int ape_consume_token() {
int machine_id;
int token_id = ape_get_token_id();
Machine current_machine = ape_parser.current_machine;
int next_state = current_machine.token_transitions[current_machine.current_state][token_id];
printf("%d: (%d,%d) -> %d\t\t(Token ID: %d)\n", current_machine.machine_id, current_machine.current_state, token_id, next_state, ape_get_token_id());
if (next_state == APE_INVALID_STATE) {
// could not read the token, let's check if we can change the machine
machine_id = machine_id_consume_token(token_id);
next_state = current_machine.machine_transitions[current_machine.current_state][machine_id];
if (machine_id != APE_INVALID_STATE) {
// change the current machine
change_ape_machine(machine_id, ape_parser.machines[machine_id].initial_state);
current_machine.current_state = next_state;
// enqueue the current machine
push_apestack(current_machine);
return ape_consume_token();
} else if (is_ape_in_final_state() && !is_apestack_empty()) { // check if there's a machine in the queue
// retrieve the machine in its original state
Machine m = pop_apestack();
change_ape_machine_pop(m.machine_id, m.current_state);
return ape_consume_token();
} else {
// could not consume the token in any machine
return 0;
}
} else {
go_to_state(next_state);
return 1;
}
return 0;
}
int run_parser( TTreeView *TreeView1 , char *file_name )
{
Config_set *config_ptr;
GRAMMAR *G_ptr;
STATE *state_ptr;
TOKEN *T_ptr , *T_tmp;
FILE *input ,*out;
char grammar_rule[1024];
char *ptr , *ptr_string;
char *ptr_lhs , *ptr_rhs;
SET *S_ptr;
int i , j;
int rule_num;
strncpy( lambda , "£f" , strlen("£f")+1 );
G_start = NULL;
G_end = NULL;
terminal_start = NULL;
terminal_end = NULL;
nonterminal_start = NULL;
nonterminal_end = NULL;
first_set_start = NULL;
first_set_end = NULL;
follow_set_start = NULL;
follow_set_end = NULL;
predict_set_start = NULL;
predict_set_end = NULL;
config_start = NULL;
config_end = NULL;
state_start = NULL;
state_end = NULL;
state_num = 0;
//scanner();
input = fopen( file_name ,"r");
scan = fopen("out.txt","r");
if ( input == NULL )
{
Application->MessageBoxA("grammar file error","error",0);
return 0;
}
// out = fopen("table.csv","w");
for ( i = 0 ; fgets( grammar_rule , 1024 , input ) ; )
{
ptr = grammar_rule;
// remove \n
ptr = strstr(grammar_rule , "\n");
if(ptr != NULL)
*ptr = '\0';
// remove rule num
ptr = strstr( grammar_rule , "." );
// get left handside
ptr_lhs = ptr + 2;
ptr = strstr( grammar_rule , " " );
*ptr = '\0';
insert_token( &nonterminal_start , &nonterminal_end , ptr_lhs ); // make nonterminal table
insert_token( &first_set_start , &first_set_end , ptr_lhs ); // make first set table with nonterminal
insert_token( &follow_set_start , &follow_set_end , ptr_lhs ); // make follow set table
insert_token( &predict_set_start , &predict_set_end , ptr_lhs ); // make predict set table
// get right handside
ptr = strstr( ++ptr , "¡÷" );
do
{
ptr = strstr( ++ptr , " " );
while(isspace(*ptr))
ptr++;
ptr_rhs = ptr;
ptr = strstr( ptr_rhs , "|" );
if ( ptr != NULL && *(ptr+1) != '|' )
{
for ( j = 1 ; isspace(*(ptr-j)) ; j++ )
*(ptr-j) = '\0';
}
insert_grammar( ++i , ptr_lhs , ptr_rhs );
} while ( ptr != NULL && *(ptr+1) != '|' );
}
// remove nonterminal in termainal table
T_ptr = nonterminal_start;
i = 0;
while ( T_ptr != NULL )
{
T_tmp = search_token( terminal_start , T_ptr->string );
delete_token( &terminal_start , T_tmp );
T_ptr = T_ptr->next;
i++;
}
num_of_nonterm = i;
