1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/extensions/spellcheck/hunspell/src/hunzip.cpp Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,227 @@
1.4 +/******* BEGIN LICENSE BLOCK *******
1.5 + * Version: MPL 1.1/GPL 2.0/LGPL 2.1
1.6 + *
1.7 + * The contents of this file are subject to the Mozilla Public License Version
1.8 + * 1.1 (the "License"); you may not use this file except in compliance with
1.9 + * the License. You may obtain a copy of the License at
1.10 + * http://www.mozilla.org/MPL/
1.11 + *
1.12 + * Software distributed under the License is distributed on an "AS IS" basis,
1.13 + * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
1.14 + * for the specific language governing rights and limitations under the
1.15 + * License.
1.16 + *
1.17 + * The Initial Developers of the Original Code are Kevin Hendricks (MySpell)
1.18 + * and László Németh (Hunspell). Portions created by the Initial Developers
1.19 + * are Copyright (C) 2002-2005 the Initial Developers. All Rights Reserved.
1.20 + *
1.21 + * Contributor(s): László Németh (nemethl@gyorsposta.hu)
1.22 + * Caolan McNamara (caolanm@redhat.com)
1.23 + *
1.24 + * Alternatively, the contents of this file may be used under the terms of
1.25 + * either the GNU General Public License Version 2 or later (the "GPL"), or
1.26 + * the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
1.27 + * in which case the provisions of the GPL or the LGPL are applicable instead
1.28 + * of those above. If you wish to allow use of your version of this file only
1.29 + * under the terms of either the GPL or the LGPL, and not to allow others to
1.30 + * use your version of this file under the terms of the MPL, indicate your
1.31 + * decision by deleting the provisions above and replace them with the notice
1.32 + * and other provisions required by the GPL or the LGPL. If you do not delete
1.33 + * the provisions above, a recipient may use your version of this file under
1.34 + * the terms of any one of the MPL, the GPL or the LGPL.
1.35 + *
1.36 + ******* END LICENSE BLOCK *******/
1.37 +
1.38 +#include <stdlib.h>
1.39 +#include <string.h>
1.40 +#include <stdio.h>
1.41 +
1.42 +#include "hunzip.hxx"
1.43 +
1.44 +#define CODELEN 65536
1.45 +#define BASEBITREC 5000
1.46 +
1.47 +#define UNCOMPRESSED '\002'
1.48 +#define MAGIC "hz0"
1.49 +#define MAGIC_ENCRYPT "hz1"
1.50 +#define MAGICLEN (sizeof(MAGIC) - 1)
1.51 +
1.52 +int Hunzip::fail(const char * err, const char * par) {
1.53 + fprintf(stderr, err, par);
1.54 + return -1;
1.55 +}
1.56 +
1.57 +Hunzip::Hunzip(const char * file, const char * key) {
1.58 + bufsiz = 0;
1.59 + lastbit = 0;
1.60 + inc = 0;
1.61 + outc = 0;
1.62 + dec = NULL;
1.63 + fin = NULL;
1.64 + filename = (char *) malloc(strlen(file) + 1);
1.65 + if (filename) strcpy(filename, file);
1.66 + if (getcode(key) == -1) bufsiz = -1;
1.67 + else bufsiz = getbuf();
1.68 +}
1.69 +
1.70 +int Hunzip::getcode(const char * key) {
1.71 + unsigned char c[2];
1.72 + int i, j, n, p;
1.73 + int allocatedbit = BASEBITREC;
1.74 + const char * enc = key;
1.75 +
1.76 + if (!filename) return -1;
1.77 +
1.78 + fin = fopen(filename, "rb");
1.79 + if (!fin) return -1;
1.80 +
1.81 + // read magic number
1.82 + if ((fread(in, 1, 3, fin) < MAGICLEN)
1.83 + || !(strncmp(MAGIC, in, MAGICLEN) == 0 ||
1.84 + strncmp(MAGIC_ENCRYPT, in, MAGICLEN) == 0)) {
1.85 + return fail(MSG_FORMAT, filename);
1.86 + }
1.87 +
1.88 + // check encryption
1.89 + if (strncmp(MAGIC_ENCRYPT, in, MAGICLEN) == 0) {
1.90 + unsigned char cs;
1.91 + if (!key) return fail(MSG_KEY, filename);
1.92 + if (fread(&c, 1, 1, fin) < 1) return fail(MSG_FORMAT, filename);
1.93 + for (cs = 0; *enc; enc++) cs ^= *enc;
1.94 + if (cs != c[0]) return fail(MSG_KEY, filename);
1.95 + enc = key;
1.96 + } else key = NULL;
