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, ichToken + cchToken));          
                ltoken.push_back(Token(Token::decimal, value));/**@todo*/
                continue;
            }
            if (lookingAtInteger((tokCommand & Token::negnums) != 0)) {
                std::string intString = script.substr(ichToken, ichToken + cchToken);
                int val = lexi_cast<int>(intString);
                ltoken.push_back(Token(Token::integer, val, intString));/**@todo*/
                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 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;
    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();
      
        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;
    }

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

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:
            distance = float(tokenDistance.intValue);
            break;
        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;
    }

}


}
Revision:	288
Committed:	Fri Feb 4 22:39:26 2005 UTC (20 years, 8 months ago) by tim
File size:	20529 byte(s)
Log Message:	half of the selection utility is working need to debug within keyword and atomproperty keyword
#	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, ichToken + cchToken));
122	ltoken.push_back(Token(Token::decimal, value));/*@todo/
123	continue;
124	}
125	if (lookingAtInteger((tokCommand & Token::negnums) != 0)) {
126	std::string intString = script.substr(ichToken, ichToken + cchToken);
127	int val = lexi_cast<int>(intString);
128	ltoken.push_back(Token(Token::integer, val, intString));/*@todo/
129	continue;
130	}
131	}
132
133	if (lookingAtLookupToken()) {
134	std::string ident = script.substr(ichToken, cchToken);
135	Token token;
136	Token* pToken = TokenMap::getInstance()->getToken(ident);
137	if (pToken != NULL) {
138	token = *pToken;
139	} else {
140	token = Token(Token::identifier, ident);
141	}
142
143	int tok = token.tok;
144
145	switch (tokCommand) {
146	case Token::nada:
147	ichCurrentCommand = ichToken;
148	//tokenCommand = token;
149	tokCommand = tok;
150	if ((tokCommand & Token::command) == 0)
151	return commandExpected();
152	break;
153
154	case Token::define:
155	if (ltoken.size() == 1) {
156	// we are looking at the variable name
157	if (tok != Token::identifier &&
158	(tok & Token::predefinedset) != Token::predefinedset)
159	return invalidExpressionToken(ident);
160	} else {
161	// we are looking at the expression
162	if (tok != Token::identifier &&
163	(tok & (Token::expression \| Token::predefinedset)) == 0)
164	return invalidExpressionToken(ident);
165	}
166
167	break;
168
169	case Token::select:
170	if (tok != Token::identifier && (tok & Token::expression) == 0)
171	return invalidExpressionToken(ident);
172	break;
173	}
174	ltoken.push_back(token);
175	continue;
176	}
177
178	if (ltoken.size() == 0) {
179	return commandExpected();
180	}
181
182	return unrecognizedToken();
183	}
184
185	return true;
186	}
187
188
189	bool SelectionCompiler::lookingAtLeadingWhitespace() {
190
191	int ichT = ichToken;
192	while (ichT < cchScript && std::isspace(script[ichT])) {
193	++ichT;
194	}
195	cchToken = ichT - ichToken;
196	return cchToken > 0;
197	}
198
199	bool SelectionCompiler::lookingAtEndOfLine() {
200	if (ichToken == cchScript)
201	return true;
202	int ichT = ichToken;
203	char ch = script[ichT];
204	if (ch == '\r') {
205	++ichT;
206	if (ichT < cchScript && script[ichT] == '\n')
207	++ichT;
208	} else if (ch == '\n') {
209	++ichT;
210	} else {
211	return false;
212	}
213	cchToken = ichT - ichToken;
214	return true;
215	}
216
217	bool SelectionCompiler::lookingAtEndOfStatement() {
