static void
collect_arguments (symbol *sym, struct obstack *argptr,
struct obstack *arguments)
{
token_data td;
token_data *tdp;
bool more_args;
bool groks_macro_args = SYMBOL_MACRO_ARGS (sym);
TOKEN_DATA_TYPE (&td) = TOKEN_TEXT;
TOKEN_DATA_TEXT (&td) = SYMBOL_NAME (sym);
tdp = (token_data *) obstack_copy (arguments, &td, sizeof td);
obstack_ptr_grow (argptr, tdp);
if (peek_token () == TOKEN_OPEN)
{
next_token (&td, NULL); /* gobble parenthesis */
do
{
more_args = expand_argument (arguments, &td);
if (!groks_macro_args && TOKEN_DATA_TYPE (&td) == TOKEN_FUNC)
{
TOKEN_DATA_TYPE (&td) = TOKEN_TEXT;
TOKEN_DATA_TEXT (&td) = (char *) "";
}
tdp = (token_data *) obstack_copy (arguments, &td, sizeof td);
obstack_ptr_grow (argptr, tdp);
}
while (more_args);
}
}
const char *
predefined_attribute (const char *key, int *ptr_argc, token_data **argv,
boolean lowercase)
{
var_stack *next;
char *cp, *sp, *lower;
int i, j, special_chars;
boolean found = FALSE;
i = 1;
while (i<*ptr_argc)
{
special_chars = 0;
sp = TOKEN_DATA_TEXT (argv[i]);
while (IS_GROUP (*sp))
{
sp++;
special_chars++;
}
cp = strchr (sp, '=');
if ((cp == NULL && strcasecmp (sp, key) == 0) ||
(cp != NULL && strncasecmp (sp, key, strlen (key)) == 0
&& *(sp + strlen (key)) == '='))
{
found = TRUE;
next = (var_stack *) xmalloc (sizeof (var_stack));
next->prev = tag_attr;
if (cp)
{
next->text = (char *) xmalloc (special_chars + strlen (cp+1) + 1);
if (special_chars)
strncpy (next->text, TOKEN_DATA_TEXT (argv[i]), special_chars);
strcpy (next->text+special_chars, cp+1);
}
else
next->text = xstrdup (key);
tag_attr = next;
if (lowercase)
for (lower=tag_attr->text; *lower != '\0'; lower++)
*lower = tolower (*lower);
/* remove this attribute from argv[]. */
for (j=i+1; j<=*ptr_argc; j++)
argv[j-1] = argv[j];
(*ptr_argc)--;
}
i++;
}
return (found ? tag_attr->text : NULL );
}
static void
print_token (const char *s, token_type t, token_data *td)
{
fprintf (stderr, "%s: ", s);
switch (t)
{ /* TOKSW */
case TOKEN_SIMPLE:
fprintf (stderr, "char:");
break;
case TOKEN_WORD:
fprintf (stderr, "word:");
break;
case TOKEN_STRING:
fprintf (stderr, "string:");
break;
case TOKEN_MACDEF:
fprintf (stderr, "macro: 0x%x\n", TOKEN_DATA_FUNC (td));
break;
case TOKEN_EOF:
fprintf (stderr, "eof\n");
break;
}
fprintf (stderr, "\t\"%s\"\n", TOKEN_DATA_TEXT (td));
}
const char *
push_string_finish (read_type expansion)
{
const char *ret = NULL;
if (next == NULL)
return NULL;
if (obstack_object_size (current_input) > 0)
{
obstack_1grow (current_input, '\0');
next->u.u_s.start = (unsigned char *) obstack_finish (current_input);
if (expansion == READ_ATTR_VERB || expansion == READ_ATTR_ASIS
|| expansion == READ_BODY)
{
TOKEN_DATA_TYPE (&token_read) = TOKEN_TEXT;
TOKEN_DATA_TEXT (&token_read) = xstrdup ((char *) next->u.u_s.start);
next->u.u_s.current = next->u.u_s.start + strlen ((char *) next->u.u_s.start);
}
else
next->u.u_s.current = next->u.u_s.start;
next->prev = isp;
isp = next;
ret = (char *) isp->u.u_s.start; /* for immediate use only */
}
else
obstack_free (current_input, next); /* people might leave garbage on it. */
next = NULL;
return ret;
}
static void
expand_token (struct obstack *obs, token_type t, token_data *td, int line)
{
symbol *sym;
switch (t)
{ /* TOKSW */
case TOKEN_EOF:
