deflate.java

纯Java实现的ZIP文件压缩解压类库,JDK中的ZIP类库源码中有一些native方法

      }
    }

    if(match_available!=0) {
      bflush=_tr_tally(0, window[strstart-1]&0xff);
      match_available = 0;
    }
    flush_block_only(flush == Z_FINISH);

    if(strm.avail_out==0){
      if(flush == Z_FINISH) return FinishStarted;
      else return NeedMore;
    }

    return flush == Z_FINISH ? FinishDone : BlockDone;
  }

  int longest_match(int cur_match){
    int chain_length = max_chain_length; // max hash chain length
    int scan = strstart;                 // current string
    int match;                           // matched string
    int len;                             // length of current match
    int best_len = prev_length;          // best match length so far
    int limit = strstart>(w_size-MIN_LOOKAHEAD) ?
      strstart-(w_size-MIN_LOOKAHEAD) : 0;
    int nice_match=this.nice_match;

    // Stop when cur_match becomes <= limit. To simplify the code,
    // we prevent matches with the string of window index 0.

    int wmask = w_mask;

    int strend = strstart + MAX_MATCH;
    byte scan_end1 = window[scan+best_len-1];
    byte scan_end = window[scan+best_len];

    // The code is optimized for HASH_BITS >= 8 and MAX_MATCH-2 multiple of 16.
    // It is easy to get rid of this optimization if necessary.

    // Do not waste too much time if we already have a good match:
    if (prev_length >= good_match) {
      chain_length >>= 2;
    }

    // Do not look for matches beyond the end of the input. This is necessary
    // to make deflate deterministic.
    if (nice_match > lookahead) nice_match = lookahead;

    do {
      match = cur_match;

      // Skip to next match if the match length cannot increase
      // or if the match length is less than 2:
      if (window[match+best_len]   != scan_end  ||
	  window[match+best_len-1] != scan_end1 ||
	  window[match]       != window[scan]     ||
	  window[++match]     != window[scan+1])      continue;

      // The check at best_len-1 can be removed because it will be made
      // again later. (This heuristic is not always a win.)
      // It is not necessary to compare scan[2] and match[2] since they
      // are always equal when the other bytes match, given that
      // the hash keys are equal and that HASH_BITS >= 8.
      scan += 2; match++;

      // We check for insufficient lookahead only every 8th comparison;
      // the 256th check will be made at strstart+258.
      do {
      } while (window[++scan] == window[++match] &&
	       window[++scan] == window[++match] &&
	       window[++scan] == window[++match] &&
	       window[++scan] == window[++match] &&
	       window[++scan] == window[++match] &&
	       window[++scan] == window[++match] &&
	       window[++scan] == window[++match] &&
	       window[++scan] == window[++match] &&
	       scan < strend);

      len = MAX_MATCH - (int)(strend - scan);
      scan = strend - MAX_MATCH;

      if(len>best_len) {
	match_start = cur_match;
	best_len = len;
	if (len >= nice_match) break;
	scan_end1  = window[scan+best_len-1];
	scan_end   = window[scan+best_len];
      }

    } while ((cur_match = (prev[cur_match & wmask]&0xffff)) > limit
	     && --chain_length != 0);

    if (best_len <= lookahead) return best_len;
    return lookahead;
  }
    
  int deflateInit(ZStream strm, int level, int bits){
    return deflateInit2(strm, level, Z_DEFLATED, bits, DEF_MEM_LEVEL,
			Z_DEFAULT_STRATEGY);
  }
  int deflateInit(ZStream strm, int level){
    return deflateInit(strm, level, MAX_WBITS);
  }
  int deflateInit2(ZStream strm, int level, int method,  int windowBits,
		   int memLevel, int strategy){
    int noheader = 0;
    //    byte[] my_version=ZLIB_VERSION;

    //
    //  if (version == null || version[0] != my_version[0]
    //  || stream_size != sizeof(z_stream)) {
    //  return Z_VERSION_ERROR;
    //  }

    strm.msg = null;

    if (level == Z_DEFAULT_COMPRESSION) level = 6;

    if (windowBits < 0) { // undocumented feature: suppress zlib header
      noheader = 1;
      windowBits = -windowBits;
    }

    if (memLevel < 1 || memLevel > MAX_MEM_LEVEL || 
	method != Z_DEFLATED ||
	windowBits < 9 || windowBits > 15 || level < 0 || level > 9 ||
        strategy < 0 || strategy > Z_HUFFMAN_ONLY) {
      return Z_STREAM_ERROR;
    }

    strm.dstate = (Deflate)this;

