1 /* 2 * CDDL HEADER START 3 * 4 * The contents of this file are subject to the terms of the 5 * Common Development and Distribution License, Version 1.0 only 6 * (the "License"). You may not use this file except in compliance 7 * with the License. 8 * 9 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE 10 * or http://www.opensolaris.org/os/licensing. 11 * See the License for the specific language governing permissions 12 * and limitations under the License. 13 * 14 * When distributing Covered Code, include this CDDL HEADER in each 15 * file and include the License file at usr/src/OPENSOLARIS.LICENSE. 16 * If applicable, add the following below this CDDL HEADER, with the 17 * fields enclosed by brackets "[]" replaced with your own identifying 18 * information: Portions Copyright [yyyy] [name of copyright owner] 19 * 20 * CDDL HEADER END 21 */ 22 /* Copyright (c) 1984, 1986, 1987, 1988, 1989 AT&T */ 23 /* All Rights Reserved */ 24 25 26 #ident "%Z%%M% %I% %E% SMI" /* SVr4.0 1.4 */ 27 #include <unistd.h> 28 #include <stdlib.h> 29 #include <stdio.h> 30 #include <locale.h> 31 #include "hash.h" 32 #include "huff.h" 33 34 int encode(long, long *); 35 36 #define S (BYTE * sizeof (long)) 37 #define B (BYTE * sizeof (unsigned)) 38 unsigned *table; 39 int hindex[NI]; 40 unsigned wp; /* word pointer */ 41 int bp = B; /* bit pointer */ 42 static int ignore; 43 static int extra; 44 45 static int 46 append(register unsigned w1, register int i) 47 { 48 while (wp < ND - 1) { 49 table[wp] |= w1>>(B-bp); 50 i -= bp; 51 if (i < 0) { 52 bp = -i; 53 return (1); 54 } 55 w1 <<= bp; 56 bp = B; 57 wp++; 58 } 59 return (0); 60 } 61 62 63 /* 64 * usage: hashin N 65 * where N is number of words in dictionary 66 * and standard input contains sorted, unique 67 * hashed words in octal 68 */ 69 70 void 71 main(int argc, char **argv) 72 { 73 long h, k, d; 74 int i; 75 long count; 76 long w1; 77 long x; 78 int t, u; 79 double z; 80 81 /* Set locale environment variables local definitions */ 82 (void) setlocale(LC_ALL, ""); 83 #if !defined(TEXT_DOMAIN) /* Should be defined by cc -D */ 84 #define TEXT_DOMAIN "SYS_TEST" /* Use this only if it wasn't */ 85 #endif 86 (void) textdomain(TEXT_DOMAIN); 87 88 k = 0; 89 u = 0; 90 if (argc != 2) { 91 (void) fprintf(stderr, gettext("%s: arg count\n"), argv[0]); 92 exit(1); 93 } 94 table = (unsigned *)malloc(ND * sizeof (*table)); 95 if (table == 0) { 96 (void) fprintf(stderr, gettext("%s: no space for table\n"), 97 argv[0]); 98 exit(1); 99 } 100 if ((atof(argv[1])) == 0.0) { 101 (void) fprintf(stderr, gettext("%s: illegal count"), argv[0]); 102 exit(1); 103 } 104 105 z = huff((1L<<HASHWIDTH)/atof(argv[1])); 106 (void) fprintf(stderr, gettext("%s: expected code widths = %f\n"), 107 argv[0], z); 108 for (count = 0; scanf("%lo", (unsigned long *)&h) == 1; ++count) { 109 if ((t = h >> (HASHWIDTH - INDEXWIDTH)) != u) { 110 if (bp != B) 111 wp++; 112 bp = B; 113 while (u < t) 114 hindex[++u] = wp; 115 k = (long)t<<(HASHWIDTH-INDEXWIDTH); 116 } 117 d = h-k; 118 k = h; 119 for (;;) { 120 for (x = d; ; x /= 2) { 121 i = encode(x, &w1); 122 if (i > 0) 123 break; 124 } 125 if (i > B) { 126 if (!(append((unsigned)(w1>>(long) (i-B)), B) && 127 append((unsigned)(w1<<(long) (B+B-i)), 128 i-B))) 129 ignore++; 130 } else 131 if (!append((unsigned)(w1<<(long)(B-i)), i)) 132 ignore++; 133 d -= x; 134 if (d > 0) 135 extra++; 136 else 137 break; 138 } 139 } 140 if (bp != B) 141 wp++; 142 while (++u < NI) 143 hindex[u] = wp; 144 whuff(); 145 (void) fwrite((char *)hindex, sizeof (*hindex), NI, stdout); 146 (void) fwrite((char *)table, sizeof (*table), wp, stdout); 147 (void) fprintf(stderr, 148 gettext("%s: %ld items, %d ignored, %d extra, %u words occupied\n"), 149 argv[0], count, ignore, extra, wp); 150 count -= ignore; 151 (void) fprintf(stderr, "%s: %f table bits/item, %f table+index bits\n", 152 argv[0], (((float)BYTE * wp) * sizeof (*table) / count), 153 (BYTE * ((float)wp * sizeof (*table) + sizeof (hindex)) / count)); 154 exit(0); 155 } 156