/* file fsamitriples.c 10/10/94 * 6/8/98 large scale reorganisation to omit globals, etc. * 13/1/98 changes for the `gen' type replacing char for generators. * This file contains the routine fsa_mitriples, which is used to construct the * general product fsa with multiple (labeled) initial-states in a * coset automatic group. * It is a variation of fsa_triples, allowing multiple initial states. * For general methodology, see comment at top of file fsalogic.c.
*/
/* readback must be true for the time being */ /* *waptr is assumed to be the word-acceptor of a coset automatic group. * (In particular, all states should be accepting.) * *diffptr is assumed to be a word-difference machine of the same automatic * group, with multiple initial states. * Both are assumed to be stored in dense-format. * This routine constructs the fsa of which the states are triples (s1,s2,d), * with s1 and s2 states of *waptr and d a state of *diffptr. * (More precisely, if *waptr has n states, then s1 and s2 may also be equal * to n+1, meaning that the end of string symbol has been read on lhs or rhs.) * The initial states are (1,1,i), where 1 is the initial state of *waptr, * and i an initial state of *diffptr. * The alphabet is 2-variable with base the alphabet of *waptr * (i.e. the same alphabet as *diffptr). * The alphabet member (g1,g2) maps (s1,s2,d) to (s1^g1,s2^g2,d^(g1,g2)) * if all three components are nonzero, and to zero otherwise. * The transition-table of the resulting fsa is output in the usual way to * file tempfilename with table-type specified by op_table_type, before * minimisation. * Short hash-tables will be used, so this routine won't work if *waptr * or *diffptr has more than MAXUSHORT states. * There are several categories of accept-states - one for each distinct * group element of word-length 0 or 1, and these are specified by * the labels of the states, which are lists of words (all words of length 1 * for the appropriate group element). * * If during the construction, a nontrivial equation between two words is * discovered as a result of encountering the identity word-difference, * then the word-acceptor *waptr must be accepting both of these words * which represent the same group-element, and must therefore be wrong. * The procedure therefore aborts without returning an fsa. * If the maxeqns is greater than zero, then a maximum of maxeqns such * equations are returned as eqnptr[i] - in order to do this, it is necessary * to store the defining transitions of the states as we proceed.
*/
fsa *fsa_mitriples(fsa *waptr, fsa *diffptr, storage_type op_table_type,
boolean destroy, char *tempfilename,
reduction_equation *eqnptr, int maxeqns, boolean eqnstop,
boolean *foundeqns, boolean readback)
{ if (kbm_print_level >= 3)
printf(" #Calling fsa_triples.\n"); if (waptr->states->size >= MAXUSHORT || diffptr->states->size >= MAXUSHORT) return fsa_mitriples_int(waptr, diffptr, op_table_type, destroy,
tempfilename, eqnptr, maxeqns, eqnstop, foundeqns,
readback); else return fsa_mitriples_short(waptr, diffptr, op_table_type, destroy,
tempfilename, eqnptr, maxeqns, eqnstop,
foundeqns, readback);
}
/* readback must be true for the time being */ static fsa *fsa_mitriples_short(fsa *waptr, fsa *diffptr,
storage_type op_table_type, boolean destroy, char *tempfilename, reduction_equation *eqnptr, int maxeqns, boolean eqnstop,
boolean *foundeqns, boolean readback)
{ int **watable, ***difftable, identity, ngens, ngens1, nswa1, ne, ns = 0, *fsarow,
nt, cstate, cswa1, cswa2, csdiff, im, i, j, k, e, len = 0, rlen, ct, bstate,
bigger, numeqns, num_init;
gen *subwd, reduced_genno[MAXGEN + 1], reduced_gen[2]; /* for calculating and storing reductions of generators in case some * generators happen to be equal to others.
*/ int labno[MAXGEN + 1], nlab; unsignedshort *ht_ptr;
boolean dense_op;
fsa *mitriples;
srec *labels;
short_hash_table ht;
FILE *tempfile;
gen g1, g2, bg1, bg2; int maxv = 65536;
reduction_struct rs_wd; int separator; struct vertexd {
gen g1;
gen g2; int state;
} * definition = 0, *newdef; /* This is used to store the defining transition for the states of *mitriples. * If definition[i] = v, then state i is defined by the transition from * state v.state, with generator (v.g1,v.g2). * State 1 does not have a definition.
