok != Token::rightparen) {
592	return rightParenthesisExpected();
593	}
594	return true;
595	}
596	return unrecognizedExpressionToken();
597	}
598
599	bool SelectionCompiler::clauseComparator() {
600	Token tokenAtomProperty = tokenNext();
601	Token tokenComparator = tokenNext();
602	if ((tokenComparator.tok & Token::comparator) == 0) {
603	return comparisonOperatorExpected();
604	}
605
606	Token tokenValue = tokenNext();
607	if (tokenValue.tok != Token::integer && tokenValue.tok != Token::decimal) {
608	return numberExpected();
609	}
610
611	float val;
612	if (tokenValue.value.type() == typeid(int)) {
613	val = boost::any_cast<int>(tokenValue.value);
614	} else if (tokenValue.value.type() == typeid(float)) {
615	val = boost::any_cast<float>(tokenValue.value);
616	} else {
617	return false;
618	}
619
620	boost::any floatVal;
621	floatVal = val;
622	return addTokenToPostfix(Token(tokenComparator.tok,
623	tokenAtomProperty.tok, floatVal));
624	}
625
626	bool SelectionCompiler::clauseWithin() {
627	tokenNext(); // WITHIN
628	if (tokenNext().tok != Token::leftparen) { // (
629	return leftParenthesisExpected();
630	}
631
632	boost::any distance;
633	Token tokenDistance = tokenNext(); // distance
634	switch(tokenDistance.tok) {
635	case Token::integer:
636	case Token::decimal:
637	distance = tokenDistance.value;
638	break;
639	default:
640	return numberOrKeywordExpected();
641	}
642
643	if (tokenNext().tok != Token::opOr) { // ,
644	return commaExpected();
645	}
646
647	if (! clauseOr()) { // expression
648	return false;
649	}
650
651	if (tokenNext().tok != Token::rightparen) { // )T
652	return rightParenthesisExpected();
653	}
654
655	return addTokenToPostfix(Token(Token::within, distance));
656	}
657
658	bool SelectionCompiler::clauseChemObjName() {
659	std::string chemObjName;
660	int tok = tokPeek();
661	if (!clauseName(chemObjName)){
662	return false;
663	}
664
665
666	tok = tokPeek();
667	//allow two dot at most
668	if (tok == Token::dot) {
669	tokenNext();
670	chemObjName += ".";
671	if (!clauseName(chemObjName)) {
672	return false;
673	}
674	tok = tokPeek();
675	if (tok == Token::dot) {
676	tokenNext();
677	chemObjName += ".";
678
679	if (!clauseName(chemObjName)) {
680	return false;
681	}
682	}
683	}
684
685	return addTokenToPostfix(Token(Token::name, chemObjName));
686	}
687
688	bool SelectionCompiler:: clauseName(std::string& name) {
689
690	int tok = tokPeek();
691
692	if (tok == Token::asterisk \|\| tok == Token::identifier \|\| tok == Token::integer) {
693
694	Token token = tokenNext();
695	if (token.value.type() == typeid(std::string)) {
696	name += boost::any_cast<std::string>(token.value);
697	} else if (token.value.type() == typeid(int)){
698	int intVal = boost::any_cast<int>(token.value);
699	char buffer[255];
700	sprintf(buffer,"%d", intVal);
701	name += buffer; /** @todo */
702	//name += toString<int>(intVal);
703	}
704	while(true){
705	tok = tokPeek();
706	switch (tok) {
707	case Token::asterisk :
708	name += "*";
709	tokenNext();
710	break;
711	case Token::identifier :
712	name += boost::any_cast<std::string>(tokenNext().value);
713	break;
714	case Token::integer :
715	name += toString(boost::any_cast<int>(tokenNext().value));
716	break;
717	case Token::dot :
718	return true;
719	default :
720	return true;
721	}
722	}
723
724	}else {
725	return false;
726	}
727
728	}
729
730	bool SelectionCompiler::clauseIndex(){
731	Token token = tokenNext();
732	if (token.tok == Token::integer) {
733	int index = boost::any_cast<int>(token.value);
734	int tok = tokPeek();
735	std::cout << "Token::to is " << Token::to << ", tok = " << tok << std::endl;
736	if (tok == Token::to) {
737	tokenNext();
738	tok = tokPeek();
739	if (tok != Token::integer) {
740	return numberExpected();
741	}
742
743	boost::any intVal = tokenNext().value;
744	int first = index;
745	if (intVal.type() != typeid(int)){
746	return false;
747	}
748	int second = boost::any_cast<int>(intVal);
749
750	return addTokenToPostfix(Token(Token::index, boost::any(std::make_pair(first, second))));
751
752	}else {
753	return addTokenToPostfix(Token(Token::index, boost::any(index)));
754	}
755	} else {
756	return numberExpected();
757	}
758	}
759
760	}