grammar *
pgen(node *n)
{
nfagrammar *gr;
grammar *g;
gr = metacompile(n);
g = maketables(gr);
translatelabels(g);
addfirstsets(g);
PyObject_FREE(gr);
return g;
}
CSA::CSA(uchar *text, ulong n, unsigned samplerate, const char *loadFromFile, const char *saveToFile) {
this->n = n;
this->samplerate = samplerate;
uchar *bwt;
if (loadFromFile != 0)
bwt = LoadFromFile(loadFromFile);
else
bwt = BWT(text);
if (saveToFile != 0)
SaveToFile(saveToFile, bwt);
ulong i,min = 0,
max;
for (i=0;i<256;i++)
C[i]=0;
for (i=0;i<n;++i)
C[(int)bwt[i]]++;
for (i=0;i<256;i++)
if (C[i]>0) {min = i; break;}
for (i=255;i>=min;--i)
if (C[i]>0) {max = i; break;}
ulong prev=C[0], temp;
C[0]=0;
for (i=1;i<256;i++) {
temp = C[i];
C[i]=C[i-1]+prev;
prev = temp;
}
this->codetable = node::makecodetable(bwt,n);
alphabetrank = new THuffAlphabetRank(bwt,n, this->codetable,0);
//if (alphabetrank->Test(bwt,n)) printf("alphabetrank ok\n");
delete [] bwt;
// Make tables
maketables();
// to avoid running out of unsigned, the sizes are computed in specific order (large/small)*small
// |class CSA| +256*|TCodeEntry|+|C[]|+|suffixes[]+positions[]|+...
//printf("FMindex takes %d B\n",
// 6*W/8+256*3*W/8+256*W/8+ (2*n/(samplerate*8))*W+sampled->SpaceRequirementInBits()/8+alphabetrank->SpaceRequirementInBits()/8+W/8);
}
int main(int argc, char **argv)
{
FILE *f;
int i = 1;
const unsigned char *tables;
const unsigned char *base_of_tables;
/* By default, the default C locale is used rather than what the building user
happens to have set. However, if the -L option is given, set the locale from
the LC_xxx environment variables. */
if (argc > 1 && strcmp(argv[1], "-L") == 0)
{
setlocale(LC_ALL, ""); /* Set from environment variables */
i++;
}
if (argc < i + 1)
{
fprintf(stderr, "dftables: one filename argument is required\n");
return 1;
}
tables = maketables();
base_of_tables = tables;
f = fopen(argv[i], "wb");
if (f == NULL)
{
fprintf(stderr, "dftables: failed to open %s for writing\n", argv[1]);
return 1;
}
/* There are several fprintf() calls here, because gcc in pedantic mode
complains about the very long string otherwise. */
fprintf(f,
"/*************************************************\n"
"* Perl-Compatible Regular Expressions *\n"
"*************************************************/\n\n"
"/* This file was automatically written by the dftables auxiliary\n"
"program. It contains character tables that are used when no external\n"
"tables are passed to PCRE2 by the application that calls it. The tables\n"
"are used only for characters whose code values are less than 256.\n\n");
/* Force config.h in z/OS */
#if defined NATIVE_ZOS
fprintf(f,
"/* For z/OS, config.h is forced */\n"
"#ifndef HAVE_CONFIG_H\n"
"#define HAVE_CONFIG_H 1\n"
"#endif\n\n");
#endif
fprintf(f,
"The following #includes are present because without them gcc 4.x may remove\n"
"the array definition from the final binary if PCRE2 is built into a static\n"
"library and dead code stripping is activated. This leads to link errors.\n"
"Pulling in the header ensures that the array gets flagged as \"someone\n"
"outside this compilation unit might reference this\" and so it will always\n"
"be supplied to the linker. */\n\n");
fprintf(f,
"#ifdef HAVE_CONFIG_H\n"