*/
if (kbm_print_level >= 3)
printf(" #Calling fsa_mitriples_short.\n");
if (!waptr->flags[DFA] || !diffptr->flags[MIDFA]) {
fprintf(stderr, "Error: fsa__mitriples only applies to (MI)DFA's.\n"); return 0;
} if (waptr->alphabet->type != IDENTIFIERS) {
fprintf(stderr, "Error in fsa_mitriples: first fsa has wrong type.\n"); return 0;
} if (waptr->num_accepting != waptr->states->size) {
fprintf(stderr, "Error in fsa_mitriples: first fsa should be a word-acceptor.\n"); return 0;
} if (diffptr->alphabet->type != PRODUCT || diffptr->alphabet->arity != 2) {
fprintf(stderr, "Error in fsa_mitriples: second fsa must be 2-variable.\n"); return 0;
} if (diffptr->states->type != WORDS) {
fprintf(
stderr, "Error in fsa_mitriples: second fsa must be word-difference type.\n"); return 0;
} if (!srec_equal(diffptr->alphabet->base, waptr->alphabet)) {
fprintf(stderr, "Error in fsa_mitriples: fsa's alphabet's don't match.\n"); return 0;
} if (waptr->states->size >= MAXUSHORT || diffptr->states->size >= MAXUSHORT) {
fprintf(stderr, "Error in fsa_mitriples: one of the fsa's has too many states.\n"); return 0;
}
tmalloc(mitriples, fsa, 1);
fsa_init(mitriples);
srec_copy(mitriples->alphabet, diffptr->alphabet);
mitriples->flags[MIDFA] = TRUE;
mitriples->flags[ACCESSIBLE] = TRUE;
mitriples->num_accepting = 0; /* In fact there will be lots of different categories of accept-states - * one for each generator - they will be recorded as labels of the states.
*/
if (maxeqns > 0) { /* We need to remember vertex definitions */
tmalloc(definition, struct vertexd, maxv);
ns = num_init;
}
*foundeqns = FALSE;
short_hash_init(&ht, TRUE, 3, 0, 0);
identity = diffptr->initial[1]; for (i = 1; i <= num_init; i++) {
ht_ptr = ht.current_ptr;
ht_ptr[0] = waptr->initial[1];
ht_ptr[1] = waptr->initial[1];
ht_ptr[2] = diffptr->initial[i];
im = short_hash_locate(&ht, 3); if (im != i) {
fprintf(stderr, "Hash-initialisation problem in fsa_mitriples.\n"); return 0;
}
}
if ((tempfile = fopen(tempfilename, "w")) == 0) {
fprintf(stderr, "Error: cannot open file %s\n", tempfilename); return 0;
} if (dense_op)
tmalloc(fsarow, int, ne) else tmalloc(fsarow, int, 2 * ne + 1)
cstate = 0; if (dense_op)
len = ne; /* The length of the fsarow output. */
nt = 0; /* Number of transitions in mitriples */
numeqns = 0; /* this becomes nonzero when we have started collecting * equations of equal words both accepted by word-acceptor.
