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 1.ca, let's check the character after the dot
    // to determine if it is an alpha
    if (ch == '.' && (ichT + 1 < cchScript) && 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) {
        name += boost::any_cast<std::string>(tokenNext().value);
        
        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:	1979
Committed:	Mon Feb 7 19:13:18 2005 UTC (19 years, 5 months ago) by tim
File size:	21799 byte(s)
Log Message:	selection library is working
#	User	Rev	Content
1	tim	1963	/*
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	tim	1965	#include "utils/StringUtils.hpp"
44	tim	1963	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	tim	1965	if (internalCompile()) {
55	tim	1963	return true;
56			}
57
58			int icharEnd;
59	tim	1965	if ((icharEnd = script.find('\r', ichCurrentCommand)) == std::string::npos &&
60			(icharEnd = script.find('\n', ichCurrentCommand)) == std::string::npos) {
61	tim	1963	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	tim	1965	//std::vector<Token> lltoken;
76			aatokenCompiled.clear();
77	tim	1963	std::vector<Token> ltoken;
78
79	tim	1965	Token tokenCommand;
80			int tokCommand = Token::nada;
81	tim	1963
82			for ( ; true; ichToken += cchToken) {
83			if (lookingAtLeadingWhitespace())
84			continue;
85	tim	1972	//if (lookingAtComment())
86			// continue;
87	tim	1965	bool endOfLine = lookingAtEndOfLine();
88	tim	1963	if (endOfLine \|\| lookingAtEndOfStatement()) {
89	tim	1965	if (tokCommand != Token::nada) {
90	tim	1963	if (! compileCommand(ltoken)) {
91			return false;
92			}
93	tim	1965	aatokenCompiled.push_back(atokenCommand);
94			lineNumbers.push_back(lineCurrent);
95			lineIndices.push_back(ichCurrentCommand);
96			ltoken.clear();
97			tokCommand = Token::nada;
98	tim	1963	}
99
100			if (ichToken < cchScript) {
101			if (endOfLine)
102			++lineCurrent;
103			continue;
104			}
105			break;
106			}
107
108	tim	1965	if (tokCommand != Token::nada) {
109	tim	1963	if (lookingAtString()) {
110			std::string str = getUnescapedStringLiteral();
111	tim	1965	ltoken.push_back(Token(Token::string, str));
112	tim	1963	continue;
113			}
114	tim	1965	//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	tim	1979	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	tim	1963	continue;
125			}
126	tim	1965	if (lookingAtInteger((tokCommand & Token::negnums) != 0)) {
127	tim	1979
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	tim	1963	continue;
132			}
133			}
134
135			if (lookingAtLookupToken()) {
136	tim	1972	std::string ident = script.substr(ichToken, cchToken);
137	tim	1965	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	tim	1963	}
144
145			int tok = token.tok;
146
147			switch (tokCommand) {
148	tim	1965	case Token::nada:
149	tim	1963	ichCurrentCommand = ichToken;
150			//tokenCommand = token;
151			tokCommand = tok;
152	tim	1965	if ((tokCommand & Token::command) == 0)
153	tim	1963	return commandExpected();
154			break;
155
156	tim	1965	case Token::define:
157	tim	1963	if (ltoken.size() == 1) {
158			// we are looking at the variable name
159	tim	1965	if (tok != Token::identifier &&
160			(tok & Token::predefinedset) != Token::predefinedset)
161	tim	1963	return invalidExpressionToken(ident);
162			} else {
163			// we are looking at the expression
164	tim	1965	if (tok != Token::identifier &&
165			(tok & (Token::expression \| Token::predefinedset)) == 0)
166	tim	1963	return invalidExpressionToken(ident);
167			}
168
169			break;
170
171	tim	1965	case Token::select:
172			if (tok != Token::identifier && (tok & Token::expression) == 0)
173	tim	1963	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	tim	1965	bool previousCharBackslash = false;
240	tim	1963	while (ichT < cchScript) {
241	tim	1965	ch = script[ichT++];
242	tim	1963	if (ch == '"' && !previousCharBackslash)
243			break;
244			previousCharBackslash = ch == '\\' ? !previousCharBackslash : false;
