You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
561 lines
12 KiB
561 lines
12 KiB
/* Copyright (C) Boris Nikolaus, Germany, 1996-1997. All rights reserved. */
|
|
|
|
#include <stdio.h>
|
|
#include <ctype.h>
|
|
#include <stdlib.h>
|
|
#include <stdarg.h>
|
|
#include <string.h>
|
|
#include "defs.h"
|
|
#include "parser.h"
|
|
|
|
#define IS_IDENT(c) (isalpha(c) || c == '_' || c == '\'' || c == '\"')
|
|
|
|
#ifdef YYDEBUG
|
|
extern int yydebug;
|
|
#endif
|
|
|
|
enum {
|
|
DeclSection,
|
|
Union,
|
|
Union1,
|
|
State,
|
|
State1,
|
|
CCode,
|
|
RuleSection,
|
|
Arg,
|
|
Action,
|
|
CSection,
|
|
End
|
|
} state;
|
|
|
|
int bufferpos;
|
|
char *buffer = 0;
|
|
int buffersize = 0;
|
|
|
|
char *line = 0;
|
|
int linesize = 0;
|
|
int linelen = 0;
|
|
char *cptr;
|
|
int meteof = 0;
|
|
int lineno;
|
|
char filename[256];
|
|
FILE *fin;
|
|
|
|
void error(LLPOS *pos, char *fmt, ...);
|
|
|
|
void
|
|
addchar(char c)
|
|
{
|
|
if (bufferpos >= buffersize) {
|
|
if (!buffersize) {
|
|
buffersize = 4096;
|
|
buffer = (char *)malloc(buffersize);
|
|
} else {
|
|
buffersize <<= 1;
|
|
buffer = (char *)realloc(buffer, buffersize);
|
|
}
|
|
if (!buffer)
|
|
error(NULL, "Out of memory");
|
|
}
|
|
buffer[bufferpos++] = c;
|
|
}
|
|
|
|
void
|
|
memmove_fwd(char *dst, char *src, int len)
|
|
{
|
|
while (len--)
|
|
*dst++ = *src++;
|
|
}
|
|
|
|
int
|
|
read_line()
|
|
{
|
|
int nread;
|
|
int left;
|
|
|
|
if (!linesize) {
|
|
linesize = 4096;
|
|
cptr = line = (char *)malloc(linesize);
|
|
linelen = 0;
|
|
}
|
|
if (!linelen && meteof)
|
|
return 0;
|
|
if (linelen && (cptr <= line || cptr > line + linelen || cptr[-1] != '\n'))
|
|
abort();
|
|
left = linelen - (int)(cptr - line);
|
|
if (left && memchr(cptr, '\n', left))
|
|
return 1;
|
|
linelen = left;
|
|
if (left)
|
|
memmove_fwd(line, cptr, left);
|
|
cptr = line;
|
|
while ((!linelen || !memchr(line, '\n', linelen)) && !meteof) {
|
|
if (linesize - linelen < 2048) {
|
|
linesize <<= 1;
|
|
line = (char *)realloc(line, linesize);
|
|
cptr = line;
|
|
}
|
|
nread = fread(line + linelen, 1, linesize - linelen, fin);
|
|
if (nread < linesize - linelen)
|
|
meteof = 1;
|
|
linelen += nread;
|
|
}
|
|
return linelen > 0;
|
|
}
|
|
|
|
int
|
|
get_line()
|
|
{
|
|
int depth = 0;
|
|
char *p;
|
|
|
|
for (;;) {
|
|
if (!read_line())
|
|
return 0;
|
|
if (!strncmp(cptr, "#line ", 6)) {
|
|
lineno = strtol(cptr + 6, &cptr, 10) - 1;
|
|
while (isspace(*cptr) && *cptr != '\n')
|
|
cptr++;
|
|
if (*cptr == '\"') {
|
|
cptr++;
|
|
for (p = filename; *cptr != '\n' && *cptr != '\"';) {
|
|
*p++ = *cptr++;
|
|
}
|
|
*p = 0;
|
|
}
|
|
cptr = strchr(cptr, '\n') + 1;
|
|
continue;
|
|
}
|
|
lineno++;
|
|
if (!strncmp(cptr, "%comment", 8)) {
|
|
cptr = strchr(cptr, '\n') + 1;