*/ while (++cstate <= ht.num_recs) { if (kbm_print_level >= 3) { if ((cstate <= 1000 && cstate % 100 == 0) ||
(cstate <= 10000 && cstate % 1000 == 0) ||
(cstate <= 100000 && cstate % 5000 == 0) || cstate % 50000 == 0)
printf(" #cstate = %d; number of states = %d.\n", cstate,
ht.num_recs);
}
ht_ptr = short_hash_rec(&ht, cstate);
cswa1 = ht_ptr[0];
cswa2 = ht_ptr[1];
csdiff = ht_ptr[2]; if (!dense_op)
len = 0;
e = 0; /* e is the num,ber of the edge corresponding to the pair (g1,g2) */ for (g1 = 1; g1 <= ngens1; g1++) for (g2 = 1; g2 <= ngens1; g2++) {
e++; /* Calculate action of generator-pair (g1,g2) on state cstate */ if (g1 == ngens1 && g2 == ngens1) continue;
ht_ptr = ht.current_ptr;
ht_ptr[2] = dense_dtarget(difftable, g1, g2, csdiff); if (ht_ptr[2] == 0)
im = 0; else {
ht_ptr[0] =
g1 > ngens
? nswa1
: cswa1 == nswa1 ? 0 : dense_target(watable, g1, cswa1); if (ht_ptr[0] == 0)
im = 0; else {
ht_ptr[1] =
g2 > ngens
? nswa1
: cswa2 == nswa1 ? 0 : dense_target(watable, g2, cswa2); if (ht_ptr[1] == 0)
im = 0; else { if (eqnstop && ht_ptr[2] == identity && g1 != g2) { /* This means that we have found a new equation between two * distinct words accepted by the word-acceptor *gpwa, and so * *gpwa must have been wrong.
*/
*foundeqns = TRUE; if (kbm_print_level > 0 && numeqns == 0)
printf("#Equation found between two words accepted by " "word-acceptor.\n"); if (maxeqns > 0) { /* We reconstruct the equation explicitly */ /* First we calculate the length of the equation */ if (kbm_print_level >= 3)
printf(" #Calculating equation number %d.\n",
numeqns + 1);
len = 1;
bg1 = g1;
bg2 = g2;
bstate = cstate;
bigger = g2 > ngens ? 1 : g1 > ngens ? 2 : 0; /* bigger=1 or 2 means resp. lhs/rhs larger in shortlex order
*/ while (bstate > num_init) {
len++;
bg1 = definition[bstate].g1;
bg2 = definition[bstate].g2;
bstate = definition[bstate].state;
} if (bigger == 0)
bigger = bg1 > bg2 ? 1 : 2;
/* Now we allocate space for it and store it - * we insert the separator at the beginning
*/
len++; /* The right hand side will be preceded by the word in the * subgroup involved in the equation - this is the label of * the state bstate of *diffptr
*/
subwd = diffptr->states->words[bstate];
rlen = len + genstrlen(subwd);
tmalloc(eqnptr[numeqns].lhs, gen, len + 1);
tmalloc(eqnptr[numeqns].rhs, gen, rlen + 1);
genstrcpy(eqnptr[numeqns].rhs, subwd);
eqnptr[numeqns].lhs[len] = eqnptr[numeqns].rhs[rlen] = 0;
bg1 = g1;
bg2 = g2;
bstate = cstate; while (1) {
len--;
rlen--; if (bigger == 1) {
eqnptr[numeqns].lhs[len] = bg1 > ngens ? 0 : bg1;
eqnptr[numeqns].rhs[rlen] = bg2 > ngens ? 0 : bg2;
} else {
eqnptr[numeqns].rhs[rlen] = bg1 > ngens ? 0 : bg1;
eqnptr[numeqns].lhs[len] = bg2 > ngens ? 0 : bg2;
} if (bstate <= num_init) break;
bg1 = definition[bstate].g1;
bg2 = definition[bstate].g2;
bstate = definition[bstate].state;
}
eqnptr[numeqns].lhs[--len] = separator;
eqnptr[numeqns].rhs[--rlen] = separator;
}
if (numeqns == 0) { /* We are no longer constructing the fsa, so we no longer need
* the file. */
fclose(tempfile);
unlink(tempfilename);
}
numeqns++; if (numeqns >= maxeqns) { /* exit */ if (kbm_print_level >= 2 && maxeqns > 0)
printf(" #Found %d new equations - aborting.\n", maxeqns);
short_hash_clear(&ht);
tfree(fsarow);
fsa_clear(mitriples);
tfree(mitriples); if (maxeqns > 0)
tfree(definition); if (destroy)
fsa_clear(waptr); return 0;
} else
eqnptr[numeqns].lhs = 0; /* to mark how many we have later */