245			}
246			cchToken = ichT - ichToken;
247	tim	1979
248
249			std::cout << "lookingAtString: encount " << script.substr(ichToken, cchToken) << std::endl;
250	tim	1963	return true;
251			}
252
253
254			std::string SelectionCompiler::getUnescapedStringLiteral() {
255	tim	1965	/** @todo */
256			std::string sb(cchToken - 2, ' ');
257
258	tim	1963	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	tim	1965	sb.append(1, ch);
301	tim	1963	}
302
303	tim	1965	return sb;
304	tim	1963	}
305
306	tim	1965	int SelectionCompiler::getHexitValue(char ch) {
307	tim	1963	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	tim	1965	bool SelectionCompiler::lookingAtDecimal(bool allowNegative) {
328	tim	1963	if (ichToken == cchScript) {
329			return false;
330			}
331
332			int ichT = ichToken;
333			if (script[ichT] == '-') {
334			++ichT;
335			}
336	tim	1965	bool digitSeen = false;
337	tim	1963	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 1.ca, let's check the character after the dot
348			// to determine if it is an alpha
349			if (ch == '.' && (ichT + 1 < cchScript) && std::isalpha(script[ichT + 1])) {
350			return false;
351			}
352
353			++ichT;
354			while (ichT < cchScript && std::isdigit(script[ichT])) {
355			++ichT;
356			digitSeen = true;
357			}
358			cchToken = ichT - ichToken;
359			return digitSeen;
360			}
361
362	tim	1965	bool SelectionCompiler::lookingAtInteger(bool allowNegative) {
363	tim	1963	if (ichToken == cchScript) {
364			return false;
365			}
366			int ichT = ichToken;
367			if (allowNegative && script[ichToken] == '-') {
368			++ichT;
369			}
370			int ichBeginDigits = ichT;
371			while (ichT < cchScript && std::isdigit(script[ichT])) {
372			++ichT;
373			}
374			if (ichBeginDigits == ichT) {
375			return false;
376			}
377			cchToken = ichT - ichToken;
378			return true;
379			}
380
381			bool SelectionCompiler::lookingAtLookupToken() {
382			if (ichToken == cchScript) {
383			return false;
384			}
385
386			int ichT = ichToken;
387			char ch;
388			switch (ch = script[ichT++]) {
389			case '(':
390			case ')':
391			case ',':
392			case '*':
393			case '-':
394			case '[':
395			case ']':
396			case '+':
397			case ':':
398			case '@':
399			case '.':
400			case '%':
401			break;
402			case '&':
403			case '\|':
404			if (ichT < cchScript && script[ichT] == ch) {
405			++ichT;
406			}
407			break;
408			case '<':
409			case '=':
410			case '>':
411			if (ichT < cchScript && ((ch = script[ichT]) == '<' \|\| ch == '=' \|\| ch == '>')) {
412			++ichT;
413			}
414			break;
415			case '/':
416			case '!':
417			if (ichT < cchScript && script[ichT] == '=') {
418			++ichT;
419			}
420			break;
421			default:
422			if ((ch < 'a' \|\| ch > 'z') && (ch < 'A' && ch > 'Z') && ch != '_') {
423			return false;
424			}
425			case '?': // include question marks in identifier for atom expressions
426	tim	1972	while (ichT < cchScript && !std::isspace(ch = script[ichT]) && (std::isalpha(ch) \|\|std::isdigit(ch) \|\|
427			ch == '_' \|\| ch == '?') ) {
428
429	tim	1963	++ichT;
430			}
431			break;
432			}
433	tim	1979
434	tim	1963	cchToken = ichT - ichToken;
435	tim	1979
436			std::cout << "lookingAtLookupToken: encount " << script.substr(ichToken, cchToken) << std::endl;
437	tim	1963	return true;
438			}
439
440	tim	1965	bool SelectionCompiler::compileCommand(const std::vector<Token>& ltoken) {
441			const Token& tokenCommand = ltoken[0];
442	tim	1963	int tokCommand = tokenCommand.tok;
443	tim	1965
444			atokenCommand = ltoken;
445			if ((tokCommand & Token::expressionCommand) != 0 && !compileExpression()) {
446	tim	1963	return false;
447			}
448	tim	1965
449	tim	1963	return true;
450			}
451
452			bool SelectionCompiler::compileExpression() {
453			/** todo */
454			int i = 1;
455			int tokCommand = atokenCommand[0].tok;
456	tim	1965	if (tokCommand == Token::define) {
457			i = 2;
458			} else if ((tokCommand & Token::embeddedExpression) != 0) {
