Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
Browse files

Starting deflate imple

  • Loading branch information...
commit 808d57f5a8cd190eedc5afe525e0c7c349cdd1a4 1 parent b65d3bc
@janjongboom janjongboom authored
View
1,690 janzip.deflate.js
@@ -0,0 +1,1690 @@
+/*
+ * Port of a script by Masanao Izumo.
+ *
+ * Only changes : wrap all the variables in a function and add the
+ * main function to JSZip (DEFLATE compression method).
+ * Everything else was written by M. Izumo.
+ *
+ * Original code can be found here: http://www.onicos.com/staff/iz/amuse/javascript/expert/inflate.txt
+ */
+
+/*
+ * Original:
+ * http://www.onicos.com/staff/iz/amuse/javascript/expert/deflate.txt
+ */
+
+module.exports = (function(){
+
+/* Copyright (C) 1999 Masanao Izumo <iz@onicos.co.jp>
+ * Version: 1.0.1
+ * LastModified: Dec 25 1999
+ */
+
+/* Interface:
+ * data = zip_deflate(src);
+ */
+
+/* constant parameters */
+var zip_WSIZE = 32768; // Sliding Window size
+var zip_STORED_BLOCK = 0;
+var zip_STATIC_TREES = 1;
+var zip_DYN_TREES = 2;
+
+/* for deflate */
+var zip_DEFAULT_LEVEL = 6;
+var zip_FULL_SEARCH = true;
+var zip_INBUFSIZ = 32768; // Input buffer size
+var zip_INBUF_EXTRA = 64; // Extra buffer
+var zip_OUTBUFSIZ = 1024 * 8;
+var zip_window_size = 2 * zip_WSIZE;
+var zip_MIN_MATCH = 3;
+var zip_MAX_MATCH = 258;
+var zip_BITS = 16;
+// for SMALL_MEM
+var zip_LIT_BUFSIZE = 0x2000;
+var zip_HASH_BITS = 13;
+// for MEDIUM_MEM
+// var zip_LIT_BUFSIZE = 0x4000;
+// var zip_HASH_BITS = 14;
+// for BIG_MEM
+// var zip_LIT_BUFSIZE = 0x8000;
+// var zip_HASH_BITS = 15;
+if(zip_LIT_BUFSIZE > zip_INBUFSIZ)
+ alert("error: zip_INBUFSIZ is too small");
+if((zip_WSIZE<<1) > (1<<zip_BITS))
+ alert("error: zip_WSIZE is too large");
+if(zip_HASH_BITS > zip_BITS-1)
+ alert("error: zip_HASH_BITS is too large");
+if(zip_HASH_BITS < 8 || zip_MAX_MATCH != 258)
+ alert("error: Code too clever");
+var zip_DIST_BUFSIZE = zip_LIT_BUFSIZE;
+var zip_HASH_SIZE = 1 << zip_HASH_BITS;
+var zip_HASH_MASK = zip_HASH_SIZE - 1;
+var zip_WMASK = zip_WSIZE - 1;
+var zip_NIL = 0; // Tail of hash chains
+var zip_TOO_FAR = 4096;
+var zip_MIN_LOOKAHEAD = zip_MAX_MATCH + zip_MIN_MATCH + 1;
+var zip_MAX_DIST = zip_WSIZE - zip_MIN_LOOKAHEAD;
+var zip_SMALLEST = 1;
+var zip_MAX_BITS = 15;
+var zip_MAX_BL_BITS = 7;
+var zip_LENGTH_CODES = 29;
+var zip_LITERALS =256;
+var zip_END_BLOCK = 256;
+var zip_L_CODES = zip_LITERALS + 1 + zip_LENGTH_CODES;
+var zip_D_CODES = 30;
+var zip_BL_CODES = 19;
+var zip_REP_3_6 = 16;
+var zip_REPZ_3_10 = 17;
+var zip_REPZ_11_138 = 18;
+var zip_HEAP_SIZE = 2 * zip_L_CODES + 1;
+var zip_H_SHIFT = parseInt((zip_HASH_BITS + zip_MIN_MATCH - 1) /
+ zip_MIN_MATCH);
+
+/* variables */
+var zip_free_queue;
+var zip_qhead, zip_qtail;
+var zip_initflag;
+var zip_outbuf = null;
+var zip_outcnt, zip_outoff;
+var zip_complete;
+var zip_window;
+var zip_d_buf;
+var zip_l_buf;
+var zip_prev;
+var zip_bi_buf;
+var zip_bi_valid;
+var zip_block_start;
+var zip_ins_h;
+var zip_hash_head;
+var zip_prev_match;
+var zip_match_available;
+var zip_match_length;
+var zip_prev_length;
+var zip_strstart;
+var zip_match_start;
+var zip_eofile;
+var zip_lookahead;
+var zip_max_chain_length;
+var zip_max_lazy_match;
+var zip_compr_level;
+var zip_good_match;
+var zip_nice_match;
+var zip_dyn_ltree;
+var zip_dyn_dtree;
+var zip_static_ltree;
+var zip_static_dtree;
+var zip_bl_tree;
+var zip_l_desc;
+var zip_d_desc;
+var zip_bl_desc;
+var zip_bl_count;
+var zip_heap;
+var zip_heap_len;
+var zip_heap_max;
+var zip_depth;
+var zip_length_code;
+var zip_dist_code;
+var zip_base_length;
+var zip_base_dist;
+var zip_flag_buf;
+var zip_last_lit;
+var zip_last_dist;
+var zip_last_flags;
+var zip_flags;
+var zip_flag_bit;
+var zip_opt_len;
+var zip_static_len;
+var zip_deflate_data;
+var zip_deflate_pos;
+
+/* objects (deflate) */
+
+var zip_DeflateCT = function() {
+ this.fc = 0; // frequency count or bit string
+ this.dl = 0; // father node in Huffman tree or length of bit string
+}
+
+var zip_DeflateTreeDesc = function() {
+ this.dyn_tree = null; // the dynamic tree
+ this.static_tree = null; // corresponding static tree or NULL
+ this.extra_bits = null; // extra bits for each code or NULL
+ this.extra_base = 0; // base index for extra_bits
+ this.elems = 0; // max number of elements in the tree
+ this.max_length = 0; // max bit length for the codes
+ this.max_code = 0; // largest code with non zero frequency
+}
+
+/* Values for max_lazy_match, good_match and max_chain_length, depending on
+ * the desired pack level (0..9). The values given below have been tuned to
+ * exclude worst case performance for pathological files. Better values may be
+ * found for specific files.