1.97 +
1.98 + // read record count
1.99 + if (fread(&c, 1, 2, fin) < 2) return fail(MSG_FORMAT, filename);
1.100 +
1.101 + if (key) {
1.102 + c[0] ^= *enc;
1.103 + if (*(++enc) == '\0') enc = key;
1.104 + c[1] ^= *enc;
1.105 + }
1.106 +
1.107 + n = ((int) c[0] << 8) + c[1];
1.108 + dec = (struct bit *) malloc(BASEBITREC * sizeof(struct bit));
1.109 + if (!dec) return fail(MSG_MEMORY, filename);
1.110 + dec[0].v[0] = 0;
1.111 + dec[0].v[1] = 0;
1.112 +
1.113 + // read codes
1.114 + for (i = 0; i < n; i++) {
1.115 + unsigned char l;
1.116 + if (fread(c, 1, 2, fin) < 2) return fail(MSG_FORMAT, filename);
1.117 + if (key) {
1.118 + if (*(++enc) == '\0') enc = key;
1.119 + c[0] ^= *enc;
1.120 + if (*(++enc) == '\0') enc = key;
1.121 + c[1] ^= *enc;
1.122 + }
1.123 + if (fread(&l, 1, 1, fin) < 1) return fail(MSG_FORMAT, filename);
1.124 + if (key) {
1.125 + if (*(++enc) == '\0') enc = key;
1.126 + l ^= *enc;
1.127 + }
1.128 + if (fread(in, 1, l/8+1, fin) < (size_t) l/8+1) return fail(MSG_FORMAT, filename);
1.129 + if (key) for (j = 0; j <= l/8; j++) {
1.130 + if (*(++enc) == '\0') enc = key;
1.131 + in[j] ^= *enc;
1.132 + }
1.133 + p = 0;
1.134 + for (j = 0; j < l; j++) {
1.135 + int b = (in[j/8] & (1 << (7 - (j % 8)))) ? 1 : 0;
1.136 + int oldp = p;
1.137 + p = dec[p].v[b];
1.138 + if (p == 0) {
1.139 + lastbit++;
1.140 + if (lastbit == allocatedbit) {
1.141 + allocatedbit += BASEBITREC;
1.142 + dec = (struct bit *) realloc(dec, allocatedbit * sizeof(struct bit));
1.143 + }
1.144 + dec[lastbit].v[0] = 0;
1.145 + dec[lastbit].v[1] = 0;
1.146 + dec[oldp].v[b] = lastbit;
1.147 + p = lastbit;
1.148 + }
1.149 + }
1.150 + dec[p].c[0] = c[0];
1.151 + dec[p].c[1] = c[1];
1.152 + }
1.153 + return 0;
1.154 +}
1.155 +
1.156 +Hunzip::~Hunzip()
1.157 +{
1.158 + if (dec) free(dec);
1.159 + if (fin) fclose(fin);
1.160 + if (filename) free(filename);
1.161 +}
1.162 +
1.163 +int Hunzip::getbuf() {
1.164 + int p = 0;
1.165 + int o = 0;
1.166 + do {
1.167 + if (inc == 0) inbits = fread(in, 1, BUFSIZE, fin) * 8;
1.168 + for (; inc < inbits; inc++) {
1.169 + int b = (in[inc / 8] & (1 << (7 - (inc % 8)))) ? 1 : 0;
1.170 + int oldp = p;
1.171 + p = dec[p].v[b];
1.172 + if (p == 0) {
1.173 + if (oldp == lastbit) {
1.174 + fclose(fin);
1.175 + fin = NULL;
1.176 + // add last odd byte
1.177 + if (dec[lastbit].c[0]) out[o++] = dec[lastbit].c[1];
1.178 + return o;
1.179 + }
1.180 + out[o++] = dec[oldp].c[0];
1.181 + out[o++] = dec[oldp].c[1];
1.182 + if (o == BUFSIZE) return o;
1.183 + p = dec[p].v[b];
1.184 + }
1.185 + }
1.186 + inc = 0;
1.187 + } while (inbits == BUFSIZE * 8);
1.188 + return fail(MSG_FORMAT, filename);
1.189 +}
1.190 +
1.191 +const char * Hunzip::getline() {
1.192 + char linebuf[BUFSIZE];
1.193 + int l = 0, eol = 0, left = 0, right = 0;
1.194 + if (bufsiz == -1) return NULL;
1.195 + while (l < bufsiz && !eol) {
1.196 + linebuf[l++] = out[outc];
1.197 + switch (out[outc]) {
1.198 + case '\t': break;
1.199 + case 31: { // escape
1.200 + if (++outc == bufsiz) {
1.201 + bufsiz = getbuf();
1.202 + outc = 0;
1.203 + }
1.204 + linebuf[l - 1] = out[outc];
1.205 + break;
1.206 + }
1.207 + case ' ': break;
1.208 + default: if (((unsigned char) out[outc]) < 47) {
1.209 + if (out[outc] > 32) {
1.210 + right = out[outc] - 31;
1.211 + if (++outc == bufsiz) {
1.212 + bufsiz = getbuf();
1.213 + outc = 0;
1.214 + }
1.215 + }
1.216 + if (out[outc] == 30) left = 9; else left = out[outc];
1.217 + linebuf[l-1] = '\n';
1.218 + eol = 1;
1.219 + }
1.220 + }
1.221 + if (++outc == bufsiz) {
1.222 + outc = 0;
1.223 + bufsiz = fin ? getbuf(): -1;
1.224 + }
1.225 + }
1.226 + if (right) strcpy(linebuf + l - 1, line + strlen(line) - right - 1);
1.227 + else linebuf[l] = '\0';
1.228 + strcpy(line + left, linebuf);
1.229 + return line;
1.230 +}
