[3365] | 1 | /**************************************************************************
|
---|
| 2 | *
|
---|
| 3 | * comp_dict.cpp -- Functions for loading the compression dictionary
|
---|
| 4 | * Copyright (C) 1994 Neil Sharman
|
---|
| 5 | *
|
---|
| 6 | * This program is free software; you can redistribute it and/or modify
|
---|
| 7 | * it under the terms of the GNU General Public License as published by
|
---|
| 8 | * the Free Software Foundation; either version 2 of the License, or
|
---|
| 9 | * (at your option) any later version.
|
---|
| 10 | *
|
---|
| 11 | * This program is distributed in the hope that it will be useful,
|
---|
| 12 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 13 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 14 | * GNU General Public License for more details.
|
---|
| 15 | *
|
---|
| 16 | * You should have received a copy of the GNU General Public License
|
---|
| 17 | * along with this program; if not, write to the Free Software
|
---|
| 18 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
| 19 | *
|
---|
| 20 | **************************************************************************/
|
---|
| 21 |
|
---|
| 22 | #include "sysfuncs.h"
|
---|
| 23 |
|
---|
| 24 | #include "huffman.h"
|
---|
| 25 | #include "local_strings.h"
|
---|
| 26 | #include "memlib.h"
|
---|
| 27 | #include "messages.h"
|
---|
| 28 |
|
---|
| 29 | #include "mg.h"
|
---|
| 30 | #include "hash.h"
|
---|
| 31 | #include "text.h"
|
---|
| 32 | #include "comp_dict.h"
|
---|
| 33 | #include "locallib.h"
|
---|
| 34 | #include "mg_files.h"
|
---|
| 35 |
|
---|
| 36 | compression_dict_header cdh;
|
---|
| 37 | compressed_text_header cth;
|
---|
| 38 | comp_frags_header cfh[2];
|
---|
| 39 |
|
---|
| 40 | dict_hash_table *ht[2];
|
---|
| 41 |
|
---|
| 42 | huff_data char_huff[2];
|
---|
| 43 | huff_data lens_huff[2];
|
---|
| 44 | u_long *char_codes[2], *lens_codes[2];
|
---|
| 45 | u_long Words_disk = 0;
|
---|
| 46 | u_long Chars_disk = 0;
|
---|
| 47 |
|
---|
| 48 |
|
---|
| 49 | static dict_hash_table *
|
---|
| 50 | ReadInWords (FILE * dict, comp_frags_header * cfh,
|
---|
| 51 | int esc)
|
---|
| 52 | {
|
---|
| 53 | int i;
|
---|
| 54 | u_char *allwords, *prev = NULL;
|
---|
| 55 | dict_hash_table *ht;
|
---|
| 56 | u_char **words;
|
---|
| 57 | u_long ht_size;
|
---|
| 58 |
|
---|
| 59 | ht_size = prime (cfh->hd.num_codes * HASH_RATIO);
|
---|
| 60 | if (!(ht = (dict_hash_table *) Xmalloc (sizeof (dict_hash_table) +
|
---|
| 61 | (ht_size - 1) * sizeof (ht->table[0]))))
|
---|
| 62 | {
|
---|
| 63 | Message ("no memory for hash_table\n");
|
---|
| 64 | return NULL;
|
---|
| 65 | }
|
---|
| 66 |
|
---|
| 67 | ht->size = ht_size;
|
---|
| 68 | ht->hd = &cfh->hd;
|
---|
| 69 |
|
---|
| 70 | if (!(ht->codes = Generate_Huffman_Codes (&cfh->hd, NULL)))
|
---|
| 71 | {
|
---|
| 72 | Message ("no memory for huffman codes\n");
|
---|
| 73 | return NULL;
|
---|
| 74 | }
|
---|
| 75 |
|
---|
| 76 | if (!(ht->words = (u_char **) Xmalloc (sizeof (u_char *) * cfh->hd.num_codes)))
|
---|
| 77 | {
|
---|
| 78 | Message ("no memory for word pointers\n");
|
---|
| 79 | return NULL;
|
---|
| 80 | }
|
---|
| 81 | words = ht->words;
|
---|
| 82 |
|
---|
| 83 | memset (ht->table, '\0', ht_size * sizeof (ht->table[0]));
|
---|
| 84 |
|
---|
| 85 | if (!(allwords = (u_char *) Xmalloc (sizeof (u_char) * cfh->uncompressed_size)))
|
---|
| 86 | {
|
---|
| 87 | Message ("no memory for words\n");
|
---|
| 88 | return NULL;
|
---|
| 89 | }
|
---|
| 90 |
|
---|
| 91 | for (i = 0; i < cfh->hd.num_codes; i++, words++)
|
---|
| 92 | {
|
---|
| 93 | register int val, copy;
|
---|
| 94 | val = fgetc (dict);
|
---|
| 95 | copy = (val >> 4) & 0xf;
|
---|
| 96 | val &= 0xf;
|
---|
| 97 |
|
---|
| 98 | *words = allwords;
|
---|
| 99 |
|
---|
| 100 | memcpy (allwords + 1, prev + 1, copy);
|
---|
| 101 | fread (allwords + copy + 1, sizeof (u_char), val, dict);
|
---|
| 102 | *allwords = val + copy;
|
---|
| 103 |
|
---|
| 104 | Words_disk += val + 1;
|