+ */
+var zip_DeflateConfiguration = function(a, b, c, d) {
+ this.good_length = a; // reduce lazy search above this match length
+ this.max_lazy = b; // do not perform lazy search above this match length
+ this.nice_length = c; // quit search above this match length
+ this.max_chain = d;
+}
+
+var zip_DeflateBuffer = function() {
+ this.next = null;
+ this.len = 0;
+ this.ptr = new Array(zip_OUTBUFSIZ);
+ this.off = 0;
+}
+
+/* constant tables */
+var zip_extra_lbits = new Array(
+ 0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0);
+var zip_extra_dbits = new Array(
+ 0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13);
+var zip_extra_blbits = new Array(
+ 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2,3,7);
+var zip_bl_order = new Array(
+ 16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15);
+var zip_configuration_table = new Array(
+ new zip_DeflateConfiguration(0, 0, 0, 0),
+ new zip_DeflateConfiguration(4, 4, 8, 4),
+ new zip_DeflateConfiguration(4, 5, 16, 8),
+ new zip_DeflateConfiguration(4, 6, 32, 32),
+ new zip_DeflateConfiguration(4, 4, 16, 16),
+ new zip_DeflateConfiguration(8, 16, 32, 32),
+ new zip_DeflateConfiguration(8, 16, 128, 128),
+ new zip_DeflateConfiguration(8, 32, 128, 256),
+ new zip_DeflateConfiguration(32, 128, 258, 1024),
+ new zip_DeflateConfiguration(32, 258, 258, 4096));
+
+
+/* routines (deflate) */
+
+var zip_deflate_start = function(level) {
+ var i;
+
+ if(!level)
+ level = zip_DEFAULT_LEVEL;
+ else if(level < 1)
+ level = 1;
+ else if(level > 9)
+ level = 9;
+
+ zip_compr_level = level;
+ zip_initflag = false;
+ zip_eofile = false;
+ if(zip_outbuf != null)
+ return;
+
+ zip_free_queue = zip_qhead = zip_qtail = null;
+ zip_outbuf = new Array(zip_OUTBUFSIZ);
+ zip_window = new Array(zip_window_size);
+ zip_d_buf = new Array(zip_DIST_BUFSIZE);
+ zip_l_buf = new Array(zip_INBUFSIZ + zip_INBUF_EXTRA);
+ zip_prev = new Array(1 << zip_BITS);
+ zip_dyn_ltree = new Array(zip_HEAP_SIZE);
+ for(i = 0; i < zip_HEAP_SIZE; i++)
+ zip_dyn_ltree[i] = new zip_DeflateCT();
+ zip_dyn_dtree = new Array(2*zip_D_CODES+1);
+ for(i = 0; i < 2*zip_D_CODES+1; i++)
+ zip_dyn_dtree[i] = new zip_DeflateCT();
+ zip_static_ltree = new Array(zip_L_CODES+2);
+ for(i = 0; i < zip_L_CODES+2; i++)
+ zip_static_ltree[i] = new zip_DeflateCT();
+ zip_static_dtree = new Array(zip_D_CODES);
+ for(i = 0; i < zip_D_CODES; i++)
+ zip_static_dtree[i] = new zip_DeflateCT();
+ zip_bl_tree = new Array(2*zip_BL_CODES+1);
+ for(i = 0; i < 2*zip_BL_CODES+1; i++)
+ zip_bl_tree[i] = new zip_DeflateCT();
+ zip_l_desc = new zip_DeflateTreeDesc();
+ zip_d_desc = new zip_DeflateTreeDesc();
+ zip_bl_desc = new zip_DeflateTreeDesc();
+ zip_bl_count = new Array(zip_MAX_BITS+1);
+ zip_heap = new Array(2*zip_L_CODES+1);
+ zip_depth = new Array(2*zip_L_CODES+1);
+ zip_length_code = new Array(zip_MAX_MATCH-zip_MIN_MATCH+1);
+ zip_dist_code = new Array(512);
+ zip_base_length = new Array(zip_LENGTH_CODES);
+ zip_base_dist = new Array(zip_D_CODES);
+ zip_flag_buf = new Array(parseInt(zip_LIT_BUFSIZE / 8));
+}
+
+var zip_deflate_end = function() {
+ zip_free_queue = zip_qhead = zip_qtail = null;
+ zip_outbuf = null;
+ zip_window = null;
+ zip_d_buf = null;
+ zip_l_buf = null;
+ zip_prev = null;
+ zip_dyn_ltree = null;
+ zip_dyn_dtree = null;
+ zip_static_ltree = null;
+ zip_static_dtree = null;
+ zip_bl_tree = null;
+ zip_l_desc = null;
+ zip_d_desc = null;
+ zip_bl_desc = null;
+ zip_bl_count = null;
+ zip_heap = null;
+ zip_depth = null;
+ zip_length_code = null;
+ zip_dist_code = null;
+ zip_base_length = null;
+ zip_base_dist = null;
+ zip_flag_buf = null;
+}
+
+var zip_reuse_queue = function(p) {
+ p.next = zip_free_queue;
+ zip_free_queue = p;
+}
+
+var zip_new_queue = function() {
+ var p;
+
+ if(zip_free_queue != null)
+ {
+ p = zip_free_queue;
+ zip_free_queue = zip_free_queue.next;
+ }
+ else
+ p = new zip_DeflateBuffer();
+ p.next = null;
+ p.len = p.off = 0;
+
+ return p;
+}
+
+var zip_head1 = function(i) {
+ return zip_prev[zip_WSIZE + i];
+}
+
+var zip_head2 = function(i, val) {
+ return zip_prev[zip_WSIZE + i] = val;
+}
+
+/* put_byte is used for the compressed output, put_ubyte for the
+ * uncompressed output. However unlzw() uses window for its
+ * suffix table instead of its output buffer, so it does not use put_ubyte
+ * (to be cleaned up).
+ */
+var zip_put_byte = function(c) {
+ zip_outbuf[zip_outoff + zip_outcnt++] = c;
+ if(zip_outoff + zip_outcnt == zip_OUTBUFSIZ)
+ zip_qoutbuf();
+}
+
+/* Output a 16 bit value, lsb first */
+var zip_put_short = function(w) {
+ w &= 0xffff;
+ if(zip_outoff + zip_outcnt < zip_OUTBUFSIZ - 2) {
+ zip_outbuf[zip_outoff + zip_outcnt++] = (w & 0xff);
+ zip_outbuf[zip_outoff + zip_outcnt++] = (w >>> 8);
+ } else {
+ zip_put_byte(w & 0xff);
+ zip_put_byte(w >>> 8);
+ }
+}
+
+/* ==========================================================================
+ * Insert string s in the dictionary and set match_head to the previous head
+ * of the hash chain (the most recent string with same hash key). Return
+ * the previous length of the hash chain.
+ * IN assertion: all calls to to INSERT_STRING are made with consecutive
+ * input characters and the first MIN_MATCH bytes of s are valid
+ * (except for the last MIN_MATCH-1 bytes of the input file).
+ */
+var zip_INSERT_STRING = function() {
+ zip_ins_h = ((zip_ins_h << zip_H_SHIFT)
+ ^ (zip_window[zip_strstart + zip_MIN_MATCH - 1] & 0xff))
+ & zip_HASH_MASK;
+ zip_hash_head = zip_head1(zip_ins_h);
+ zip_prev[zip_strstart & zip_WMASK] = zip_hash_head;
+ zip_head2(zip_ins_h, zip_strstart);
+}
+
+/* Send a code of the given tree. c and tree must not have side effects */
+var zip_SEND_CODE = function(c, tree) {
+ zip_send_bits(tree[c].fc, tree[c].dl);
+}
+
+/* Mapping from a distance to a distance code. dist is the distance - 1 and
+ * must not have side effects. dist_code[256] and dist_code[257] are never
+ * used.
+ */
+var zip_D_CODE = function(dist) {
+ return (dist < 256 ? zip_dist_code[dist]
+ : zip_dist_code[256 + (dist>>7)]) & 0xff;
+}
+
+/* ==========================================================================
+ * Compares to subtrees, using the tree depth as tie breaker when
+ * the subtrees have equal frequency. This minimizes the worst case length.
+ */
+var zip_SMALLER = function(tree, n, m) {
+ return tree[n].fc < tree[m].fc ||
+ (tree[n].fc == tree[m].fc && zip_depth[n] <= zip_depth[m]);
+}
+
+/* ==========================================================================
+ * read string data
+ */
+var zip_read_buff = function(buff, offset, n) {
+ var i;
+ for(i = 0; i < n && zip_deflate_pos < zip_deflate_data.length; i++)
+ buff[offset + i] =
+ zip_deflate_data.charCodeAt(zip_deflate_pos++) & 0xff;
+ return i;
+}
+
+/* ==========================================================================
+ * Initialize the "longest match" routines for a new file
+ */
+var zip_lm_init = function() {
+ var j;
+
+ /* Initialize the hash table. */
+ for(j = 0; j < zip_HASH_SIZE; j++)
+// zip_head2(j, zip_NIL);
+ zip_prev[zip_WSIZE + j] = 0;
+ /* prev will be initialized on the fly */
+
+ /* Set the default configuration parameters:
+ */
+ zip_max_lazy_match = zip_configuration_table[zip_compr_level].max_lazy;
+ zip_good_match = zip_configuration_table[zip_compr_level].good_length;
+ if(!zip_FULL_SEARCH)
+ zip_nice_match = zip_configuration_table[zip_compr_level].nice_length;
+ zip_max_chain_length = zip_configuration_table[zip_compr_level].max_chain;
+
+ zip_strstart = 0;
+ zip_block_start = 0;
+
+ zip_lookahead = zip_read_buff(zip_window, 0, 2 * zip_WSIZE);
+ if(zip_lookahead <= 0) {
+ zip_eofile = true;
+ zip_lookahead = 0;
+ return;
+ }
+ zip_eofile = false;
+ /* Make sure that we always have enough lookahead. This is important
+ * if input comes from a device such as a tty.
+ */
+ while(zip_lookahead < zip_MIN_LOOKAHEAD && !zip_eofile)
+ zip_fill_window();
+
+ /* If lookahead < MIN_MATCH, ins_h is garbage, but this is
+ * not important since only literal bytes will be emitted.