    this.noheader = noheader;
    w_bits = windowBits;
    w_size = 1 << w_bits;
    w_mask = w_size - 1;

    hash_bits = memLevel + 7;
    hash_size = 1 << hash_bits;
    hash_mask = hash_size - 1;
    hash_shift = ((hash_bits+MIN_MATCH-1)/MIN_MATCH);

    window = new byte[w_size*2];
    prev = new short[w_size];
    head = new short[hash_size];

    lit_bufsize = 1 << (memLevel + 6); // 16K elements by default

    // We overlay pending_buf and d_buf+l_buf. This works since the average
    // output size for (length,distance) codes is <= 24 bits.
    pending_buf = new byte[lit_bufsize*4];
    pending_buf_size = lit_bufsize*4;

    d_buf = lit_bufsize/2;
    l_buf = (1+2)*lit_bufsize;

    this.level = level;

//System.out.println("level="+level);

    this.strategy = strategy;
    this.method = (byte)method;

    return deflateReset(strm);
  }

  int deflateReset(ZStream strm){
    strm.total_in = strm.total_out = 0;
    strm.msg = null; //
    strm.data_type = Z_UNKNOWN;

    pending = 0;
    pending_out = 0;

    if(noheader < 0) {
      noheader = 0; // was set to -1 by deflate(..., Z_FINISH);
    }
    status = (noheader!=0) ? BUSY_STATE : INIT_STATE;
    strm.adler=strm._adler.adler32(0, null, 0, 0);

    last_flush = Z_NO_FLUSH;

    tr_init();
    lm_init();
    return Z_OK;
  }

  int deflateEnd(){
    if(status!=INIT_STATE && status!=BUSY_STATE && status!=FINISH_STATE){
      return Z_STREAM_ERROR;
    }
    // Deallocate in reverse order of allocations:
    pending_buf=null;
    head=null;
    prev=null;
    window=null;
    // free
    // dstate=null;
    return status == BUSY_STATE ? Z_DATA_ERROR : Z_OK;
  }

  int deflateParams(ZStream strm, int _level, int _strategy){
    int err=Z_OK;

    if(_level == Z_DEFAULT_COMPRESSION){
      _level = 6;
    }
    if(_level < 0 || _level > 9 || 
       _strategy < 0 || _strategy > Z_HUFFMAN_ONLY) {
      return Z_STREAM_ERROR;
    }

    if(config_table[level].func!=config_table[_level].func &&
       strm.total_in != 0) {
      // Flush the last buffer:
      err = strm.deflate(Z_PARTIAL_FLUSH);
    }

    if(level != _level) {
      level = _level;
      max_lazy_match   = config_table[level].max_lazy;
      good_match       = config_table[level].good_length;
      nice_match       = config_table[level].nice_length;
      max_chain_length = config_table[level].max_chain;
    }
    strategy = _strategy;
    return err;
  }

  int deflateSetDictionary (ZStream strm, byte[] dictionary, int dictLength){
    int length = dictLength;
    int index=0;

    if(dictionary == null || status != INIT_STATE)
      return Z_STREAM_ERROR;

    strm.adler=strm._adler.adler32(strm.adler, dictionary, 0, dictLength);

    if(length < MIN_MATCH) return Z_OK;
    if(length > w_size-MIN_LOOKAHEAD){
      length = w_size-MIN_LOOKAHEAD;
      index=dictLength-length; // use the tail of the dictionary
    }
    System.arraycopy(dictionary, index, window, 0, length);
    strstart = length;
    block_start = length;

    // Insert all strings in the hash table (except for the last two bytes).
    // s->lookahead stays null, so s->ins_h will be recomputed at the next
    // call of fill_window.

    ins_h = window[0]&0xff;
    ins_h=(((ins_h)<<hash_shift)^(window[1]&0xff))&hash_mask;

    for(int n=0; n<=length-MIN_MATCH; n++){
      ins_h=(((ins_h)<<hash_shift)^(window[(n)+(MIN_MATCH-1)]&0xff))&hash_mask;
      prev[n&w_mask]=head[ins_h];
      head[ins_h]=(short)n;
    }
    return Z_OK;
  }

  int deflate(ZStream strm, int flush){
    int old_flush;

    if(flush>Z_FINISH || flush<0){
      return Z_STREAM_ERROR;
    }

    if(strm.next_out == null ||
       (strm.next_in == null && strm.avail_in != 0) ||
       (status == FINISH_STATE && flush != Z_FINISH)) {
      strm.msg=z_errmsg[Z_NEED_DICT-(Z_STREAM_ERROR)];
      return Z_STREAM_ERROR;
    }
    if(strm.avail_out == 0){
      strm.msg=z_errmsg[Z_NEED_DICT-(Z_BUF_ERROR)];
      return Z_BUF_ERROR;
    }

    this.strm = strm; // just in case
    old_flush = last_flush;
    last_flush = flush;

