/* Global directive code */
void printTreeRec(rdp_tree_node_data *node, int level) {
for (int i = 0; i < level; i++) text_printf(" ");
text_printf("%s\n", node->id);
for (void *edge = graph_next_out_edge(node); edge != NULL;
edge = graph_next_out_edge(edge))
printTreeRec((rdp_tree_node_data*) graph_destination(edge), level+1);
}
static int rdp_check_nullable(void * base)
{
int bad = 0;
rdp_data * temp =(rdp_data *) symbol_next_symbol_in_scope(base);
set_ work = SET_NULL;
while (temp != NULL)
{
if (temp->contains_null &&(temp->kind != K_CODE))
{
set_assign_set(& work, & temp->first);
set_intersect_set(& work, & temp->follow);
if (set_cardinality(& work)!= 0)
{
text_message(TEXT_ERROR, "LL(1) violation - rule\n %s ::= ", temp->id);
rdp_print_sub_item(temp, 1);
text_printf(".\n contains null but first and follow sets both include: ");
set_print_set(& work, rdp_token_string, 78);
text_printf("\n");
temp->ll1_violation = 1;
bad = 1;
}
}
temp =(rdp_data *) symbol_next_symbol_in_scope(temp);
}
return bad;
}
/* Check if we have a nullable alternate inside a nullable iterator */
static int rdp_check_nested_nullable(void * base)
{
int bad = 0;
rdp_data * temp =(rdp_data *) symbol_next_symbol_in_scope(base);
while (temp != NULL)
{
if (set_includes_element(& rdp_production_set, temp->kind)&& temp->kind != K_SEQUENCE)
{
rdp_list * inner = temp->list;
while (inner != NULL)
{
if (temp->lo == 0 && inner->production->contains_null)
{
text_message(TEXT_ERROR, "LL(1) violation - rule \'%s\'\n is nullable but contains the nullable subrule\n", temp->id);
text_printf(" %s ::= ", inner->production->id);
rdp_print_sub_item(inner->production, 1);
text_printf(".\n");
bad = 1;
temp->ll1_violation = 1;
inner->production->ll1_violation = 1;
}
inner = inner->next;
}
}
temp =(rdp_data *) symbol_next_symbol_in_scope(temp);
}
return bad;
}
void text_dump(void) /* debugging routine to dump text space */
{
char * p = text_bot - 1;
while (p <= text_top)
{
while (++p <= text_top && * p != 0)
text_printf("%c", isprint(* p)? * p: '.');
text_printf("\n");
}
}
void scan_vcg_print_edge(const void * edge)
{
static unsigned long edge_count = 0;
/* Set user inserted edges to red */
if (*((int*) edge) == 0)
text_printf("color:red");
#if defined(show_vcg_atom_numbers)
text_printf("label:\"%lu\" ", graph_atom_number(edge));
#endif
text_printf(" horizontal_order:%lu", 1000 - ++edge_count);
}
void text_print_time(void)
{
char line[80];
time_t timer = time(NULL);
strftime(line, 80, "%b %d %Y %H:%M:%S", localtime(& timer));
text_printf("%s", line);
}
static int rdp_check_disjoint(void * base)
{
int bad = 0;
set_ work = SET_NULL;
rdp_data * temp =(rdp_data *) symbol_next_symbol_in_scope(base);
while (temp != NULL)
{
if (set_includes_element(& rdp_production_set, temp->kind)&& temp->kind != K_SEQUENCE)
{
rdp_list * left = temp->list;
while (left != NULL)
{
rdp_list * right = left->next;
while (right != NULL)
{
/* First check for disjoint on epsilon */
if (left->production->contains_null && right->production->contains_null)
{
text_message(TEXT_ERROR, "LL(1) violation - rule \'%s\'\n", temp->id);
text_printf(" productions %s ::= ", left->production->id);