+ */
+ zip_ins_h = 0;
+ for(j = 0; j < zip_MIN_MATCH - 1; j++) {
+// UPDATE_HASH(ins_h, window[j]);
+ zip_ins_h = ((zip_ins_h << zip_H_SHIFT) ^ (zip_window[j] & 0xff)) & zip_HASH_MASK;
+ }
+}
+
+/* ==========================================================================
+ * Set match_start to the longest match starting at the given string and
+ * return its length. Matches shorter or equal to prev_length are discarded,
+ * in which case the result is equal to prev_length and match_start is
+ * garbage.
+ * IN assertions: cur_match is the head of the hash chain for the current
+ * string (strstart) and its distance is <= MAX_DIST, and prev_length >= 1
+ */
+var zip_longest_match = function(cur_match) {
+ var chain_length = zip_max_chain_length; // max hash chain length
+ var scanp = zip_strstart; // current string
+ var matchp; // matched string
+ var len; // length of current match
+ var best_len = zip_prev_length; // best match length so far
+
+ /* Stop when cur_match becomes <= limit. To simplify the code,
+ * we prevent matches with the string of window index 0.
+ */
+ var limit = (zip_strstart > zip_MAX_DIST ? zip_strstart - zip_MAX_DIST : zip_NIL);
+
+ var strendp = zip_strstart + zip_MAX_MATCH;
+ var scan_end1 = zip_window[scanp + best_len - 1];
+ var scan_end = zip_window[scanp + best_len];
+
+ /* Do not waste too much time if we already have a good match: */
+ if(zip_prev_length >= zip_good_match)
+ chain_length >>= 2;
+
+// Assert(encoder->strstart <= window_size-MIN_LOOKAHEAD, "insufficient lookahead");
+
+ do {
+// Assert(cur_match < encoder->strstart, "no future");
+ matchp = cur_match;
+
+ /* Skip to next match if the match length cannot increase
+ * or if the match length is less than 2:
+ */
+ if(zip_window[matchp + best_len] != scan_end ||
+ zip_window[matchp + best_len - 1] != scan_end1 ||
+ zip_window[matchp] != zip_window[scanp] ||
+ zip_window[++matchp] != zip_window[scanp + 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.
+ */
+ scanp += 2;
+ matchp++;
+
+ /* We check for insufficient lookahead only every 8th comparison;
+ * the 256th check will be made at strstart+258.
+ */
+ do {
+ } while(zip_window[++scanp] == zip_window[++matchp] &&
+ zip_window[++scanp] == zip_window[++matchp] &&
+ zip_window[++scanp] == zip_window[++matchp] &&
+ zip_window[++scanp] == zip_window[++matchp] &&
+ zip_window[++scanp] == zip_window[++matchp] &&
+ zip_window[++scanp] == zip_window[++matchp] &&
+ zip_window[++scanp] == zip_window[++matchp] &&
+ zip_window[++scanp] == zip_window[++matchp] &&
+ scanp < strendp);
+
+ len = zip_MAX_MATCH - (strendp - scanp);
+ scanp = strendp - zip_MAX_MATCH;
+
+ if(len > best_len) {
+ zip_match_start = cur_match;
+ best_len = len;
+ if(zip_FULL_SEARCH) {
+ if(len >= zip_MAX_MATCH) break;
+ } else {
+ if(len >= zip_nice_match) break;
+ }
+
+ scan_end1 = zip_window[scanp + best_len-1];
+ scan_end = zip_window[scanp + best_len];
+ }
+ } while((cur_match = zip_prev[cur_match & zip_WMASK]) > limit
+ && --chain_length != 0);
+
+ return best_len;
+}
+
+/* ==========================================================================
+ * Fill the window when the lookahead becomes insufficient.
+ * Updates strstart and lookahead, and sets eofile if end of input file.
+ * IN assertion: lookahead < MIN_LOOKAHEAD && strstart + lookahead > 0
+ * OUT assertions: at least one byte has been read, or eofile is set;
+ * file reads are performed for at least two bytes (required for the
+ * translate_eol option).
+ */
+var zip_fill_window = function() {
+ var n, m;
+
+ // Amount of free space at the end of the window.
+ var more = zip_window_size - zip_lookahead - zip_strstart;
+
+ /* If the window is almost full and there is insufficient lookahead,
+ * move the upper half to the lower one to make room in the upper half.
+ */
+ if(more == -1) {
+ /* Very unlikely, but possible on 16 bit machine if strstart == 0
+ * and lookahead == 1 (input done one byte at time)
+ */
+ more--;
+ } else if(zip_strstart >= zip_WSIZE + zip_MAX_DIST) {
+ /* By the IN assertion, the window is not empty so we can't confuse
+ * more == 0 with more == 64K on a 16 bit machine.
+ */
+// Assert(window_size == (ulg)2*WSIZE, "no sliding with BIG_MEM");
+
+// System.arraycopy(window, WSIZE, window, 0, WSIZE);
+ for(n = 0; n < zip_WSIZE; n++)
+ zip_window[n] = zip_window[n + zip_WSIZE];
+
+ zip_match_start -= zip_WSIZE;
+ zip_strstart -= zip_WSIZE; /* we now have strstart >= MAX_DIST: */
+ zip_block_start -= zip_WSIZE;
+
+ for(n = 0; n < zip_HASH_SIZE; n++) {
+ m = zip_head1(n);
+ zip_head2(n, m >= zip_WSIZE ? m - zip_WSIZE : zip_NIL);
+ }
+ for(n = 0; n < zip_WSIZE; n++) {
+ /* If n is not on any hash chain, prev[n] is garbage but
+ * its value will never be used.
+ */
+ m = zip_prev[n];
+ zip_prev[n] = (m >= zip_WSIZE ? m - zip_WSIZE : zip_NIL);
+ }
+ more += zip_WSIZE;
+ }
+ // At this point, more >= 2
+ if(!zip_eofile) {
+ n = zip_read_buff(zip_window, zip_strstart + zip_lookahead, more);
+ if(n <= 0)
+ zip_eofile = true;
+ else
+ zip_lookahead += n;
+ }
+}
+
+/* ==========================================================================
+ * Processes a new input file and return its compressed length. This
+ * function does not perform lazy evaluationof matches and inserts
+ * new strings in the dictionary only for unmatched strings or for short
+ * matches. It is used only for the fast compression options.
+ */
+var zip_deflate_fast = function() {
+ while(zip_lookahead != 0 && zip_qhead == null) {
+ var flush; // set if current block must be flushed
+
+ /* Insert the string window[strstart .. strstart+2] in the
+ * dictionary, and set hash_head to the head of the hash chain:
+ */
+ zip_INSERT_STRING();
+
+ /* Find the longest match, discarding those <= prev_length.
+ * At this point we have always match_length < MIN_MATCH
+ */
+ if(zip_hash_head != zip_NIL &&
+ zip_strstart - zip_hash_head <= zip_MAX_DIST) {
+ /* To simplify the code, we prevent matches with the string
+ * of window index 0 (in particular we have to avoid a match
+ * of the string with itself at the start of the input file).
+ */
+ zip_match_length = zip_longest_match(zip_hash_head);
+ /* longest_match() sets match_start */
+ if(zip_match_length > zip_lookahead)
+ zip_match_length = zip_lookahead;
+ }
+ if(zip_match_length >= zip_MIN_MATCH) {
+// check_match(strstart, match_start, match_length);
+
+ flush = zip_ct_tally(zip_strstart - zip_match_start,
+ zip_match_length - zip_MIN_MATCH);
+ zip_lookahead -= zip_match_length;
+
+ /* Insert new strings in the hash table only if the match length
+ * is not too large. This saves time but degrades compression.
+ */
+ if(zip_match_length <= zip_max_lazy_match) {
+ zip_match_length--; // string at strstart already in hash table
+ do {
+ zip_strstart++;
+ zip_INSERT_STRING();
+ /* strstart never exceeds WSIZE-MAX_MATCH, so there are
+ * always MIN_MATCH bytes ahead. If lookahead < MIN_MATCH
+ * these bytes are garbage, but it does not matter since
+ * the next lookahead bytes will be emitted as literals.
