sl2p 프로그램 소스

복잡프로그래밍2010. 12. 16. 18:21

sl2p 프로그램 소스

이 프로그램 만들때 pglex랑

초기버젼 sl2p 를 사용하여 만들었다.

그냥 생각나는 대로 마구 짜서 효율성이고 뭐고 없다.;;

직접 타이핑 한 부분은 fprintf 남발된 부분 제외하면 100라인 정도 밖에 안될 것 같다.

/*
이 프로그램은 sl2p 프로그램입니다.

: 
grammar 파일을 읽어들여 context free 
구조가 맞는지 확인하고
특히 strong LL(1) grammar 로 생각하여
그에 맞게 순환 프로시져를 작성해 줍니다.

작성자: 성대아싸
*/

// !!usr!! 
#define SCANFUNC pg_lex()
// !!usr!!

//terminal set
typedef enum{
	NON=1,
	PRO,
	END,
	UNI,
	TER
}symbol;
//

// global
symbol lookahead;

//rules
void gram();
void body(int);
void sub();
void body_(int);
//

//default define
#ifndef __STDIO_H__
	#include <stdio.h>
#endif
#ifndef __STDLIB_H__
	#include <stdlib.h>
#endif
#ifndef __STRING_H__
	#include <string.h>
#endif
//

// default setting
#define BUFFER_SIZE 512
#define TOKEN_SIZE 512
#define PG_EXIT_SCAN 0
//
// pg_ variables //
char pg_buffer[BUFFER_SIZE];
char* pg_head=0;
char* pg_prev;
char pg_token[TOKEN_SIZE];
char* pg_save;
FILE* pg_input=0;
// pg_ variables //
//

//generated by regex 0: ;
int R0_match(){
R0_STATE0: 
	switch(*pg_head){
		case 0: goto R0_EXIT0;
		case 59: pg_head++; goto R0_STATE1;
		default:
	R0_EXIT0:
	return 0;
	}
R0_STATE1: //final
	switch(*pg_head){
		case 0: goto R0_EXIT1;
		default:
	R0_EXIT1:
	return 1;
	}
}

//generated by regex 1: ->
int R1_match(){
R1_STATE0: 
	switch(*pg_head){
		case 0: goto R1_EXIT0;
		case 45: pg_head++; goto R1_STATE1;
		default:
	R1_EXIT0:
	return 0;
	}
R1_STATE1: 
	switch(*pg_head){
		case 0: goto R1_EXIT1;
		case 62: pg_head++; goto R1_STATE2;
		default:
	R1_EXIT1:
	return 0;
	}
R1_STATE2: //final
	switch(*pg_head){
		case 0: goto R1_EXIT2;
		default:
	R1_EXIT2:
	return 1;
	}
}

//generated by regex 2: \|
int R2_match(){
R2_STATE0: 
	switch(*pg_head){
		case 0: goto R2_EXIT0;
		case 124: pg_head++; goto R2_STATE1;
		default:
	R2_EXIT0:
	return 0;
	}
R2_STATE1: //final
	switch(*pg_head){
		case 0: goto R2_EXIT1;
		default:
	R2_EXIT1:
	return 1;
	}
}

//generated by regex 3: //.*\n
int R3_match(){
R3_STATE0: 
	switch(*pg_head){
		case 0: goto R3_EXIT0;
		case 47: pg_head++; goto R3_STATE1;
		default:
	R3_EXIT0:
	return 0;
	}
R3_STATE1: 
	switch(*pg_head){
		case 0: goto R3_EXIT1;
		case 47: pg_head++; goto R3_STATE2;
		default:
	R3_EXIT1:
	return 0;
	}
R3_STATE2: 
	switch(*pg_head){
		case 0: goto R3_EXIT2;
		case 10: pg_head++; goto R3_STATE3;
		default:
		if(((-128<=*pg_head) && (*pg_head<=127))
		){
			pg_head++; goto R3_STATE4;
		}
	R3_EXIT2:
	return 0;
	}
R3_STATE3: //final
	switch(*pg_head){
		case 0: goto R3_EXIT3;
		default:
	R3_EXIT3:
	return 1;
	}
R3_STATE4: 
	switch(*pg_head){
		case 0: goto R3_EXIT4;
		case 10: pg_head++; goto R3_STATE3;
		default:
		if(((-128<=*pg_head) && (*pg_head<=127))
		){
			pg_head++; goto R3_STATE4;
		}
	R3_EXIT4:
	return 0;
	}
}