|
|
depth++;
|
|
continue;
|
|
}
|
|
if (!strncmp(cptr, "%endcomment", 11)) {
|
|
cptr = strchr(cptr, '\n') + 1;
|
|
depth--;
|
|
continue;
|
|
}
|
|
if (!depth)
|
|
return 1;
|
|
cptr = strchr(cptr, '\n') + 1;
|
|
}
|
|
}
|
|
|
|
int
|
|
read_char()
|
|
{
|
|
for (;;) {
|
|
if (!cptr) {
|
|
if (!get_line())
|
|
return EOF;
|
|
continue;
|
|
}
|
|
if (*cptr == '\n') {
|
|
cptr++;
|
|
if (!get_line())
|
|
return EOF;
|
|
continue;
|
|
}
|
|
if (isspace(*cptr)) {
|
|
cptr++;
|
|
continue;
|
|
}
|
|
if (*cptr == '/' && cptr[1] == '/') {
|
|
cptr = strchr(cptr, '\n') + 1;
|
|
continue;
|
|
}
|
|
if (*cptr == '/' && cptr[1] == '*') {
|
|
cptr += 2;
|
|
do {
|
|
while (*cptr != '*') {
|
|
if (*cptr++ == '\n') {
|
|
if (!get_line())
|
|
error(NULL, "Unexpected EOF");
|
|
}
|
|
}
|
|
} while (*++cptr != '/');
|
|
cptr++;
|
|
continue;
|
|
}
|
|
break;
|
|
}
|
|
return *cptr;
|
|
}
|
|
|
|
void
|
|
get_identifier(LLPOS *pos)
|
|
{
|
|
int quote;
|
|
|
|
bufferpos = 0;
|
|
addchar(*cptr);
|
|
if (*cptr == '\'' || *cptr == '\"') {
|
|
quote = *cptr++;
|
|
while (*cptr != quote) {
|
|
if (*cptr == '\n')
|
|
error(pos, "Missing closing quote");
|
|
addchar(*cptr);
|
|
if (*cptr++ == '\\')
|
|
addchar(*cptr++);
|
|
}
|
|
addchar(*cptr++);
|
|
} else {
|
|
cptr++;
|
|
while (isalnum(*cptr) || *cptr == '_')
|
|
addchar(*cptr++);
|
|
}
|
|
addchar(0);
|
|
}
|
|
|
|
int
|
|
get_token(LLSTYPE *lval, LLPOS *pos)
|
|
{
|
|
int c;
|
|
int depth;
|
|
|
|
c = read_char();
|
|
pos->line = lineno;
|
|
pos->column = cptr ? (int)(cptr - line) - 1 : 0;
|
|
pos->file = strdup(filename);
|
|
|
|
switch (state) {
|
|
case DeclSection:
|
|
if (!strncmp(cptr, "%%", 2)) {
|
|
state = RuleSection;
|
|
cptr += 2;
|
|
return PERCENT_PERCENT;
|
|
}
|
|
if (!strncmp(cptr, "%token", 6)) {
|
|
cptr += 6;
|
|
return PERCENT_TOKEN;
|
|
}
|
|
if (!strncmp(cptr, "%type", 5)) {
|
|
cptr += 5;
|
|
return PERCENT_TYPE;
|
|
}
|
|
if (!strncmp(cptr, "%external", 9)) {
|
|
cptr += 9;
|
|
return PERCENT_EXTERNAL;
|
|
}
|
|
if (!strncmp(cptr, "%start", 6)) {
|
|
cptr += 6;
|
|
return PERCENT_START;
|
|
}
|
|
if (!strncmp(cptr, "%union", 6)) {
|
|
state = Union;
|
|
cptr += 6;
|
|
return PERCENT_UNION;
|
|
}
|
|
if (!strncmp(cptr, "%state", 6)) {
|
|
state = State;
|
|
cptr += 6;
|
|
return PERCENT_STATE;
|
|
}
|
|
if (!strncmp(cptr, "%prefix", 7)) {
|
|
cptr += 7;
|
|
return PERCENT_PREFIX;
|
|
}
|
|
if (!strncmp(cptr, "%module", 7)) {
|
|
cptr += 7;
|
|
return PERCENT_MODULE;
|
|
}
|
|
if (!strncmp(cptr, "%{", 2)) { /*}*/
|
|
state = CCode;
|
|
cptr += 2;
|
|
return PERCENT_LBRACE;
|
|
}
|
|
if (!strncmp(cptr, "%}", 2)) { /*{*/
|
|
cptr += 2;
|
|
return PERCENT_RBRACE;
|
|
}
|
|
if (c == '<' || c == '>' || c == '=') {
|
|
return *cptr++;
|
|
}
|
|
if (IS_IDENT(c)) {