+ */
+ } while(--zip_match_length != 0);
+ zip_strstart++;
+ } else {
+ zip_strstart += zip_match_length;
+ zip_match_length = 0;
+ zip_ins_h = zip_window[zip_strstart] & 0xff;
+// UPDATE_HASH(ins_h, window[strstart + 1]);
+ zip_ins_h = ((zip_ins_h<<zip_H_SHIFT) ^ (zip_window[zip_strstart + 1] & 0xff)) & zip_HASH_MASK;
+
+//#if MIN_MATCH != 3
+// Call UPDATE_HASH() MIN_MATCH-3 more times
+//#endif
+
+ }
+ } else {
+ /* No match, output a literal byte */
+ flush = zip_ct_tally(0, zip_window[zip_strstart] & 0xff);
+ zip_lookahead--;
+ zip_strstart++;
+ }
+ if(flush) {
+ zip_flush_block(0);
+ zip_block_start = zip_strstart;
+ }
+
+ /* Make sure that we always have enough lookahead, except
+ * at the end of the input file. We need MAX_MATCH bytes
+ * for the next match, plus MIN_MATCH bytes to insert the
+ * string following the next match.
+ */
+ while(zip_lookahead < zip_MIN_LOOKAHEAD && !zip_eofile)
+ zip_fill_window();
+ }
+}
+
+var zip_deflate_better = function() {
+ /* Process the input block. */
+ while(zip_lookahead != 0 && zip_qhead == null) {
+ /* Insert the string window[strstart .. strstart+2] in the
+ * dictionary, and set hash_head to the head of the hash chain:
+ */
+ zip_INSERT_STRING();
+
+ /* Find the longest match, discarding those <= prev_length.
+ */
+ zip_prev_length = zip_match_length;
+ zip_prev_match = zip_match_start;
+ zip_match_length = zip_MIN_MATCH - 1;
+
+ if(zip_hash_head != zip_NIL &&
+ zip_prev_length < zip_max_lazy_match &&
+ zip_strstart - zip_hash_head <= zip_MAX_DIST) {
+ /* To simplify the code, we prevent matches with the string
+ * of window index 0 (in particular we have to avoid a match
+ * of the string with itself at the start of the input file).
+ */
+ zip_match_length = zip_longest_match(zip_hash_head);
+ /* longest_match() sets match_start */
+ if(zip_match_length > zip_lookahead)
+ zip_match_length = zip_lookahead;
+
+ /* Ignore a length 3 match if it is too distant: */
+ if(zip_match_length == zip_MIN_MATCH &&
+ zip_strstart - zip_match_start > zip_TOO_FAR) {
+ /* If prev_match is also MIN_MATCH, match_start is garbage
+ * but we will ignore the current match anyway.
+ */
+ zip_match_length--;
+ }
+ }
+ /* If there was a match at the previous step and the current
+ * match is not better, output the previous match:
+ */
+ if(zip_prev_length >= zip_MIN_MATCH &&
+ zip_match_length <= zip_prev_length) {
+ var flush; // set if current block must be flushed
+
+// check_match(strstart - 1, prev_match, prev_length);
+ flush = zip_ct_tally(zip_strstart - 1 - zip_prev_match,
+ zip_prev_length - zip_MIN_MATCH);
+
+ /* Insert in hash table all strings up to the end of the match.
+ * strstart-1 and strstart are already inserted.
+ */
+ zip_lookahead -= zip_prev_length - 1;
+ zip_prev_length -= 2;
+ do {
+ zip_strstart++;
+ zip_INSERT_STRING();
+ /* strstart never exceeds WSIZE-MAX_MATCH, so there are
+ * always MIN_MATCH bytes ahead. If lookahead < MIN_MATCH
+ * these bytes are garbage, but it does not matter since the
+ * next lookahead bytes will always be emitted as literals.
+ */
+ } while(--zip_prev_length != 0);
+ zip_match_available = 0;
+ zip_match_length = zip_MIN_MATCH - 1;
+ zip_strstart++;
+ if(flush) {
+ zip_flush_block(0);
+ zip_block_start = zip_strstart;
+ }
+ } else if(zip_match_available != 0) {
+ /* If there was no match at the previous position, output a
+ * single literal. If there was a match but the current match
+ * is longer, truncate the previous match to a single literal.
+ */
+ if(zip_ct_tally(0, zip_window[zip_strstart - 1] & 0xff)) {
+ zip_flush_block(0);
+ zip_block_start = zip_strstart;
+ }
+ zip_strstart++;
+ zip_lookahead--;
+ } else {
+ /* There is no previous match to compare with, wait for
+ * the next step to decide.
+ */
+ zip_match_available = 1;
+ zip_strstart++;
+ zip_lookahead--;
+ }
+
+ /* Make sure that we always have enough lookahead, except
+ * at the end of the input file. We need MAX_MATCH bytes
+ * for the next match, plus MIN_MATCH bytes to insert the
+ * string following the next match.
+ */
+ while(zip_lookahead < zip_MIN_LOOKAHEAD && !zip_eofile)
+ zip_fill_window();
+ }
+}
+
+var zip_init_deflate = function() {
+ if(zip_eofile)
+ return;
+ zip_bi_buf = 0;
+ zip_bi_valid = 0;
+ zip_ct_init();
+ zip_lm_init();
+
+ zip_qhead = null;
+ zip_outcnt = 0;
+ zip_outoff = 0;
+
+ if(zip_compr_level <= 3)
+ {
+ zip_prev_length = zip_MIN_MATCH - 1;
+ zip_match_length = 0;
+ }
+ else
+ {
+ zip_match_length = zip_MIN_MATCH - 1;
+ zip_match_available = 0;
+ }
+
+ zip_complete = false;
+}
+
+/* ==========================================================================
+ * Same as above, but achieves better compression. We use a lazy
+ * evaluation for matches: a match is finally adopted only if there is
+ * no better match at the next window position.
+ */
+var zip_deflate_internal = function(buff, off, buff_size) {
+ var n;
+
+ if(!zip_initflag)
+ {
+ zip_init_deflate();
+ zip_initflag = true;
+ if(zip_lookahead == 0) { // empty
+ zip_complete = true;
+ return 0;
+ }
+ }
+
+ if((n = zip_qcopy(buff, off, buff_size)) == buff_size)
+ return buff_size;
+
+ if(zip_complete)
+ return n;
+
+ if(zip_compr_level <= 3) // optimized for speed
+ zip_deflate_fast();
+ else
+ zip_deflate_better();
+ if(zip_lookahead == 0) {
+ if(zip_match_available != 0)
+ zip_ct_tally(0, zip_window[zip_strstart - 1] & 0xff);
+ zip_flush_block(1);
+ zip_complete = true;
+ }
+ return n + zip_qcopy(buff, n + off, buff_size - n);
+}
+
+var zip_qcopy = function(buff, off, buff_size) {
+ var n, i, j;
+
+ n = 0;
+ while(zip_qhead != null && n < buff_size)
+ {
+ i = buff_size - n;
+ if(i > zip_qhead.len)
+ i = zip_qhead.len;
+// System.arraycopy(qhead.ptr, qhead.off, buff, off + n, i);
+ for(j = 0; j < i; j++)
+ buff[off + n + j] = zip_qhead.ptr[zip_qhead.off + j];
+
+ zip_qhead.off += i;
+ zip_qhead.len -= i;
+ n += i;
+ if(zip_qhead.len == 0) {
+ var p;
+ p = zip_qhead;
+ zip_qhead = zip_qhead.next;
+ zip_reuse_queue(p);
+ }
+ }
+
+ if(n == buff_size)
+ return n;
+
+ if(zip_outoff < zip_outcnt) {
+ i = buff_size - n;
+ if(i > zip_outcnt - zip_outoff)
+ i = zip_outcnt - zip_outoff;
+ // System.arraycopy(outbuf, outoff, buff, off + n, i);
+ for(j = 0; j < i; j++)
+ buff[off + n + j] = zip_outbuf[zip_outoff + j];
+ zip_outoff += i;
+ n += i;
+ if(zip_outcnt == zip_outoff)
+ zip_outcnt = zip_outoff = 0;
+ }
+ return n;
+}
+
+/* ==========================================================================
+ * Allocate the match buffer, initialize the various tables and save the
+ * location of the internal file attribute (ascii/binary) and method
+ * (DEFLATE/STORE).