459			// look for the open parenthesis
460			while (i < atokenCommand.size() &&
461			atokenCommand[i].tok != Token::leftparen)
462	tim	1963	++i;
463			}
464	tim	1965
465			if (i >= atokenCommand.size()) {
466			return true;
467			}
468	tim	1963	return compileExpression(i);
469			}
470
471
472	tim	1965	bool SelectionCompiler::addTokenToPostfix(const Token& token) {
473	tim	1963	ltokenPostfix.push_back(token);
474			return true;
475			}
476
477			bool SelectionCompiler::compileExpression(int itoken) {
478	tim	1965	ltokenPostfix.clear();
479			for (int i = 0; i < itoken; ++i) {
480	tim	1963	addTokenToPostfix(atokenCommand[i]);
481	tim	1965	}
482
483	tim	1963	atokenInfix = atokenCommand;
484			itokenInfix = itoken;
485
486	tim	1965	addTokenToPostfix(Token::tokenExpressionBegin);
487	tim	1963	if (!clauseOr()) {
488			return false;
489			}
490
491	tim	1965	addTokenToPostfix(Token::tokenExpressionEnd);
492			if (itokenInfix != atokenInfix.size()) {
493	tim	1963	return endOfExpressionExpected();
494			}
495
496			atokenCommand = ltokenPostfix;
497			return true;
498			}
499
500			Token SelectionCompiler::tokenNext() {
501	tim	1965	if (itokenInfix == atokenInfix.size()) {
502			return Token();
503			}
504			return atokenInfix[itokenInfix++];
505	tim	1963	}
506
507	tim	1965	boost::any SelectionCompiler::valuePeek() {
508			if (itokenInfix == atokenInfix.size()) {
509			return boost::any();
510	tim	1963	} else {
511			return atokenInfix[itokenInfix].value;
512			}
513			}
514
515			int SelectionCompiler::tokPeek() {
516	tim	1965	if (itokenInfix == atokenInfix.size()) {
517	tim	1963	return 0;
518			}else {
519			return atokenInfix[itokenInfix].tok;
520			}
521			}
522
523			bool SelectionCompiler::clauseOr() {
524			if (!clauseAnd()) {
525			return false;
526			}
527
528	tim	1965	while (tokPeek() == Token::opOr) {
529	tim	1963	Token tokenOr = tokenNext();
530			if (!clauseAnd()) {
531			return false;
532			}
533			addTokenToPostfix(tokenOr);
534			}
535			return true;
536			}
537
538			bool SelectionCompiler::clauseAnd() {
539			if (!clauseNot()) {
540			return false;
541			}
542
543	tim	1965	while (tokPeek() == Token::opAnd) {
544	tim	1963	Token tokenAnd = tokenNext();
545			if (!clauseNot()) {
546			return false;
547			}
548			addTokenToPostfix(tokenAnd);
549			}
550			return true;
551			}
552
553			bool SelectionCompiler::clauseNot() {
554	tim	1965	if (tokPeek() == Token::opNot) {
555	tim	1963	Token tokenNot = tokenNext();
556			if (!clauseNot()) {
557			return false;
558			}
559			return addTokenToPostfix(tokenNot);
560			}
561			return clausePrimitive();
562			}
563
564			bool SelectionCompiler::clausePrimitive() {
565			int tok = tokPeek();
566			switch (tok) {
567	tim	1965	case Token::within:
568	tim	1963	return clauseWithin();
569	tim	1967
570			case Token::asterisk:
571			case Token::identifier:
572			return clauseChemObjName();
573	tim	1979
574			case Token::integer :
575			return clauseIndex();
576	tim	1963	default:
577	tim	1965	if ((tok & Token::atomproperty) == Token::atomproperty) {
578	tim	1963	return clauseComparator();
579			}
580	tim	1965	if ((tok & Token::predefinedset) != Token::predefinedset) {
581	tim	1963	break;
582			}
583			// fall into the code and below and just add the token
584	tim	1965	case Token::all:
585			case Token::none:
586	tim	1963	return addTokenToPostfix(tokenNext());
587	tim	1965	case Token::leftparen:
588	tim	1963	tokenNext();
589			if (!clauseOr()) {
590			return false;
591			}
592	tim	1965	if (tokenNext().tok != Token::rightparen) {
593	tim	1963	return rightParenthesisExpected();
594			}
595			return true;
596			}
597			return unrecognizedExpressionToken();
598			}
599
600			bool SelectionCompiler::clauseComparator() {
601			Token tokenAtomProperty = tokenNext();
602			Token tokenComparator = tokenNext();
603	tim	1965	if ((tokenComparator.tok & Token::comparator) == 0) {
604	tim	1963	return comparisonOperatorExpected();
605			}
606
