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"
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;
    std::vector<Token> ltoken;

    //Token tokenCommand = null;
    int tokCommand = Token.nada;

    for ( ; true; ichToken += cchToken) {
        if (lookingAtLeadingWhitespace())
            continue;
        if (lookingAtComment())
            continue;
        boolean endOfLine = lookingAtEndOfLine();
        if (endOfLine || lookingAtEndOfStatement()) {
            if (tokCommand != Token.nada) {
                if (! compileCommand(ltoken)) {
                    return false;
                }
            lltoken.push_back(atokenCommand);
            /** @todo*/
            int iCommand = lltoken.size();
            lineNumbers[iCommand] = lineCurrent;
            lineIndices[iCommand] = (short) 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, new Float(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(new Token(Token.integer, val, intString));/**@todo*/
                continue;
            }
        }
      
        if (lookingAtLookupToken()) {
            std::string ident = script.subst(ichToken, ichToken + cchToken);

            /**@todo */
            Token token = (Token) Token.map.get(ident);
            if (token == NULL) {

            }
            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.set &&
                        (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();
    }

    aatokenCompiled.push_back(lltoken);
    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;
    boolean 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() {
    StringBuffer sb = new StringBuffer(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(ch);
    }

    return "" + sb;
}

static 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(boolean allowNegative) {
    if (ichToken == cchScript) {
        return false;
    }
    
    int ichT = ichToken;
    if (script[ichT] == '-') {
        ++ichT;
    }
    boolean 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(boolean 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::isalpha(ch = script[ichT]) ||std::isdigit(ch) ||
                ch == '_' || ch == '?') ||(ch == '^' && ichT > ichToken && std::isdigit(script[ichT - 1]))) {
                // hack for insertion codes embedded in an atom expression :-(
                // select c3^a
                ++ichT;
            }
        break;
    }
    cchToken = ichT - ichToken;
    return true;
}

bool SelectionCompiler::compileCommand(Vector ltoken) {
    /** todo */
    Token tokenCommand = (Token)ltoken.firstElement();
    int tokCommand = tokenCommand.tok;
        
    atokenCommand = new Token[ltoken.size()];
    ltoken.copyInto(atokenCommand);
    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.length &&
             atokenCommand[i].tok != Token.leftparen)
        ++i;
    }
    if (i >= atokenCommand.length)
      return true;
    return compileExpression(i);
  }

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

bool SelectionCompiler::compileExpression(int itoken) {
    ltokenPostfix = new Vector();
    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.length) {
        return endOfExpressionExpected();
    }

    atokenCommand = ltokenPostfix;
    return true;
}

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

Object SelectionCompiler::valuePeek() {
    if (itokenInfix == atokenInfix.length) {
        return null;
    } else {
        return atokenInfix[itokenInfix].value;
    }
}

int SelectionCompiler::tokPeek() {
    if (itokenInfix == atokenInfix.length) {
        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.hyphen: // selecting a negative residue spec
        case Token.integer:
        case Token.seqcode:
        case Token.asterisk:
        case Token.leftsquare:
        case Token.identifier:
        case Token.x:
        case Token.y:
        case Token.z:
        case Token.colon:
            return clauseResidueSpec();
        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) {
        return integerExpected();
    }
    int val = tokenValue.intValue;
    // note that a comparator instruction is a complicated instruction
    // int intValue is the tok of the property you are comparing
    // the value against which you are comparing is stored as an Integer
    // in the object value
    return addTokenToPostfix(new Token(tokenComparator.tok,
                       tokenAtomProperty.tok,
                       new Integer(val)));
}

bool SelectionCompiler::clauseWithin() {
    tokenNext();                             // WITHIN
    if (tokenNext().tok != Token.leftparen) {  // (
        return leftParenthesisExpected();
    }
    