+ */
+var zip_ct_init = function() {
+ var n; // iterates over tree elements
+ var bits; // bit counter
+ var length; // length value
+ var code; // code value
+ var dist; // distance index
+
+ if(zip_static_dtree[0].dl != 0) return; // ct_init already called
+
+ zip_l_desc.dyn_tree = zip_dyn_ltree;
+ zip_l_desc.static_tree = zip_static_ltree;
+ zip_l_desc.extra_bits = zip_extra_lbits;
+ zip_l_desc.extra_base = zip_LITERALS + 1;
+ zip_l_desc.elems = zip_L_CODES;
+ zip_l_desc.max_length = zip_MAX_BITS;
+ zip_l_desc.max_code = 0;
+
+ zip_d_desc.dyn_tree = zip_dyn_dtree;
+ zip_d_desc.static_tree = zip_static_dtree;
+ zip_d_desc.extra_bits = zip_extra_dbits;
+ zip_d_desc.extra_base = 0;
+ zip_d_desc.elems = zip_D_CODES;
+ zip_d_desc.max_length = zip_MAX_BITS;
+ zip_d_desc.max_code = 0;
+
+ zip_bl_desc.dyn_tree = zip_bl_tree;
+ zip_bl_desc.static_tree = null;
+ zip_bl_desc.extra_bits = zip_extra_blbits;
+ zip_bl_desc.extra_base = 0;
+ zip_bl_desc.elems = zip_BL_CODES;
+ zip_bl_desc.max_length = zip_MAX_BL_BITS;
+ zip_bl_desc.max_code = 0;
+
+ // Initialize the mapping length (0..255) -> length code (0..28)
+ length = 0;
+ for(code = 0; code < zip_LENGTH_CODES-1; code++) {
+ zip_base_length[code] = length;
+ for(n = 0; n < (1<<zip_extra_lbits[code]); n++)
+ zip_length_code[length++] = code;
+ }
+ // Assert (length == 256, "ct_init: length != 256");
+
+ /* Note that the length 255 (match length 258) can be represented
+ * in two different ways: code 284 + 5 bits or code 285, so we
+ * overwrite length_code[255] to use the best encoding:
+ */
+ zip_length_code[length-1] = code;
+
+ /* Initialize the mapping dist (0..32K) -> dist code (0..29) */
+ dist = 0;
+ for(code = 0 ; code < 16; code++) {
+ zip_base_dist[code] = dist;
+ for(n = 0; n < (1<<zip_extra_dbits[code]); n++) {
+ zip_dist_code[dist++] = code;
+ }
+ }
+ // Assert (dist == 256, "ct_init: dist != 256");
+ dist >>= 7; // from now on, all distances are divided by 128
+ for( ; code < zip_D_CODES; code++) {
+ zip_base_dist[code] = dist << 7;
+ for(n = 0; n < (1<<(zip_extra_dbits[code]-7)); n++)
+ zip_dist_code[256 + dist++] = code;
+ }
+ // Assert (dist == 256, "ct_init: 256+dist != 512");
+
+ // Construct the codes of the static literal tree
+ for(bits = 0; bits <= zip_MAX_BITS; bits++)
+ zip_bl_count[bits] = 0;
+ n = 0;
+ while(n <= 143) { zip_static_ltree[n++].dl = 8; zip_bl_count[8]++; }
+ while(n <= 255) { zip_static_ltree[n++].dl = 9; zip_bl_count[9]++; }
+ while(n <= 279) { zip_static_ltree[n++].dl = 7; zip_bl_count[7]++; }
+ while(n <= 287) { zip_static_ltree[n++].dl = 8; zip_bl_count[8]++; }
+ /* Codes 286 and 287 do not exist, but we must include them in the
+ * tree construction to get a canonical Huffman tree (longest code
+ * all ones)
+ */
+ zip_gen_codes(zip_static_ltree, zip_L_CODES + 1);
+
+ /* The static distance tree is trivial: */
+ for(n = 0; n < zip_D_CODES; n++) {
+ zip_static_dtree[n].dl = 5;
+ zip_static_dtree[n].fc = zip_bi_reverse(n, 5);
+ }
+
+ // Initialize the first block of the first file:
+ zip_init_block();
+}
+
+/* ==========================================================================
+ * Initialize a new block.
+ */
+var zip_init_block = function() {
+ var n; // iterates over tree elements
+
+ // Initialize the trees.
+ for(n = 0; n < zip_L_CODES; n++) zip_dyn_ltree[n].fc = 0;
+ for(n = 0; n < zip_D_CODES; n++) zip_dyn_dtree[n].fc = 0;
+ for(n = 0; n < zip_BL_CODES; n++) zip_bl_tree[n].fc = 0;
+
+ zip_dyn_ltree[zip_END_BLOCK].fc = 1;
+ zip_opt_len = zip_static_len = 0;
+ zip_last_lit = zip_last_dist = zip_last_flags = 0;
+ zip_flags = 0;
+ zip_flag_bit = 1;
+}
+
+/* ==========================================================================
+ * Restore the heap property by moving down the tree starting at node k,
+ * exchanging a node with the smallest of its two sons if necessary, stopping
+ * when the heap property is re-established (each father smaller than its
+ * two sons).
+ */
+var zip_pqdownheap = function(
+ tree, // the tree to restore
+ k) { // node to move down
+ var v = zip_heap[k];
+ var j = k << 1; // left son of k
+
+ while(j <= zip_heap_len) {
+ // Set j to the smallest of the two sons:
+ if(j < zip_heap_len &&
+ zip_SMALLER(tree, zip_heap[j + 1], zip_heap[j]))
+ j++;
+
+ // Exit if v is smaller than both sons
+ if(zip_SMALLER(tree, v, zip_heap[j]))
+ break;
+
+ // Exchange v with the smallest son
+ zip_heap[k] = zip_heap[j];
+ k = j;
+
+ // And continue down the tree, setting j to the left son of k
+ j <<= 1;
+ }
+ zip_heap[k] = v;
+}
+
+/* ==========================================================================
+ * Compute the optimal bit lengths for a tree and update the total bit length
+ * for the current block.
+ * IN assertion: the fields freq and dad are set, heap[heap_max] and
+ * above are the tree nodes sorted by increasing frequency.
+ * OUT assertions: the field len is set to the optimal bit length, the
+ * array bl_count contains the frequencies for each bit length.
+ * The length opt_len is updated; static_len is also updated if stree is
+ * not null.
+ */
+var zip_gen_bitlen = function(desc) { // the tree descriptor
+ var tree = desc.dyn_tree;
+ var extra = desc.extra_bits;
+ var base = desc.extra_base;
+ var max_code = desc.max_code;
+ var max_length = desc.max_length;
+ var stree = desc.static_tree;
+ var h; // heap index
+ var n, m; // iterate over the tree elements
+ var bits; // bit length
+ var xbits; // extra bits
+ var f; // frequency
+ var overflow = 0; // number of elements with bit length too large
+
+ for(bits = 0; bits <= zip_MAX_BITS; bits++)
+ zip_bl_count[bits] = 0;
+
+ /* In a first pass, compute the optimal bit lengths (which may
+ * overflow in the case of the bit length tree).
+ */
+ tree[zip_heap[zip_heap_max]].dl = 0; // root of the heap
+
+ for(h = zip_heap_max + 1; h < zip_HEAP_SIZE; h++) {
+ n = zip_heap[h];
+ bits = tree[tree[n].dl].dl + 1;
+ if(bits > max_length) {
+ bits = max_length;
+ overflow++;
+ }
+ tree[n].dl = bits;
+ // We overwrite tree[n].dl which is no longer needed
+
+ if(n > max_code)
+ continue; // not a leaf node
+
+ zip_bl_count[bits]++;
+ xbits = 0;
+ if(n >= base)
+ xbits = extra[n - base];
+ f = tree[n].fc;
+ zip_opt_len += f * (bits + xbits);
+ if(stree != null)
+ zip_static_len += f * (stree[n].dl + xbits);
+ }
+ if(overflow == 0)
+ return;
+
+ // This happens for example on obj2 and pic of the Calgary corpus
+
+ // Find the first bit length which could increase:
+ do {
+ bits = max_length - 1;
+ while(zip_bl_count[bits] == 0)
+ bits--;
+ zip_bl_count[bits]--; // move one leaf down the tree
+ zip_bl_count[bits + 1] += 2; // move one overflow item as its brother
+ zip_bl_count[max_length]--;
+ /* The brother of the overflow item also moves one step up,
+ * but this does not affect bl_count[max_length]
+ */
+ overflow -= 2;
+ } while(overflow > 0);
+
+ /* Now recompute all bit lengths, scanning in increasing frequency.