|
|
get_identifier(pos);
|
|
lval->_string = strdup(buffer);
|
|
return IDENTIFIER;
|
|
}
|
|
break;
|
|
case Union:
|
|
if (c == '{') { /*}*/
|
|
state = Union1;
|
|
cptr++;
|
|
return c;
|
|
}
|
|
break;
|
|
case Union1:
|
|
bufferpos = 0;
|
|
if (c == '}') { /*{*/
|
|
state = DeclSection;
|
|
cptr++;
|
|
return c;
|
|
}
|
|
if (c == ';') {
|
|
cptr++;
|
|
return c;
|
|
}
|
|
while (*cptr != ';') {
|
|
addchar(*cptr++);
|
|
if (*cptr == '\n') {
|
|
cptr++;
|
|
if (!get_line())
|
|
error(pos, "Unexpected EOF");
|
|
}
|
|
}
|
|
addchar(0);
|
|
lval->_string = strdup(buffer);
|
|
return TAGDEF;
|
|
case State:
|
|
if (c == '{') { /*}*/
|
|
state = State1;
|
|
cptr++;
|
|
return c;
|
|
}
|
|
break;
|
|
case State1:
|
|
bufferpos = 0;
|
|
if (c == '}') { /*{*/
|
|
state = DeclSection;
|
|
cptr++;
|
|
return c;
|
|
}
|
|
if (c == ';') {
|
|
cptr++;
|
|
return c;
|
|
}
|
|
while (*cptr != ';') {
|
|
addchar(*cptr++);
|
|
if (*cptr == '\n') {
|
|
cptr++;
|
|
if (!get_line())
|
|
error(pos, "Unexpected EOF");
|
|
}
|
|
}
|
|
addchar(0);
|
|
lval->_string = strdup(buffer);
|
|
return TAGDEF;
|
|
case CCode:
|
|
bufferpos = 0;
|
|
for (;;) {
|
|
if (*cptr == '%' && cptr[1] == '}') /*{*/
|
|
break;
|
|
addchar(*cptr);
|
|
if (*cptr++ == '\n') {
|
|
if (!get_line())
|
|
error(pos, "Unexpected EOF");
|
|
}
|
|
}
|
|
state = DeclSection;
|
|
addchar(0);
|
|
lval->_ccode = strdup(buffer);
|
|
return CCODE;
|
|
case RuleSection:
|
|
if (IS_IDENT(c)) {
|
|
get_identifier(pos);
|
|
lval->_string = strdup(buffer);
|
|
return IDENTIFIER;
|
|
}
|
|
if (c == '(') {
|
|
state = Arg;
|
|
cptr++;
|
|
return c;
|
|
}
|
|
if (c == ':' || c == ';' || c == '|' || c == '}' || c == '+' ||
|
|
c == '*' || c == '?' || c == '[' || c == ']') { /*{*/
|
|
cptr++;
|
|
return c;
|
|
}
|
|
if (c == '%' && cptr[1] == '%') {
|
|
state = CSection;
|
|
cptr += 2;
|
|
return PERCENT_PERCENT;
|
|
}
|
|
if (c == '{') {
|
|
state = Action;
|
|
cptr++;
|
|
return c;
|
|
}
|
|
break;
|
|
case Arg:
|
|
if (c == ')') {
|
|
state = RuleSection;
|
|
cptr++;
|
|
return c;
|
|
}
|
|
if (c == ',') {
|
|
cptr++;
|
|
return c;
|
|
}
|
|
bufferpos = 0;
|
|
depth = 0;
|
|
for (;;) {
|
|
switch (*cptr) {
|
|
case '(':
|
|
depth++;
|
|
addchar(*cptr++);
|
|
continue;
|
|
case ')':
|
|
if (depth > 0) {
|
|
depth--;
|
|
addchar(*cptr++);
|
|
continue;
|
|
}
|
|
break;
|
|
case ',':
|
|
if (depth > 0) {
|
|
addchar(*cptr++);
|
|
continue;
|
|
}
|
|
break;
|
|
case '\n':
|
|
error(pos, "Unterminated argument");
|
|
/*NOTREACHED*/
|
|
default:
|
|
addchar(*cptr++);
|
|
continue;
|
|
}
|
|
break;
|
|
}
|
|
addchar(0);
|
|
lval->_string = strdup(buffer);
|
|
return ARG;
|
|
case Action:
|
|
bufferpos = 0;
|
|
depth = 0;
|
|
for (;;) {
|
|
switch (*cptr) {
|
|
case '{': /*}*/
|