218	if (ichToken == cchScript \|\| script[ichToken] != ';')
219	return false;
220	cchToken = 1;
221	return true;
222	}
223
224	bool SelectionCompiler::lookingAtString() {
225	if (ichToken == cchScript)
226	return false;
227	if (script[ichToken] != '"')
228	return false;
229	// remove support for single quote
230	// in order to use it in atom expressions
231	// char chFirst = script.charAt(ichToken);
232	// if (chFirst != '"' && chFirst != '\'')
233	// return false;
234	int ichT = ichToken + 1;
235	// while (ichT < cchScript && script.charAt(ichT++) != chFirst)
236	char ch;
237	bool previousCharBackslash = false;
238	while (ichT < cchScript) {
239	ch = script[ichT++];
240	if (ch == '"' && !previousCharBackslash)
241	break;
242	previousCharBackslash = ch == '\\' ? !previousCharBackslash : false;
243	}
244	cchToken = ichT - ichToken;
245	return true;
246	}
247
248
249	std::string SelectionCompiler::getUnescapedStringLiteral() {
250	/** @todo */
251	std::string sb(cchToken - 2, ' ');
252
253	int ichMax = ichToken + cchToken - 1;
254	int ich = ichToken + 1;
255
256	while (ich < ichMax) {
257	char ch = script[ich++];
258	if (ch == '\\' && ich < ichMax) {
259	ch = script[ich++];
260	switch (ch) {
261	case 'b':
262	ch = '\b';
263	break;
264	case 'n':
265	ch = '\n';
266	break;
267	case 't':
268	ch = '\t';
269	break;
270	case 'r':
271	ch = '\r';
272	// fall into
273	case '"':
274	case '\\':
275	case '\'':
276	break;
277	case 'x':
278	case 'u':
279	int digitCount = ch == 'x' ? 2 : 4;
280	if (ich < ichMax) {
281	int unicode = 0;
282	for (int k = digitCount; --k >= 0 && ich < ichMax; ) {
283	char chT = script[ich];
284	int hexit = getHexitValue(chT);
285	if (hexit < 0)
286	break;
287	unicode <<= 4;
288	unicode += hexit;
289	++ich;
290	}
291	ch = (char)unicode;
292	}
293	}
294	}
295	sb.append(1, ch);
296	}
297
298	return sb;
299	}
300
301	int SelectionCompiler::getHexitValue(char ch) {
302	if (ch >= '0' && ch <= '9')
303	return ch - '0';
304	else if (ch >= 'a' && ch <= 'f')
305	return 10 + ch - 'a';
306	else if (ch >= 'A' && ch <= 'F')
307	return 10 + ch - 'A';
308	else
309	return -1;
310	}
311
312	bool SelectionCompiler::lookingAtSpecialString() {
313	int ichT = ichToken;
314	char ch = script[ichT];
315	while (ichT < cchScript && ch != ';' && ch != '\r' && ch != '\n') {
316	++ichT;
317	}
318	cchToken = ichT - ichToken;
319	return cchToken > 0;
320	}
321
322	bool SelectionCompiler::lookingAtDecimal(bool allowNegative) {
323	if (ichToken == cchScript) {
324	return false;
325	}
326
327	int ichT = ichToken;
328	if (script[ichT] == '-') {
329	++ichT;
330	}
331	bool digitSeen = false;
332	char ch = 'X';
333	while (ichT < cchScript && std::isdigit(ch = script[ichT])) {
334	++ichT;
335	digitSeen = true;
336	}
337
338	if (ichT == cchScript \|\| ch != '.') {
339	return false;
340	}
341
342	// to support 1.ca, let's check the character after the dot
343	// to determine if it is an alpha
344	if (ch == '.' && (ichT + 1 < cchScript) && std::isalpha(script[ichT + 1])) {
345	return false;
346	}
347
348	++ichT;
349	while (ichT < cchScript && std::isdigit(script[ichT])) {
350	++ichT;
351	digitSeen = true;
352	}
353	cchToken = ichT - ichToken;
354	return digitSeen;
355	}
356
357	bool SelectionCompiler::lookingAtInteger(bool allowNegative) {
358	if (ichToken == cchScript) {
359	return false;
360	}
361	int ichT = ichToken;
362	if (allowNegative && script[ichToken] == '-') {
363	++ichT;
364	}
365	int ichBeginDigits = ichT;
366	while (ichT < cchScript && std::isdigit(script[ichT])) {
367	++ichT;
368	}
369	if (ichBeginDigits == ichT) {
370	return false;
371	}
372	cchToken = ichT - ichToken;
373	return true;
374	}
375
376	bool SelectionCompiler::lookingAtLookupToken() {