+ * h is still equal to HEAP_SIZE. (It is simpler to reconstruct all
+ * lengths instead of fixing only the wrong ones. This idea is taken
+ * from 'ar' written by Haruhiko Okumura.)
+ */
+ for(bits = max_length; bits != 0; bits--) {
+ n = zip_bl_count[bits];
+ while(n != 0) {
+ m = zip_heap[--h];
+ if(m > max_code)
+ continue;
+ if(tree[m].dl != bits) {
+ zip_opt_len += (bits - tree[m].dl) * tree[m].fc;
+ tree[m].fc = bits;
+ }
+ n--;
+ }
+ }
+}
+
+ /* ==========================================================================
+ * Generate the codes for a given tree and bit counts (which need not be
+ * optimal).
+ * IN assertion: the array bl_count contains the bit length statistics for
+ * the given tree and the field len is set for all tree elements.
+ * OUT assertion: the field code is set for all tree elements of non
+ * zero code length.
+ */
+var zip_gen_codes = function(tree, // the tree to decorate
+ max_code) { // largest code with non zero frequency
+ var next_code = new Array(zip_MAX_BITS+1); // next code value for each bit length
+ var code = 0; // running code value
+ var bits; // bit index
+ var n; // code index
+
+ /* The distribution counts are first used to generate the code values
+ * without bit reversal.
+ */
+ for(bits = 1; bits <= zip_MAX_BITS; bits++) {
+ code = ((code + zip_bl_count[bits-1]) << 1);
+ next_code[bits] = code;
+ }
+
+ /* Check that the bit counts in bl_count are consistent. The last code
+ * must be all ones.
+ */
+// Assert (code + encoder->bl_count[MAX_BITS]-1 == (1<<MAX_BITS)-1,
+// "inconsistent bit counts");
+// Tracev((stderr,"\ngen_codes: max_code %d ", max_code));
+
+ for(n = 0; n <= max_code; n++) {
+ var len = tree[n].dl;
+ if(len == 0)
+ continue;
+ // Now reverse the bits
+ tree[n].fc = zip_bi_reverse(next_code[len]++, len);
+
+// Tracec(tree != static_ltree, (stderr,"\nn %3d %c l %2d c %4x (%x) ",
+// n, (isgraph(n) ? n : ' '), len, tree[n].fc, next_code[len]-1));
+ }
+}
+
+/* ==========================================================================
+ * Construct one Huffman tree and assigns the code bit strings and lengths.
+ * Update the total bit length for the current block.
+ * IN assertion: the field freq is set for all tree elements.
+ * OUT assertions: the fields len and code are set to the optimal bit length
+ * and corresponding code. The length opt_len is updated; static_len is
+ * also updated if stree is not null. The field max_code is set.
+ */
+var zip_build_tree = function(desc) { // the tree descriptor
+ var tree = desc.dyn_tree;
+ var stree = desc.static_tree;
+ var elems = desc.elems;
+ var n, m; // iterate over heap elements
+ var max_code = -1; // largest code with non zero frequency
+ var node = elems; // next internal node of the tree
+
+ /* Construct the initial heap, with least frequent element in
+ * heap[SMALLEST]. The sons of heap[n] are heap[2*n] and heap[2*n+1].
+ * heap[0] is not used.
+ */
+ zip_heap_len = 0;
+ zip_heap_max = zip_HEAP_SIZE;
+
+ for(n = 0; n < elems; n++) {
+ if(tree[n].fc != 0) {
+ zip_heap[++zip_heap_len] = max_code = n;
+ zip_depth[n] = 0;
+ } else
+ tree[n].dl = 0;
+ }
+
+ /* The pkzip format requires that at least one distance code exists,
+ * and that at least one bit should be sent even if there is only one
+ * possible code. So to avoid special checks later on we force at least
+ * two codes of non zero frequency.
+ */
+ while(zip_heap_len < 2) {
+ var xnew = zip_heap[++zip_heap_len] = (max_code < 2 ? ++max_code : 0);
+ tree[xnew].fc = 1;
+ zip_depth[xnew] = 0;
+ zip_opt_len--;
+ if(stree != null)
+ zip_static_len -= stree[xnew].dl;
+ // new is 0 or 1 so it does not have extra bits
+ }
+ desc.max_code = max_code;
+
+ /* The elements heap[heap_len/2+1 .. heap_len] are leaves of the tree,
+ * establish sub-heaps of increasing lengths:
+ */
+ for(n = zip_heap_len >> 1; n >= 1; n--)
+ zip_pqdownheap(tree, n);
+
+ /* Construct the Huffman tree by repeatedly combining the least two
+ * frequent nodes.
+ */
+ do {
+ n = zip_heap[zip_SMALLEST];
+ zip_heap[zip_SMALLEST] = zip_heap[zip_heap_len--];
+ zip_pqdownheap(tree, zip_SMALLEST);
+
+ m = zip_heap[zip_SMALLEST]; // m = node of next least frequency
+
+ // keep the nodes sorted by frequency
+ zip_heap[--zip_heap_max] = n;
+ zip_heap[--zip_heap_max] = m;
+
+ // Create a new node father of n and m
+ tree[node].fc = tree[n].fc + tree[m].fc;
+// depth[node] = (char)(MAX(depth[n], depth[m]) + 1);
+ if(zip_depth[n] > zip_depth[m] + 1)
+ zip_depth[node] = zip_depth[n];
+ else
+ zip_depth[node] = zip_depth[m] + 1;
+ tree[n].dl = tree[m].dl = node;
+
+ // and insert the new node in the heap
+ zip_heap[zip_SMALLEST] = node++;
+ zip_pqdownheap(tree, zip_SMALLEST);
+
+ } while(zip_heap_len >= 2);
+
+ zip_heap[--zip_heap_max] = zip_heap[zip_SMALLEST];
+
+ /* At this point, the fields freq and dad are set. We can now
+ * generate the bit lengths.
+ */
+ zip_gen_bitlen(desc);
+
+ // The field len is now set, we can generate the bit codes
+ zip_gen_codes(tree, max_code);
+}
+
+/* ==========================================================================
+ * Scan a literal or distance tree to determine the frequencies of the codes
+ * in the bit length tree. Updates opt_len to take into account the repeat
+ * counts. (The contribution of the bit length codes will be added later
+ * during the construction of bl_tree.)
+ */
+var zip_scan_tree = function(tree,// the tree to be scanned
+ max_code) { // and its largest code of non zero frequency
+ var n; // iterates over all tree elements
+ var prevlen = -1; // last emitted length
+ var curlen; // length of current code
+ var nextlen = tree[0].dl; // length of next code
+ var count = 0; // repeat count of the current code
+ var max_count = 7; // max repeat count
+ var min_count = 4; // min repeat count
+
+ if(nextlen == 0) {
+ max_count = 138;
+ min_count = 3;
+ }
+ tree[max_code + 1].dl = 0xffff; // guard
+
+ for(n = 0; n <= max_code; n++) {
+ curlen = nextlen;
+ nextlen = tree[n + 1].dl;
+ if(++count < max_count && curlen == nextlen)
+ continue;
+ else if(count < min_count)
+ zip_bl_tree[curlen].fc += count;
+ else if(curlen != 0) {
+ if(curlen != prevlen)
+ zip_bl_tree[curlen].fc++;
+ zip_bl_tree[zip_REP_3_6].fc++;
+ } else if(count <= 10)
+ zip_bl_tree[zip_REPZ_3_10].fc++;
+ else
+ zip_bl_tree[zip_REPZ_11_138].fc++;
+ count = 0; prevlen = curlen;
+ if(nextlen == 0) {
+ max_count = 138;
+ min_count = 3;
+ } else if(curlen == nextlen) {
+ max_count = 6;
+ min_count = 3;
+ } else {
+ max_count = 7;
+ min_count = 4;
+ }
+ }
+}
+
+ /* ==========================================================================
+ * Send a literal or distance tree in compressed form, using the codes in
+ * bl_tree.
