/* * $Header: /cvsroot/arc/arc/arcsq.c,v 1.2 2003/10/31 02:22:36 highlandsun Exp $ */ /* * ARC - Archive utility - ARCSQ * * Version 3.10, created on 01/30/86 at 20:10:46 * * (C) COPYRIGHT 1985 by System Enhancement Associates; ALL RIGHTS RESERVED * * By: Thom Henderson * * Description: This file contains the routines used to squeeze a file when * placing it in an archive. * * Language: Computer Innovations Optimizing C86 * * Programming notes: Most of the routines used for the Huffman squeezing * algorithm were lifted from the SQ program by Dick Greenlaw, as adapted to * CI-C86 by Robert J. Beilstein. */ #include #include "arc.h" #include "proto.h" /* stuff for Huffman squeezing */ #define TRUE 1 #define FALSE 0 #define ERROR (-1) #define SPEOF 256 /* special endfile token */ #define NOCHILD (-1) /* marks end of path through tree */ #define NUMVALS 257 /* 256 data values plus SPEOF */ #define NUMNODES (NUMVALS+NUMVALS-1) /* number of nodes */ #define MAXCOUNT (u_short) 65535/* biggest unsigned integer */ /* * The following array of structures are the nodes of the binary trees. The * first NUMVALS nodes become the leaves of the final tree and represent the * values of the data bytes being encoded and the special endfile, SPEOF. The * remaining nodes become the internal nodes of the final tree. */ struct nd { /* shared by unsqueezer */ u_short weight; /* number of appearances */ short tdepth; /* length on longest path in tree */ short lchild, rchild; /* indices to next level */ } node[NUMNODES]; /* use large buffer */ static int dctreehd; /* index to head of final tree */ /* * This is the encoding table: The bit strings have first bit in low bit. * Note that counts were scaled so code fits unsigned integer. */ static int codelen[NUMVALS]; /* number of bits in code */ static u_short code[NUMVALS]; /* code itself, right adjusted */ static u_short tcode; /* temporary code value */ static long valcount[NUMVALS]; /* actual count of times seen */ /* Variables used by encoding process */ int curin; /* value currently being encoded */ static int cbitsrem; /* # of code string bits left */ static u_short ccode; /* current code right justified */ static VOID scale(), heap(), adjust(), bld_tree(), init_enc(); static int cmptrees(), buildenc(); extern u_char *outbuf, *outbeg, *outend; VOID init_sq() { /* prepare for scanning pass */ int i; /* node index */ /* * Initialize all nodes to single element binary trees with zero * weight and depth. */ for (i = 0; i < NUMNODES; ++i) { node[i].weight = 0; node[i].tdepth = 0; node[i].lchild = NOCHILD; node[i].rchild = NOCHILD; } for (i = 0; i < NUMVALS; i++) valcount[i] = 0; } VOID hufb_tab(buf, len) /* add a byte to the tables */ u_char *buf; u_int len; { for (; len > 0; len--) { /* Build frequency info in tree */ if (node[*buf].weight != MAXCOUNT) node[*buf].weight++; /* bump weight counter */ valcount[*buf++]++; /* bump byte counter */ } } long pred_sq() { /* predict size of squeezed file */ int i; int btlist[NUMVALS]; /* list of intermediate * b-trees */ int listlen;/* length of btlist */ u_short ceiling;/* limit for scaling */ long size = 0; /* predicted size */ int numnodes; /* # of nodes in simplified tree */ node[SPEOF].weight = 1; /* signal end of input */ valcount[SPEOF] = 1; ceiling = MAXCOUNT; /* Keep trying to scale and encode */ do { scale(ceiling); ceiling /= 2; /* in case we rescale */ /* * Build list of single node binary trees having leaves for * the input values with non-zero counts */ for (i = listlen = 0; i < NUMVALS; ++i) { if (node[i].weight != 0) { node[i].tdepth = 0; btlist[listlen++] = i; } } /* * Arrange list of trees into a heap with the entry indexing * the node with the least weight at the top. */ heap(btlist, listlen); /* Convert the list of trees to a single decoding tree */ bld_tree(btlist, listlen); /* Initialize the encoding table */ init_enc(); /* * Try to build encoding table. Fail if any code is > 16 bits * long. */ } while (buildenc(0, dctreehd) == ERROR); /* Initialize encoding variables */ cbitsrem = 0; /* force initial read */ curin = 0; /* anything but endfile */ for (i = 0; i < NUMVALS; i++) /* add bits for each code */ size += valcount[i] * codelen[i]; size = (size + 7) / 8; /* reduce to number of bytes */ numnodes = dctreehd < NUMVALS ? 