rdp_print_sub_item(left->production, 1);
text_printf(".\n and %s ::= ", right->production->id);
rdp_print_sub_item(right->production, 1);
text_printf(".\n are both nullable \n");
left->production->ll1_violation = 1;
right->production->ll1_violation = 1;
bad = 1;
}
set_assign_set(& work, & left->production->first);
set_intersect_set(& work, & right->production->first);
if (set_cardinality(& work)!= 0)
{
text_message(TEXT_ERROR, "LL(1) violation - rule \'%s\'\n", temp->id);
text_printf(" productions %s ::= ", left->production->id);
rdp_print_sub_item(left->production, 1);
text_printf(".\n and %s ::= ", right->production->id);
rdp_print_sub_item(right->production, 1);
text_printf(".\n share these start tokens: ");
set_print_set(& work, rdp_token_string, 78);
text_printf("\n");
left->production->ll1_violation = 1;
right->production->ll1_violation = 1;
bad = 1;
}
right = right->next;
}
left = left->next;
}
}
temp =(rdp_data *) symbol_next_symbol_in_scope(temp);
}
return bad;
}
void hist_print(histogram_node *this_hist)
{
histogram_node *this_histogram_node = this_hist;
while (this_histogram_node->next != NULL)
{
if (this_histogram_node->value != 0)
text_printf("%lu: %lu\n", this_histogram_node->bucket, this_histogram_node->value);
this_histogram_node = this_histogram_node->next;
}
}
int scan_test_set(const char * production, set_ * valid, set_ * stop)
{
if (!set_includes_element(valid, SCAN_CAST->token))
{
if (stop != NULL)
{
if (production != NULL)
text_message(TEXT_ERROR_ECHO, "In rule \'%s\', scanned ", production);
else
text_message(TEXT_ERROR_ECHO, "Scanned ");
set_print_element(SCAN_CAST->token, scan_token_names);
text_printf(" whilst expecting %s", set_cardinality(valid)== 1 ? "": "one of ");
set_print_set(valid, scan_token_names, 60);
text_printf("\n");
scan_skip(stop);
}
return 0;
}
else
return 1;
}
int scan_test(const char * production, const int valid, set_ * stop)
{
if (valid != SCAN_CAST->token)
{
if (stop != NULL)
{
if (production != NULL)
text_message(TEXT_ERROR_ECHO, "In rule \'%s\', scanned ", production);
else
text_message(TEXT_ERROR_ECHO, "Scanned ");
set_print_element(SCAN_CAST->token, scan_token_names);
text_printf(" whilst expecting ");
set_print_element(valid, scan_token_names);
text_printf("\n");
scan_skip(stop);
}
return 0;
}
else
return 1;
}
void hist_report_differences(histogram_node *hista, histogram_node *histb)
{
bool same = true;
bool go_on = true;
while (go_on)
{
while(hista->bucket < histb->bucket)
{
text_printf("\n%lu:\t%lu\t-\n", hista->bucket, hista->value);
same = false;
hista = hista->next;
}
while(histb->bucket < hista->bucket)
{
text_printf("\n%lu:\t-\t%lu\n", histb->bucket, histb->value);
same = false;
histb = histb->next;
}
if (histb->bucket == hista->bucket)
if (hista->bucket != ULONG_MAX)
{
if (hista->value != histb->value)
{
text_printf("\n%lu:\t%lu\t%lu\n", hista->bucket, hista->value, histb->value);
same = false;
}
hista = hista->next;
histb = histb->next;
}
else
go_on = false;
}
text_printf("%s\n", same ? "--- same ----" :
"- different -");
}
void pretty_close(char * sourcefilename, char * outputfilename)
{
unsigned long useful_lexeme_count = lexeme_count - comment_count - eoln_count;
