+/*----------------------------------------------------------------------------
+ * jacu - Just Another Compression Utility
+ *----------------------------------------------------------------------------
+ * This file is part of jacu.
+ *
+ * jacu is free software; you can redistribute it and/or modify it under the
+ * terms of the GNU General Public License as published by the Free Software
+ * Foundation; either version 2 of the License, or (at your option) any later
+ * version.
+ *
+ * jacu is distributed in the hope that it will be useful, but WITHOUT ANY
+ * WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+ * FOR A PARTICULAR PURPOSE. See the GNU General Public License for more
+ * details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with jacu; if not, write to the Free Software Foundation, Inc., 59 Temple
+ * Place, Suite 330, Boston, MA 02111-1307 USA
+ *----------------------------------------------------------------------------
+ */
+
#include "bs.h"
#include <stdlib.h>
+#include <ctype.h>
/* Block Sorting Optimizado en memoria! */
+typedef struct _dict_ {
+ char *word;
+ int len;
+ int hits;
+} t_Diccionario;
+
+#define PALABRA(a) {a, sizeof(a)-1, 0}
+
+#define DICT_SIZE sizeof(dic)/sizeof(t_Diccionario)
+
+/* ASCII Numero 3 como caracter de escape */
+#define ESCAPE_CHARACTER 0x3
+
+/* Diccionario */
+t_Diccionario dic[] = {
+ /* Preposiciones (no todas) */
+ PALABRA(" ante"),
+ PALABRA(" bajo"),
+ PALABRA(" cabe"),
+ PALABRA(" con"),
+ PALABRA(" contra"),
+ /* PALABRA(" de"), XXX No se si conviene, solo ahorra 1 byte y puede romper localidad */
+ PALABRA(" desde"),
+ PALABRA(" hasta"),
+ PALABRA(" hacia"),
+ PALABRA(" para"),
+ PALABRA(" por"),
+ PALABRA(" según"),
+ PALABRA(" sin"),
+ PALABRA(" sobre"),
+ PALABRA(" tras"),
+ PALABRA(" mediante"),
+ PALABRA(" durante"),
+
+ /* Articulos */
+ PALABRA(" el "),
+ PALABRA(" la "),
+ PALABRA(" ella "),
+ PALABRA(" del "),
+ PALABRA(" los "),
+ PALABRA(" las "),
+ PALABRA(" lo "),
+ PALABRA(" que"),
+ PALABRA(" una "),
+ PALABRA(" también"),
+ PALABRA(" tambien"),
+ PALABRA(" cuando"),
+ PALABRA(" pero"),
+ PALABRA(" todo"),
+ /* Otras de test */
+ PALABRA(" si "),
+ PALABRA(" mas "),
+ PALABRA(" más "),
+ PALABRA(" porque"),
+ PALABRA(" entonces"),
+ PALABRA(" siempre"),
+ PALABRA(" segundo"),
+ PALABRA(" programa"),
+ PALABRA(" existe"),
+ PALABRA(" recien"),
+ PALABRA(" máximo"),
+ PALABRA(" mínimo"),
+ PALABRA(" casi"),
+ PALABRA(" sección"),
+ PALABRA(" informe"),
+ PALABRA(" Informe"),
+ PALABRA(" acción"),
+ PALABRA(" perdedor"),
+ PALABRA(" existencia"),
+ PALABRA(" aire"),
+ PALABRA(" árbol"),
+ PALABRA(" prueba"),
+ PALABRA(" muestra"),
+ PALABRA(" animal"),
+ PALABRA(" suerte"),
+ PALABRA(" portador"),
+ PALABRA(" molesto"),
+ PALABRA(" cielo"),
+ PALABRA(" impulso"),
+ PALABRA(" alcohol"),
+ PALABRA(" seguido"),
+ PALABRA(" permiso"),
+ PALABRA(" cuarto"),
+ PALABRA(" brillante"),
+ PALABRA(" tener"),
+ PALABRA(" ningún"),
+ PALABRA(" fácil"),
+ /* Algunas cosas de ingles para mejorar otros textos */
+ PALABRA(" as "),
+ PALABRA(" will "),
+ PALABRA(" with "),
+ PALABRA(" this "),
+ PALABRA(" program "),
+ PALABRA(" are "),