//generated by regex 4: '[a-zA-Z_]+'
int R4_match(){
R4_STATE0: 
	switch(*pg_head){
		case 0: goto R4_EXIT0;
		case 39: pg_head++; goto R4_STATE1;
		default:
	R4_EXIT0:
	return 0;
	}
R4_STATE1: 
	switch(*pg_head){
		case 0: goto R4_EXIT1;
		default:
		if(((97<=*pg_head) && (*pg_head<=122))
		||((65<=*pg_head) && (*pg_head<=90))
		||((95<=*pg_head) && (*pg_head<=95))
		){
			pg_head++; goto R4_STATE2;
		}
	R4_EXIT1:
	return 0;
	}
R4_STATE2: 
	switch(*pg_head){
		case 0: goto R4_EXIT2;
		case 39: pg_head++; goto R4_STATE3;
		default:
		if(((97<=*pg_head) && (*pg_head<=122))
		||((65<=*pg_head) && (*pg_head<=90))
		||((95<=*pg_head) && (*pg_head<=95))
		){
			pg_head++; goto R4_STATE2;
		}
	R4_EXIT2:
	return 0;
	}
R4_STATE3: //final
	switch(*pg_head){
		case 0: goto R4_EXIT3;
		default:
	R4_EXIT3:
	return 1;
	}
}

//generated by regex 5: [a-zA-Z_]+
int R5_match(){
R5_STATE0: 
	switch(*pg_head){
		case 0: goto R5_EXIT0;
		default:
		if(((97<=*pg_head) && (*pg_head<=122))
		||((65<=*pg_head) && (*pg_head<=90))
		||((95<=*pg_head) && (*pg_head<=95))
		){
			pg_head++; goto R5_STATE1;
		}
	R5_EXIT0:
	return 0;
	}
R5_STATE1: //final
	switch(*pg_head){
		case 0: goto R5_EXIT1;
		default:
		if(((97<=*pg_head) && (*pg_head<=122))
		||((65<=*pg_head) && (*pg_head<=90))
		||((95<=*pg_head) && (*pg_head<=95))
		){
			pg_head++; goto R5_STATE1;
		}
	R5_EXIT1:
	return 1;
	}
}

//generated by regex 6: (\s|\t|\n)+
int R6_match(){
R6_STATE0: 
	switch(*pg_head){
		case 0: goto R6_EXIT0;
		case 32: pg_head++; goto R6_STATE1;
		case 9: pg_head++; goto R6_STATE2;
		case 10: pg_head++; goto R6_STATE3;
		default:
	R6_EXIT0:
	return 0;
	}
R6_STATE1: //final
	switch(*pg_head){
		case 0: goto R6_EXIT1;
		case 32: pg_head++; goto R6_STATE1;
		case 9: pg_head++; goto R6_STATE2;
		case 10: pg_head++; goto R6_STATE3;
		default:
	R6_EXIT1:
	return 1;
	}
R6_STATE2: //final
	switch(*pg_head){
		case 0: goto R6_EXIT2;
		case 32: pg_head++; goto R6_STATE1;
		case 9: pg_head++; goto R6_STATE2;
		case 10: pg_head++; goto R6_STATE3;
		default:
	R6_EXIT2:
	return 1;
	}
R6_STATE3: //final
	switch(*pg_head){
		case 0: goto R6_EXIT3;
		case 32: pg_head++; goto R6_STATE1;
		case 9: pg_head++; goto R6_STATE2;
		case 10: pg_head++; goto R6_STATE3;
		default:
	R6_EXIT3:
	return 1;
	}
}