|
depth++;
|
|
addchar(*cptr++);
|
|
continue;
|
|
case '}': /*{*/
|
|
if (depth > 0) {
|
|
depth--;
|
|
addchar(*cptr++);
|
|
continue;
|
|
}
|
|
break;
|
|
case '\'':
|
|
case '\"':
|
|
c = *cptr++;
|
|
addchar((char)c);
|
|
while (*cptr != c) {
|
|
addchar(*cptr);
|
|
if (*cptr == '\n') {
|
|
cptr++;
|
|
get_line();
|
|
continue;
|
|
}
|
|
if (*cptr == '\\')
|
|
addchar(*++cptr);
|
|
cptr++;
|
|
}
|
|
addchar((char)c);
|
|
cptr++;
|
|
continue;
|
|
case '\n':
|
|
addchar(*cptr++);
|
|
if (!get_line())
|
|
error(pos, "Unexpected EOF");
|
|
continue;
|
|
default:
|
|
addchar(*cptr++);
|
|
continue;
|
|
}
|
|
break;
|
|
}
|
|
state = RuleSection;
|
|
addchar(0);
|
|
lval->_ccode = strdup(buffer);
|
|
return CCODE;
|
|
case CSection:
|
|
bufferpos = 0;
|
|
if (c != EOF) {
|
|
do {
|
|
do {
|
|
addchar(*cptr);
|
|
} while (*cptr++ != '\n');
|
|
} while (get_line());
|
|
}
|
|
state = End;
|
|
addchar(0);
|
|
lval->_ccode = strdup(buffer);
|
|
return CCODE;
|
|
case End:
|
|
return EOF;
|
|
}
|
|
error(pos, "Syntax error");
|
|
return EOF;
|
|
/*NOTREACHED*/
|
|
}
|
|
|
|
void
|
|
open_file(char *file)
|
|
{
|
|
strcpy(filename, file);
|
|
lineno = 0;
|
|
fin = fopen(filename, "r");
|
|
if (!fin) {
|
|
perror(file);
|
|
exit(1);
|
|
}
|
|
}
|
|
|
|
int
|
|
llgettoken(int *token, LLSTYPE *lval, LLPOS *pos)
|
|
{
|
|
*token = get_token(lval, pos);
|
|
return *token != EOF;
|
|
}
|
|
|
|
void
|
|
llprinttoken(LLTERM *token, char *identifier, FILE *f)
|
|
{
|
|
switch (token->token) {
|
|
case IDENTIFIER:
|
|
printf("scanner: delivering token IDENTIFIER(%s)\n", token->lval._string);
|
|
break;
|
|
case ARG:
|
|
printf("scanner: delivering token ARG(%s)\n", token->lval._string);
|
|
break;
|
|
case CCODE:
|
|
printf("scanner: delivering token CCODE(%s)\n", token->lval._ccode);
|
|
break;
|
|
case TAGDEF:
|
|
printf("scanner: delivering token TAGDEF(%s)\n", token->lval._string);
|
|
break;
|
|
case PERCENT_PERCENT:
|
|
printf("scanner: delivering token %%%%\n");
|
|
break;
|
|
case PERCENT_TOKEN:
|
|
printf("scanner: delivering token %%token\n");
|
|
break;
|
|
case PERCENT_TYPE:
|
|
printf("scanner: delivering token %%type\n");
|
|
break;
|
|
case PERCENT_UNION:
|
|
printf("scanner: delivering token %%union\n");
|
|
break;
|
|
case PERCENT_START:
|
|
printf("scanner: delivering token %%start\n");
|
|
break;
|
|
case PERCENT_PREFIX:
|
|
printf("scanner: delivering token %%prefix\n");
|
|
break;
|
|
case PERCENT_LBRACE:
|
|
printf("scanner: delivering token %%{\n");
|
|
break;
|
|
case PERCENT_RBRACE:
|
|
printf("scanner: delivering token %%}\n");
|
|
break;
|
|
case EOF:
|
|
printf("scanner: delivering token EOF\n");
|
|
break;
|
|
default:
|
|
printf("scanner: delivering token %c\n", token->token);
|
|
break;
|
|
}
|
|
}
|