case TOKEN_MACDEF:
break;
case TOKEN_OPEN:
case TOKEN_COMMA:
case TOKEN_CLOSE:
case TOKEN_SIMPLE:
case TOKEN_STRING:
shipout_text (obs, TOKEN_DATA_TEXT (td), strlen (TOKEN_DATA_TEXT (td)),
line);
break;
case TOKEN_WORD:
sym = lookup_symbol (TOKEN_DATA_TEXT (td), SYMBOL_LOOKUP);
if (sym == NULL || SYMBOL_TYPE (sym) == TOKEN_VOID
|| (SYMBOL_TYPE (sym) == TOKEN_FUNC
&& SYMBOL_BLIND_NO_ARGS (sym)
&& peek_token () != TOKEN_OPEN))
{
#ifdef ENABLE_CHANGEWORD
shipout_text (obs, TOKEN_DATA_ORIG_TEXT (td),
strlen (TOKEN_DATA_ORIG_TEXT (td)), line);
#else
shipout_text (obs, TOKEN_DATA_TEXT (td),
strlen (TOKEN_DATA_TEXT (td)), line);
#endif
}
else
expand_macro (sym);
break;
default:
M4ERROR ((warning_status, 0,
"INTERNAL ERROR: bad token type in expand_token ()"));
abort ();
}
}
boolean
bad_argc (token_data *name, int argc, int min, int max)
{
boolean isbad = FALSE;
if (min > 0 && argc < min)
{
if (!suppress_warnings)
MP4HERROR ((warning_status, 0,
_("Warning:%s:%d: Too few arguments to built-in `%s'"),
CURRENT_FILE_LINE, TOKEN_DATA_TEXT (name)));
isbad = TRUE;
}
else if (max > 0 && argc > max && !suppress_warnings)
{
MP4HERROR ((warning_status, 0,
_("Warning:%s:%d: Excess arguments to built-in `%s' ignored"),
CURRENT_FILE_LINE, TOKEN_DATA_TEXT (name)));
}
return isbad;
}
boolean
numeric_arg (token_data *macro, const char *arg, boolean warn, int *valuep)
{
char *endp;
if (*arg == 0 || (*valuep = strtol (skip_space (arg), &endp, 10),
*skip_space (endp) != 0))
{
if (warn)
MP4HERROR ((warning_status, 0,
_("Warning:%s:%d: Argument `%s' non-numeric in the <%s> tag"),
CURRENT_FILE_LINE, arg, TOKEN_DATA_TEXT (macro)));
return FALSE;
}
return TRUE;
}
void
trace_pre (const char *name, int id, int argc, token_data **argv)
{
int i;
const builtin *bp;
trace_header (id);
trace_format ("%s", name);
if (argc > 1 && (debug_level & DEBUG_TRACE_ARGS))
{
trace_format ("(");
for (i = 1; i < argc; i++)
{
if (i != 1)
trace_format (", ");
switch (TOKEN_DATA_TYPE (argv[i]))
{
case TOKEN_TEXT:
trace_format ("%l%S%r", TOKEN_DATA_TEXT (argv[i]));
break;
case TOKEN_FUNC:
bp = find_builtin_by_addr (TOKEN_DATA_FUNC (argv[i]));
if (bp == NULL)
{
M4ERROR ((warning_status, 0, "\
INTERNAL ERROR: builtin not found in builtin table! (trace_pre ())"));
abort ();
}
trace_format ("<%s>", bp->name);
break;
case TOKEN_VOID:
default:
M4ERROR ((warning_status, 0,
"INTERNAL ERROR: bad token data type (trace_pre ())"));
abort ();
}
}
token_type
next_token (token_data *td)
{
int ch;
int quote_level;
token_type type;
#ifdef ENABLE_CHANGEWORD
int startpos;
char *orig_text = 0;
#endif
obstack_free (&token_stack, token_bottom);
obstack_1grow (&token_stack, '\0');
token_bottom = obstack_finish (&token_stack);
ch = peek_input ();
if (ch == CHAR_EOF)
{
return TOKEN_EOF;
#ifdef DEBUG_INPUT
fprintf (stderr, "next_token -> EOF\n");
#endif
}
if (ch == CHAR_MACRO)
{
init_macro_token (td);
(void) next_char ();
return TOKEN_MACDEF;
}
(void) next_char ();
if (MATCH (ch, bcomm.string))
{
obstack_grow (&token_stack, bcomm.string, bcomm.length);
while ((ch = next_char ()) != CHAR_EOF && !MATCH (ch, ecomm.string))
obstack_1grow (&token_stack, ch);
if (ch != CHAR_EOF)
obstack_grow (&token_stack, ecomm.string, ecomm.length);
type = TOKEN_STRING;
}
#ifdef ENABLE_CHANGEWORD
