Datum
unaccent_init(PG_FUNCTION_ARGS)
{
List *dictoptions = (List *) PG_GETARG_POINTER(0);
SuffixChar *rootSuffixTree = NULL;
bool fileloaded = false;
ListCell *l;
foreach(l, dictoptions)
{
DefElem *defel = (DefElem *) lfirst(l);
if (pg_strcasecmp("Rules", defel->defname) == 0)
{
if (fileloaded)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("multiple Rules parameters")));
rootSuffixTree = initSuffixTree(defGetString(defel));
fileloaded = true;
}
else
{
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("unrecognized Unaccent parameter: \"%s\"",
defel->defname)));
}
}
/**
* Compresses the input text and writes the compressed data to a file.
* @param[in] filename name and path of the file to compress.
* @param[in] compressed name and path of the compressed output file.
* @param[in] algorithm the algorithm that will be used to build the suffix tree (Ukkonnen or Kurtz).
* @param[in] see if see will be used.
*/
static void zip(char *filename, char *compressed, BOOL algorithm, int parts, BOOL see) {
Uchar *origText, *prevText = NULL;
Uint origTextLen, partTextLen, currentTextLen;
FILE *compressed_file;
int i, part;
fsmTree_t stree = NULL, prevTree = NULL;
BOOL alloc = False;
#ifdef WIN32
HANDLE hndl;
origText = (Uchar *) file2String(filename, &origTextLen, &hndl);
#else
origText = (Uchar *) file2String(filename, &origTextLen);
#endif
if(origText == NULL) {
fprintf(stderr,"Cannot open file %s\n", filename);
exit(EXIT_FAILURE);
}
/*if(textLen > MAXTEXTLEN)
{
fprintf(stderr,"Sorry, textlen = %lu is larger than maximal textlen = %lu\n",
(Showuint) textLen,(Showuint) MAXTEXTLEN);
exit(EXIT_FAILURE);
}*/
if (!compressed) {
CALLOC(compressed, Uchar, strlen(filename) + 5);
strcpy(compressed, filename);
strcat(compressed, ".ctx");
alloc = True;
}
compressed_file = fopen(compressed, "wb");
if (!compressed_file) {
printf( "Could not open output file");
exit(1);
}
if (alloc) FREE(compressed);
buildAlpha(origText, origTextLen);
printf ("Alphasize: %ld\n", alphasize);
printf("Algorithm %d\n", algorithm);
setMaxCount();
/* write magic number */
putc(MAGIC >> 8, compressed_file);
putc(MAGIC, compressed_file);
/* write # of parts */
putc(parts, compressed_file);
initialize_output_bitstream();
initialize_arithmetic_encoder();
writeAlphabet(compressed_file);
currentTextLen = 0;
for (part = 1; part <= parts; part++) {
printf("---------- part %d ---------------\n", part);
if (part != parts) {
partTextLen = floor(origTextLen / parts);
}
else {
partTextLen = origTextLen - (floor(origTextLen / parts) * (parts - 1));
}
if (part > 1) {
prevText = text;
prevTree = stree;
}
textlen = partTextLen;
CALLOC(text, Uchar, textlen);
reversestring(origText + currentTextLen, textlen, text);
if (algorithm == UKKONEN) {
suffixTree_t tree = initSuffixTree();
buildSuffixTree(tree);
printf("Tree built\n");
pruneSuffixTree(tree);
stree = fsmSuffixTree(tree);
}
else {
stree = buildSTree();
printf("Tree built\n");
}
/*if (part > 1) {
copyStatistics(prevTree, stree, prevText);
FREE(prevText);
freeFsmTree(prevTree);
}*/
DEBUGCODE(printf("gamma hits: %d gamma Misses: %d\n", getHits(), getMisses()));
printf("height: %ld\n", getHeight(stree));
/* write textlen */
for (i=3; i>=0; i--) {
writeByte(textlen >> (8 * i), compressed_file);
}
printf ("Textlen: %ld\n", textlen);
writeFsmTree(stree, compressed_file);
printf("FSM...\n");
makeFsm(stree);
DEBUGCODE(printFsmTree(stree));
printf("Encoding...\n");
encode(stree, compressed_file, origText + currentTextLen, partTextLen, see);
currentTextLen += partTextLen;
}
FREE(text);
freeFsmTree(stree);
flush_arithmetic_encoder(compressed_file);
flush_output_bitstream(compressed_file);
#ifdef WIN32
freetextspace(origText, hndl);
#else
freetextspace(origText, origTextLen);
#endif
fclose(compressed_file);
}
