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. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 *
 * 2. 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.
 *
 * SUPPORT OPEN SCIENCE!  If you use OpenMD or its source code in your
 * research, please cite the appropriate papers when you publish your
 * work.  Good starting points are:
 *                                                                      
 * [1]  Meineke, et al., J. Comp. Chem. 26, 252-271 (2005).             
 * [2]  Fennell & Gezelter, J. Chem. Phys. 124, 234104 (2006).          
 * [3]  Sun, Lin & Gezelter, J. Chem. Phys. 128, 234107 (2008).          
 * [4]  Kuang & Gezelter,  J. Chem. Phys. 133, 164101 (2010).
 * [5]  Vardeman, Stocker & Gezelter, J. Chem. Theory Comput. 7, 834 (2011).
 */

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

  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;
    }
      
    std::size_t 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)) {
        if (lookingAtDecimal((tokCommand) != 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)) {
        if (lookingAtInteger((tokCommand) != 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.empty()) {
        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 isInteger(script.substr(ichToken, cchToken).c_str());
  }

  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 */
    unsigned 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:
    case Token::hull:
      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 (unsigned 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();
      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:	2069
Committed:	Thu Mar 5 16:30:23 2015 UTC (10 years, 7 months ago) by gezelter
File size:	18526 byte(s)
Log Message:	More g++ warning fixes
#	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. Redistributions of source code must retain the above copyright
10	* notice, this list of conditions and the following disclaimer.
11	*
12	* 2. Redistributions in binary form must reproduce the above copyright
13	* notice, this list of conditions and the following disclaimer in the
14	* documentation and/or other materials provided with the
15	* distribution.
16	*
17	* This software is provided "AS IS," without a warranty of any
18	* kind. All express or implied conditions, representations and
19	* warranties, including any implied warranty of merchantability,
20	* fitness for a particular purpose or non-infringement, are hereby
21	* excluded. The University of Notre Dame and its licensors shall not
22	* be liable for any damages suffered by licensee as a result of
23	* using, modifying or distributing the software or its
24	* derivatives. In no event will the University of Notre Dame or its
25	* licensors be liable for any lost revenue, profit or data, or for
26	* direct, indirect, special, consequential, incidental or punitive
27	* damages, however caused and regardless of the theory of liability,
28	* arising out of the use of or inability to use software, even if the
29	* University of Notre Dame has been advised of the possibility of
30	* such damages.
31	*
32	* SUPPORT OPEN SCIENCE! If you use OpenMD or its source code in your
33	* research, please cite the appropriate papers when you publish your
34	* work. Good starting points are:
35	*
36	* [1] Meineke, et al., J. Comp. Chem. 26, 252-271 (2005).
37	* [2] Fennell & Gezelter, J. Chem. Phys. 124, 234104 (2006).
38	* [3] Sun, Lin & Gezelter, J. Chem. Phys. 128, 234107 (2008).
39	* [4] Kuang & Gezelter, J. Chem. Phys. 133, 164101 (2010).
40	* [5] Vardeman, Stocker & Gezelter, J. Chem. Theory Comput. 7, 834 (2011).
41	*/
42
43	#include "selection/SelectionCompiler.hpp"
44	#include "utils/StringUtils.hpp"
45	namespace OpenMD {
46
47	bool SelectionCompiler::compile(const std::string& filename,
48	const std::string& script) {
49
50	this->filename = filename;
51	this->script = script;
52	lineNumbers.clear();
53	lineIndices.clear();
54	aatokenCompiled.clear();
55
56	if (internalCompile()) {
57	return true;
58	}
59
60	std::size_t icharEnd;
61	if ((icharEnd = script.find('\r', ichCurrentCommand)) == std::string::npos &&
62	(icharEnd = script.find('\n', ichCurrentCommand)) == std::string::npos) {
63	icharEnd = script.size();