int pg_lex(){
	int pg_sel,pg_rec,pg_i;
	int (*pg_rex_funcs[7])()={
		R0_match,
		R1_match,
		R2_match,
		R3_match,
		R4_match,
		R5_match,
		R6_match
	};
	if(!pg_input)pg_input = stdin;
	if(!pg_head)pg_head = pg_buffer;
	while(1){pg_sel=-1,pg_rec=-1;
		if(!*pg_head){
			if(fgets(pg_buffer,BUFFER_SIZE,pg_input)!=NULL)
				pg_head = pg_buffer;
			else
				return PG_EXIT_SCAN;
		}
		//matching
		for(pg_i=0;pg_i<7;pg_i++){
			pg_prev=pg_head;//save
			if(pg_rex_funcs[pg_i]()){
				if(pg_rec<pg_head-pg_prev){
					pg_sel=pg_i;
					pg_rec=pg_head-pg_prev;
				}
			}
			pg_head=pg_prev;//repair
		}
		//no match
		if(pg_sel==-1){
			puts("error");exit(1);
		}
		//token update
		for(pg_i=0;pg_i<pg_rec; pg_i++){
			pg_token[pg_i]=*pg_head++;
		}
		pg_token[pg_i]=0;
		//run usr code
		switch(pg_sel){
			case 0:
			// !! usr code !!//
						return END;
			// !!==========!!//
			break;
			case 1:
			// !! usr code !!//
						return PRO;
			// !!==========!!//
			break;
			case 2:
			// !! usr code !!//
						return UNI;
			// !!==========!!//
			break;
			case 3:
			// !! usr code !!//
						//skip
			// !!==========!!//
			break;
			case 4:
			// !! usr code !!//
					return TER;
			// !!==========!!//
			break;
			case 5:
			// !! usr code !!//
					return NON;
			// !!==========!!//
			break;
			case 6:
			// !! usr code !!//
					//skip
			// !!==========!!//
			break;
		}
		pg_head=pg_prev+pg_rec;
	}
}

// using funcs
void match(symbol target){
	if(target!=lookahead){
		printf("syntax error!\n");exit(1);	}
	lookahead = SCANFUNC;
}

void parse(){
	lookahead = SCANFUNC;
	gram();
}
//

// sym tab
char symtab[1024][128];
int st_num;
// rules
int rules[1024][16];
int rule_row;
int rule_col;
int max_col;
// info
int cnt_non;
int cnt_ter;
int info_chk[1024][3];
//

void r_input_table(int in){
	rules[rule_row][rule_col] = in;
	rule_col++;
	if(max_col < rule_col)
		max_col = rule_col;
}
int st_get_index(){
	int i;
	for(i=1; i<=st_num; i++){
		if(strcmp(symtab[i],pg_token)==0)
			return i;
	}
	st_num = i;
	strcpy(symtab[i],pg_token);
	return i;
}
//-----------------------

void gram(){
	int head;//
	while(1){
		if(lookahead==NON){
			head = st_get_index();//
			cnt_non++;
			match(NON);
			match(PRO);
			body(head);//
			match(END);
			continue;
		}
		break;
	}
}
void body(int head){//
	rule_row++;//
			rule_col=0;
			r_input_table(head);//
			info_chk[head][0]++;
			sub();
			body_(head);//
}
void body_(int head){//
	while(1){
		if(lookahead==UNI){
			rule_row++;//
			rule_col=0;
			r_input_table(head);//
			info_chk[head][0]++;
			match(UNI);
			sub();
			continue;
		}
		break;
	}
}
void sub(){
	while(1){
		if(lookahead==NON){
			r_input_table(st_get_index());
			match(NON);
			continue;
		}
		else if(lookahead==TER){
			r_input_table(-1*st_get_index());
			cnt_ter++;
			match(TER);
			continue;
		}
		break;
	}
}

//
void st_print(){
	int i=1;
	puts("===== symbol table =====");
	for(; i<=st_num; i++){
		printf("%3d: %s\n", i, symtab[i]);
	}
	puts("========================");
	puts("");
}
void r_print(){
	int i,j;
	puts("===== rule table =====");
	for(i=0; i<=rule_row; i++){
		for(j=0; j<max_col; j++){
			printf("%5d,",rules[i][j]);
		}
		printf("\n");
	}
	puts("=======================");
	puts("");
}
void info_print(){
	puts("===== info =====");
	printf("NON: %d, TER: %d, \n", cnt_non, cnt_ter);
	puts("================");
}

int chk_left_recur(){
	int i;
	for(i=1; i<=rule_row; i++){
		if(rules[i][0] == rules[i][1]){
			printf("!! left recur %d !!\n", i);
			return 1;
		}
	}
	return 0;
}
void set_tail(){
	int i,j,tail;
	for(i=1; i<=rule_row; i++){
		tail =0;
		for(j=1; j<16; j++){
			if(j>1&&rules[i][j]==0){
				tail = rules[i][j-1];
				break;
			}
		}
		if(rules[i][0] == tail){
			info_chk[rules[i][0]][1] = 1;
		}
	}
}
int chk_uni(int i){
	if(info_chk[i][0]>1){
		return 1;
	}else{
		return 0;
	}
}
void print_tail(){
	int i;
	printf("tail recur: ");
	for(i=0; i<1024; i++){
		if(info_chk[i][1]==1){
			printf("%d, ", i);
		}
	}
	printf("\n");
}
void set_epsilon(){
	int i;
	for(i=1; i<= rule_row; i++){
		if(rules[i][1] == 0){
			info_chk[rules[i][0]][2] = 1;
		}
	}
}
void print_epsilon(){
	int i;
	printf("has epsilon: ");
	for(i=0; i<1024; i++){
		if(info_chk[i][2]==1){
			printf("%d, ", i);
		}
	}
	printf("\n");
}