}
im = short_hash_locate(&ht, 3); if (im == -1) return 0; if (maxeqns > 0 && im > ns) {
ns++; if (ns == maxv) { /* need room for more definitions */ if (kbm_print_level >= 3)
printf( " #Allocating more space for vertex definitions.\n");
tmalloc(newdef, struct vertexd, 2 * maxv); for (i = 1; i < maxv; i++)
newdef[i] = definition[i];
tfree(definition);
definition = newdef;
maxv *= 2;
}
definition[ns].g1 = g1;
definition[ns].g2 = g2;
definition[ns].state = cstate;
}
}
}
}
if (dense_op)
fsarow[e - 1] = im; elseif (im > 0) {
fsarow[++len] = e;
fsarow[++len] = im;
} if (im > 0)
nt++;
} /* for (g1=1;g1<=ngens1; ... */ if (!dense_op)
fsarow[0] = len++; if (numeqns == 0)
fwrite((void *)fsarow, sizeof(int), (size_t)len, tempfile);
} /*while (++cstate <= ht.num_recs) */
if (numeqns > 0) {
short_hash_clear(&ht);
tfree(fsarow);
fsa_clear(mitriples);
tfree(mitriples);
tfree(definition); if (destroy)
fsa_clear(waptr); if (kbm_print_level >= 2)
printf(" #Found %d new equations - aborting with algorithm complete.\n",
numeqns); return 0;
}
if (kbm_print_level >= 3) {
printf(" #Calculated transitions - %d states, %d transitions.\n", ns,
nt);
printf(" #Now calculating state labels.\n");
}
tmalloc(mitriples->states->setToLabels, setToLabelsType, ns + 1);
mitriples->states->setToLabels[0] = 0; /* First we calculate the array reduced_genno, to record if any generators * are equal to a lower one.
*/
reduced_genno[0] = 0;
rs_wd.wd_fsa = diffptr; for (i = 1; i <= ngens; i++) {
reduced_gen[0] = i;
reduced_gen[1] = 0;
diff_reduce(reduced_gen, &rs_wd);
reduced_genno[i] = reduced_gen[0]; /* also OK if equal to null-string */
} for (i = 0; i <= ngens; i++)
labno[i] = 0;
labels = mitriples->states->labels;
labels->type = LISTOFWORDS; for (i = 1; i <= ngens; i++) {
tmalloc(labels->alphabet[i], char,
stringlen(waptr->alphabet->names[i]) + 1);
strcpy(labels->alphabet[i], waptr->alphabet->names[i]);
}
labels->alphabet_size = ngens;
tmalloc(labels->wordslist, gen **, ngens + num_init + 1); /* All states whose label has length <=1 will be classed as accept states, * since they are accept-states for some generator. * We need to mark them specifically in the "mi" case, since the information * is needed by the minimisation function midfa_labeled_minimize".
*/
tmalloc(mitriples->is_accepting, boolean, ns + 1); for (i = 1; i <= ns; i++)
mitriples->is_accepting[i] = FALSE;
mitriples->num_accepting = 0;
nlab = 0; for (i = 1; i <= ns; i++) {
ht_ptr = short_hash_rec(&ht, i);
csdiff = ht_ptr[2];
len = genstrlen(diffptr->states->words[csdiff]); if (len <= 1) {
mitriples->is_accepting[i] = TRUE;
mitriples->num_accepting++;
j = (len == 0) ? 0 : diffptr->states->words[csdiff][0]; if (labno[j] > 0)
mitriples->states->setToLabels[i] = labno[j]; else { /* new label - first see how many generators reduce to this */
nlab++;
ct = 0; for (k = 0; k <= ngens; k++) if (reduced_genno[k] == reduced_genno[j]) {
ct++;
labno[k] = nlab;
}
tmalloc(labels->wordslist[nlab], gen *, ct + 1);
ct = 0; for (k = 0; k <= ngens; k++) if (reduced_genno[k] == reduced_genno[j]) { if (k == 0) {
tmalloc(labels->wordslist[nlab][ct], gen, 1);
labels->wordslist[nlab][ct][0] = 0;
} else {
tmalloc(labels->wordslist[nlab][ct], gen, 2);
labels->wordslist[nlab][ct][0] = k;
labels->wordslist[nlab][ct][1] = 0;
}
ct++;
}
labels->wordslist[nlab][ct] = 0;
mitriples->states->setToLabels[i] = nlab;
}
} elseif (i <= num_init) {
nlab++;
tmalloc(labels->wordslist[nlab], gen *, 2);