for ( T_ptr = terminal_start , i =0 ; T_ptr != NULL ; T_ptr = T_ptr->next , i++ )
{
insert_token( &first_set_start , &first_set_end , T_ptr->string ); // make first set table
}
num_of_term = ++i;
fill_frist_set();
fill_follow_set();
i = 0;
view = fopen("state.txt","w");
//view = stdout;
out = fopen("go_to_table.csv","w");
build_CFSM();
// build goto table
parser_table = (PARSER**) malloc( sizeof(PARSER) * state_num );
for( i = 0 ; i < state_num ; i++ )
{
parser_table[i] = (PARSER*) malloc( sizeof(PARSER) * ( num_of_term + num_of_nonterm ));
for( j = 0 ; j < num_of_term + num_of_nonterm ; j++ )
{
parser_table[i][j].go_to = go_to_state(i,j+1);
if( j+1 != make_id("$") )
parser_table[i][j].action = SHIFT;
else
parser_table[i][j].action = ACCEPT;
if( parser_table[i][j].go_to == 0 )
parser_table[i][j].action = ERROR;
}
}
// comput lalr lookahead
build_LALR_lookahead();
fprintf(out,"\t,");
for( j = 0 ; j < num_of_term + num_of_nonterm ; j++ )
{
if( j+1 != make_id(",") )
fprintf(out," %s,",idtostr(j+1));
else
fprintf(out," ' ,");
}
fprintf( out ,"\n");
// build action table
for( i = 0 , state_ptr = state_start ; i < state_num ; i++ , state_ptr = state_ptr->next )
{
fprintf(out,"state%d,",i);
for( j = 0 ; j < num_of_term + num_of_nonterm ; j++ )
{
for( config_ptr = state_ptr->config_set ; config_ptr != NULL ; config_ptr = config_ptr->set_next )
{
if( config_ptr->dot == NULL && search_set( config_ptr->lookahead , idtostr(j+1) ) == TRUE )
{
if(parser_table[i][j].go_to == 0)
{
//printf("hit\n");
parser_table[i][j].go_to = config_ptr->rule->rule;
parser_table[i][j].action = REDUCE;
}
//fprintf(out,"R%02d,",parser_table[i][j].go_to);
}
}
if(parser_table[i][j].go_to > 0)
{
if( parser_table[i][j].action == SHIFT )
fprintf( out , "S%02d,",parser_table[i][j].go_to);
else
fprintf( out , "R%02d,",parser_table[i][j].go_to);
}
else
fprintf( out , ",");
}
fprintf( out ,"\n");
}
fclose(out);
//printf( "%d\n", go_to_state(4,5) );
for ( i = 0 ; i < state_num ; i++ )
{
view_state(i);
fprintf(view,"\n");
//getch();
//system("cls");
}
shift_reduce_driver( TreeView1 );
free_token(&terminal_start);
free_token(&nonterminal_start);
free_grammar();
free_state(&state_start);
//free_config(&config_start);
//system("pause");
return 0;
}
void cql_escape_sequences_remover_t::push_character(const char c) {
switch (_state)
{
case ESC_SEQ_STATE_OUTSIDE:
if(c == ESCAPE)
go_to_state(ESC_SEQ_STATE_AFTER_ESCAPE);
else if((unsigned char)c == CSI)
go_to_state(ESC_SEQ_STATE_AFTER_ESCAPE_BRACKET);
else if(is_control_character(c))
go_to_state(ESC_SEQ_STATE_OUTSIDE);
else {
_buffer.push_back(c);
}
break;
case ESC_SEQ_STATE_AFTER_ESCAPE:
if(c == '[')
go_to_state(ESC_SEQ_STATE_AFTER_ESCAPE_BRACKET);
else if(c == ']')
go_to_state(ESC_SEQ_STATE_SKIP_TO_SEQ_END);
else if(strchr("%#()", c) != 0)
go_to_state(ESC_SEQ_STATE_SKIP_NEXT);
else {
/* current character is skipped */
go_to_state(ESC_SEQ_STATE_OUTSIDE);
}
break;
case ESC_SEQ_STATE_AFTER_ESCAPE_BRACKET:
if(c == '[')
go_to_state(ESC_SEQ_STATE_SKIP_NEXT);
else
go_to_state(ESC_SEQ_STATE_SKIP_TO_SEQ_END);
break;
case ESC_SEQ_STATE_SKIP_NEXT:
/* current character is skipped */
go_to_state(ESC_SEQ_STATE_OUTSIDE);
break;
case ESC_SEQ_STATE_SKIP_NEXT_2:
/* current character is skipped */
go_to_state(ESC_SEQ_STATE_SKIP_NEXT);
break;
case ESC_SEQ_STATE_SKIP_TO_SEQ_END:
if(strchr("ABCDEFGHIJKLMPXacdefghlmnqrsu`];", c) != 0)
go_to_state(ESC_SEQ_STATE_OUTSIDE);
/* current character is skipped */
break;
}
}