char * backup_filename = text_force_filetype(sourcefilename, "bak");
fclose(outputfile);
remove(backup_filename);
if (rename(sourcefilename, backup_filename)!= 0)
text_message(TEXT_FATAL, "unable to rename \'%s\' to \'%s\'\n", sourcefilename, backup_filename);
if (rename(outputfilename, sourcefilename)!= 0)
text_message(TEXT_FATAL, "unable to rename \'%s\' to \'%s\'\n", outputfilename, sourcefilename);
text_printf("%s,%lu,%lu,%.2lf\n", sourcefilename,
last_line,
useful_lexeme_count,
(double) useful_lexeme_count /(double) last_line);
}
int main(int argc, char *argv[])
{
clock_t rdp_finish_time, rdp_start_time = clock();
int
rdp_symbol_statistics = 0, /* show symbol_ table statistics flag */
rdp_line_echo_all = 0, /* make a listing on all passes flag */
rdp_filter = 0, /* filter flag */
rdp_line_echo = 0, /* make listing flag */
rdp_lexicalise = 0; /* print lexicalised output flag */
unsigned long rdp_textsize = 35000l; /* size of scanner text array */
unsigned long rdp_tabwidth = 8l; /* tab expansion width */
char* rdp_vcg_filename = NULL; /* filename for -V option */
rdp_tree_node_data* rdp_tree = (rdp_tree_node_data*) graph_insert_graph("RDP derivation tree"); /* hook for derivation tree */
rdp_tree_node_data* rdp_tree_root;
arg_message("Minitree compiler V1.50 (c) Adrian Johnstone 1997\n" RDP_STAMP "\n\n""Usage: minitree [options] source[.m]");
arg_message("");
arg_boolean('f', "Filter mode (read from stdin and write to stdout)", &rdp_filter);
arg_boolean('l', "Make a listing", &rdp_line_echo);
arg_boolean('L', "Print lexicalised source file", &rdp_lexicalise);
arg_string ('o', "Write output to filename", &rdp_outputfilename);
arg_boolean('s', "Echo each scanner symbol as it is read", &rdp_symbol_echo);
arg_boolean('S', "Print summary symbol table statistics", &rdp_symbol_statistics);
arg_numeric('t', "Tab expansion width (default 8)", &rdp_tabwidth);
arg_numeric('T', "Text buffer size in bytes for scanner (default 20000)", &rdp_textsize);
arg_boolean('v', "Set verbose mode", &rdp_verbose);
arg_string ('V', "Write derivation tree to filename in VCG format", &rdp_vcg_filename);
rdp_sourcefilenames = arg_process(argc, argv);
/* Fix up filetypes */
for (rdp_sourcefilenumber = 0; rdp_sourcefilenames[rdp_sourcefilenumber] != NULL; rdp_sourcefilenumber++)
rdp_sourcefilenames[rdp_sourcefilenumber] = text_default_filetype(rdp_sourcefilenames[rdp_sourcefilenumber], "m");
if (rdp_filter)
{
rdp_sourcefilenames[0] = "-";
rdp_outputfilename = "-";
rdp_sourcefilenames[1] = NULL; /* make sure no further filenames are taken from the array */
}
if ((rdp_sourcefilename = rdp_sourcefilenames[0]) == NULL)
arg_help("no source files specified");
if (rdp_sourcefilenames[1] != NULL)
text_message(TEXT_FATAL, "multiple source files not allowed\n");
text_init(rdp_textsize, 25, 100, (int) rdp_tabwidth);
scan_init(0, 0, 0, rdp_symbol_echo, rdp_tokens);
if (rdp_lexicalise)
scan_lexicalise();
mini = symbol_new_table("mini", 101, 31, symbol_compare_string, symbol_hash_string, symbol_print_string);
rdp_set_initialise();
rdp_load_keywords();
if (rdp_verbose)