tmalloc(labels->wordslist[nlab][0], gen,
genstrlen(diffptr->states->words[csdiff]) + 1);
genstrcpy(labels->wordslist[nlab][0], diffptr->states->words[csdiff]);
labels->wordslist[nlab][1] = 0;
mitriples->states->setToLabels[i] = nlab;
} else
mitriples->states->setToLabels[i] = 0;
}
labels->size = nlab;
short_hash_clear(&ht);
tfree(fsarow); if (maxeqns > 0)
tfree(definition); /* Now read the transition table back in */ if (readback) {
tempfile = fopen(tempfilename, "r");
compressed_transitions_read(mitriples, tempfile);
fclose(tempfile);
unlink(tempfilename);
}
tmalloc(mitriples->accepting, int, mitriples->num_accepting + 1);
ct = 0; for (i = 1; i <= ns; i++) if (mitriples->is_accepting[i])
mitriples->accepting[++ct] = i;
tfree(mitriples->is_accepting); if (destroy) {
fsa_clear(waptr);
fsa_clear(diffptr);
}
return mitriples;
}
/* readback must be true for the time being */ static fsa *fsa_mitriples_int(fsa *waptr, fsa *diffptr,
storage_type op_table_type, boolean destroy, char *tempfilename, reduction_equation *eqnptr, int maxeqns, boolean eqnstop, boolean *foundeqns,
boolean readback)
{ int **watable, ***difftable, identity, ngens, ngens1, nswa1, ne, ns = 0, *fsarow,
nt, cstate, cswa1, cswa2, csdiff, im, i, j, k, e, len = 0, rlen, ct, bstate,
bigger, numeqns, num_init;
gen *subwd, reduced_genno[MAXGEN + 1], reduced_gen[2]; /* for calculating and storing reductions of generators in case some * generators happen to be equal to others.
*/ int labno[MAXGEN + 1], nlab; int *ht_ptr;
boolean dense_op;
fsa *mitriples;
srec *labels;
hash_table ht;
FILE *tempfile;
gen g1, g2, bg1, bg2; int maxv = 65536;
reduction_struct rs_wd; int separator; struct vertexd {
gen g1;
gen g2; int state;
} * definition = 0, *newdef; /* This is used to store the defining transition for the states of *mitriples. * If definition[i] = v, then state i is defined by the transition from * state v.state, with generator (v.g1,v.g2). * State 1 does not have a definition.
*/
if (kbm_print_level >= 3)
printf(" #Calling fsa_mitriples_int.\n");
if (!waptr->flags[DFA] || !diffptr->flags[MIDFA]) {
fprintf(stderr, "Error: fsa__mitriples only applies to (MI)DFA's.\n"); return 0;
} if (waptr->alphabet->type != IDENTIFIERS) {
fprintf(stderr, "Error in fsa_mitriples: first fsa has wrong type.\n"); return 0;
} if (waptr->num_accepting != waptr->states->size) {
fprintf(stderr, "Error in fsa_mitriples: first fsa should be a word-acceptor.\n"); return 0;
} if (diffptr->alphabet->type != PRODUCT || diffptr->alphabet->arity != 2) {
fprintf(stderr, "Error in fsa_mitriples: second fsa must be 2-variable.\n"); return 0;
} if (diffptr->states->type != WORDS) {
fprintf(
stderr, "Error in fsa_mitriples: second fsa must be word-difference type.\n"); return 0;
} if (!srec_equal(diffptr->alphabet->base, waptr->alphabet)) {
fprintf(stderr, "Error in fsa_mitriples: fsa's alphabet's don't match.\n"); return 0;
} if (waptr->states->size >= MAXUSHORT || diffptr->states->size >= MAXUSHORT) {
fprintf(stderr, "Error in fsa_mitriples: one of the fsa's has too many states.\n"); return 0;
}
tmalloc(mitriples, fsa, 1);
fsa_init(mitriples);
srec_copy(mitriples->alphabet, diffptr->alphabet);
mitriples->flags[MIDFA] = TRUE;
mitriples->flags[ACCESSIBLE] = TRUE;
mitriples->num_accepting = 0; /* In fact there will be lots of different categories of accept-states - * one for each generator - they will be recorded as labels of the states.
