Method advance_pattern_mink

lib/matcher.cpp:2091–2187

Minimal k byte long patterns (MIN>=k) using bitap hashed pairs and PMk+PMx

/// Minimal k byte long patterns (MIN>=k) using bitap hashed pairs and PMk+PMx
bool Matcher::advance_pattern_mink(size_t loc)
{
  const Pattern::Bitap *tap = pat_->tap_;
  const uint16_t min = pat_->min_;
  const uint16_t sub = (min < Pattern::Const::PM_M + 2 ? Pattern::Const::PM_M + 2 - min : 0);
  const uint32_t mask = 1 << (min - 1);
  uint32_t state1 = ~0;
  uint32_t state2 = ~0;
  while (true)
  {
    const char *s = buf_ + loc;
    const char *e = buf_ + end_ - sub;
    uint8_t c0 = static_cast<uint8_t>(*s);
    while (s < e)
    {
      uint8_t c1 = static_cast<uint8_t>(*++s);
      state2 = (state1 << 1) | tap[Pattern::bihash(c0, c1)];
      c0 = static_cast<uint8_t>(*++s);
      state1 = (state2 << 1) | tap[Pattern::bihash(c1, c0)];
      if ((state2 & mask) == 0 && pat_->predict_match_min(s - min - 1))
      {
        size_t k = s - buf_ - min - 1;
        set_current(k);
        return true;
      }
      if ((state1 & mask) == 0 && pat_->predict_match_min(s - min))
      {
        size_t k = s - buf_ - min;
        set_current(k);
        return true;
      }
    }
    loc = s - buf_;
    size_t m = std::min<size_t>(min - 1, loc); // to clamp loc - min + 1
    set_current_and_peek_more(loc - m); // clamp loc - min + 1
    loc = cur_ + m;
    if (loc + sub >= end_ && eof_)
    {
      // keep going, we may have a match in the last bytes
      s = buf_ + loc;
      e = buf_ + end_ - 2;
      c0 = static_cast<uint8_t>(s[0]);
      while (s < e)
      {
        uint8_t c1 = static_cast<uint8_t>(*++s);
        state2 = (state1 << 1) | tap[Pattern::bihash(c0, c1)];
        c0 = static_cast<uint8_t>(*++s);
        state1 = (state2 << 1) | tap[Pattern::bihash(c1, c0)];
        if ((state2 & mask) == 0)
        {
          size_t k = s - buf_ - min - 1;
          set_current(k);
          return true;
        }
        if ((state1 & mask) == 0)
        {
          size_t k = s - buf_ - min;
          set_current(k);