+ PALABRA(" The "),
+ PALABRA(" it "),
+ PALABRA(" which "),
+ PALABRA(" for "),
+ PALABRA(" be "),
+ PALABRA(" that "),
+ PALABRA(" is "),
+ PALABRA(" and "),
+ PALABRA(" to "),
+ PALABRA(" of "),
+ PALABRA(" the ")
+};
+
typedef struct _bs_decode_t_ {
char c;
- unsigned long int pos;
+ Uint32 sig;
} t_BlockSortDecode;
char es_menor(char *data, t_BlockSort *bs, int i, int j);
s1 = (t_BlockSortDecode *)d1;
s2 = (t_BlockSortDecode *)d2;
+ /* Hack si es el mismo caracter, dejo como mejor la de menor sig */
+ if (s1->c == s2->c) {
+ return (s1->sig - s2->sig);
+ }
+
return (s1->c - s2->c);
}
s1 = (t_BlockSortData *)d1;
s2 = (t_BlockSortData *)d2;
- return es_menor(s1->bs->data, s1->bs, s1->pos_inicial, s2->pos_inicial);
+ return es_menor(s1->bs->data, s1->bs, s1->pos_inicial, s2->pos_inicial);
}
char es_menor(char *data, t_BlockSort *bs, int i, int j)
{
- unsigned long int pi, pj, k;
+ Uint32 pi, pj, k;
for(k=0; k<bs->len; k++) {
pi = (i+k)%bs->len;
qsort(bs->array, bs->len, sizeof(t_BlockSortData), __compare);
}
-void print_(char *data, unsigned long int pos, unsigned long int len)
-{
- unsigned long int i;
-
- for(i=0; i<len; i++)
- printf("%c", data[(pos+i)%len]);
- printf("\n");
-}
-
int generar_salida(char *data, t_BlockSort *bs, char *salida)
{
- unsigned long int i, k;
+ Uint32 i, k;
char *out;
- /* Dejo lugar para guardar el k y el tamaño de este bloque */
- out = salida + sizeof(unsigned long int)*2;
+ /* Dejo lugar para guardar el K */
+ out = salida + sizeof(Uint32);
k=-1;
for(i=0; i<bs->len; i++) {
- /* print_(data, bs->array[i].pos_inicial, bs->len); */
out[i] = data[bs->array[i].pos_final];
if (bs->array[i].ord == 1) k = i;
}
return k;
}
-void bs_solve(char *in, char *out, t_BlockSort *bs, unsigned long int *k, unsigned int leido)
+void bs_solve(char *in, char *out, t_BlockSort *bs, Uint32 *k, Uint32 leido)
{
unsigned int l;
l = bs->len;
ordenar_array(in, bs);
(*k) = generar_salida(in, bs, out);
- /* Guardo el k y el tamaño en el array */
- memcpy(out, &leido, sizeof(unsigned long int));
- memcpy(out+sizeof(unsigned long int), k, sizeof(unsigned long int));
-
+ /* Guardo el k y el tamaño en el array */
+ memcpy(out, k, sizeof(Uint32));
bs->len = l;
}
-void bs_restore(char *dst, char *c, unsigned long int k, unsigned long int len)
+void bs_restore(char *dst, char *c, Uint32 k, Uint32 len)
{
- unsigned long int i, current;
+ Uint32 i, current;
t_BlockSortDecode *in;
in = malloc(sizeof(t_BlockSortDecode)*len);
for(i=0; i<len; i++) {
in[i].c = c[i];
- in[i].pos = i;
+ in[i].sig = i;
}
qsort(in, len, sizeof(t_BlockSortDecode), _compare);
i=0;
do {
dst[i++] = in[current].c;
- current = in[current].pos;
+ current = in[current].sig;
} while (current != k);
free(in);
}
-t_BlockSort *bs_create(unsigned long int len)
+t_BlockSort *bs_create(Uint32 len)
{
t_BlockSort *tmp;
free(bs);
}
+
+void bs_EOL(char *data, Uint32 pagesize, Uint32 *j)
+{
+ /* Trato de agregar 4 espacios antes y 4 despues de un \n */
+ int i = (*j);
+
+ /* Verifico poder hacerlo */
+ if ((i+9) >= pagesize) return; /* No pude! */
+
+ data[i++] = ' ';
+ data[i++] = ' ';
+ data[i++] = ' ';
+ data[i++] = ' ';
+ data[i++] = ' ';