607			Token tokenValue = tokenNext();
608	tim	1972	if (tokenValue.tok != Token::integer && tokenValue.tok != Token::decimal) {
609			return numberExpected();
610	tim	1963	}
611	tim	1972
612			float val;
613			if (tokenValue.value.type() == typeid(int)) {
614			val = boost::any_cast<int>(tokenValue.value);
615			} else if (tokenValue.value.type() == typeid(float)) {
616			val = boost::any_cast<float>(tokenValue.value);
617			} else {
618			return false;
619			}
620
621	tim	1979	boost::any floatVal;
622			floatVal = val;
623	tim	1965	return addTokenToPostfix(Token(tokenComparator.tok,
624	tim	1979	tokenAtomProperty.tok, floatVal));
625	tim	1963	}
626
627			bool SelectionCompiler::clauseWithin() {
628			tokenNext(); // WITHIN
629	tim	1965	if (tokenNext().tok != Token::leftparen) { // (
630	tim	1963	return leftParenthesisExpected();
631			}
632
633	tim	1965	boost::any distance;
634	tim	1963	Token tokenDistance = tokenNext(); // distance
635			switch(tokenDistance.tok) {
636	tim	1965	case Token::integer:
637			case Token::decimal:
638	tim	1963	distance = tokenDistance.value;
639			break;
640			default:
641			return numberOrKeywordExpected();
642			}
643
644	tim	1965	if (tokenNext().tok != Token::opOr) { // ,
645	tim	1963	return commaExpected();
646			}
647
648			if (! clauseOr()) { // expression
649			return false;
650			}
651
652	tim	1965	if (tokenNext().tok != Token::rightparen) { // )T
653	tim	1963	return rightParenthesisExpected();
654			}
655
656	tim	1965	return addTokenToPostfix(Token(Token::within, distance));
657	tim	1963	}
658
659	tim	1967	bool SelectionCompiler::clauseChemObjName() {
660			std::string chemObjName;
661			int tok = tokPeek();
662			if (!clauseName(chemObjName)){
663			return false;
664			}
665	tim	1963
666
667	tim	1967	tok = tokPeek();
668			//allow two dot at most
669			if (tok == Token::dot) {
670	tim	1972	tokenNext();
671			chemObjName += ".";
672	tim	1967	if (!clauseName(chemObjName)) {
673			return false;
674			}
675			tok = tokPeek();
676			if (tok == Token::dot) {
677	tim	1972	tokenNext();
678			chemObjName += ".";
679
680	tim	1967	if (!clauseName(chemObjName)) {
681			return false;
682			}
683			}
684			}
685
686			return addTokenToPostfix(Token(Token::name, chemObjName));
687	tim	1963	}
688
689	tim	1967	bool SelectionCompiler:: clauseName(std::string& name) {
690
691			int tok = tokPeek();
692
693			if (tok == Token::asterisk \|\| tok == Token::identifier) {
694			name += boost::any_cast<std::string>(tokenNext().value);
695
696			while(true){
697			tok = tokPeek();
698			switch (tok) {
699			case Token::asterisk :
700			name += "*";
701			tokenNext();
702			break;
703			case Token::identifier :
704			name += boost::any_cast<std::string>(tokenNext().value);
705			break;
706			case Token::integer :
707			name += toString(boost::any_cast<int>(tokenNext().value));
708			break;
709			case Token::dot :
710			return true;
711			default :
712			return true;
713			}
714			}
715
716			}else {
717			return false;
718			}
719
720	tim	1963	}
721
722	tim	1979	bool SelectionCompiler::clauseIndex(){
723			Token token = tokenNext();
724			if (token.tok == Token::integer) {
725			int index = boost::any_cast<int>(token.value);
726			int tok = tokPeek();
727			std::cout << "Token::to is " << Token::to << ", tok = " << tok << std::endl;
728			if (tok == Token::to) {
729			tokenNext();
730			tok = tokPeek();
731			if (tok != Token::integer) {
732			return numberExpected();
733			}
734
735			boost::any intVal = tokenNext().value;
736			int first = index;
737			if (intVal.type() != typeid(int)){
738			return false;
739			}
740			int second = boost::any_cast<int>(intVal);
741	tim	1967
742	tim	1979	return addTokenToPostfix(Token(Token::index, boost::any(std::make_pair(first, second))));
743
744			}else {
745			return addTokenToPostfix(Token(Token::index, boost::any(index)));
746			}
747			} else {
748			return numberExpected();
749			}
750	tim	1963	}
751	tim	1979
752			}