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));        
                ltoken.push_back(Token(Token::decimal, boost::any(value)));
                continue;
            }
            if (lookingAtInteger((tokCommand & Token::negnums) != 0)) {

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

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

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

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

    return true;
  }


  bool SelectionCompiler::lookingAtLeadingWhitespace() {

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

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

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

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

    return true;
  }

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

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

    return sb;
}

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

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

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

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

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

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

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

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

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

                ++ichT;
            }
        break;
    }

    cchToken = ichT - ichToken;

    return true;
}

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

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

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

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

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

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

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

    atokenCommand = ltokenPostfix;
    return true;
}

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

    return false;
        
}

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

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

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

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

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

}
Revision:	2147
Committed:	Tue Apr 5 23:09:48 2005 UTC (19 years, 3 months ago) by tim
File size:	20736 byte(s)
Log Message:	support '+' and '-' in atom type name;clean the code a little bit
#	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	ltoken.push_back(Token(Token::decimal, boost::any(value)));
123	continue;
124	}
125	if (lookingAtInteger((tokCommand & Token::negnums) != 0)) {
126
127	int val = lexi_cast<int>(script.substr(ichToken, cchToken));
128	ltoken.push_back(Token(Token::integer, boost::any(val)));
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
246	return true;
247	}
248
249
250	std::string SelectionCompiler::getUnescapedStringLiteral() {
251	/** @todo */
252	std::string sb(cchToken - 2, ' ');
253
254	int ichMax = ichToken + cchToken - 1;
255	int ich = ichToken + 1;
256
257	while (ich < ichMax) {
258	char ch = script[ich++];
259	if (ch == '\\' && ich < ichMax) {
260	ch = script[ich++];
261	switch (ch) {
262	case 'b':
263	ch = '\b';
264	break;
265	case 'n':
266	ch = '\n';
267	break;
268	case 't':
269	ch = '\t';
270	break;
271	case 'r':
272	ch = '\r';
273	// fall into
274	case '"':
275	case '\\':
276	case '\'':
277	break;
278	case 'x':
279	case 'u':
280	int digitCount = ch == 'x' ? 2 : 4;
281	if (ich < ichMax) {
282	int unicode = 0;
283	for (int k = digitCount; --k >= 0 && ich < ichMax; ) {
284	char chT = script[ich];
285	int hexit = getHexitValue(chT);
286	if (hexit < 0)
287	break;
288	unicode <<= 4;
289	unicode += hexit;
290	++ich;
291	}
292	ch = (char)unicode;
293	}
294	}
295	}
296	sb.append(1, ch);
297	}
298
299	return sb;
300	}
301
302	int SelectionCompiler::getHexitValue(char ch) {
303	if (ch >= '0' && ch <= '9')
304	return ch - '0';
305	else if (ch >= 'a' && ch <= 'f')
306	return 10 + ch - 'a';
307	else if (ch >= 'A' && ch <= 'F')
308	return 10 + ch - 'A';
309	else
310	return -1;
311	}
312
313	bool SelectionCompiler::lookingAtSpecialString() {