377	if (ichToken == cchScript) {
378	return false;
379	}
380
381	int ichT = ichToken;
382	char ch;
383	switch (ch = script[ichT++]) {
384	case '(':
385	case ')':
386	case ',':
387	case '*':
388	case '-':
389	case '[':
390	case ']':
391	case '+':
392	case ':':
393	case '@':
394	case '.':
395	case '%':
396	break;
397	case '&':
398	case '\|':
399	if (ichT < cchScript && script[ichT] == ch) {
400	++ichT;
401	}
402	break;
403	case '<':
404	case '=':
405	case '>':
406	if (ichT < cchScript && ((ch = script[ichT]) == '<' \|\| ch == '=' \|\| ch == '>')) {
407	++ichT;
408	}
409	break;
410	case '/':
411	case '!':
412	if (ichT < cchScript && script[ichT] == '=') {
413	++ichT;
414	}
415	break;
416	default:
417	if ((ch < 'a' \|\| ch > 'z') && (ch < 'A' && ch > 'Z') && ch != '_') {
418	return false;
419	}
420	case '?': // include question marks in identifier for atom expressions
421	while (ichT < cchScript && !std::isspace(ch = script[ichT]) && (std::isalpha(ch) \|\|std::isdigit(ch) \|\|
422	ch == '_' \|\| ch == '?') ) {
423
424	++ichT;
425	}
426	break;
427	}
428	cchToken = ichT - ichToken;
429	return true;
430	}
431
432	bool SelectionCompiler::compileCommand(const std::vector<Token>& ltoken) {
433	const Token& tokenCommand = ltoken[0];
434	int tokCommand = tokenCommand.tok;
435
436	atokenCommand = ltoken;
437	if ((tokCommand & Token::expressionCommand) != 0 && !compileExpression()) {
438	return false;
439	}
440
441	return true;
442	}
443
444	bool SelectionCompiler::compileExpression() {
445	/** todo */
446	int i = 1;
447	int tokCommand = atokenCommand[0].tok;
448	if (tokCommand == Token::define) {
449	i = 2;
450	} else if ((tokCommand & Token::embeddedExpression) != 0) {
451	// look for the open parenthesis
452	while (i < atokenCommand.size() &&
453	atokenCommand[i].tok != Token::leftparen)
454	++i;
455	}
456
457	if (i >= atokenCommand.size()) {
458	return true;
459	}
460	return compileExpression(i);
461	}
462
463
464	bool SelectionCompiler::addTokenToPostfix(const Token& token) {
465	ltokenPostfix.push_back(token);
466	return true;
467	}
468
469	bool SelectionCompiler::compileExpression(int itoken) {
470	ltokenPostfix.clear();
471	for (int i = 0; i < itoken; ++i) {
472	addTokenToPostfix(atokenCommand[i]);
473	}
474
475	atokenInfix = atokenCommand;
476	itokenInfix = itoken;
477
478	addTokenToPostfix(Token::tokenExpressionBegin);
479	if (!clauseOr()) {
480	return false;
481	}
482
483	addTokenToPostfix(Token::tokenExpressionEnd);
484	if (itokenInfix != atokenInfix.size()) {
485	return endOfExpressionExpected();
486	}
487
488	atokenCommand = ltokenPostfix;
489	return true;
490	}
491
492	Token SelectionCompiler::tokenNext() {
493	if (itokenInfix == atokenInfix.size()) {
494	return Token();
495	}
496	return atokenInfix[itokenInfix++];
497	}
498
499	boost::any SelectionCompiler::valuePeek() {
500	if (itokenInfix == atokenInfix.size()) {
501	return boost::any();
502	} else {
503	return atokenInfix[itokenInfix].value;
504	}
505	}
506
507	int SelectionCompiler::tokPeek() {
508	if (itokenInfix == atokenInfix.size()) {
509	return 0;
510	}else {
511	return atokenInfix[itokenInfix].tok;
512	}
513	}
514
515	bool SelectionCompiler::clauseOr() {
516	if (!clauseAnd()) {
517	return false;
518	}
519
520	while (tokPeek() == Token::opOr) {
521	Token tokenOr = tokenNext();
522	if (!clauseAnd()) {
523	return false;
524	}
525	addTokenToPostfix(tokenOr);
526	}
527	return true;
528	}
529
530	bool SelectionCompiler::clauseAnd() {
531	if (!clauseNot()) {
532	return false;
533	}
534
535	while (tokPeek() == Token::opAnd) {
536	Token tokenAnd = tokenNext();
537	if (!clauseNot()) {
538	return false;
539	}
540	addTokenToPostfix(tokenAnd);
541	}
542	return true;
543	}
544
545	bool SelectionCompiler::clauseNot() {