+ data[i++] = ' ';
+ data[i++] = ' ';
+ data[i++] = ' ';
+ data[i++] = ' ';
+
+ (*j) += 9;
+}
+
+char check_hint(char c)
+{
+ static int current_pos = 0;
+ char hits = 0;
+ int i;
+ char hit_pos = -1;
+
+ for(i=0; i<DICT_SIZE; i++) {
+ /* Veo de no pasarme */
+ if (current_pos < dic[i].len) {
+ /* Si la palabra tiene suficientes hints */
+ if (dic[i].hits == current_pos) {
+ if (dic[i].word[current_pos] == c) {
+ dic[i].hits++;
+ hits++;
+ if (dic[i].hits == dic[i].len) {
+ hit_pos = i;
+ }
+ } else {
+ /* Esta palabra ya no tiene posibilidades */
+ dic[i].hits = 0;
+ }
+ }
+ } else {
+ dic[i].hits = 0;
+ }
+ }
+
+ current_pos++;
+
+ if (hits == 0) {
+ current_pos = 0;
+ return -1; /* No hay hints ! */
+ }
+ if (hits > 1) return -2; /* Tengo mas de 1 hint! */
+
+ if (hit_pos == -1) return -2; /* Tengo 1 solo hit pero no es completo */
+
+ /* Tengo 1 solo hint !!!! */
+ current_pos = 0;
+ return hit_pos;
+}
+
+void bs_clean_dic()
+{
+ int i;
+ for(i=0; i<DICT_SIZE; i++) {
+ dic[i].hits = 0;
+ }
+}
+
+int bs_readblock(FILE *fp, char *data, Uint32 pagesize, int usar_dic)
+{
+ Uint32 i=0;
+ char c, hint;
+ char buffer[100];
+ int j, buffer_pos=0;
+
+ while ((!feof(fp)) && ((i+buffer_pos) < pagesize)) {
+ c = fgetc(fp);
+
+ if (usar_dic != 1) {
+ data[i++] = c;
+ continue;
+ }
+
+ if (c == ESCAPE_CHARACTER) {
+ data[i++] = c;
+ data[i++] = 0xF;
+ bs_clean_dic();
+ continue;
+ }
+
+ hint = check_hint(c);
+
+ switch (hint) {
+ case -1:
+ /* No hay hints, vacio el buffer y luego saco el c */
+ for(j=0; j<buffer_pos; j++)
+ data[i++] = buffer[j];
+
+ data[i++] = c;
+ buffer_pos = 0;
+ bs_clean_dic();
+ break;
+ case -2:
+ /* Tengo mas de 1 hit, tengo posibilidades, guardo este char en el buffer */
+ buffer[buffer_pos++] = c;
+ break;
+ default:
+ /* Me retornaron un numero positivo!!, eso quiere decir que ese numero
+ * es la posicion dentro del diccionario de la palabra que puedo reemplazar
+ */
+ data[i++] = ESCAPE_CHARACTER;
+ /* Trato de hacer que el caracter sea comun en textos, para que no me rompa
+ * la localidad
+ */
+ data[i++] = hint;
+ bs_clean_dic();
+ /* El buffer no lo necesito mas */
+ buffer_pos = 0;
+ }
+ }
+
+ for(j=0; j<buffer_pos; j++)
+ data[i++] = buffer[j];
+ /* Saco un EOF que lee de mas */
+ if (i<pagesize) i--;
+
+ return i;
+}
+
+char *bs_finalblock(char *data, Uint32 len, Uint32 *new_size)
+{
+ Uint32 size=0, i, j;
+ char *out = NULL;
+ unsigned char tmp;
+
+ for(i=0; i < len; i++) {
+ if (data[i] == ESCAPE_CHARACTER) {
+ i++;
+ if (data[i] == 0xF) {
+ /* Solo tengo que dejar el ESCAPE_CHARACTER */
+ size++;
+ out = (char *)realloc(out, size*sizeof(char));
+ out[size-1] = data[i];
+ /* Salteo ese caracter */
+ } else {
+ /* Va una palabra!! */
+ tmp = data[i];
+ size += dic[(size_t)tmp].len;
+ out = (char *)realloc(out, size*sizeof(char));
+ for(j=0; j<dic[(size_t)tmp].len; j++) {
+ out[size-dic[(size_t)tmp].len+j] = dic[(size_t)tmp].word[j];
+ }
+ /* Salteo el caracter siguiente al de escape */
+ }
+ } else {
+ size++;
+ out = (char *)realloc(out, size*sizeof(char));
+ out[size-1] = data[i];
+ }
+ }
+
+ (*new_size) = size;
+ return out;
+}
+
projects / z.facultad / 75.06 / jacu.git / blobdiff