else if (default_word_regexp && (isalpha (ch) || ch == '_'))
#else
else if (isalpha (ch) || ch == '_')
#endif
{
obstack_1grow (&token_stack, ch);
while ((ch = peek_input ()) != CHAR_EOF && (isalnum (ch) || ch == '_'))
{
obstack_1grow (&token_stack, ch);
(void) next_char ();
}
type = TOKEN_WORD;
}
#ifdef ENABLE_CHANGEWORD
else if (!default_word_regexp && strchr (word_start, ch))
{
obstack_1grow (&token_stack, ch);
while (1)
{
ch = peek_input ();
if (ch == CHAR_EOF)
break;
obstack_1grow (&token_stack, ch);
startpos = re_search (&word_regexp, obstack_base (&token_stack),
obstack_object_size (&token_stack), 0, 0,
®s);
if (startpos != 0 ||
regs.end [0] != obstack_object_size (&token_stack))
{
*(((char *) obstack_base (&token_stack)
+ obstack_object_size (&token_stack)) - 1) = '\0';
break;
}
next_char ();
}
obstack_1grow (&token_stack, '\0');
orig_text = obstack_finish (&token_stack);
if (regs.start[1] != -1)
obstack_grow (&token_stack,orig_text + regs.start[1],
regs.end[1] - regs.start[1]);
else
obstack_grow (&token_stack, orig_text,regs.end[0]);
type = TOKEN_WORD;
}
#endif /* ENABLE_CHANGEWORD */
else if (!MATCH (ch, lquote.string))
{
type = TOKEN_SIMPLE;
obstack_1grow (&token_stack, ch);
}
else
{
quote_level = 1;
while (1)
{
ch = next_char ();
if (ch == CHAR_EOF)
M4ERROR ((EXIT_FAILURE, 0,
"ERROR: EOF in string"));
if (MATCH (ch, rquote.string))
{
if (--quote_level == 0)
break;
obstack_grow (&token_stack, rquote.string, rquote.length);
}
else if (MATCH (ch, lquote.string))
{
quote_level++;
obstack_grow (&token_stack, lquote.string, lquote.length);
}
else
obstack_1grow (&token_stack, ch);
}
type = TOKEN_STRING;
}
obstack_1grow (&token_stack, '\0');
TOKEN_DATA_TYPE (td) = TOKEN_TEXT;
TOKEN_DATA_TEXT (td) = obstack_finish (&token_stack);
#ifdef ENABLE_CHANGEWORD
if (orig_text == NULL)
orig_text = TOKEN_DATA_TEXT (td);
TOKEN_DATA_ORIG_TEXT (td) = orig_text;
#endif
#ifdef DEBUG_INPUT
fprintf (stderr, "next_token -> %d (%s)\n", type, TOKEN_DATA_TEXT (td));
#endif
return type;
}
token_type
next_token (token_data *td)
{
int ch;
int quote_level;
token_type type;
obstack_free (&token_stack, token_bottom);
obstack_1grow (&token_stack, '\0');
token_bottom = obstack_finish (&token_stack);
ch = peek_input ();
if (ch == CHAR_EOF)
{
return TOKEN_EOF;
#ifdef DEBUG_INPUT
fprintf (stderr, "next_token -> EOF\n");
#endif
}
if (ch == CHAR_MACRO)
{
init_macro_token (td);
(void) next_char ();
return TOKEN_MACDEF;
}
(void) next_char ();
if (MATCH (ch, bcomm))
{
obstack_grow (&token_stack, bcomm, len_bcomm);
while ((ch = next_char ()) != CHAR_EOF && !MATCH (ch, ecomm))
obstack_1grow (&token_stack, ch);
if (ch != CHAR_EOF)
obstack_grow (&token_stack, ecomm, len_ecomm);
type = TOKEN_STRING;
}
else if (isalpha (ch) || ch == '_')
{
obstack_1grow (&token_stack, ch);
while ((ch = peek_input ()) != CHAR_EOF && (isalnum (ch) || ch == '_'))
{
obstack_1grow (&token_stack, ch);
(void) next_char ();
}
type = TOKEN_WORD;
}
else if (!MATCH (ch, lquote))
{
type = TOKEN_SIMPLE;
obstack_1grow (&token_stack, ch);
}
else
{
quote_level = 1;
while (1)
{
ch = next_char ();
if (ch == CHAR_EOF)
fatal ("EOF in string");
if (MATCH (ch, rquote))
{
if (--quote_level == 0)
break;
obstack_grow (&token_stack, rquote, len_rquote);
}
else if (MATCH (ch, lquote))
{
quote_level++;