char o_type[128] = "void";
FILE* output;

void print_body(int i, int level){
	int j, k;
	for(j=1; rules[i][j]!=0; j++){
		for(k=0; k<level; k++)
			fprintf(output,"\t");
		if(rules[i][j]>0){
			if(rules[i][j+1]==0 && rules[i][0] == rules[i][j])
				fprintf(output,"continue;\n");
			else
				fprintf(output,"%s();\n", symtab[rules[i][j]]);
		}else
		if(rules[i][j]<0){
			fprintf(output,"match(%s);\n", symtab[-1*rules[i][j]]+1);
		}else{
			//fprintf(output,"break;\n");
		}
	}
}

void print_uni(int non, int level){
	int i,j,k,chk_else=0;
	if(info_chk[non][0]>1){
		for(i=1; i<= rule_row; i++){
			if(rules[i][0] == non){
				if(rules[i][1]<0){
					if(chk_else){
						for(k=0; k<level; k++)
							fprintf(output,"\t");
						fprintf(output,"else\n");
					}else
						chk_else=1;
					for(k=0; k<level; k++)
						fprintf(output,"\t");
					fprintf(output,"if(lookahead==%s){\n",symtab[-1*rules[i][1]]+1);
					
					print_body(i,level+1);

for(k=0; k<level; k++)
						fprintf(output,"\t");
					fprintf(output,"}\n");
				}
			}
		}
		if(info_chk[non][2]==0){
			for(k=0; k<level; k++)
				fprintf(output,"\t");
			fprintf(output,"else{\n");
			for(k=0; k<level; k++)
				fprintf(output,"\t");
			fprintf(output,"\tputs(\"parse error\");exit(1);\n");
			//fprintf(output,"\tparse_error();\n");
			for(k=0; k<level; k++)
				fprintf(output,"\t");
			fprintf(output,"}\n");
		}
		if(info_chk[non][2]==1 && info_chk[non][1]==1){
			for(k=0; k<level; k++)
				fprintf(output,"\t");
			fprintf(output,"break;\n");
		}
	}else{
		for(i=1; i<= rule_row; i++){
			if(rules[i][0] == non){
				print_body(i,level);				
			}
		}
	}
}

void code_tail(int non, int level){
	int i;
	if(info_chk[non][1] == 1){
		for(i=0; i<level; i++)
			fprintf(output,"\t");
		fprintf(output,"while(1){\n");
		
		print_uni(non,level+1);

for(i=0; i<level; i++)
			fprintf(output,"\t");
		fprintf(output,"}\n");
	}else{
		print_uni(non,level);
	}
}
void print_code(int non){
	int i,j, f_chk=0;
	if(info_chk[non][0]==0){
		return ;
	}
	fprintf(output,"/* ");
	fprintf(output,"rule: <%s>\n",symtab[non]);
	for(i=1; i<=rule_row; i++){
		if(rules[i][0]==non){
			if(f_chk==0){
				fprintf(output,"\t->\t");f_chk=1;
			}else{
				fprintf(output,"\t|\t");
			}
			for(j=1; rules[i][j]!=0; j++){
				if(rules[i][j]>0){
					fprintf(output,"<%s>\t", symtab[rules[i][j]]);
				}else if(rules[i][j] < 0){
					fprintf(output,"%s'\t", symtab[-1*rules[i][j]]);
				}
			}
			fprintf(output,"\n");
		}
	}
	fprintf(output,";\n");
	fprintf(output,"*/\n");
	fprintf(output,"%s %s(){\n",o_type, symtab[non]);

code_tail(non,1);

fprintf(output,"}\n//\n\n");
}

void mod_ter_sym(){
	int i;
	for(i=1; i<=st_num; i++){
		if(symtab[i][0]=='\''){
			symtab[i][strlen(symtab[i])-1] = 0;
		}
	}
}

int main(void){
	int i;
	int enum_first=0;
	pg_input = fopen("gramfile.txt", "rt");
	if(pg_input == NULL){
		puts("file open error"); exit(1);
	}
	//=====================