0 : dctreehd - (NUMVALS - 1); size += sizeof(short) + 2 * numnodes * sizeof(short); return size; } /* * The count of number of occurrances of each input value have already been * prevented from exceeding MAXCOUNT. Now we must scale them so that their * sum doesn't exceed ceiling and yet no non-zero count can become zero. This * scaling prevents errors in the weights of the interior nodes of the * Huffman tree and also ensures that the codes will fit in an unsigned * integer. Rescaling is used if necessary to limit the code length. */ static VOID scale(ceil) u_short ceil; /* upper limit on total weight */ { register int i; int ovflw, divisor; u_short w, sum; unsigned char increased; /* flag */ do { for (i = sum = ovflw = 0; i < NUMVALS; ++i) { if (node[i].weight > (ceil - sum)) ++ovflw; sum += node[i].weight; } divisor = ovflw + 1; /* Ensure no non-zero values are lost */ increased = FALSE; for (i = 0; i < NUMVALS; ++i) { w = node[i].weight; if (w < divisor && w != 0) { /* Don't fail to provide * a code if it's used * at all */ node[i].weight = divisor; increased = TRUE; } } } while (increased); /* Scaling factor chosen, now scale */ if (divisor > 1) for (i = 0; i < NUMVALS; ++i) node[i].weight /= divisor; } /* * heap() and adjust() maintain a list of binary trees as a heap with the top * indexing the binary tree on the list which has the least weight or, in * case of equal weights, least depth in its longest path. The depth part is * not strictly necessary, but tends to avoid long codes which might provoke * rescaling. */ static VOID heap(list, length) int list[], length; { register int i; for (i = (length - 2) / 2; i >= 0; --i) adjust(list, i, length - 1); } /* Make a heap from a heap with a new top */ static VOID adjust(list, top, bottom) int list[], top, bottom; { register int k, temp; k = 2 * top + 1; /* left child of top */ temp = list[top]; /* remember root node of top tree */ if (k <= bottom) { if (k < bottom && cmptrees(list[k], list[k + 1])) ++k; /* k indexes "smaller" child (in heap of trees) of top */ /* now make top index "smaller" of old top and smallest child */ if (cmptrees(temp, list[k])) { list[top] = list[k]; list[k] = temp; /* Make the changed list a heap */ adjust(list, k, bottom); /* recursive */ } } } /* * Compare two trees, if a > b return true, else return false. Note * comparison rules in previous comments. */ static int cmptrees(a, b) int a, b; /* root nodes of trees */ { if (node[a].weight > node[b].weight) return TRUE; if (node[a].weight == node[b].weight) if (node[a].tdepth > node[b].tdepth) return TRUE; return FALSE; } /* * HUFFMAN ALGORITHM: develops the single element trees into a single binary * tree by forming subtrees rooted in interior nodes having weights equal to * the sum of weights of all their descendents and having depth counts * indicating the depth of their longest paths. * * When all trees have been formed into a single tree satisfying the heap * property (on weight, with depth as a tie breaker) then the binary code * assigned to a leaf (value to be encoded) is then the series of left (0) * and right (1) paths leading from the root to the leaf. Note that trees are * removed from the heaped list by moving the last element over the top * element and reheaping the shorter list. */ #define MAXCHAR(a,b) ((a > b) ? a : b) static VOID bld_tree(list, len) int list[]; int len; { register int freenode; /* next free node in tree */ register struct nd *frnp; /* free node pointer */ int lch, rch; /* temps for left, right children */ /* * Initialize index to next available (non-leaf) node. Lower numbered * nodes correspond to leaves (data values). */ freenode = NUMVALS; while (len > 1) { /* Take from list two btrees with least * weight and build an interior node pointing * to them. This forms a new tree. */ lch = list[0]; /* This one will be left child */ /* delete top (least) tree from the list of trees */ list[0] = list[--len]; adjust(list, 0, len - 1); /* Take new top (least) tree. Reuse list slot later */ rch = list[0]; /* This one will be right child */ /* * Form new tree from the two least trees using a free node * as root. Put the new tree in the list. */ frnp = &node[freenode]; /* address of next free node */ list[0] = freenode++; /* put at top for now */ frnp->lchild = lch; frnp->rchild = rch; frnp->weight = node[lch].weight + node[rch].weight; frnp->tdepth = 1 + MAXCHAR(node[lch].tdepth, node[rch].tdepth); /* reheap list to get least tree at top */ adjust(list, 0, len - 1); } dctreehd = list[0]; /* head of final tree */ } static VOID init_enc() { register int i; /* Initialize encoding table */ for (i = 0; i < NUMVALS; ++i) codelen[i] = 0; } /* * Recursive routine to walk