source: indexers/trunk/mgpp/text/GSDLQueryLex.cpp@ 18340

/**************************************************************************
 *
 * GSDLQueryLex.cpp -- Lexical analyser for a simple query language
 * Copyright (C) 2000  Rodger McNab
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
 *
 **************************************************************************/

#include "GSDLQueryLex.h"
#include "unitool.h"
#include "words.h"

inline void AddNChar (UCArray::const_iterator &here,
              UCArray &text,
              int len) {
  if (text.capacity() < text.size() + len + 1) {
    text.reserve(text.size() + len + 1);
  }
  while (len > 0) {
    text.push_back (*here++);
    --len;
  }
}

static bool ParseInteger (UCArray::const_iterator &here,
              UCArray::const_iterator end,
              LexEl &el) {
  el.Clear();

  // this version of end is used in unitool
  //  UCArray::const_iterator endMinus1 = end-1;
  const unsigned char* endMinus1 = &*(end - 1);

  int charLen;
  unsigned short c; // one character lookahead
  charLen = parse_utf8_char (&*here, endMinus1, &c);

  // check for positive or negative
  bool neg = false;
  if (c == '+') {
    AddNChar (here, el.text, charLen);
    charLen = parse_utf8_char (&*here, endMinus1, &c);
  } else if (c == '-') {
    neg = true;
    AddNChar (here, el.text, charLen);
    charLen = parse_utf8_char (&*here, endMinus1, &c);
  }

  // read in number part
  int numeric=0;
  el.num = 0;
  el.lexType = IntegerE;
  /* stop integers at 4 digits */
  while (c >= '0' && c <= '9'&& ++numeric<=MAXNUMERIC) {
    el.num = el.num*10 + c - '0';
    AddNChar (here, el.text, charLen);
    charLen = parse_utf8_char (&*here, endMinus1, &c);
  }

  if (neg) el.num *= -1;

  return (!el.text.empty());
}

static bool ParsePotentialInteger(UCArray::const_iterator &here,
                  UCArray::const_iterator end,
                  LexEl &el) {
  el.Clear();

  // this version of end is used in unitool
  //UCArray::const_iterator endMinus1 = end-1;
  const unsigned char* endMinus1 = &*(end - 1);

  int charLen=0;
  int length=0;
  unsigned short c; // one character lookahead
  charLen = parse_utf8_char (&*here, endMinus1, &c);

  // read in number part
  int numeric=0;
  el.num = 0;
  el.lexType = IntegerE;
  /* stop integers at 4 digits */
  while (c >= '0' && c <= '9'&& ++numeric<=MAXNUMERIC) {
    el.num = el.num*10 + c - '0';
    AddNChar (here, el.text, charLen);
    length += charLen;
    charLen = parse_utf8_char (&*here, endMinus1, &c);
  }
  // check the next character -if it is a letter, then have a term, not an integer
  if (!is_unicode_letter(c)) {
    // this was just an integer
    return (!el.text.empty());
  }
  // else its a term
  el.lexType = TermE;
  el.num = 0;
  /* this bit taken from ParseIndexWord in words.h*/
  while (length+charLen<=MAXSTEMLEN && charLen > 0 &&
     (is_unicode_letter(c) || (is_unicode_digit(c) &&
                   ++numeric <= MAXNUMERIC))) {
    AddNChar (here, el.text, charLen);
    length += charLen;
    charLen = parse_utf8_char (&*here, endMinus1, &c);
  }

  return (!el.text.empty());
}
static bool ParseTerm (UCArray::const_iterator &here,
               UCArray::const_iterator end,
               UCArray &text) {
  if (here == end)
    return false;

  //UCArray::const_iterator endMinus1 = end-1;
  const unsigned char* endMinus1 = &*(end - 1);
  const unsigned char* new_here = ParseIndexWord (&*here, endMinus1, text);
  here += (new_here - &*here); // advance iterator by number of chars advanced
  return !text.empty();
}


bool ParseLexEl (UCArray::const_iterator &here,
         UCArray::const_iterator end,
         LexEl &el) {
  el.Clear();