    Object distance;
    Token tokenDistance = tokenNext();       // distance
    switch(tokenDistance.tok) {
        case Token.integer:
            distance = new Float((tokenDistance.intValue * 4) / 1000f);
            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(new Token(Token.within, distance));
}

bool SelectionCompiler:: clauseChemObject() {
}

bool SelectionCompiler:: clauseMolecule() {
}

bool SelectionCompiler:: clauseMolName() {
}

bool SelectionCompiler:: clauseMolIndex() {
}

bool SelectionCompiler:: clauseName() {
}

bool SelectionCompiler:: clauseIndex() {
}

bool SelectionCompiler:: clauseStuntDoubleName() {
}

bool SelectionCompiler:: clauseStuntDoubleIndex() {
}

}
Revision:	279
Committed:	Wed Feb 2 12:51:25 2005 UTC (20 years, 8 months ago) by tim
File size:	19659 byte(s)
Log Message:	add SelectionCompiler.cpp
#	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	namespace oopse {
44
45	bool SelectionCompiler::compile(const std::string& filename, const std::string& script) {
46
47	this->filename = filename;
48	this->script = script;
49	lineNumbers.clear();
50	lineIndices.clear();
51	aatokenCompiled.clear();
52
53	if (internalcompile()) {
54	return true;
55	}
56
57	int icharEnd;
58	if ((icharEnd = script.find('\r', ichCurrentCommand)) == std::string:npos &&
59	(icharEnd = script.find('\n', ichCurrentCommand)) == std::string:npos) {
60	icharEnd = script.size();
61	}
62	errorLine = script.substr(ichCurrentCommand, icharEnd);
63	return false;
64	}
65
66	bool SelectionCompiler::internalCompile(){
67
68	cchScript = script.size();
69	ichToken = 0;
70	lineCurrent = 1;
71
72	error = false;
73
74	std::vector<Token> lltoken;
75	std::vector<Token> ltoken;
76
77	//Token tokenCommand = null;
78	int tokCommand = Token.nada;
79
80	for ( ; true; ichToken += cchToken) {
81	if (lookingAtLeadingWhitespace())
82	continue;
83	if (lookingAtComment())
84	continue;
85	boolean endOfLine = lookingAtEndOfLine();
86	if (endOfLine \|\| lookingAtEndOfStatement()) {
87	if (tokCommand != Token.nada) {
88	if (! compileCommand(ltoken)) {
89	return false;
90	}
91	lltoken.push_back(atokenCommand);
92	/** @todo*/
93	int iCommand = lltoken.size();
94	lineNumbers[iCommand] = lineCurrent;
95	lineIndices[iCommand] = (short) 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, new Float(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(new Token(Token.integer, val, intString));/*@todo/
129	continue;
130	}
131	}
132
133	if (lookingAtLookupToken()) {
134	std::string ident = script.subst(ichToken, ichToken + cchToken);
135
136	/*@todo /
137	Token token = (Token) Token.map.get(ident);
138	if (token == NULL) {
139
140	}
141	Token token(Token.identifier, ident);
142
143
144	int tok = token.tok;
145
146	switch (tokCommand) {
147	case Token.nada:
148	ichCurrentCommand = ichToken;
149	//tokenCommand = token;
150	tokCommand = tok;
151	if ((tokCommand & Token.command) == 0)
152	return commandExpected();
153	break;
154
155	case Token.define:
156	if (ltoken.size() == 1) {
157	// we are looking at the variable name
158	if (tok != Token.identifier &&
159	(tok & Token.predefinedset) != Token.predefinedset)
160	return invalidExpressionToken(ident);
161	} else {
162	// we are looking at the expression
163	if (tok != Token.identifier && tok != Token.set &&
164	(tok & (Token.expression \| Token.predefinedset)) == 0)
165	return invalidExpressionToken(ident);
166	}
167
168	break;
169
170	case Token.select:
171	if (tok != Token.identifier && (tok & Token.expression) == 0)