---|
| 105 |
|
---|
| 106 | /* insert into the hash table */
|
---|
| 107 | if (i < cfh->hd.num_codes - esc)
|
---|
| 108 | {
|
---|
| 109 | register u_char **wptr;
|
---|
| 110 | register unsigned long tsize = ht->size;
|
---|
| 111 | register unsigned long hashval, step;
|
---|
| 112 |
|
---|
| 113 | HASH (hashval, step, allwords, tsize);
|
---|
| 114 |
|
---|
| 115 | wptr = ht->table[hashval];
|
---|
| 116 | for (; wptr;)
|
---|
| 117 | {
|
---|
| 118 | hashval += step;
|
---|
| 119 | if (hashval >= tsize)
|
---|
| 120 | hashval -= tsize;
|
---|
| 121 | wptr = ht->table[hashval];
|
---|
| 122 | }
|
---|
| 123 | ht->table[hashval] = words;
|
---|
| 124 | }
|
---|
| 125 | prev = allwords;
|
---|
| 126 | allwords += *allwords + 1;
|
---|
| 127 | }
|
---|
| 128 | return ht;
|
---|
| 129 | }
|
---|
| 130 |
|
---|
| 131 | inline int LoadWords(FILE *dict, int which)
|
---|
| 132 | {
|
---|
| 133 | if (Read_cfh (dict, &cfh[which], NULL, &Words_disk) == -1)
|
---|
| 134 | return 1;
|
---|
| 135 |
|
---|
| 136 | if (!(ht[which] = ReadInWords (dict, &cfh[which], 0)))
|
---|
| 137 | return 1;
|
---|
| 138 |
|
---|
| 139 | return 0;
|
---|
| 140 | }
|
---|
| 141 |
|
---|
| 142 | inline int LoadHuffmanCodes(FILE *dict, int which)
|
---|
| 143 | {
|
---|
| 144 | if (Read_Huffman_Data (dict, &char_huff[which], NULL,
|
---|
| 145 | &Chars_disk) == -1)
|
---|
| 146 | return 1;
|
---|
| 147 | if (!(char_codes[which] =
|
---|
| 148 | Generate_Huffman_Codes (&char_huff[which], NULL)))
|
---|
| 149 | return 2;
|
---|
| 150 | if (Read_Huffman_Data (dict, &lens_huff[which], NULL,
|
---|
| 151 | &Chars_disk) == -1)
|
---|
| 152 | return 3;
|
---|
| 153 | if (!(lens_codes[which] =
|
---|
| 154 | Generate_Huffman_Codes (&lens_huff[which], NULL)))
|
---|
| 155 | return 4;
|
---|
| 156 |
|
---|
| 157 | return 0;
|
---|
| 158 | }
|
---|
| 159 |
|
---|
| 160 | int
|
---|
| 161 | LoadCompressionDictionary (char *dict_file_name)
|
---|
| 162 | {
|
---|
| 163 | FILE *dict;
|
---|
| 164 | int which;
|
---|
| 165 | if (!(dict = open_named_file (dict_file_name, "rb", MAGIC_DICT, MG_MESSAGE))) /* [RPAP - Feb 97: WIN32 Port] */
|
---|
| 166 | return COMPERROR;
|
---|
| 167 |
|
---|
| 168 | Words_disk = sizeof (u_long);
|
---|
| 169 |
|
---|
| 170 | if (Read_cdh (dict, &cdh, NULL, &Words_disk) == -1)
|
---|
| 171 | goto error;
|
---|
| 172 |
|
---|
| 173 | for (which = 0; which < 2; which++)
|
---|
| 174 | switch (cdh.dict_type)
|
---|
| 175 | {
|
---|
| 176 | case MG_COMPLETE_DICTIONARY:
|
---|
| 177 | {
|
---|
| 178 | if (LoadWords(dict, which) != 0)
|
---|
| 179 | goto error;
|
---|
| 180 | }
|
---|
| 181 | break;
|
---|
| 182 | case MG_PARTIAL_DICTIONARY:
|
---|
| 183 | {
|
---|
| 184 | if (cdh.num_words[which])
|
---|
| 185 | {
|
---|
| 186 | if (LoadWords(dict, which) != 0)
|
---|
| 187 | goto error;
|
---|
| 188 | }
|
---|
| 189 | else
|
---|
| 190 | ht[which] = NULL;
|
---|
| 191 | if (LoadHuffmanCodes(dict, which) != 0)
|
---|
| 192 | goto error;
|
---|
| 193 | }
|
---|
| 194 | break;
|
---|
| 195 | case MG_SEED_DICTIONARY:
|
---|
| 196 | {
|
---|
| 197 | if (cdh.num_words[which])
|
---|
| 198 | {
|
---|
| 199 | if (LoadWords(dict, which) != 0)
|
---|
| 200 | goto error;
|
---|
| 201 | }
|
---|
| 202 | else
|
---|
| 203 | ht[which] = NULL;
|
---|
| 204 | switch (cdh.novel_method)
|
---|
| 205 | {
|
---|
| 206 | case MG_NOVEL_HUFFMAN_CHARS:
|
---|
| 207 | if (LoadHuffmanCodes(dict, which) != 0)
|
---|
| 208 | goto error;
|
---|
| 209 | break;
|
---|
| 210 | case MG_NOVEL_DELTA:
|
---|
| 211 | break;
|
---|
| 212 | case MG_NOVEL_HYBRID:
|
---|
| 213 | break;
|
---|
| 214 | default:
|
---|
| 215 | FatalError (1, "Bad novel method");
|
---|
| 216 | }
|
---|
| 217 | }
|
---|
| 218 | break;
|
---|
| 219 | default:
|
---|
| 220 | FatalError (1, "Bad dictionary kind\n");
|
---|
| 221 | }
|
---|
| 222 |
|
---|
| 223 | return (COMPALLOK);
|
---|
| 224 |
|
---|
| 225 |
|
---|
| 226 | error:
|
---|
| 227 | fclose (dict);
|
---|
| 228 | return (COMPERROR);
|
---|
| 229 | }
|
---|