  // strange things can happen if here == end == 0
  if (here == end)
    return false;
  
  // this version of end is used in unitool
  //UCArray::const_iterator endMinus1 = end-1;
  const unsigned char* endMinus1 = &*(end - 1);

  // ignore all white space
  int charLen;
  unsigned short c; // one character lookahead
  charLen = parse_utf8_char (&*here, endMinus1, &c);
  while (here != end && is_unicode_space (c)) {
    here += charLen;
    if (here == end) break;
    charLen = parse_utf8_char (&*here, endMinus1, &c);
  }
  if (here == end) return false;

  if (c == '(') {
    el.lexType = OpenBracketE;
    AddNChar (here, el.text, charLen);
    return true;
    
  } else if (c == ')') {
    el.lexType = CloseBracketE;
    AddNChar (here, el.text, charLen);
    return true;

  } else if (c =='[') {
    el.lexType = OpenSquareBracketE;
    AddNChar (here, el.text, charLen);
    return true;
  
  } else if (c ==']') {
    el.lexType = CloseSquareBracketE;
    AddNChar (here, el.text, charLen);
    return true;
  
  } else if (c == '\"') {
    el.lexType = QuoteE;
    AddNChar (here, el.text, charLen);
    return true;
    
  } else if (c == '/') {
    el.lexType = TermWeightE;
    AddNChar (here, el.text, charLen);
    return true;
    
  } else if (c == '#') {
    el.lexType = StemMethodE;
    AddNChar (here, el.text, charLen);
    return true;
    
  } else if (c == '*') {
    el.lexType = StarE;
    AddNChar (here, el.text, charLen);
    return true;
    
  } else if (c == '^') {
    el.lexType = RangeE;
    AddNChar (here, el.text, charLen);
    return true;
    
  } else if (c == '@') {
    el.lexType = AtE;
    AddNChar (here, el.text, charLen);
    return true;
    
  } else if (c == ':') {
    el.lexType = TagE;
    AddNChar (here, el.text, charLen);
    return true;
    
  } else if (c=='&') {
    el.lexType = AndOpE;
    AddNChar (here, el.text, charLen);
    return true;
    
  } else if (c == '|') {
    el.lexType = OrOpE;
    AddNChar (here, el.text, charLen);
    return true;
    
  } else if (c == '!') {
    el.lexType = NotOpE;
    AddNChar (here, el.text, charLen);
    return true;
    
  } else if (c == '+' || c == '-' ) {
    return  ParseInteger (here, end, el);
  }

  else if (c >= '0' && c <= '9') {
    return ParsePotentialInteger (here, end, el);
  }

  // assume it is a term of some sort
  if (!ParseTerm (here, end, el.text))  {
    // parse term returns false if it hasn't parsed anything that is a term
    // here should be the same as it was before 
    el.lexType = UnknownE;
    AddNChar (here, el.text, charLen);
    return true;
  }
  //return false;

  //UCArray AND; SetCStr (AND, "AND");
  //if (el.text == AND) {
  if (UCArrayCStrEquals(el.text, "AND")) {
    el.lexType = AndOpE;
    return true;
  }
  //UCArray OR; SetCStr (OR, "OR");
  //if (el.text == OR) {
  if (UCArrayCStrEquals(el.text, "OR")) {
    el.lexType = OrOpE;
    return true;
  }
  //UCArray NOT; SetCStr (NOT, "NOT");
  //if (el.text == NOT) {
  if (UCArrayCStrEquals(el.text, "NOT")) {
    el.lexType = NotOpE;
    return true;
  }
  UCArray NEAR; SetCStr (NEAR, "NEAR", 4);
  if (PrefixLen(el.text, NEAR)==4) {
    el.lexType = NearOpE;
    return true;
  }
  UCArray WITHIN; SetCStr (WITHIN, "WITHIN", 6);
  if (PrefixLen(el.text, WITHIN)==6) {
    el.lexType = WithinOpE;
    return true;
  }
  
  el.lexType = TermE;
  return true;
}