172	return invalidExpressionToken(ident);
173	break;
174	}
175	ltoken.push_back(token);
176	continue;
177	}
178
179	if (ltoken.size() == 0) {
180	return commandExpected();
181	}
182
183	return unrecognizedToken();
184	}
185
186	aatokenCompiled.push_back(lltoken);
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	boolean previousCharBackslash = false;
240	while (ichT < cchScript) {
241	ch = script.[ichT++];
242	if (ch == '"' && !previousCharBackslash)
243	break;
244	previousCharBackslash = ch == '\\' ? !previousCharBackslash : false;
245	}
246	cchToken = ichT - ichToken;
247	return true;
248	}
249
250
251	std::string SelectionCompiler::getUnescapedStringLiteral() {
252	StringBuffer sb = new StringBuffer(cchToken - 2);
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(ch);
296	}
297
298	return "" + sb;
299	}
300
301	static 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(boolean allowNegative) {
323	if (ichToken == cchScript) {
324	return false;
325	}
326
327	int ichT = ichToken;
328	if (script[ichT] == '-') {
329	++ichT;
330	}
331	boolean 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(boolean 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::isalpha(ch = script[ichT]) \|\|std::isdigit(ch) \|\|
422	ch == '_' \|\| ch == '?') \|\|(ch == '^' && ichT > ichToken && std::isdigit(script[ichT - 1]))) {
423	// hack for insertion codes embedded in an atom expression :-(
424	// select c3^a
425	++ichT;
426	}
427	break;
428	}
429	cchToken = ichT - ichToken;
430	return true;
431	}
432
433	bool SelectionCompiler::compileCommand(Vector ltoken) {
434	/** todo */
435	Token tokenCommand = (Token)ltoken.firstElement();
436	int tokCommand = tokenCommand.tok;
437
438	atokenCommand = new Token[ltoken.size()];
439	ltoken.copyInto(atokenCommand);
440	if ((tokCommand & Token.expressionCommand) != 0 && !compileExpression()) {
441	return false;
442	}
443	return true;
444	}
445
446	bool SelectionCompiler::compileExpression() {
447	/** todo */
448	int i = 1;
449	int tokCommand = atokenCommand[0].tok;
450	if (tokCommand == Token.define)
451	i = 2;
452	else if ((tokCommand & Token.embeddedExpression) != 0) {
453	// look for the open parenthesis
454	while (i < atokenCommand.length &&
455	atokenCommand[i].tok != Token.leftparen)
456	++i;
457	}
458	if (i >= atokenCommand.length)
459	return true;
460	return compileExpression(i);
461	}
462
463
464	bool SelectionCompiler::addTokenToPostfix(Token token) {
465	ltokenPostfix.push_back(token);
466	return true;
467	}
468
469	bool SelectionCompiler::compileExpression(int itoken) {
470	ltokenPostfix = new Vector();
471	for (int i = 0; i < itoken; ++i)
472	addTokenToPostfix(atokenCommand[i]);
473
474	atokenInfix = atokenCommand;
475	itokenInfix = itoken;
476
477	addTokenToPostfix(Token.tokenExpressionBegin);
478	if (!clauseOr()) {
479	return false;
480	}
481
482	addTokenToPostfix(Token.tokenExpressionEnd);
483	if (itokenInfix != atokenInfix.length) {
484	return endOfExpressionExpected();
485	}
486
487	atokenCommand = ltokenPostfix;
488	return true;
489	}
490
491	Token SelectionCompiler::tokenNext() {
492	if (itokenInfix == atokenInfix.length)
493	return null;
494	return atokenInfix[itokenInfix++];
495	}
496
497	Object SelectionCompiler::valuePeek() {
498	if (itokenInfix == atokenInfix.length) {
499	return null;
500	} else {
501	return atokenInfix[itokenInfix].value;
502	}
503	}
504
505	int SelectionCompiler::tokPeek() {
506	if (itokenInfix == atokenInfix.length) {
507	return 0;
508	}else {
509	return atokenInfix[itokenInfix].tok;
510	}
511	}