+ */
+var zip_send_tree = function(tree, // the tree to be scanned
+ max_code) { // and its largest code of non zero frequency
+ var n; // iterates over all tree elements
+ var prevlen = -1; // last emitted length
+ var curlen; // length of current code
+ var nextlen = tree[0].dl; // length of next code
+ var count = 0; // repeat count of the current code
+ var max_count = 7; // max repeat count
+ var min_count = 4; // min repeat count
+
+ /* tree[max_code+1].dl = -1; */ /* guard already set */
+ if(nextlen == 0) {
+ max_count = 138;
+ min_count = 3;
+ }
+
+ for(n = 0; n <= max_code; n++) {
+ curlen = nextlen;
+ nextlen = tree[n+1].dl;
+ if(++count < max_count && curlen == nextlen) {
+ continue;
+ } else if(count < min_count) {
+ do { zip_SEND_CODE(curlen, zip_bl_tree); } while(--count != 0);
+ } else if(curlen != 0) {
+ if(curlen != prevlen) {
+ zip_SEND_CODE(curlen, zip_bl_tree);
+ count--;
+ }
+ // Assert(count >= 3 && count <= 6, " 3_6?");
+ zip_SEND_CODE(zip_REP_3_6, zip_bl_tree);
+ zip_send_bits(count - 3, 2);
+ } else if(count <= 10) {
+ zip_SEND_CODE(zip_REPZ_3_10, zip_bl_tree);
+ zip_send_bits(count-3, 3);
+ } else {
+ zip_SEND_CODE(zip_REPZ_11_138, zip_bl_tree);
+ zip_send_bits(count-11, 7);
+ }
+ count = 0;
+ prevlen = curlen;
+ if(nextlen == 0) {
+ max_count = 138;
+ min_count = 3;
+ } else if(curlen == nextlen) {
+ max_count = 6;
+ min_count = 3;
+ } else {
+ max_count = 7;
+ min_count = 4;
+ }
+ }
+}
+
+/* ==========================================================================
+ * Construct the Huffman tree for the bit lengths and return the index in
+ * bl_order of the last bit length code to send.
+ */
+var zip_build_bl_tree = function() {
+ var max_blindex; // index of last bit length code of non zero freq
+
+ // Determine the bit length frequencies for literal and distance trees
+ zip_scan_tree(zip_dyn_ltree, zip_l_desc.max_code);
+ zip_scan_tree(zip_dyn_dtree, zip_d_desc.max_code);
+
+ // Build the bit length tree:
+ zip_build_tree(zip_bl_desc);
+ /* opt_len now includes the length of the tree representations, except
+ * the lengths of the bit lengths codes and the 5+5+4 bits for the counts.
+ */
+
+ /* Determine the number of bit length codes to send. The pkzip format
+ * requires that at least 4 bit length codes be sent. (appnote.txt says
+ * 3 but the actual value used is 4.)
+ */
+ for(max_blindex = zip_BL_CODES-1; max_blindex >= 3; max_blindex--) {
+ if(zip_bl_tree[zip_bl_order[max_blindex]].dl != 0) break;
+ }
+ /* Update opt_len to include the bit length tree and counts */
+ zip_opt_len += 3*(max_blindex+1) + 5+5+4;
+// Tracev((stderr, "\ndyn trees: dyn %ld, stat %ld",
+// encoder->opt_len, encoder->static_len));
+
+ return max_blindex;
+}
+
+/* ==========================================================================
+ * Send the header for a block using dynamic Huffman trees: the counts, the
+ * lengths of the bit length codes, the literal tree and the distance tree.
+ * IN assertion: lcodes >= 257, dcodes >= 1, blcodes >= 4.
+ */
+var zip_send_all_trees = function(lcodes, dcodes, blcodes) { // number of codes for each tree
+ var rank; // index in bl_order
+
+// Assert (lcodes >= 257 && dcodes >= 1 && blcodes >= 4, "not enough codes");
+// Assert (lcodes <= L_CODES && dcodes <= D_CODES && blcodes <= BL_CODES,
+// "too many codes");
+// Tracev((stderr, "\nbl counts: "));
+ zip_send_bits(lcodes-257, 5); // not +255 as stated in appnote.txt
+ zip_send_bits(dcodes-1, 5);
+ zip_send_bits(blcodes-4, 4); // not -3 as stated in appnote.txt
+ for(rank = 0; rank < blcodes; rank++) {
+// Tracev((stderr, "\nbl code %2d ", bl_order[rank]));
+ zip_send_bits(zip_bl_tree[zip_bl_order[rank]].dl, 3);
+ }
+
+ // send the literal tree
+ zip_send_tree(zip_dyn_ltree,lcodes-1);
+
+ // send the distance tree
+ zip_send_tree(zip_dyn_dtree,dcodes-1);
+}
+
+/* ==========================================================================
+ * Determine the best encoding for the current block: dynamic trees, static
+ * trees or store, and output the encoded block to the zip file.
+ */
+var zip_flush_block = function(eof) { // true if this is the last block for a file
+ var opt_lenb, static_lenb; // opt_len and static_len in bytes
+ var max_blindex; // index of last bit length code of non zero freq
+ var stored_len; // length of input block
+
+ stored_len = zip_strstart - zip_block_start;
+ zip_flag_buf[zip_last_flags] = zip_flags; // Save the flags for the last 8 items
+
+ // Construct the literal and distance trees
+ zip_build_tree(zip_l_desc);
+// Tracev((stderr, "\nlit data: dyn %ld, stat %ld",
+// encoder->opt_len, encoder->static_len));
+
+ zip_build_tree(zip_d_desc);
+// Tracev((stderr, "\ndist data: dyn %ld, stat %ld",
+// encoder->opt_len, encoder->static_len));
+ /* At this point, opt_len and static_len are the total bit lengths of
+ * the compressed block data, excluding the tree representations.
+ */
+
+ /* Build the bit length tree for the above two trees, and get the index
+ * in bl_order of the last bit length code to send.
+ */
+ max_blindex = zip_build_bl_tree();
+
+ // Determine the best encoding. Compute first the block length in bytes
+ opt_lenb = (zip_opt_len +3+7)>>3;
+ static_lenb = (zip_static_len+3+7)>>3;
+
+// Trace((stderr, "\nopt %lu(%lu) stat %lu(%lu) stored %lu lit %u dist %u ",
+// opt_lenb, encoder->opt_len,
+// static_lenb, encoder->static_len, stored_len,
+// encoder->last_lit, encoder->last_dist));
+
+ if(static_lenb <= opt_lenb)
+ opt_lenb = static_lenb;
+ if(stored_len + 4 <= opt_lenb // 4: two words for the lengths
+ && zip_block_start >= 0) {
+ var i;
+
+ /* The test buf != NULL is only necessary if LIT_BUFSIZE > WSIZE.
+ * Otherwise we can't have processed more than WSIZE input bytes since
+ * the last block flush, because compression would have been
+ * successful. If LIT_BUFSIZE <= WSIZE, it is never too late to
+ * transform a block into a stored block.
+ */
+ zip_send_bits((zip_STORED_BLOCK<<1)+eof, 3); /* send block type */
+ zip_bi_windup(); /* align on byte boundary */
+ zip_put_short(stored_len);
+ zip_put_short(~stored_len);
+
+ // copy block
+/*
+ p = &window[block_start];
+ for(i = 0; i < stored_len; i++)
+ put_byte(p[i]);
+*/
+ for(i = 0; i < stored_len; i++)
+ zip_put_byte(zip_window[zip_block_start + i]);
+
+ } else if(static_lenb == opt_lenb) {
+ zip_send_bits((zip_STATIC_TREES<<1)+eof, 3);
+ zip_compress_block(zip_static_ltree, zip_static_dtree);
+ } else {
+ zip_send_bits((zip_DYN_TREES<<1)+eof, 3);
+ zip_send_all_trees(zip_l_desc.max_code+1,
+ zip_d_desc.max_code+1,
+ max_blindex+1);
+ zip_compress_block(zip_dyn_ltree, zip_dyn_dtree);
+ }
+
+ zip_init_block();
+
+ if(eof != 0)
+ zip_bi_windup();
+}
+
+/* ==========================================================================
+ * Save the match info and tally the frequency counts. Return true if
+ * the current block must be flushed.