"#include \"config.h\"\n"
"#endif\n\n"
"#include \"pcre2_internal.h\"\n\n");
fprintf(f,
"const uint8_t PRIV(default_tables)[] = {\n\n"
"/* This table is a lower casing table. */\n\n");
fprintf(f, " ");
for (i = 0; i < 256; i++)
{
if ((i & 7) == 0 && i != 0) fprintf(f, "\n ");
fprintf(f, "%3d", *tables++);
if (i != 255) fprintf(f, ",");
}
fprintf(f, ",\n\n");
fprintf(f, "/* This table is a case flipping table. */\n\n");
fprintf(f, " ");
for (i = 0; i < 256; i++)
{
if ((i & 7) == 0 && i != 0) fprintf(f, "\n ");
fprintf(f, "%3d", *tables++);
if (i != 255) fprintf(f, ",");
}
fprintf(f, ",\n\n");
fprintf(f,
"/* This table contains bit maps for various character classes.\n"
"Each map is 32 bytes long and the bits run from the least\n"
"significant end of each byte. The classes that have their own\n"
"maps are: space, xdigit, digit, upper, lower, word, graph\n"
"print, punct, and cntrl. Other classes are built from combinations. */\n\n");
fprintf(f, " ");
for (i = 0; i < cbit_length; i++)
{
if ((i & 7) == 0 && i != 0)
{
if ((i & 31) == 0) fprintf(f, "\n");
fprintf(f, "\n ");
}
fprintf(f, "0x%02x", *tables++);
if (i != cbit_length - 1) fprintf(f, ",");
}
fprintf(f, ",\n\n");
fprintf(f,
"/* This table identifies various classes of character by individual bits:\n"
" 0x%02x white space character\n"
" 0x%02x letter\n"
" 0x%02x decimal digit\n"
" 0x%02x hexadecimal digit\n"
" 0x%02x alphanumeric or '_'\n"
" 0x%02x regular expression metacharacter or binary zero\n*/\n\n",
ctype_space, ctype_letter, ctype_digit, ctype_xdigit, ctype_word,
ctype_meta);
fprintf(f, " ");
for (i = 0; i < 256; i++)
{
if ((i & 7) == 0 && i != 0)
{
fprintf(f, " /* ");
if (isprint(i-8)) fprintf(f, " %c -", i-8);
else fprintf(f, "%3d-", i-8);
if (isprint(i-1)) fprintf(f, " %c ", i-1);
else fprintf(f, "%3d", i-1);
fprintf(f, " */\n ");
}
fprintf(f, "0x%02x", *tables++);
if (i != 255) fprintf(f, ",");
}
fprintf(f, "};/* ");
if (isprint(i-8)) fprintf(f, " %c -", i-8);
else fprintf(f, "%3d-", i-8);
if (isprint(i-1)) fprintf(f, " %c ", i-1);
else fprintf(f, "%3d", i-1);
fprintf(f, " */\n\n/* End of pcre2_chartables.c */\n");
fclose(f);
free((void *)base_of_tables);
return 0;
}
int
main(int argc, char *argv[])
{
int ch;
FILE *f;
struct quotahandle *qh;
int dflag = 0;
const char *volume = NULL;
const char *dumpfile = NULL;
while ((ch = getopt(argc, argv, "d")) != -1) {
switch (ch) {
case 'd': dflag = 1; break;
default: usage(); break;
}
}
if (optind >= argc) {
usage();
}
volume = argv[optind++];
if (optind < argc) {
dumpfile = argv[optind++];
}
if (optind < argc) {
usage();
}
qh = quota_open(volume);
if (qh == NULL) {
err(EXIT_FAILURE, "quota_open: %s", volume);
}
if (dumpfile != NULL) {
f = fopen(dumpfile, "r");
if (f == NULL) {
err(EXIT_FAILURE, "%s", dumpfile);
}
} else {
f = stdin;
dumpfile = "<stdin>";
}
maketables(qh);
if (dflag) {
struct qklist *seenkeys, *dropkeys;
seenkeys = qklist_create();
dropkeys = qklist_create();
readdumpfile(qh, f, dumpfile, seenkeys);
qklist_sort(seenkeys);
scankeys(qh, seenkeys, dropkeys);
purge(qh, dropkeys);
qklist_destroy(dropkeys);
qklist_destroy(seenkeys);
} else {
readdumpfile(qh, f, dumpfile, NULL);
}
if (f != stdin) {
fclose(f);
}
quota_close(qh);
return EXIT_SUCCESS;
}