text_printf("\nMinitree compiler V1.50 (c) Adrian Johnstone 1997\n" RDP_STAMP "\n\n");
for (rdp_pass = 1; rdp_pass <= RDP_PASSES; rdp_pass++)
{
rdp_tree_update = rdp_pass == RDP_PASSES;
text_echo(rdp_line_echo_all || (rdp_line_echo && rdp_pass == RDP_PASSES));
for (rdp_sourcefilenumber = 0; (rdp_sourcefilename = rdp_sourcefilenames[rdp_sourcefilenumber]) != NULL; rdp_sourcefilenumber++)
{
if (text_open(rdp_sourcefilename) == NULL)
arg_help("unable to open source file");
text_get_char();
scan_();
program(rdp_tree_root = rdp_add_node("program", rdp_tree)); /* call parser at top level */
if (text_total_errors() != 0)
text_message(TEXT_FATAL, "error%s detected in source file ''\n", text_total_errors() == 1 ? "" : "s", rdp_sourcefilename); /* crash quietly */
graph_epsilon_prune_rdp_tree(rdp_tree_root, sizeof(rdp_tree_edge_data));
}
}
rdp_sourcefilename = rdp_sourcefilenames[0]; /* Reset filename to first file in the list */
graph_set_root(rdp_tree, rdp_tree_root);
if (rdp_vcg_filename != NULL)
{
FILE *rdp_vcg_file;
if (*rdp_vcg_filename == '\0') /* No filename supplied */
rdp_vcg_filename = "rdparser";
rdp_vcg_file = fopen((rdp_vcg_filename = text_default_filetype(rdp_vcg_filename, "vcg")), "w");
if (rdp_vcg_file == NULL)
text_message(TEXT_FATAL, "unable to open VCG file '%s' for write\n", rdp_vcg_filename);
if (rdp_verbose)
text_message(TEXT_INFO, "Dumping derivation tree to VCG file '%s'\n", rdp_vcg_filename);
text_redirect(rdp_vcg_file);
graph_vcg(rdp_tree, NULL, scan_vcg_print_node, scan_vcg_print_edge);
text_redirect(stdout);
fclose(rdp_vcg_file);
}
code_generate(rdp_sourcefilename, rdp_outputfilename, rdp_tree);
if (rdp_symbol_statistics)
{
symbol_print_all_table_statistics(11);
symbol_print_all_table();
}
text_print_total_errors();
if (rdp_verbose)
{
rdp_finish_time = clock();
text_message(TEXT_INFO, "%.3f CPU seconds used\n", ((double) (rdp_finish_time-rdp_start_time)) / CLOCKS_PER_SEC);
}
return rdp_error_return;
}
int main(int argc, char *argv[])
{
clock_t rdp_finish_time, rdp_start_time = clock();
int
rdp_symbol_statistics = 0, /* show symbol_ table statistics flag */
rdp_line_echo_all = 0, /* make a listing on all passes flag */
rdp_filter = 0, /* filter flag */
rdp_line_echo = 0, /* make listing flag */
rdp_lexicalise = 0; /* print lexicalised output flag */
unsigned long rdp_textsize = 35000l; /* size of scanner text array */
unsigned long rdp_tabwidth = 8l; /* tab expansion width */
char* rdp_vcg_filename = NULL; /* filename for -V option */
arg_message("rdparser\n" RDP_STAMP "\n\n""Usage: regex [options] source");
arg_message("");
arg_boolean('f', "Filter mode (read from stdin and write to stdout)", &rdp_filter);
arg_boolean('l', "Make a listing", &rdp_line_echo);
arg_boolean('L', "Print lexicalised source file", &rdp_lexicalise);
arg_string ('o', "Write output to filename", &rdp_outputfilename);
arg_boolean('s', "Echo each scanner symbol as it is read", &rdp_symbol_echo);
arg_boolean('S', "Print summary symbol table statistics", &rdp_symbol_statistics);
arg_numeric('t', "Tab expansion width (default 8)", &rdp_tabwidth);
arg_numeric('T', "Text buffer size in bytes for scanner (default 20000)", &rdp_textsize);