the indicated subtree and level and maintain the * current path code in bstree. When a leaf is found the entire code string * and length are put into the encoding table entry for the leaf's data value. * * Returns ERROR if codes are too long. */ static int buildenc(level, root) int level; /* level of tree being examined, from zero */ int root; /* root of subtree is also data value if leaf */ { register int l, r; l = node[root].lchild; r = node[root].rchild; if (l == NOCHILD && r == NOCHILD) { /* Leaf. Previous path * determines bit string code * of length level (bits 0 to * level - 1). Ensures unused * code bits are zero. */ codelen[root] = level; code[root] = tcode & (((u_short) ~ 0) >> (16 - level)); return (level > 16) ? ERROR : 0; } else { if (l != NOCHILD) { /* Clear path bit and continue deeper */ tcode &= ~(1 << level); if (buildenc(level + 1, l) == ERROR) return ERROR; /* pass back bad statuses */ } if (r != NOCHILD) { /* Set path bit and continue deeper */ tcode |= 1 << level; if (buildenc(level + 1, r) == ERROR) return ERROR; /* pass back bad statuses */ } } return 0; /* it worked if we reach here */ } #define OUT_INT(n) \ *outbeg++ = n & 0xff; /* first the low byte */ \ *outbeg++ = n >> 8; /* then the high byte */ /* Write out the header of the compressed file */ long head_sq() { register int l, r; int i, k; int numnodes; /* # of nodes in simplified tree */ outbeg = outbuf; /* * Write out a simplified decoding tree. Only the interior nodes are * written. When a child is a leaf index (representing a data value) * it is recoded as -(index + 1) to distinguish it from interior * indexes which are recoded as positive indexes in the new tree. * * Note that this tree will be empty for an empty file. */ numnodes = dctreehd < NUMVALS ? 0 : dctreehd - (NUMVALS - 1); OUT_INT(numnodes) for (k = 0, i = dctreehd; k < numnodes; ++k, --i) { l = node[i].lchild; r = node[i].rchild; l = l < NUMVALS ? -(l + 1) : dctreehd - l; r = r < NUMVALS ? -(r + 1) : dctreehd - r; OUT_INT(l) OUT_INT(r) } return sizeof(short) + numnodes * 2 * sizeof(short); } /* * This routine is used to perform the actual squeeze operation. It can only * be called after the file has been scanned. It returns the true length of * the squeezed entry. * * There are two unsynchronized bit-byte relationships here. The input stream * bytes are converted to bit strings of various lengths via the static * variables named c... These bit strings are concatenated without padding to * become the stream of encoded result bytes, which this function returns one * at a time. The EOF (end of file) is converted to SPEOF for convenience and * encoded like any other input value. True EOF is returned after that. */ long huf_buf(pbuf, plen, len, ob) u_char *pbuf; /* ncr'd input buffer */ u_int plen; /* length of pack buffer */ u_int len; /* length of input buffer */ FILE *ob; /* output file */ { int rbyte; /* Result byte value */ int need; /* number of bits */ long size = 0; if (len == 0) /* Account for EOF/SPEOF */ plen++; while (plen != 0) { if (outbeg > outend) { putb_pak(outbuf, (u_int) (outbeg-outbuf), ob); outbeg = outbuf; } rbyte = 0; need = 8; /* build one byte per call */ /* * Loop to build a byte of encoded data. */ loop: if (cbitsrem >= need) { /* if current code is big enough */ if (need == 0) { *outbeg++ = rbyte; size++; continue; } rbyte |= ccode << (8 - need); /* take what we need */ ccode >>= need; /* and leave the rest */ cbitsrem -= need; *outbeg++ = rbyte & 0xff; size++; continue; } /* We need more than current code */ if (cbitsrem > 0) { rbyte |= ccode << (8 - need); /* take what there is */ need -= cbitsrem; } /* No more bits in current code string */ if (curin == SPEOF) { /* The end of file token has been * encoded. Save result byte. */ cbitsrem = 0; if (need != 8) { *outbeg++ = rbyte; size++; } break; } /* Get an input byte */ if (plen == 1 && len == 0) curin = SPEOF; /* convenient for encoding */ else if (plen == 0) { /* Here we may have the situation where 8-need > cbitsrem. That is, we have more information in the local variable than in the 'bits remaining' global variable. Just returning would cause us bit loss. What we can do, is put everything in the local variable back on the small 16 bits queue. */ cbitsrem = 8-need; ccode = rbyte; break; } else curin = *pbuf++; plen--; ccode = code[curin]; /* get the new byte's code */ cbitsrem = codelen[curin]; goto loop; } if (len == 0) putb_pak(outbuf, (u_int) (outbeg - outbuf), ob); return (size); }