*/
if (maxeqns > 0) { /* We need to remember vertex definitions */
tmalloc(definition, struct vertexd, maxv);
ns = num_init;
}
*foundeqns = FALSE;
hash_init(&ht, TRUE, 3, 0, 0);
identity = diffptr->initial[1]; for (i = 1; i <= num_init; i++) {
ht_ptr = ht.current_ptr;
ht_ptr[0] = waptr->initial[1];
ht_ptr[1] = waptr->initial[1];
ht_ptr[2] = diffptr->initial[i];
im = hash_locate(&ht, 3); if (im != i) {
fprintf(stderr, "Hash-initialisation problem in fsa_mitriples.\n"); return 0;
}
}
if ((tempfile = fopen(tempfilename, "w")) == 0) {
fprintf(stderr, "Error: cannot open file %s\n", tempfilename); return 0;
} if (dense_op)
tmalloc(fsarow, int, ne) else tmalloc(fsarow, int, 2 * ne + 1)
cstate = 0; if (dense_op)
len = ne; /* The length of the fsarow output. */
nt = 0; /* Number of transitions in mitriples */
numeqns = 0; /* this becomes nonzero when we have started collecting * equations of equal words both accepted by word-acceptor.
*/ while (++cstate <= ht.num_recs) { if (kbm_print_level >= 3) { if ((cstate <= 1000 && cstate % 100 == 0) ||
(cstate <= 10000 && cstate % 1000 == 0) ||
(cstate <= 100000 && cstate % 5000 == 0) || cstate % 50000 == 0)
printf(" #cstate = %d; number of states = %d.\n", cstate,
ht.num_recs);
}
ht_ptr = hash_rec(&ht, cstate);
cswa1 = ht_ptr[0];
cswa2 = ht_ptr[1];
csdiff = ht_ptr[2]; if (!dense_op)
len = 0;
e = 0; /* e is the num,ber of the edge corresponding to the pair (g1,g2) */ for (g1 = 1; g1 <= ngens1; g1++) for (g2 = 1; g2 <= ngens1; g2++) {
e++; /* Calculate action of generator-pair (g1,g2) on state cstate */ if (g1 == ngens1 && g2 == ngens1) continue;
ht_ptr = ht.current_ptr;
ht_ptr[2] = dense_dtarget(difftable, g1, g2, csdiff); if (ht_ptr[2] == 0)
im = 0; else {
ht_ptr[0] =
g1 > ngens
? nswa1
: cswa1 == nswa1 ? 0 : dense_target(watable, g1, cswa1); if (ht_ptr[0] == 0)
im = 0; else {
ht_ptr[1] =
g2 > ngens
? nswa1
: cswa2 == nswa1 ? 0 : dense_target(watable, g2, cswa2); if (ht_ptr[1] == 0)
im = 0; else { if (eqnstop && ht_ptr[2] == identity && g1 != g2) { /* This means that we have found a new equation between two * distinct words accepted by the word-acceptor *gpwa, and so * *gpwa must have been wrong.