314	int ichT = ichToken;
315	char ch = script[ichT];
316	while (ichT < cchScript && ch != ';' && ch != '\r' && ch != '\n') {
317	++ichT;
318	}
319	cchToken = ichT - ichToken;
320	return cchToken > 0;
321	}
322
323	bool SelectionCompiler::lookingAtDecimal(bool allowNegative) {
324	if (ichToken == cchScript) {
325	return false;
326	}
327
328	int ichT = ichToken;
329	if (script[ichT] == '-') {
330	++ichT;
331	}
332	bool digitSeen = false;
333	char ch = 'X';
334	while (ichT < cchScript && std::isdigit(ch = script[ichT])) {
335	++ichT;
336	digitSeen = true;
337	}
338
339	if (ichT == cchScript \|\| ch != '.') {
340	return false;
341	}
342
343	// to support DMPC.1, let's check the character before the dot
344	if (ch == '.' && (ichT > 0) && 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	break;
390	case '&':
391	case '\|':
392	if (ichT < cchScript && script[ichT] == ch) {
393	++ichT;
394	}
395	break;
396	case '<':
397	case '=':
398	case '>':
399	if (ichT < cchScript && ((ch = script[ichT]) == '<' \|\| ch == '=' \|\| ch == '>')) {
400	++ichT;
401	}
402	break;
403	case '/':
404	case '!':
405	if (ichT < cchScript && script[ichT] == '=') {
406	++ichT;
407	}
408	break;
409	default:
410	if ((ch < 'a' \|\| ch > 'z') && (ch < 'A' && ch > 'Z') && ch != '_') {
411	return false;
412	}
413	case '*':
414	case '?': // include question marks in identifier for atom expressions
415	while (ichT < cchScript && !std::isspace(ch = script[ichT]) &&
416	(std::isalpha(ch) \|\|std::isdigit(ch) \|\| ch == '_' \|\| ch == '.' \|\| ch == '*' \|\| ch == '?' \|\| ch == '+' \|\| ch == '-' \|\| ch == '[' \|\| ch == ']') ){
417
418	++ichT;
419	}
420	break;
421	}
422
423	cchToken = ichT - ichToken;
424
425	return true;
426	}
427
428	bool SelectionCompiler::compileCommand(const std::vector<Token>& ltoken) {
429	const Token& tokenCommand = ltoken[0];
430	int tokCommand = tokenCommand.tok;
431
432	atokenCommand = ltoken;
433	if ((tokCommand & Token::expressionCommand) != 0 && !compileExpression()) {
434	return false;
435	}
436
437	return true;
438	}
439
440	bool SelectionCompiler::compileExpression() {
441	/** todo */
442	int i = 1;
443	int tokCommand = atokenCommand[0].tok;
444	if (tokCommand == Token::define) {
445	i = 2;
446	} else if ((tokCommand & Token::embeddedExpression) != 0) {
447	// look for the open parenthesis
448	while (i < atokenCommand.size() &&
449	atokenCommand[i].tok != Token::leftparen)
450	++i;
451	}
452
453	if (i >= atokenCommand.size()) {
454	return true;
455	}
456	return compileExpression(i);
457	}
458
459
460	bool SelectionCompiler::addTokenToPostfix(const Token& token) {
461	ltokenPostfix.push_back(token);
462	return true;
463	}
464
465	bool SelectionCompiler::compileExpression(int itoken) {
466	ltokenPostfix.clear();
467	for (int i = 0; i < itoken; ++i) {
468	addTokenToPostfix(atokenCommand[i]);
469	}
470
471	atokenInfix = atokenCommand;
472	itokenInfix = itoken;
473
474	addTokenToPostfix(Token::tokenExpressionBegin);
475	if (!clauseOr()) {
476	return false;
477	}
478
479	addTokenToPostfix(Token::tokenExpressionEnd);
480	if (itokenInfix != atokenInfix.size()) {
481	return endOfExpressionExpected();
482	}
483
484	atokenCommand = ltokenPostfix;
485	return true;
486	}
487
488	Token SelectionCompiler::tokenNext() {
489	if (itokenInfix == atokenInfix.size()) {
490	return Token();
491	}
492	return atokenInfix[itokenInfix++];
493	}
494
495	boost::any SelectionCompiler::valuePeek() {
496	if (itokenInfix == atokenInfix.size()) {
497	return boost::any();
498	} else {
499	return atokenInfix[itokenInfix].value;
500	}
501	}
502
503	int SelectionCompiler::tokPeek() {
504	if (itokenInfix == atokenInfix.size()) {
505	return 0;
506	}else {
507	return atokenInfix[itokenInfix].tok;
508	}
509	}
510
511	bool SelectionCompiler::clauseOr() {
512	if (!clauseAnd()) {
513	return false;
514	}
515
516	while (tokPeek() == Token::opOr) {
517	Token tokenOr = tokenNext();
518	if (!clauseAnd()) {
519	return false;
520	}
521	addTokenToPostfix(tokenOr);
522	}
523	return true;
524	}
525
526	bool SelectionCompiler::clauseAnd() {
527	if (!clauseNot()) {
528	return false;
529	}
530
531	while (tokPeek() == Token::opAnd) {
532	Token tokenAnd = tokenNext();
533	if (!clauseNot()) {
