extensions/spellcheck/hunspell/src/hunzip.cpp

   1 /******* BEGIN LICENSE BLOCK *******
   2  * Version: MPL 1.1/GPL 2.0/LGPL 2.1
   3  *
   4  * The contents of this file are subject to the Mozilla Public License Version
   5  * 1.1 (the "License"); you may not use this file except in compliance with
   6  * the License. You may obtain a copy of the License at
   7  * http://www.mozilla.org/MPL/
   8  *
   9  * Software distributed under the License is distributed on an "AS IS" basis,
  10  * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
  11  * for the specific language governing rights and limitations under the
  12  * License.
  13  *
  14  * The Initial Developers of the Original Code are Kevin Hendricks (MySpell)
  15  * and László Németh (Hunspell). Portions created by the Initial Developers
  16  * are Copyright (C) 2002-2005 the Initial Developers. All Rights Reserved.
  17  *
  18  * Contributor(s): László Németh (nemethl@gyorsposta.hu)
  19  *
  20  * Alternatively, the contents of this file may be used under the terms of
  21  * either the GNU General Public License Version 2 or later (the "GPL"), or
  22  * the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
  23  * in which case the provisions of the GPL or the LGPL are applicable instead
  24  * of those above. If you wish to allow use of your version of this file only
  25  * under the terms of either the GPL or the LGPL, and not to allow others to
  26  * use your version of this file under the terms of the MPL, indicate your
  27  * decision by deleting the provisions above and replace them with the notice
  28  * and other provisions required by the GPL or the LGPL. If you do not delete
  29  * the provisions above, a recipient may use your version of this file under
  30  * the terms of any one of the MPL, the GPL or the LGPL.
  31  *
  32  ******* END LICENSE BLOCK *******/
  33
  34 #ifndef MOZILLA_CLIENT
  35 #include <cstdlib>
  36 #include <cstring>
  37 #include <cstdio>
  38 #else
  39 #include <stdlib.h>
  40 #include <string.h>
  41 #include <stdio.h>
  42 #endif
  43
  44 #include "hunzip.hxx"
  45
  46 #define CODELEN  65536
  47 #define BASEBITREC 5000
  48
  49 #define UNCOMPRESSED '\002'
  50 #define MAGIC "hz0"
  51 #define MAGIC_ENCRYPT "hz1"
  52 #define MAGICLEN (sizeof(MAGIC) - 1)
  53
  54 int Hunzip::fail(const char * err, const char * par) {
  55     fprintf(stderr, err, par);
  56     return -1;
  57 }
  58
  59 Hunzip::Hunzip(const char * file, const char * key) {
  60     bufsiz = 0;
  61     lastbit = 0;
  62     inc = 0;
  63     outc = 0;
  64     dec = NULL;
  65     filename = (char *) malloc(strlen(file) + 1);
  66     if (filename) strcpy(filename, file);
  67     if (getcode(key) == -1) bufsiz = -1;
  68     else bufsiz = getbuf();
  69 }
  70
  71 int Hunzip::getcode(const char * key) {
  72     unsigned char c[2];
  73     int i, j, n, p;
  74     int allocatedbit = BASEBITREC;
  75     const char * enc = key;
  76
  77     fin = fopen(filename, "rb");
  78     if (!fin) return -1;
  79
  80     // read magic number
  81     if ((fread(in, 1, 3, fin) < MAGICLEN)
  82         || !(strncmp(MAGIC, in, MAGICLEN) == 0 ||
  83                 strncmp(MAGIC_ENCRYPT, in, MAGICLEN) == 0)) {
  84             return fail(MSG_FORMAT, filename);
  85     }
  86
  87     // check encryption
  88     if (strncmp(MAGIC_ENCRYPT, in, MAGICLEN) == 0) {
  89         unsigned char cs;
  90         if (!key) return fail(MSG_KEY, filename);
  91         if (fread(&c, 1, 1, fin) < 1) return fail(MSG_FORMAT, filename);
  92         for (cs = 0; *enc; enc++) cs ^= *enc;
  93         if (cs != c[0]) return fail(MSG_KEY, filename);
  94         enc = key;
  95     } else key = NULL;
  96
  97     // read record count
  98     if (fread(&c, 1, 2, fin) < 2) return fail(MSG_FORMAT, filename);
  99
 100     if (key) {
 101         c[0] ^= *enc;
 102         if (*(++enc) == '\0') enc = key;
 103         c[1] ^= *enc;
 104     }
 105
 106     n = ((int) c[0] << 8) + c[1];
 107     dec = (struct bit *) malloc(BASEBITREC * sizeof(struct bit));
 108     if (!dec) return fail(MSG_MEMORY, filename);
 109     dec[0].v[0] = 0;
 110     dec[0].v[1] = 0;