*/
*foundeqns = TRUE; if (kbm_print_level > 0 && numeqns == 0)
printf("#Equation found between two words accepted by " "word-acceptor.\n"); if (maxeqns > 0) { /* We reconstruct the equation explicitly */ /* First we calculate the length of the equation */ if (kbm_print_level >= 3)
printf(" #Calculating equation number %d.\n",
numeqns + 1);
len = 1;
bg1 = g1;
bg2 = g2;
bstate = cstate;
bigger = g2 > ngens ? 1 : g1 > ngens ? 2 : 0; /* bigger=1 or 2 means resp. lhs/rhs larger in shortlex order
*/ while (bstate > num_init) {
len++;
bg1 = definition[bstate].g1;
bg2 = definition[bstate].g2;
bstate = definition[bstate].state;
} if (bigger == 0)
bigger = bg1 > bg2 ? 1 : 2;
/* Now we allocate space for it and store it - * we insert the separator at the beginning
*/
len++; /* The right hand side will be preceded by the word in the * subgroup involved in the equation - this is the label of * the state bstate of *diffptr
*/
subwd = diffptr->states->words[bstate];
rlen = len + genstrlen(subwd);
tmalloc(eqnptr[numeqns].lhs, gen, len + 1);
tmalloc(eqnptr[numeqns].rhs, gen, rlen + 1);
genstrcpy(eqnptr[numeqns].rhs, subwd);
eqnptr[numeqns].lhs[len] = eqnptr[numeqns].rhs[rlen] = 0;
bg1 = g1;
bg2 = g2;
bstate = cstate; while (1) {
len--;
rlen--; if (bigger == 1) {
eqnptr[numeqns].lhs[len] = bg1 > ngens ? 0 : bg1;
eqnptr[numeqns].rhs[rlen] = bg2 > ngens ? 0 : bg2;
} else {
eqnptr[numeqns].rhs[rlen] = bg1 > ngens ? 0 : bg1;
eqnptr[numeqns].lhs[len] = bg2 > ngens ? 0 : bg2;
} if (bstate <= num_init) break;
bg1 = definition[bstate].g1;
bg2 = definition[bstate].g2;
bstate = definition[bstate].state;
}
eqnptr[numeqns].lhs[--len] = separator;
eqnptr[numeqns].rhs[--rlen] = separator;
}
if (numeqns == 0) { /* We are no longer constructing the fsa, so we no longer need
* the file. */
fclose(tempfile);
unlink(tempfilename);
}
numeqns++; if (numeqns >= maxeqns) { /* exit */ if (kbm_print_level >= 2 && maxeqns > 0)
printf(" #Found %d new equations - aborting.\n", maxeqns);
hash_clear(&ht);
tfree(fsarow);
fsa_clear(mitriples);
tfree(mitriples); if (maxeqns > 0)
tfree(definition); if (destroy)
fsa_clear(waptr); return 0;
} else
eqnptr[numeqns].lhs = 0; /* to mark how many we have later */
}
im = hash_locate(&ht, 3); if (im == -1) return 0; if (maxeqns > 0 && im > ns) {
ns++; if (ns == maxv) { /* need room for more definitions */ if (kbm_print_level >= 3)
printf( " #Allocating more space for vertex definitions.\n");
tmalloc(newdef, struct vertexd, 2 * maxv); for (i = 1; i < maxv; i++)
newdef[i] = definition[i];
tfree(definition);
definition = newdef;
maxv *= 2;
}
definition[ns].g1 = g1;
definition[ns].g2 = g2;
definition[ns].state = cstate;
}
}
}
}
if (dense_op)
fsarow[e - 1] = im; elseif (im > 0) {
fsarow[++len] = e;
fsarow[++len] = im;
} if (im > 0)
nt++;
} /* for (g1=1;g1<=ngens1; ... */ if (!dense_op)
fsarow[0] = len++; if (numeqns == 0)
fwrite((void *)fsarow, sizeof(int), (size_t)len, tempfile);
} /*while (++cstate <= ht.num_recs) */
if (numeqns > 0) {
hash_clear(&ht);
tfree(fsarow);
fsa_clear(mitriples);
tfree(mitriples);
tfree(definition); if (destroy)
fsa_clear(waptr); if (kbm_print_level >= 2)
printf(" #Found %d new equations - aborting with algorithm complete.\n",
numeqns); return 0;
}
if (kbm_print_level >= 3) {
printf(" #Calculated transitions - %d states, %d transitions.\n", ns,
nt);
printf(" #Now calculating state labels.\n");
}
tmalloc(mitriples->states->setToLabels, setToLabelsType, ns + 1);
mitriples->states->setToLabels[0] = 0; /* First we calculate the array reduced_genno, to record if any generators * are equal to a lower one.
