void read_changes()
{
int i;
scanf("%d\n", &nmergers);
for (i = 0; i < nmergers; i++) {
read_quoted_string((mergers[i][0]));
read_quoted_string((mergers[i][1]));
}
}
static char *
read_string (struct obstack *ob, FILE *infile, int star_if_braced)
{
char *stringbuf;
int saw_paren = 0;
int c;
c = read_skip_spaces (infile);
if (c == '(')
{
saw_paren = 1;
c = read_skip_spaces (infile);
}
if (c == '"')
stringbuf = read_quoted_string (ob, infile);
else if (c == '{')
{
if (star_if_braced)
obstack_1grow (ob, '*');
stringbuf = read_braced_string (ob, infile);
}
else
fatal_with_file_and_line (infile, "expected `\"' or `{', found `%c'", c);
if (saw_paren)
{
c = read_skip_spaces (infile);
if (c != ')')
fatal_expected_char (infile, ')', c);
}
return stringbuf;
}
static char *next_token (const char **buffer) /* {{{ */
{
const char *ptr = *buffer;
char *ret;
while (isspace ((int) (*ptr)))
ptr++;
if (ptr[0] == 0)
return (NULL);
else if (ptr[0] == '"')
{
ret = read_quoted_string (&ptr);
if (ret != NULL)
{
*buffer = ptr;
return (ret);
}
}
ret = read_unquoted_word (&ptr);
if (ret != NULL)
*buffer = ptr;
return (ret);
} /* }}} char *next_token */
static const char *read_token_or_quoted_string(const char *s, char **token)
{
while (is_space_char(*s))
s++;
if (*s == '"')
return read_quoted_string(s, token);
else
return read_token(s, token);
}
static struct scanner_token scan_next(struct scanner_input *I)
{
char c = 1;
int i, num_invalid_chars = 0;
struct scanner_token T;
T.line = I->line_number;
T.value = NULL;
while (c != 0)
{
c = next_char(I);
if (isalpha(c)) /* An identifier */
{
char name[512]; /* Arbitrary limit to name length */
name[0] = c;
i = 1;
c = next_char(I);
while (isalnum(c) || c == '_')
{
name[i++] = c;
c = next_char(I);
}
name[i] = 0;
put_back(I, c);
if ( strcmp(name, "class") == 0 ) { T.type = TOK_CLASS; }
else if ( strcmp(name, "end") == 0 ) { T.type = TOK_END; }
else if ( strcmp(name, "return") == 0 ) { T.type = TOK_RETURN; }
else
{
T.type = TOK_IDENTIFIER;
T.value = GC_strdup(name);
}
return T;
}
/* FIXME: Need to check that there is only 1 '.' and 'e' in a number.
* Also maybe make 'e' an actual operator instead because that would be cool */
if (isdigit(c))
{
char num[512];
int ndots = 0;
num[0] = c;
i = 1;
c = next_char(I);
while (isdigit(c) || c == '.')
{
num[i++] = c;
if (c == '.')
ndots++;
c = next_char(I);
}
num[i] = 0;
put_back(I, c);
if (ndots > 1)
die("Invalid number: Too many decimal points", 1);
T.type = (ndots == 0) ? TOK_INT : TOK_DECIMAL;
/* It must be strdup'd to stop it being destroyed as soon as this loop returns */
T.value = GC_strdup(num);
return T;
}
if (c == '=')
{
T.type = TOK_ASSIGN;
/* Check it is =, not == */
c = next_char(I);
if (c == '=') /* It must be comparison: == */
T.type = TOK_ISEQUAL;
else
put_back(I, c);
return T;
}
if (c == '<')
{
T.type = TOK_LT;
c = next_char(I);
if (c == '=')
T.type = TOK_LTE;
else
put_back(I, c);
return T;
}
if (c == '>')
{
T.type = TOK_GT;
c = next_char(I);
if (c == '=')
T.type = TOK_GTE;
else
put_back(I, c);
return T;
}
if (c == '\"')
{
char *s = read_quoted_string(I);
T.value = s;
T.type = TOK_STRING;
return T;
}
if (c == '#')
{
while (c != '\n')
c = next_char(I);
}
if (isspace(c)) /* Ignore whitespace */
continue;
switch (c)
{
case 0: T.type = TOK_EOF; return T;
case '.': T.type = TOK_DOT; return T;
case '+': T.type = TOK_ADD; return T;
case '-': T.type = TOK_SUB; return T;
case '*': T.type = TOK_MUL; return T;
case '/': T.type = TOK_DIV; return T;
case '(': T.type = TOK_OPEN_BRACKET; return T;
case ')': T.type = TOK_CLOSED_BRACKET;return T;
//case '{': T.type = TOK_STARTBLOCK; return T;
//case '}': T.type = TOK_ENDBLOCK; return T;
case ',': T.type = TOK_COMMA; return T;