+ */
+var zip_ct_tally = function(
+ dist, // distance of matched string
+ lc) { // match length-MIN_MATCH or unmatched char (if dist==0)
+ zip_l_buf[zip_last_lit++] = lc;
+ if(dist == 0) {
+ // lc is the unmatched char
+ zip_dyn_ltree[lc].fc++;
+ } else {
+ // Here, lc is the match length - MIN_MATCH
+ dist--; // dist = match distance - 1
+// Assert((ush)dist < (ush)MAX_DIST &&
+// (ush)lc <= (ush)(MAX_MATCH-MIN_MATCH) &&
+// (ush)D_CODE(dist) < (ush)D_CODES, "ct_tally: bad match");
+
+ zip_dyn_ltree[zip_length_code[lc]+zip_LITERALS+1].fc++;
+ zip_dyn_dtree[zip_D_CODE(dist)].fc++;
+
+ zip_d_buf[zip_last_dist++] = dist;
+ zip_flags |= zip_flag_bit;
+ }
+ zip_flag_bit <<= 1;
+
+ // Output the flags if they fill a byte
+ if((zip_last_lit & 7) == 0) {
+ zip_flag_buf[zip_last_flags++] = zip_flags;
+ zip_flags = 0;
+ zip_flag_bit = 1;
+ }
+ // Try to guess if it is profitable to stop the current block here
+ if(zip_compr_level > 2 && (zip_last_lit & 0xfff) == 0) {
+ // Compute an upper bound for the compressed length
+ var out_length = zip_last_lit * 8;
+ var in_length = zip_strstart - zip_block_start;
+ var dcode;
+
+ for(dcode = 0; dcode < zip_D_CODES; dcode++) {
+ out_length += zip_dyn_dtree[dcode].fc * (5 + zip_extra_dbits[dcode]);
+ }
+ out_length >>= 3;
+// Trace((stderr,"\nlast_lit %u, last_dist %u, in %ld, out ~%ld(%ld%%) ",
+// encoder->last_lit, encoder->last_dist, in_length, out_length,
+// 100L - out_length*100L/in_length));
+ if(zip_last_dist < parseInt(zip_last_lit/2) &&
+ out_length < parseInt(in_length/2))
+ return true;
+ }
+ return (zip_last_lit == zip_LIT_BUFSIZE-1 ||
+ zip_last_dist == zip_DIST_BUFSIZE);
+ /* We avoid equality with LIT_BUFSIZE because of wraparound at 64K
+ * on 16 bit machines and because stored blocks are restricted to
+ * 64K-1 bytes.
+ */
+}
+
+ /* ==========================================================================
+ * Send the block data compressed using the given Huffman trees
+ */
+var zip_compress_block = function(
+ ltree, // literal tree
+ dtree) { // distance tree
+ var dist; // distance of matched string
+ var lc; // match length or unmatched char (if dist == 0)
+ var lx = 0; // running index in l_buf
+ var dx = 0; // running index in d_buf
+ var fx = 0; // running index in flag_buf
+ var flag = 0; // current flags
+ var code; // the code to send
+ var extra; // number of extra bits to send
+
+ if(zip_last_lit != 0) do {
+ if((lx & 7) == 0)
+ flag = zip_flag_buf[fx++];
+ lc = zip_l_buf[lx++] & 0xff;
+ if((flag & 1) == 0) {
+ zip_SEND_CODE(lc, ltree); /* send a literal byte */
+// Tracecv(isgraph(lc), (stderr," '%c' ", lc));
+ } else {
+ // Here, lc is the match length - MIN_MATCH
+ code = zip_length_code[lc];
+ zip_SEND_CODE(code+zip_LITERALS+1, ltree); // send the length code
+ extra = zip_extra_lbits[code];
+ if(extra != 0) {
+ lc -= zip_base_length[code];
+ zip_send_bits(lc, extra); // send the extra length bits
+ }
+ dist = zip_d_buf[dx++];
+ // Here, dist is the match distance - 1
+ code = zip_D_CODE(dist);
+// Assert (code < D_CODES, "bad d_code");
+
+ zip_SEND_CODE(code, dtree); // send the distance code
+ extra = zip_extra_dbits[code];
+ if(extra != 0) {
+ dist -= zip_base_dist[code];
+ zip_send_bits(dist, extra); // send the extra distance bits
+ }
+ } // literal or match pair ?
+ flag >>= 1;
+ } while(lx < zip_last_lit);
+
+ zip_SEND_CODE(zip_END_BLOCK, ltree);
+}
+
+/* ==========================================================================
+ * Send a value on a given number of bits.
+ * IN assertion: length <= 16 and value fits in length bits.
+ */
+var zip_Buf_size = 16; // bit size of bi_buf
+var zip_send_bits = function(
+ value, // value to send
+ length) { // number of bits
+ /* If not enough room in bi_buf, use (valid) bits from bi_buf and
+ * (16 - bi_valid) bits from value, leaving (width - (16-bi_valid))
+ * unused bits in value.
+ */
+ if(zip_bi_valid > zip_Buf_size - length) {
+ zip_bi_buf |= (value << zip_bi_valid);
+ zip_put_short(zip_bi_buf);
+ zip_bi_buf = (value >> (zip_Buf_size - zip_bi_valid));
+ zip_bi_valid += length - zip_Buf_size;
+ } else {
+ zip_bi_buf |= value << zip_bi_valid;
+ zip_bi_valid += length;
+ }
+}
+
+/* ==========================================================================
+ * Reverse the first len bits of a code, using straightforward code (a faster
+ * method would use a table)
+ * IN assertion: 1 <= len <= 15
+ */
+var zip_bi_reverse = function(
+ code, // the value to invert
+ len) { // its bit length
+ var res = 0;
+ do {
+ res |= code & 1;
+ code >>= 1;
+ res <<= 1;
+ } while(--len > 0);
+ return res >> 1;
+}
+
+/* ==========================================================================
+ * Write out any remaining bits in an incomplete byte.
+ */
+var zip_bi_windup = function() {
+ if(zip_bi_valid > 8) {
+ zip_put_short(zip_bi_buf);
+ } else if(zip_bi_valid > 0) {
+ zip_put_byte(zip_bi_buf);
+ }
+ zip_bi_buf = 0;
+ zip_bi_valid = 0;
+}
+
+var zip_qoutbuf = function() {
+ if(zip_outcnt != 0) {
+ var q, i;
+ q = zip_new_queue();
+ if(zip_qhead == null)
+ zip_qhead = zip_qtail = q;
+ else
+ zip_qtail = zip_qtail.next = q;
+ q.len = zip_outcnt - zip_outoff;
+// System.arraycopy(zip_outbuf, zip_outoff, q.ptr, 0, q.len);
+ for(i = 0; i < q.len; i++)
+ q.ptr[i] = zip_outbuf[zip_outoff + i];
+ zip_outcnt = zip_outoff = 0;
+ }
+}
+
+var zip_deflate = function(str, level) {
+ var i, j;
+
+ zip_deflate_data = str;
+ zip_deflate_pos = 0;
+ if(typeof level == "undefined")
+ level = zip_DEFAULT_LEVEL;
+ zip_deflate_start(level);
+
+ var buff = new Array(1024);
+ var aout = [];
+ while((i = zip_deflate_internal(buff, 0, buff.length)) > 0) {
+ var cbuf = new Array(i);
+ for(j = 0; j < i; j++){
+ cbuf[j] = String.fromCharCode(buff[j]);
+ }
+ aout[aout.length] = cbuf.join("");
+ }
+ zip_deflate_data = null; // G.C.
+ return aout.join("");
+}
+
+//
+// end of the script of Masanao Izumo.
+//
+
+// we add the compression method for JSZip
+return {
+ indicator : [ 0x08, 0x00 ],
+ compress : function (content) {
+ var base64 = content.toString("base64");
+ console.log(base64);
+ return zip_deflate(base64);
+ }
+}
+
+})();
View
2  janzip.js
@@ -65,7 +65,7 @@ var Zip = function () {
header.writeInt16(dt.time);
header.writeInt16(dt.date);
// crc32
- header.writeInt32(crc32(file.data));
+ header.writeInt32(crc32(compressedData));
// compressed size
header.writeInt32(compressedData.length);
// uncompressed size
View
1  jszip.js
@@ -314,6 +314,7 @@ JSZip.compressions = {
"STORE" : {
magic : "\x00\x00",
compress : function (content) {
+ console.log(content);
return content; // no compression
}
}
View
1  server.js
@@ -20,6 +20,7 @@ var fs = require("fs");
var archive = new zip();
archive.add("smile.txt", new Buffer("Smile!", "utf8"));
var buff = archive.toBuffer();
+
fs.writeFile("./test_new.zip", buff, function () {
console.log("im too");
});
View
BIN  test_new.zip
Binary file not shown
View
BIN  test_original.zip
Binary file not shown
Please sign in to comment.
Something went wrong with that request. Please try again.