534	return false;
535	}
536	addTokenToPostfix(tokenAnd);
537	}
538	return true;
539	}
540
541	bool SelectionCompiler::clauseNot() {
542	if (tokPeek() == Token::opNot) {
543	Token tokenNot = tokenNext();
544	if (!clauseNot()) {
545	return false;
546	}
547	return addTokenToPostfix(tokenNot);
548	}
549	return clausePrimitive();
550	}
551
552	bool SelectionCompiler::clausePrimitive() {
553	int tok = tokPeek();
554	switch (tok) {
555	case Token::within:
556	return clauseWithin();
557
558	case Token::asterisk:
559	case Token::identifier:
560	return clauseChemObjName();
561
562	case Token::integer :
563	return clauseIndex();
564	default:
565	if ((tok & Token::atomproperty) == Token::atomproperty) {
566	return clauseComparator();
567	}
568	if ((tok & Token::predefinedset) != Token::predefinedset) {
569	break;
570	}
571	// fall into the code and below and just add the token
572	case Token::all:
573	case Token::none:
574	return addTokenToPostfix(tokenNext());
575	case Token::leftparen:
576	tokenNext();
577	if (!clauseOr()) {
578	return false;
579	}
580	if (tokenNext().tok != Token::rightparen) {
581	return rightParenthesisExpected();
582	}
583	return true;
584	}
585	return unrecognizedExpressionToken();
586	}
587
588	bool SelectionCompiler::clauseComparator() {
589	Token tokenAtomProperty = tokenNext();
590	Token tokenComparator = tokenNext();
591	if ((tokenComparator.tok & Token::comparator) == 0) {
592	return comparisonOperatorExpected();
593	}
594
595	Token tokenValue = tokenNext();
596	if (tokenValue.tok != Token::integer && tokenValue.tok != Token::decimal) {
597	return numberExpected();
598	}
599
600	float val;
601	if (tokenValue.value.type() == typeid(int)) {
602	val = boost::any_cast<int>(tokenValue.value);
603	} else if (tokenValue.value.type() == typeid(float)) {
604	val = boost::any_cast<float>(tokenValue.value);
605	} else {
606	return false;
607	}
608
609	boost::any floatVal;
610	floatVal = val;
611	return addTokenToPostfix(Token(tokenComparator.tok,
612	tokenAtomProperty.tok, floatVal));
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	case Token::decimal:
626	distance = tokenDistance.value;
627	break;
628	default:
629	return numberOrKeywordExpected();
630	}
631
632	if (tokenNext().tok != Token::opOr) { // ,
633	return commaExpected();
634	}
635
636	if (! clauseOr()) { // expression
637	return false;
638	}
639
640	if (tokenNext().tok != Token::rightparen) { // )T
641	return rightParenthesisExpected();
642	}
643
644	return addTokenToPostfix(Token(Token::within, distance));
645	}
646
647	bool SelectionCompiler::clauseChemObjName() {
648	Token token = tokenNext();
649	if (token.tok == Token::identifier && token.value.type() == typeid(std::string)) {
650
651	std::string name = boost::any_cast<std::string>(token.value);
652	if (isNameValid(name)) {
653	return addTokenToPostfix(Token(Token::name, name));
654	} else {
655	return compileError("invalid name: " + name);
656	}
657	}
658
659	return false;
660
661	}
662
663	bool SelectionCompiler::isNameValid(const std::string& name) {
664	int nbracket = 0;
665	int ndot = 0;
666	for (int i =0 ; i < name.size(); ++i) {
667	switch(name[i]) {
668
669	case '[' :
670	++nbracket;
671	break;
672	case ']' :
673	--nbracket;
674	break;
675	case '.' :
676	++ndot;
677	break;
678	}
679	}
680
681	//only allow 3 dots at most
682	return (ndot <=3 && nbracket == 0) ? true : false;
683	}
684
685	bool SelectionCompiler::clauseIndex(){
686	Token token = tokenNext();
687	if (token.tok == Token::integer) {
688	int index = boost::any_cast<int>(token.value);
689	int tok = tokPeek();
690	std::cout << "Token::to is " << Token::to << ", tok = " << tok << std::endl;
691	if (tok == Token::to) {
692	tokenNext();
693	tok = tokPeek();
694	if (tok != Token::integer) {
695	return numberExpected();
696	}
697
698	boost::any intVal = tokenNext().value;
699	int first = index;
700	if (intVal.type() != typeid(int)){
701	return false;
702	}
703	int second = boost::any_cast<int>(intVal);
704
705	return addTokenToPostfix(Token(Token::index, boost::any(std::make_pair(first, second))));
706
707	}else {
708	return addTokenToPostfix(Token(Token::index, boost::any(index)));
709	}
710	} else {
711	return numberExpected();
712	}
713	}
714
715	}