64	}
65	errorLine = script.substr(ichCurrentCommand, icharEnd);
66	return false;
67	}
68
69	bool SelectionCompiler::internalCompile(){
70
71	cchScript = script.size();
72	ichToken = 0;
73	lineCurrent = 1;
74
75	error = false;
76
77	//std::vector<Token> lltoken;
78	aatokenCompiled.clear();
79	std::vector<Token> ltoken;
80
81	Token tokenCommand;
82	int tokCommand = Token::nada;
83
84	for ( ; true; ichToken += cchToken) {
85	if (lookingAtLeadingWhitespace())
86	continue;
87	//if (lookingAtComment())
88	// continue;
89	bool endOfLine = lookingAtEndOfLine();
90	if (endOfLine \|\| lookingAtEndOfStatement()) {
91	if (tokCommand != Token::nada) {
92	if (! compileCommand(ltoken)) {
93	return false;
94	}
95	aatokenCompiled.push_back(atokenCommand);
96	lineNumbers.push_back(lineCurrent);
97	lineIndices.push_back(ichCurrentCommand);
98	ltoken.clear();
99	tokCommand = Token::nada;
100	}
101
102	if (ichToken < cchScript) {
103	if (endOfLine)
104	++lineCurrent;
105	continue;
106	}
107	break;
108	}
109
110	if (tokCommand != Token::nada) {
111	if (lookingAtString()) {
112	std::string str = getUnescapedStringLiteral();
113	ltoken.push_back(Token(Token::string, str));
114	continue;
115	}
116	//if ((tokCommand & Token::specialstring) != 0 &&
117	// lookingAtSpecialString()) {
118	// std::string str = script.substr(ichToken, ichToken + cchToken);
119	// ltoken.push_back(Token(Token::string, str));
120	// continue;
121	//}
122	//if (lookingAtDecimal((tokCommand & Token::negnums) != 0)) {
123	if (lookingAtDecimal((tokCommand) != 0)) {
124	float value = lexi_cast<float>(script.substr(ichToken, cchToken));
125	ltoken.push_back(Token(Token::decimal, boost::any(value)));
126	continue;
127	}
128	//if (lookingAtInteger((tokCommand & Token::negnums) != 0)) {
129	if (lookingAtInteger((tokCommand) != 0)) {
130
131	int val = lexi_cast<int>(script.substr(ichToken, cchToken));
132	ltoken.push_back(Token(Token::integer, boost::any(val)));
133	continue;
134	}
135	}
136
137	if (lookingAtLookupToken()) {
138	std::string ident = script.substr(ichToken, cchToken);
139	Token token;
140	Token* pToken = TokenMap::getInstance()->getToken(ident);
141	if (pToken != NULL) {
142	token = *pToken;
143	} else {
144	token = Token(Token::identifier, ident);
145	}
146
147	int tok = token.tok;
148
149	switch (tokCommand) {
150	case Token::nada:
151	ichCurrentCommand = ichToken;
152	//tokenCommand = token;
153	tokCommand = tok;
154	if ((tokCommand & Token::command) == 0)
155	return commandExpected();
156	break;
157
158	case Token::define:
159	if (ltoken.size() == 1) {
160	// we are looking at the variable name
161	if (tok != Token::identifier &&
162	(tok & Token::predefinedset) != Token::predefinedset)
163	return invalidExpressionToken(ident);
164	} else {
165	// we are looking at the expression
166	if (tok != Token::identifier &&
167	(tok & (Token::expression \| Token::predefinedset)) == 0)
168	return invalidExpressionToken(ident);
169	}
170
171	break;
172
173	case Token::select:
174	if (tok != Token::identifier && (tok & Token::expression) == 0)
175	return invalidExpressionToken(ident);
176	break;
177	}
178	ltoken.push_back(token);
179	continue;
180	}
181
182	if (ltoken.empty()) {
183	return commandExpected();
184	}
185
186	return unrecognizedToken();
187	}
188
189	return true;
190	}
191
192
193	bool SelectionCompiler::lookingAtLeadingWhitespace() {
194
195	int ichT = ichToken;
196	while (ichT < cchScript && std::isspace(script[ichT])) {
197	++ichT;
198	}
199	cchToken = ichT - ichToken;
200	return cchToken > 0;
201	}
202
203	bool SelectionCompiler::lookingAtEndOfLine() {
204	if (ichToken == cchScript)
205	return true;
206	int ichT = ichToken;
207	char ch = script[ichT];
208	if (ch == '\r') {
209	++ichT;
210	if (ichT < cchScript && script[ichT] == '\n')
211	++ichT;
212	} else if (ch == '\n') {
213	++ichT;
214	} else {
215	return false;
216	}
217	cchToken = ichT - ichToken;
218	return true;
219	}
220
221	bool SelectionCompiler::lookingAtEndOfStatement() {
222	if (ichToken == cchScript \|\| script[ichToken] != ';')
223	return false;
224	cchToken = 1;
225	return true;
226	}
227
228	bool SelectionCompiler::lookingAtString() {
229	if (ichToken == cchScript)
230	return false;
231	if (script[ichToken] != '"')
232	return false;