546	if (tokPeek() == Token::opNot) {
547	Token tokenNot = tokenNext();
548	if (!clauseNot()) {
549	return false;
550	}
551	return addTokenToPostfix(tokenNot);
552	}
553	return clausePrimitive();
554	}
555
556	bool SelectionCompiler::clausePrimitive() {
557	int tok = tokPeek();
558	switch (tok) {
559	case Token::within:
560	return clauseWithin();
561
562	case Token::asterisk:
563	case Token::identifier:
564	return clauseChemObjName();
565
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	return addTokenToPostfix(tokenNext());
577	case Token::leftparen:
578	tokenNext();
579	if (!clauseOr()) {
580	return false;
581	}
582	if (tokenNext().tok != Token::rightparen) {
583	return rightParenthesisExpected();
584	}
585	return true;
586	}
587	return unrecognizedExpressionToken();
588	}
589
590	bool SelectionCompiler::clauseComparator() {
591	Token tokenAtomProperty = tokenNext();
592	Token tokenComparator = tokenNext();
593	if ((tokenComparator.tok & Token::comparator) == 0) {
594	return comparisonOperatorExpected();
595	}
596
597	Token tokenValue = tokenNext();
598	if (tokenValue.tok != Token::integer && tokenValue.tok != Token::decimal) {
599	return numberExpected();
600	}
601
602	float val;
603	if (tokenValue.value.type() == typeid(int)) {
604	val = boost::any_cast<int>(tokenValue.value);
605	} else if (tokenValue.value.type() == typeid(float)) {
606	val = boost::any_cast<float>(tokenValue.value);
607	} else {
608	return false;
609	}
610
611	return addTokenToPostfix(Token(tokenComparator.tok,
612	tokenAtomProperty.tok, boost::any(val)));
613	}
614
615	bool SelectionCompiler::clauseWithin() {
616	tokenNext(); // WITHIN
617	if (tokenNext().tok != Token::leftparen) { // (
618	return leftParenthesisExpected();
619	}
620
621	boost::any distance;
622	Token tokenDistance = tokenNext(); // distance
623	switch(tokenDistance.tok) {
624	case Token::integer:
625	distance = float(tokenDistance.intValue);
626	break;
627	case Token::decimal:
628	distance = tokenDistance.value;
629	break;
630	default:
631	return numberOrKeywordExpected();
632	}
633
634	if (tokenNext().tok != Token::opOr) { // ,
635	return commaExpected();
636	}
637
638	if (! clauseOr()) { // expression
639	return false;
640	}
641
642	if (tokenNext().tok != Token::rightparen) { // )T
643	return rightParenthesisExpected();
644	}
645
646	return addTokenToPostfix(Token(Token::within, distance));
647	}
648
649	bool SelectionCompiler::clauseChemObjName() {
650	std::string chemObjName;
651	int tok = tokPeek();
652	if (!clauseName(chemObjName)){
653	return false;
654	}
655
656
657	tok = tokPeek();
658	//allow two dot at most
659	if (tok == Token::dot) {
660	tokenNext();
661	chemObjName += ".";
662	if (!clauseName(chemObjName)) {
663	return false;
664	}
665	tok = tokPeek();
666	if (tok == Token::dot) {
667	tokenNext();
668	chemObjName += ".";
669
670	if (!clauseName(chemObjName)) {
671	return false;
672	}
673	}
674	}
675
676	return addTokenToPostfix(Token(Token::name, chemObjName));
677	}
678
679	bool SelectionCompiler:: clauseName(std::string& name) {
680
681	int tok = tokPeek();
682
683	if (tok == Token::asterisk \|\| tok == Token::identifier) {
684	name += boost::any_cast<std::string>(tokenNext().value);
685
686	while(true){
687	tok = tokPeek();
688	switch (tok) {
689	case Token::asterisk :
690	name += "*";
691	tokenNext();
692	break;
693	case Token::identifier :
694	name += boost::any_cast<std::string>(tokenNext().value);
695	break;
696	case Token::integer :
697	name += toString(boost::any_cast<int>(tokenNext().value));
698	break;
699	case Token::dot :
700	return true;
701	default :
702	return true;
703	}
704	}
705
706	}else {
707	return false;
708	}
709
710	}
711
712
713	}