    // Write the zlib header
    if(status == INIT_STATE) {
      int header = (Z_DEFLATED+((w_bits-8)<<4))<<8;
      int level_flags=((level-1)&0xff)>>1;

      if(level_flags>3) level_flags=3;
      header |= (level_flags<<6);
      if(strstart!=0) header |= PRESET_DICT;
      header+=31-(header % 31);

      status=BUSY_STATE;
      putShortMSB(header);


      // Save the adler32 of the preset dictionary:
      if(strstart!=0){
        putShortMSB((int)(strm.adler>>>16));
        putShortMSB((int)(strm.adler&0xffff));
      }
      strm.adler=strm._adler.adler32(0, null, 0, 0);
    }

    // Flush as much pending output as possible
    if(pending != 0) {
      strm.flush_pending();
      if(strm.avail_out == 0) {
	//System.out.println("  avail_out==0");
	// Since avail_out is 0, deflate will be called again with
	// more output space, but possibly with both pending and
	// avail_in equal to zero. There won't be anything to do,
	// but this is not an error situation so make sure we
	// return OK instead of BUF_ERROR at next call of deflate:
	last_flush = -1;
	return Z_OK;
      }

      // Make sure there is something to do and avoid duplicate consecutive
      // flushes. For repeated and useless calls with Z_FINISH, we keep
      // returning Z_STREAM_END instead of Z_BUFF_ERROR.
    }
    else if(strm.avail_in==0 && flush <= old_flush &&
	    flush != Z_FINISH) {
      strm.msg=z_errmsg[Z_NEED_DICT-(Z_BUF_ERROR)];
      return Z_BUF_ERROR;
    }

    // User must not provide more input after the first FINISH:
    if(status == FINISH_STATE && strm.avail_in != 0) {
      strm.msg=z_errmsg[Z_NEED_DICT-(Z_BUF_ERROR)];
      return Z_BUF_ERROR;
    }

    // Start a new block or continue the current one.
    if(strm.avail_in!=0 || lookahead!=0 ||
       (flush != Z_NO_FLUSH && status != FINISH_STATE)) {
      int bstate=-1;
      switch(config_table[level].func){
      case STORED: 
	bstate = deflate_stored(flush);
	break;
      case FAST: 
	bstate = deflate_fast(flush);
	break;
      case SLOW: 
	bstate = deflate_slow(flush);
	break;
      default:
      }

      if (bstate==FinishStarted || bstate==FinishDone) {
	status = FINISH_STATE;
      }
      if (bstate==NeedMore || bstate==FinishStarted) {
	if(strm.avail_out == 0) {
	  last_flush = -1; // avoid BUF_ERROR next call, see above
	}
	return Z_OK;
	// If flush != Z_NO_FLUSH && avail_out == 0, the next call
	// of deflate should use the same flush parameter to make sure
	// that the flush is complete. So we don't have to output an
	// empty block here, this will be done at next call. This also
	// ensures that for a very small output buffer, we emit at most
	// one empty block.
      }

      if (bstate==BlockDone) {
	if(flush == Z_PARTIAL_FLUSH) {
	  _tr_align();
	} 
	else { // FULL_FLUSH or SYNC_FLUSH
	  _tr_stored_block(0, 0, false);
	  // For a full flush, this empty block will be recognized
	  // as a special marker by inflate_sync().
	  if(flush == Z_FULL_FLUSH) {
	    //state.head[s.hash_size-1]=0;
	    for(int i=0; i<hash_size/*-1*/; i++)  // forget history
	      head[i]=0;
	  }
	}
	strm.flush_pending();
	if(strm.avail_out == 0) {
	  last_flush = -1; // avoid BUF_ERROR at next call, see above
	  return Z_OK;
	}
      }
    }

    if(flush!=Z_FINISH) return Z_OK;
    if(noheader!=0) return Z_STREAM_END;

    // Write the zlib trailer (adler32)
    putShortMSB((int)(strm.adler>>>16));
    putShortMSB((int)(strm.adler&0xffff));
    strm.flush_pending();

    // If avail_out is zero, the application will call deflate again
    // to flush the rest.
    noheader = -1; // write the trailer only once!
    return pending != 0 ? Z_OK : Z_STREAM_END;
  }
}