233	// remove support for single quote
234	// in order to use it in atom expressions
235	// char chFirst = script.charAt(ichToken);
236	// if (chFirst != '"' && chFirst != '\'')
237	// return false;
238	int ichT = ichToken + 1;
239	// while (ichT < cchScript && script.charAt(ichT++) != chFirst)
240	char ch;
241	bool previousCharBackslash = false;
242	while (ichT < cchScript) {
243	ch = script[ichT++];
244	if (ch == '"' && !previousCharBackslash)
245	break;
246	previousCharBackslash = ch == '\\' ? !previousCharBackslash : false;
247	}
248	cchToken = ichT - ichToken;
249
250	return true;
251	}
252
253
254	std::string SelectionCompiler::getUnescapedStringLiteral() {
255	/** @todo */
256	std::string sb(cchToken - 2, ' ');
257
258	int ichMax = ichToken + cchToken - 1;
259	int ich = ichToken + 1;
260
261	while (ich < ichMax) {
262	char ch = script[ich++];
263	if (ch == '\\' && ich < ichMax) {
264	ch = script[ich++];
265	switch (ch) {
266	case 'b':
267	ch = '\b';
268	break;
269	case 'n':
270	ch = '\n';
271	break;
272	case 't':
273	ch = '\t';
274	break;
275	case 'r':
276	ch = '\r';
277	// fall into
278	case '"':
279	case '\\':
280	case '\'':
281	break;
282	case 'x':
283	case 'u':
284	int digitCount = ch == 'x' ? 2 : 4;
285	if (ich < ichMax) {
286	int unicode = 0;
287	for (int k = digitCount; --k >= 0 && ich < ichMax; ) {
288	char chT = script[ich];
289	int hexit = getHexitValue(chT);
290	if (hexit < 0)
291	break;
292	unicode <<= 4;
293	unicode += hexit;
294	++ich;
295	}
296	ch = (char)unicode;
297	}
298	}
299	}
300	sb.append(1, ch);
301	}
302
303	return sb;
304	}
305
306	int SelectionCompiler::getHexitValue(char ch) {
307	if (ch >= '0' && ch <= '9')
308	return ch - '0';
309	else if (ch >= 'a' && ch <= 'f')
310	return 10 + ch - 'a';
311	else if (ch >= 'A' && ch <= 'F')
312	return 10 + ch - 'A';
313	else
314	return -1;
315	}
316
317	bool SelectionCompiler::lookingAtSpecialString() {
318	int ichT = ichToken;
319	char ch = script[ichT];
320	while (ichT < cchScript && ch != ';' && ch != '\r' && ch != '\n') {
321	++ichT;
322	}
323	cchToken = ichT - ichToken;
324	return cchToken > 0;
325	}
326
327	bool SelectionCompiler::lookingAtDecimal(bool allowNegative) {
328	if (ichToken == cchScript) {
329	return false;
330	}
331
332	int ichT = ichToken;
333	if (script[ichT] == '-') {
334	++ichT;
335	}
336	bool digitSeen = false;
337	char ch = 'X';
338	while (ichT < cchScript && std::isdigit(ch = script[ichT])) {
339	++ichT;
340	digitSeen = true;
341	}
342
343	if (ichT == cchScript \|\| ch != '.') {
344	return false;
345	}
346
347	// to support DMPC.1, let's check the character before the dot
348	if (ch == '.' && (ichT > 0) && std::isalpha(script[ichT - 1])) {
349	return false;
350	}
351
352	++ichT;
353	while (ichT < cchScript && std::isdigit(script[ichT])) {
354	++ichT;
355	digitSeen = true;
356	}
357	cchToken = ichT - ichToken;
358	return digitSeen;
359	}
360
361	bool SelectionCompiler::lookingAtInteger(bool allowNegative) {
362	if (ichToken == cchScript) {
363	return false;
364	}
365	int ichT = ichToken;
366	if (allowNegative && script[ichToken] == '-') {
367	++ichT;
368	}
369	int ichBeginDigits = ichT;
370	while (ichT < cchScript && std::isdigit(script[ichT])) {
371	++ichT;
372	}
373	if (ichBeginDigits == ichT) {
374	return false;
375	}
376	cchToken = ichT - ichToken;
377	return isInteger(script.substr(ichToken, cchToken).c_str());
378	}
379
380	bool SelectionCompiler::lookingAtLookupToken() {
381	if (ichToken == cchScript) {
382	return false;
383	}
384
385	int ichT = ichToken;
386	char ch;
387	switch (ch = script[ichT++]) {
388	case '(':
389	case ')':
390	case ',':
391	case '[':
392	case ']':
393	break;
394	case '&':
395	case '\|':
396	if (ichT < cchScript && script[ichT] == ch) {
397	++ichT;
398	}
399	break;
400	case '<':
401	case '=':
402	case '>':
403	if (ichT < cchScript && ((ch = script[ichT]) == '<' \|\| ch == '=' \|\| ch == '>')) {
404	++ichT;
405	}
406	break;
407	case '/':
408	case '!':
409	if (ichT < cchScript && script[ichT] == '=') {
410	++ichT;
411	}
412	break;
413	default:
414	if ((ch < 'a' \|\| ch > 'z') && (ch < 'A' && ch > 'Z') && ch != '_') {
415	return false;
416	}
417	case '*':
418	case '?': // include question marks in identifier for atom expressions