 111
 112     // read codes
 113     for (i = 0; i < n; i++) {
 114         unsigned char l;
 115         if (fread(c, 1, 2, fin) < 2) return fail(MSG_FORMAT, filename);
 116         if (key) {
 117             if (*(++enc) == '\0') enc = key;
 118             c[0] ^= *enc;
 119             if (*(++enc) == '\0') enc = key;
 120             c[1] ^= *enc;
 121         }
 122         if (fread(&l, 1, 1, fin) < 1) return fail(MSG_FORMAT, filename);
 123         if (key) {
 124             if (*(++enc) == '\0') enc = key;
 125             l ^= *enc;
 126         }
 127         if (fread(in, 1, l/8+1, fin) < (size_t) l/8+1) return fail(MSG_FORMAT, filename);
 128         if (key) for (j = 0; j <= l/8; j++) {
 129             if (*(++enc) == '\0') enc = key;
 130             in[j] ^= *enc;
 131         }
 132         p = 0;
 133         for (j = 0; j < l; j++) {
 134             int b = (in[j/8] & (1 << (7 - (j % 8)))) ? 1 : 0;
 135             int oldp = p;
 136             p = dec[p].v[b];
 137             if (p == 0) {
 138                 lastbit++;
 139                 if (lastbit == allocatedbit) {
 140                     allocatedbit += BASEBITREC;
 141                     dec = (struct bit *) realloc(dec, allocatedbit * sizeof(struct bit));
 142                 }
 143                 dec[lastbit].v[0] = 0;
 144                 dec[lastbit].v[1] = 0;
 145                 dec[oldp].v[b] = lastbit;
 146                 p = lastbit;
 147             }
 148         }
 149         dec[p].c[0] = c[0];
 150         dec[p].c[1] = c[1];
 151     }
 152     return 0;
 153 }
 154
 155 Hunzip::~Hunzip()
 156 {
 157     if (dec) free(dec);
 158     if (fin) fclose(fin);
 159     if (filename) free(filename);
 160 }
 161
 162 int Hunzip::getbuf() {
 163     int p = 0;
 164     int o = 0;
 165     do {
 166         if (inc == 0) inbits = fread(in, 1, BUFSIZE, fin) * 8;
 167         for (; inc < inbits; inc++) {
 168             int b = (in[inc / 8] & (1 << (7 - (inc % 8)))) ? 1 : 0;
 169             int oldp = p;
 170             p = dec[p].v[b];
 171             if (p == 0) {
 172                 if (oldp == lastbit) {
 173                     fclose(fin);
 174                     fin = NULL;
 175                     // add last odd byte
 176                     if (dec[lastbit].c[0]) out[o++]  = dec[lastbit].c[1];
 177                     return o;
 178                 }
 179                 out[o++] = dec[oldp].c[0];
 180                 out[o++] = dec[oldp].c[1];
 181                 if (o == BUFSIZE) return o;
 182                 p = dec[p].v[b];
 183             }
 184         }
 185         inc = 0;
 186     } while (inbits == BUFSIZE * 8);
 187     return fail(MSG_FORMAT, filename);
 188 }
 189
 190 const char * Hunzip::getline() {
 191     char linebuf[BUFSIZE];
 192     int l = 0, eol = 0, left = 0, right = 0;
 193     if (bufsiz == -1) return NULL;
 194     while (l < bufsiz && !eol) {
 195         linebuf[l++] = out[outc];
 196         switch (out[outc]) {
 197             case '\t': break;
 198             case 31: { // escape
 199                 if (++outc == bufsiz) {
 200                     bufsiz = getbuf();
 201                     outc = 0;
 202                 }
 203                 linebuf[l - 1] = out[outc];
 204                 break;
 205             }
 206             case ' ': break;
 207             default: if (((unsigned char) out[outc]) < 47) {
 208                 if (out[outc] > 32) {
 209                     right = out[outc] - 31;
 210                     if (++outc == bufsiz) {
 211                         bufsiz = getbuf();
 212                         outc = 0;
 213                     }
 214                 }
 215                 if (out[outc] == 30) left = 9; else left = out[outc];
 216                 linebuf[l-1] = '\n';
 217                 eol = 1;
 218             }
 219         }
 220         if (++outc == bufsiz) {
 221             outc = 0;
 222             bufsiz = fin ? getbuf(): -1;
 223         }
 224     }
 225     if (right) strcpy(linebuf + l - 1, line + strlen(line) - right - 1);
 226     else linebuf[l] = '\0';
 227     strcpy(line + left, linebuf);
 228     return line;
 229 }