512
513	bool SelectionCompiler::clauseOr() {
514	if (!clauseAnd()) {
515	return false;
516	}
517
518	while (tokPeek() == Token.opOr) {
519	Token tokenOr = tokenNext();
520	if (!clauseAnd()) {
521	return false;
522	}
523	addTokenToPostfix(tokenOr);
524	}
525	return true;
526	}
527
528	bool SelectionCompiler::clauseAnd() {
529	if (!clauseNot()) {
530	return false;
531	}
532
533	while (tokPeek() == Token.opAnd) {
534	Token tokenAnd = tokenNext();
535	if (!clauseNot()) {
536	return false;
537	}
538	addTokenToPostfix(tokenAnd);
539	}
540	return true;
541	}
542
543	bool SelectionCompiler::clauseNot() {
544	if (tokPeek() == Token.opNot) {
545	Token tokenNot = tokenNext();
546	if (!clauseNot()) {
547	return false;
548	}
549	return addTokenToPostfix(tokenNot);
550	}
551	return clausePrimitive();
552	}
553
554	bool SelectionCompiler::clausePrimitive() {
555	int tok = tokPeek();
556	switch (tok) {
557	case Token.within:
558	return clauseWithin();
559	case Token.hyphen: // selecting a negative residue spec
560	case Token.integer:
561	case Token.seqcode:
562	case Token.asterisk:
563	case Token.leftsquare:
564	case Token.identifier:
565	case Token.x:
566	case Token.y:
567	case Token.z:
568	case Token.colon:
569	return clauseResidueSpec();
570	default:
571	if ((tok & Token.atomproperty) == Token.atomproperty) {
572	return clauseComparator();
573	}
574	if ((tok & Token.predefinedset) != Token.predefinedset) {
575	break;
576	}
577	// fall into the code and below and just add the token
578	case Token.all:
579	case Token.none:
580	return addTokenToPostfix(tokenNext());
581	case Token.leftparen:
582	tokenNext();
583	if (!clauseOr()) {
584	return false;
585	}
586	if (tokenNext().tok != Token.rightparen) {
587	return rightParenthesisExpected();
588	}
589	return true;
590	}
591	return unrecognizedExpressionToken();
592	}
593
594	bool SelectionCompiler::clauseComparator() {
595	Token tokenAtomProperty = tokenNext();
596	Token tokenComparator = tokenNext();
597	if ((tokenComparator.tok & Token.comparator) == 0) {
598	return comparisonOperatorExpected();
599	}
600
601	Token tokenValue = tokenNext();
602	if (tokenValue.tok != Token.integer) {
603	return integerExpected();
604	}
605	int val = tokenValue.intValue;
606	// note that a comparator instruction is a complicated instruction
607	// int intValue is the tok of the property you are comparing
608	// the value against which you are comparing is stored as an Integer
609	// in the object value
610	return addTokenToPostfix(new Token(tokenComparator.tok,
611	tokenAtomProperty.tok,
612	new Integer(val)));
613	}
614
615	bool SelectionCompiler::clauseWithin() {
616	tokenNext(); // WITHIN
617	if (tokenNext().tok != Token.leftparen) { // (
618	return leftParenthesisExpected();
619	}
620
621	Object distance;
622	Token tokenDistance = tokenNext(); // distance
623	switch(tokenDistance.tok) {
624	case Token.integer:
625	distance = new Float((tokenDistance.intValue * 4) / 1000f);
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(new Token(Token.within, distance));
647	}
648
649	bool SelectionCompiler:: clauseChemObject() {
650	}
651
652	bool SelectionCompiler:: clauseMolecule() {
653	}
654
655	bool SelectionCompiler:: clauseMolName() {
656	}
657
658	bool SelectionCompiler:: clauseMolIndex() {
659	}
660
661	bool SelectionCompiler:: clauseName() {
662	}
663
664	bool SelectionCompiler:: clauseIndex() {
665	}
666
667	bool SelectionCompiler:: clauseStuntDoubleName() {
668	}
669
670	bool SelectionCompiler:: clauseStuntDoubleIndex() {
671	}
672
673	}