419	while (ichT < cchScript && !std::isspace(ch = script[ichT]) &&
420	(std::isalpha(ch) \|\|std::isdigit(ch) \|\| ch == '_' \|\| ch == '.' \|\| ch == '*' \|\| ch == '?' \|\| ch == '+' \|\| ch == '-' \|\| ch == '[' \|\| ch == ']') ){
421
422	++ichT;
423	}
424	break;
425	}
426
427	cchToken = ichT - ichToken;
428
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	unsigned 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	case Token::integer :
567	return clauseIndex();
568	default:
569	if ((tok & Token::atomproperty) == Token::atomproperty) {
570	return clauseComparator();
571	}
572	if ((tok & Token::predefinedset) != Token::predefinedset) {
573	break;
574	}
575	// fall into the code and below and just add the token
576	case Token::all:
577	case Token::none:
578	case Token::hull:
579	return addTokenToPostfix(tokenNext());
580	case Token::leftparen:
581	tokenNext();
582	if (!clauseOr()) {
583	return false;
584	}
585	if (tokenNext().tok != Token::rightparen) {
586	return rightParenthesisExpected();
587	}
588	return true;
589	}
590	return unrecognizedExpressionToken();
591	}
592
593	bool SelectionCompiler::clauseComparator() {
594	Token tokenAtomProperty = tokenNext();
595	Token tokenComparator = tokenNext();
596	if ((tokenComparator.tok & Token::comparator) == 0) {
597	return comparisonOperatorExpected();
598	}
599
600	Token tokenValue = tokenNext();
601	if (tokenValue.tok != Token::integer && tokenValue.tok != Token::decimal) {
602	return numberExpected();
603	}
604
605	float val;
606	if (tokenValue.value.type() == typeid(int)) {
607	val = boost::any_cast<int>(tokenValue.value);
608	} else if (tokenValue.value.type() == typeid(float)) {
609	val = boost::any_cast<float>(tokenValue.value);
610	} else {
611	return false;
612	}
613
614	boost::any floatVal;
615	floatVal = val;
616	return addTokenToPostfix(Token(tokenComparator.tok,
617	tokenAtomProperty.tok, floatVal));
618	}
619
620	bool SelectionCompiler::clauseWithin() {
621	tokenNext(); // WITHIN
622	if (tokenNext().tok != Token::leftparen) { // (
623	return leftParenthesisExpected();
624	}
625
626	boost::any distance;
627	Token tokenDistance = tokenNext(); // distance
628	switch(tokenDistance.tok) {
629	case Token::integer:
630	case Token::decimal:
631	distance = tokenDistance.value;
632	break;
633	default:
634	return numberOrKeywordExpected();
635	}
636
637	if (tokenNext().tok != Token::opOr) { // ,
638	return commaExpected();
639	}
640
641	if (! clauseOr()) { // expression
642	return false;
643	}
644
645	if (tokenNext().tok != Token::rightparen) { // )T
646	return rightParenthesisExpected();
647	}
648
649	return addTokenToPostfix(Token(Token::within, distance));
650	}
651
652	bool SelectionCompiler::clauseChemObjName() {
653	Token token = tokenNext();
654	if (token.tok == Token::identifier && token.value.type() == typeid(std::string)) {
655
656	std::string name = boost::any_cast<std::string>(token.value);
657	if (isNameValid(name)) {
658	return addTokenToPostfix(Token(Token::name, name));
659	} else {
660	return compileError("invalid name: " + name);
661	}
662	}
663
664	return false;
665
666	}
667
668	bool SelectionCompiler::isNameValid(const std::string& name) {
669	int nbracket = 0;
670	int ndot = 0;
671	for (unsigned int i = 0 ; i < name.size(); ++i) {
672	switch(name[i]) {
673
674	case '[' :
675	++nbracket;
676	break;
677	case ']' :
678	--nbracket;
679	break;
680	case '.' :
681	++ndot;
682	break;
683	}
684	}
685
686	//only allow 3 dots at most
687	return (ndot <=3 && nbracket == 0) ? true : false;
688	}
689
690	bool SelectionCompiler::clauseIndex(){
691	Token token = tokenNext();
692	if (token.tok == Token::integer) {
693	int index = boost::any_cast<int>(token.value);
694	int tok = tokPeek();
695	if (tok == Token::to) {
696	tokenNext();
697	tok = tokPeek();
698	if (tok != Token::integer) {
699	return numberExpected();
700	}
701
702	boost::any intVal = tokenNext().value;
703	int first = index;
704	if (intVal.type() != typeid(int)){
705	return false;
706	}
707	int second = boost::any_cast<int>(intVal);
708
709	return addTokenToPostfix(Token(Token::index, boost::any(std::make_pair(first, second))));
710
711	}else {
712	return addTokenToPostfix(Token(Token::index, boost::any(index)));
713	}
714	} else {
715	return numberExpected();
716	}
717	}
718
719	}