*/
reduced_genno[0] = 0;
rs_wd.wd_fsa = diffptr; for (i = 1; i <= ngens; i++) {
reduced_gen[0] = i;
reduced_gen[1] = 0;
diff_reduce(reduced_gen, &rs_wd);
reduced_genno[i] = reduced_gen[0]; /* also OK if equal to null-string */
} for (i = 0; i <= ngens; i++)
labno[i] = 0;
labels = mitriples->states->labels;
labels->type = LISTOFWORDS; for (i = 1; i <= ngens; i++) {
tmalloc(labels->alphabet[i], char,
stringlen(waptr->alphabet->names[i]) + 1);
strcpy(labels->alphabet[i], waptr->alphabet->names[i]);
}
labels->alphabet_size = ngens;
tmalloc(labels->wordslist, gen **, ngens + num_init + 1); /* All states whose label has length <=1 will be classed as accept states, * since they are accept-states for some generator. * We need to mark them specifically in the "mi" case, since the information * is needed by the minimisation function midfa_labeled_minimize".
*/
tmalloc(mitriples->is_accepting, boolean, ns + 1); for (i = 1; i <= ns; i++)
mitriples->is_accepting[i] = FALSE;
mitriples->num_accepting = 0;
nlab = 0; for (i = 1; i <= ns; i++) {
ht_ptr = hash_rec(&ht, i);
csdiff = ht_ptr[2];
len = genstrlen(diffptr->states->words[csdiff]); if (len <= 1) {
mitriples->is_accepting[i] = TRUE;
mitriples->num_accepting++;
j = (len == 0) ? 0 : diffptr->states->words[csdiff][0]; if (labno[j] > 0)
mitriples->states->setToLabels[i] = labno[j]; else { /* new label - first see how many generators reduce to this */
nlab++;
ct = 0; for (k = 0; k <= ngens; k++) if (reduced_genno[k] == reduced_genno[j]) {
ct++;
labno[k] = nlab;
}
tmalloc(labels->wordslist[nlab], gen *, ct + 1);
ct = 0; for (k = 0; k <= ngens; k++) if (reduced_genno[k] == reduced_genno[j]) { if (k == 0) {
tmalloc(labels->wordslist[nlab][ct], gen, 1);
labels->wordslist[nlab][ct][0] = 0;
} else {
tmalloc(labels->wordslist[nlab][ct], gen, 2);
labels->wordslist[nlab][ct][0] = k;
labels->wordslist[nlab][ct][1] = 0;
}
ct++;
}
labels->wordslist[nlab][ct] = 0;
mitriples->states->setToLabels[i] = nlab;
}
} elseif (i <= num_init) {
nlab++;
tmalloc(labels->wordslist[nlab], gen *, 2);
tmalloc(labels->wordslist[nlab][0], gen,
genstrlen(diffptr->states->words[csdiff]) + 1);
genstrcpy(labels->wordslist[nlab][0], diffptr->states->words[csdiff]);
labels->wordslist[nlab][1] = 0;
mitriples->states->setToLabels[i] = nlab;
} else
mitriples->states->setToLabels[i] = 0;
}
labels->size = nlab;
hash_clear(&ht);
tfree(fsarow); if (maxeqns > 0)
tfree(definition); /* Now read the transition table back in */ if (readback) {
tempfile = fopen(tempfilename, "r");
compressed_transitions_read(mitriples, tempfile);
fclose(tempfile);
unlink(tempfilename);
}
tmalloc(mitriples->accepting, int, mitriples->num_accepting + 1);
ct = 0; for (i = 1; i <= ns; i++) if (mitriples->is_accepting[i])
mitriples->accepting[++ct] = i;
tfree(mitriples->is_accepting); if (destroy) {
fsa_clear(waptr);
fsa_clear(diffptr);
}
Die Informationen auf dieser Webseite wurden
nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit,
noch Qualität der bereit gestellten Informationen zugesichert.
Bemerkung:
Die farbliche Syntaxdarstellung und die Messung sind noch experimentell.