arg_boolean('v', "Set verbose mode", &rdp_verbose);
arg_string ('V', "Write derivation tree to filename in VCG format", &rdp_vcg_filename);
rdp_sourcefilenames = arg_process(argc, argv);
/* Fix up filetypes */
for (rdp_sourcefilenumber = 0; rdp_sourcefilenames[rdp_sourcefilenumber] != NULL; rdp_sourcefilenumber++)
rdp_sourcefilenames[rdp_sourcefilenumber] = text_default_filetype(rdp_sourcefilenames[rdp_sourcefilenumber], "");
if (rdp_filter)
{
rdp_sourcefilenames[0] = "-";
rdp_outputfilename = "-";
rdp_sourcefilenames[1] = NULL; /* make sure no further filenames are taken from the array */
}
if ((rdp_sourcefilename = rdp_sourcefilenames[0]) == NULL)
arg_help("no source files specified");
if (rdp_sourcefilenames[1] != NULL)
text_message(TEXT_FATAL, "multiple source files not allowed\n");
text_init(rdp_textsize, 25, 100, (int) rdp_tabwidth);
scan_init(0, 0, 0, rdp_symbol_echo, rdp_tokens);
if (rdp_lexicalise)
scan_lexicalise();
rdp_set_initialise();
rdp_load_keywords();
if (rdp_verbose)
text_printf("\nrdparser\n" RDP_STAMP "\n\n");
for (rdp_pass = 1; rdp_pass <= RDP_PASSES; rdp_pass++)
{
text_echo(rdp_line_echo_all || (rdp_line_echo && rdp_pass == RDP_PASSES));
for (rdp_sourcefilenumber = 0; (rdp_sourcefilename = rdp_sourcefilenames[rdp_sourcefilenumber]) != NULL; rdp_sourcefilenumber++)
{
if (text_open(rdp_sourcefilename) == NULL)
arg_help("unable to open source file");
text_get_char();
scan_();
reg(); /* call parser at top level */
if (text_total_errors() != 0)
text_message(TEXT_FATAL, "error%s detected in source file ''\n", text_total_errors() == 1 ? "" : "s", rdp_sourcefilename); /* crash quietly */
}
}
rdp_sourcefilename = rdp_sourcefilenames[0]; /* Reset filename to first file in the list */
if (rdp_symbol_statistics)
{
symbol_print_all_table_statistics(11);
symbol_print_all_table();
}
text_print_total_errors();
if (rdp_verbose)
{
rdp_finish_time = clock();
text_message(TEXT_INFO, "%.3f CPU seconds used\n", ((double) (rdp_finish_time-rdp_start_time)) / CLOCKS_PER_SEC);
}
return rdp_error_return;
}
void scan_vcg_print_node(const void * node)
{
text_printf("label:\"");
#if defined(show_vcg_atom_numbers)
text_printf("%lu: ", graph_atom_number(node));
#endif
switch (((scan_data *) node)->token)
{
case SCAN_P_IGNORE:
case SCAN_P_ID: text_print_C_string(((scan_data *) node)->id); break;
case SCAN_P_INTEGER:
text_printf("INTEGER: %lu", ((scan_data *) node)->data.i); break;
case SCAN_P_REAL: text_printf("REAL: %lf", ((scan_data *) node)->data.r); break;
case SCAN_P_STRING:
case SCAN_P_STRING_ESC:
case SCAN_P_COMMENT:
case SCAN_P_COMMENT_VISIBLE:
case SCAN_P_COMMENT_NEST:
case SCAN_P_COMMENT_NEST_VISIBLE:
case SCAN_P_COMMENT_LINE:
case SCAN_P_COMMENT_LINE_VISIBLE: text_printf("\\\""); text_print_C_string(((scan_data *) node)->id); text_printf("\\\""); break;
case SCAN_P_EOF: text_printf("EOF"); break;
case SCAN_P_EOLN: text_printf("EOLN"); break;
default: text_printf("\'");
text_print_C_string(((scan_data *) node)->id);
text_printf("\'");
break;
}
text_printf("\" horizontal_order:%lu", graph_atom_number(node));
if (((scan_data *) node)->token == SCAN_P_IGNORE) /* make nonterminals round */
text_printf(" shape:ellipse ");
}