case ';': T.type = TOK_SEMICOLON; return T;
default:
if (num_invalid_chars++ < 1)
fprintf(stderr,
"Syntax error, line %lu: Invalid character \'%c\' (%d)\n",
I->line_number, c, c);
break;
}
}
printf("EOF\n");
T.type = TOK_EOF;
return T;
}
void
AsxTokenizer::move_next ()
{
char c;
while ((c = read_char ()) != -1) {
if (in_data) {
bool entirely_whitespace;
current_token->set_type (TOKEN_DATA);
current_token->set_value (read_data (c, entirely_whitespace));
current_token->set_entirely_whitespace (entirely_whitespace);
return;
}
if (c == OP_OPEN_ELEMENT) {
if (is_comment_open (c)) {
read_comment ();
continue;
}
current_token->set_type (TOKEN_OPEN_ELEMENT);
current_token->set_value (NULL);
in_data = false;
return;
}
if (c == OP_CLOSE_ELEMENT) {
current_token->set_type (TOKEN_CLOSE_ELEMENT);
current_token->set_value (NULL);
in_data = true;
return;
}
if (c == OP_ASSIGNMENT) {
current_token->set_type (TOKEN_ASSIGNMENT);
current_token->set_value (NULL);
return;
}
if (c == OP_SLASH) {
current_token->set_type (TOKEN_SLASH);
current_token->set_value (NULL);
return;
}
if (g_ascii_isspace (c)) {
current_token->set_type (TOKEN_WHITESPACE);
current_token->set_value (NULL);
return;
}
if (is_quote_char (c)) {
current_token->set_type (TOKEN_QUOTED_STRING);
current_token->set_value (read_quoted_string (c));
return;
}
if (is_name_char (c)) {
current_token->set_type (TOKEN_NAME);
current_token->set_value (read_name (c));
return;
}
}
current_token->set_type (TOKEN_EOF);
current_token->set_value (NULL);
}
Token InputReader::get_next_token() {
skip_white_spaces();
Token token, raw_token;
std::string qstr;
std::string value;
get_raw_token(raw_token);
switch(raw_token.type_) {
case null_token:
token.type_ = null_token; // this means no more tokens left
break;
case digit_token:
case negative_token:
float_t n_value;
input_stream_.unget();
if(!read_number(n_value)) {
std::cerr << "fatal error: failed while reading a number" << std::endl;
token.type_ = error_token;
} else {
token.type_ = number_token;
token.dvalue_ = n_value;
} // if-else
break;
case object_begin_token:
token.type_ = object_begin_token;
structure_stack_.push(object_begin_token);
break;
case object_end_token:
token.type_ = object_end_token;
if(structure_stack_.top() != object_begin_token) {
std::cerr << "fatal error: mismatched object encapsulators" << std::endl;
token.type_ = error_token;
} else {
structure_stack_.pop();
} // if-else
break;
case array_begin_token:
token.type_ = array_begin_token;
structure_stack_.push(array_begin_token);
break;
case array_end_token:
token.type_ = array_end_token;
if(structure_stack_.top() != array_begin_token) {
std::cerr << "fatal error: mismatched array encapsulators" << std::endl;
token.type_ = error_token;
} else {
structure_stack_.pop();
} // if-else
break;
case string_begin_end_token: // will always be begin since
// end will be removed while reading
//the whole string earlier
if(!read_quoted_string(qstr)) {
std::cerr << "fatal error: premature EOF reached while reading string" << std::endl;
token.type_ = error_token;
} else {
token.type_ = string_token;
token.svalue_ = qstr;
} // if-else
break;
case character_token:
input_stream_.unget();
read_keyword(value);
token.type_ = process_keyword_token(value);
if(token.type_ == error_token) {
std::cerr << "fatal error: unknown keyword '" << value << "'" << std::endl;
} // if
token.svalue_ = value;
break;
case assignment_token:
token.type_ = assignment_token;
break;
case separator_token:
token.type_ = separator_token;
break;
case comment_token:
skip_comments();
token.type_ = comment_token;
break;
default:
std::cerr << "fatal error: unknown token" << std::endl;
token.type_ = error_token;
} // switch
return token;
} // InputReader::get_next_token()