parse(); // parsing grammar

//st_print();
	//r_print();
	//info_print();
	if(chk_left_recur()){
		printf("left recur!!\n");
		exit(1);
	}
	set_tail();
	//print_tail();
	set_epsilon();
	//print_epsilon();
	mod_ter_sym();
	//===================

output = fopen("sl2pout.c","wt");

fprintf(output,"// !!usr!! \n");
	fprintf(output,"#define SCANFUNC pg_lex()\n");
	fprintf(output,"// !!usr!! \n\n\n\n\n");

fprintf(output,"/**********************************/\n");
	fprintf(output,"/*   generated by sl2p, def part  */\n\n");

fprintf(output,"//default define\n");
fprintf(output,"#ifndef __STDIO_H__\n");
	fprintf(output,"\t#include <stdio.h>\n");
fprintf(output,"#endif\n");
fprintf(output,"#ifndef __STDLIB_H__\n");
	fprintf(output,"\t#include <stdlib.h>\n");
fprintf(output,"#endif\n");
fprintf(output,"#ifndef __STRING_H__\n");
	fprintf(output,"\t#include <string.h>\n");
fprintf(output,"#endif\n");
fprintf(output,"//\n\n");

// get enum item
	fprintf(output,"// terminal symbols\n");
	fprintf(output,"enum{");
	for(i=1; i<=st_num; i++){
		if(symtab[i][0] == '\''){
			if(enum_first==0){
				fprintf(output,"\n\t%s=1", symtab[i]+1);
				enum_first=1;
			}else{
				fprintf(output,",\n\t%s",symtab[i]+1);
			}
				
		}
	}
	fprintf(output,"\n}symbol;\n\n");

fprintf(output,"symbol lookahead;\n\n");
	
	// get rules list
	fprintf(output,"// rule func list\n");
	for(i=1; i<=st_num; i++){
		if(symtab[i][0] != '\''){
			fprintf(output,"%s %s();\n",o_type,symtab[i]);
		}
	}
	fprintf(output,"//\n");
	fprintf(output,"void match(symbol);\n\n\n");
	fprintf(output,"// !! using this function !!//\n");
	fprintf(output,"void parse();\n");
	fprintf(output,"//\n\n\n");

fprintf(output,"/*   generated by sl2p , def part  */\n");
	fprintf(output,"/**********************************/\n\n\n\n\n");

fprintf(output,"/**********************************/\n");
	fprintf(output,"/*   generated by sl2p, rules     */\n\n");

for(i=1; i<=rule_row; i++){
		print_code(i);
	}

// using funcs
	fprintf(output,"// matching func\n");
	fprintf(output,"void match(symbol target){\n");
	fprintf(output,"\tif(target!=lookahead){\n");
	fprintf(output,"\t\tprintf(\"syntax error!\");exit(1);\n");
	fprintf(output,"\t}\n");
	fprintf(output,"\tlookahead = SCANFUNC;\n");
	fprintf(output,"}\n");
	fprintf(output,"//\n\n");

fprintf(output,"// !! using this func !!\n");
	fprintf(output,"void parse(){\n");
	fprintf(output,"\tlookahead = SCANFUNC;\n");
	fprintf(output,"\t%s();\n",symtab[1]);
	fprintf(output,"}\n");
	fprintf(output,"// !!using this func!! \n");
	fprintf(output,"//\n\n");
	fprintf(output,"/*   generated by sl2p, rules     */\n");
	fprintf(output,"/**********************************/\n\n\n\n\n");
//

fclose(output);

//=====================
	fclose(pg_input);
	return 0;
}

저작자표시

'복잡프로그래밍' 카테고리의 다른 글

pglexer 사용법 (2)	2011.09.02
pglexer 코드 (0)	2011.09.02
sl2p 프로그램 소스 (0)	2010.12.16
sl2p 프로그램 (0)	2010.12.16
pg_lex 프로그램 (0)	2010.10.27
pglex 사용예제 (0)	2010.10.09

Posted by 멍충한아싸

Trackback 0 Comment 0

댓글을 달아 주세요

«이전 1 ··· 96 97 98 99 100 101 102 103 104 ··· 240 다음»

일	월	화	수	목	금	토
					1	2
3	4	5	6	7	8	9
10	11	12	13	14	15	16
17	18	19	20	21	22	23
24	25	26	27	28	29	30
31

성대아싸의 멍충한 프로그래밍