lexeme *st_eoln_eof(lexer_info *linfo, buffer *buf)
{
lexeme *lex = NULL;
#ifdef LEXER_DEBUG
print_state("ST_EOLN_EOF", linfo->c);
#endif
switch (linfo->c) {
case '\n':
lex = make_lex(LEX_EOLINE);
break;
case EOF:
lex = make_lex(LEX_EOFILE);
break;
default:
fprintf(stderr, "Lexer: error in ST_EOLN_EOF;");
print_state("ST_EOLN_EOF", linfo->c);
exit(ES_LEXER_INCURABLE_ERROR);
}
deferred_get_char(linfo);
linfo->state = ST_START;
return lex;
}
lexeme *st_one_sym_lex(lexer_info *linfo, buffer *buf)
{
lexeme *lex = NULL;
#ifdef LEXER_DEBUG
print_state("ST_ONE_SYM_LEX", linfo->c);
#endif
switch (linfo->c) {
case '<':
lex = make_lex(LEX_INPUT);
break;
case ';':
lex = make_lex(LEX_SEMICOLON);
break;
case '(':
lex = make_lex(LEX_BRACKET_OPEN);
break;
case ')':
lex = make_lex(LEX_BRACKET_CLOSE);
break;
case '`':
lex = make_lex(LEX_REVERSE);
break;
default:
fprintf(stderr, "Lexer: error in ST_ONE_SYM_LEX;");
print_state("ST_ONE_SYM_LEX", linfo->c);
exit(ES_LEXER_INCURABLE_ERROR);
}
/* We don't need buffer */
deferred_get_char(linfo);
linfo->state = ST_START;
return lex;
}
void init_lexicons()
{
int i,j;
for(i=0;i<GMAXWORDS;i++)
fwords[i] = GUNKNOWN;
for(i=0;i<GMAXWORDS;i++)
for(j=0;j<GMAXNTS;j++)
tagdict[i][j] = 0;
make_lex(200003,&wordlex);
make_lex(1003,&nt_lex);
TRACEWORD = add_word("*TRACE*",&wordlex);
fwords[TRACEWORD] = 1;
}
lexeme *st_error(lexer_info *linfo, buffer *buf)
{
lexeme *lex = NULL;
lex = make_lex(LEX_ERROR);
print_state("ST_ERROR", linfo->c);
clear_buffer(buf);
linfo->get_next_char = 0;
linfo->state = ST_START;
/* TODO: read to '\n' or EOF (flush read buffer) */
return lex;
}
lexeme *st_one_two_sym_lex(lexer_info *linfo, buffer *buf)
{
lexeme *lex = NULL;
#ifdef LEXER_DEBUG
print_state("ST_ONE_TWO_SYM_LEX", linfo->c);
#endif
if (buf->count_sym == 0) {
add_to_buffer(buf, linfo->c);
deferred_get_char(linfo);
} else if (buf->count_sym == 1) {
/* TODO: make it more pretty */
char prev_c = get_last_from_buffer(buf);
clear_buffer(buf);
switch (prev_c) {
case '>':
lex = (prev_c == linfo->c) ?
make_lex(LEX_APPEND) :
make_lex(LEX_OUTPUT);
break;
case '|':
lex = (prev_c == linfo->c) ?
make_lex(LEX_OR) :
make_lex(LEX_PIPE);
break;
case '&':
lex = (prev_c == linfo->c) ?
make_lex(LEX_AND) :
make_lex(LEX_BACKGROUND);
break;
default:
fprintf(stderr, "Lexer: error (type 1) in ST_ONE_TWO_SYM_LEX;");
print_state("ST_ONE_TWO_SYM_LEX", linfo->c);
exit(ES_LEXER_INCURABLE_ERROR);
}
if (prev_c == linfo->c)
deferred_get_char(linfo);
linfo->state = ST_START;
return lex;
} else {
fprintf(stderr, "Lexer: error (type 2) in ST_ONE_TWO_SYM_LEX;");
print_state("ST_ONE_TWO_SYM_LEX", linfo->c);
exit(ES_LEXER_INCURABLE_ERROR);
}
return lex;
}
lexeme *st_word(lexer_info *linfo, buffer *buf)
{
lexeme *lex = NULL;
#ifdef LEXER_DEBUG
print_state("ST_WORD", linfo->c);
#endif
switch (linfo->c) {
case EOF:
case '\n':
case ' ':
case '\t':
case '<':
case ';':
case '(':
case ')':
case '>':
case '|':
case '&':
case '`':
linfo->state = ST_START;
lex = make_lex(LEX_WORD);
lex->str = convert_to_string(buf, 1);
return lex;
case '\\':
deferred_get_char(linfo);
linfo->state = ST_BACKSLASH;
break;
case '\"':
deferred_get_char(linfo);
linfo->state = ST_IN_QUOTES;
break;
default:
add_to_buffer(buf, linfo->c);
deferred_get_char(linfo);
}
return lex;
}