Method advance_pattern_min2

lib/matcher.cpp:1971–2027

Minimal 2 byte long patterns with min=2 using bitap hashed pairs then PM

/// Minimal 2 byte long patterns with min=2 using bitap hashed pairs then PM
bool Matcher::advance_pattern_min2(size_t loc)
{
  const Pattern::Bitap *tap = pat_->tap_;
  uint32_t state = ~0;
  while (true)
  {
    const char *s = buf_ + loc;
    const char *e = buf_ + end_ - (Pattern::Const::PM_M - 1);
    uint8_t c0 = static_cast<uint8_t>(*s);
    while (s < e)
    {
      uint8_t c1 = static_cast<uint8_t>(*++s);
      state = (state << 1) | tap[Pattern::bihash(c0, c1)];
      c0 = c1;
      if ((state & 2) == 0 && pat_->predict_match(s - 2))
      {
        size_t k = s - buf_ - 2;
        set_current(k);
        return true;
      }
    }
    loc = s - buf_;
    size_t m = std::min<size_t>(1, loc); // to clamp loc - 1
    set_current_and_peek_more(loc - m); // clamp loc - 1
    loc = cur_ + m;
    if (loc + (Pattern::Const::PM_M - 1) >= end_ && eof_)
    {
      // keep going, we may have matches in the last bytes
      s = buf_ + loc;
      e = buf_ + end_ - 1;
      c0 = static_cast<uint8_t>(s[0]);
      while (s < e)
      {
        uint8_t c1 = static_cast<uint8_t>(*++s);
        state = (state << 1) | tap[Pattern::bihash(c0, c1)];
        c0 = c1;
        if ((state & 2) == 0)
        {
          size_t k = s - buf_ - 2;
          set_current(k);
          return true;
        }
      }
      uint8_t c1 = 0; // reached the end
      state = (state << 1) | tap[Pattern::bihash(c0, c1)];
      if ((state & 2) == 0)
      {
        size_t k = s - buf_ - 1;
        set_current(k);
        return true;
      }
      loc = s - buf_;
      set_current(loc);
      return false;
    }
  }
}