obstack_grow (&token_stack, lquote, len_lquote);
}
else
obstack_1grow (&token_stack, ch);
}
type = TOKEN_STRING;
}
obstack_1grow (&token_stack, '\0');
TOKEN_DATA_TYPE (td) = TOKEN_TEXT;
TOKEN_DATA_TEXT (td) = obstack_finish (&token_stack);
#ifdef DEBUG_INPUT
fprintf (stderr, "next_token -> %d (%s)\n", type, TOKEN_DATA_TEXT (td));
#endif
return type;
}
token_type
next_token (token_data *td, read_type expansion, boolean in_string)
{
int ch;
token_type type = TOKEN_NONE;
int quote_level;
if (TOKEN_DATA_TYPE (&token_read) != TOKEN_VOID)
{
type = TOKEN_STRING;
obstack_grow (&token_stack, TOKEN_DATA_TEXT (&token_read),
strlen (TOKEN_DATA_TEXT (&token_read)));
xfree ((voidstar) TOKEN_DATA_TEXT (&token_read));
TOKEN_DATA_TYPE (&token_read) = TOKEN_VOID;
}
while (type == TOKEN_NONE)
{
obstack_free (&token_stack, token_bottom);
obstack_1grow (&token_stack, '\0');
token_bottom = obstack_finish (&token_stack);
ch = peek_input ();
if (ch == CHAR_EOF) /* EOF */
{
#ifdef DEBUG_INPUT
fprintf (stderr, "next_token -> EOF\n");
#endif
return TOKEN_EOF;
}
if (ch == CHAR_MACRO) /* MACRO TOKEN */
{
init_macro_token (td);
(void) next_char ();
#ifdef DEBUG_INPUT
print_token("next_token", TOKEN_MACDEF, td);
#endif
return TOKEN_MACDEF;
}
(void) next_char ();
if (IS_TAG(ch)) /* ESCAPED WORD */
{
if (lquote.length > 0 && MATCH (ch, lquote.string))
{
if (visible_quotes || expansion == READ_ATTR_VERB
|| expansion == READ_ATTR_ASIS || expansion == READ_BODY)
obstack_grow (&token_stack, lquote.string, lquote.length);
while ((ch = next_char ()) != CHAR_EOF)
{
if (rquote.length > 0 && MATCH (ch, rquote.string))
break;
obstack_1grow (&token_stack, ch);
}
if (visible_quotes || expansion == READ_ATTR_VERB
|| expansion == READ_ATTR_ASIS || expansion == READ_BODY)
{
obstack_grow (&token_stack, rquote.string, rquote.length);
type = TOKEN_STRING;
}
else
type = TOKEN_QUOTED;
}
else
{
obstack_1grow (&token_stack, ch);
if ((ch = peek_input ()) != CHAR_EOF)
{
if (ch == '/')
{
obstack_1grow (&token_stack, '/');
(void) next_char ();
ch = peek_input ();
}
if (IS_ALPHA(ch))
{
ch = next_char ();
obstack_1grow (&token_stack, ch);
while ((ch = next_char ()) != CHAR_EOF && IS_ALNUM(ch))
{
obstack_1grow (&token_stack, ch);
}
if (ch == '*')
{
obstack_1grow (&token_stack, ch);
ch = peek_input ();
}
else
unget_input(ch);
if (IS_SPACE(ch) || IS_CLOSE(ch) || IS_SLASH (ch))
type = TOKEN_WORD;
else
type = TOKEN_STRING;
}
else
type = TOKEN_STRING;
}
else
type = TOKEN_SIMPLE; /* escape before eof */
}
}
else if (IS_CLOSE(ch))
{
obstack_1grow (&token_stack, ch);
type = TOKEN_SIMPLE;
}
else if (IS_ENTITY(ch)) /* entity */
{
obstack_1grow (&token_stack, ch);
if ((ch = peek_input ()) != CHAR_EOF)
{
if (IS_ALPHA(ch))
{
ch = next_char ();
obstack_1grow (&token_stack, ch);
while ((ch = next_char ()) != CHAR_EOF && IS_ALNUM(ch))
{
obstack_1grow (&token_stack, ch);
}
if (ch == ';')
{
obstack_1grow (&token_stack, ch);
type = TOKEN_ENTITY;
}
else
{
type = TOKEN_STRING;
unget_input(ch);
}
}
else
type = TOKEN_STRING;
}
else
type = TOKEN_SIMPLE; /* escape before eof */
}
else if (eolcomm.length > 0 && MATCH (ch, eolcomm.string))
skip_line ();
else if (expansion == READ_BODY)
{
if (ch == '"')
obstack_1grow (&token_stack, CHAR_QUOTE);
else
obstack_1grow (&token_stack, ch);
while ((ch = next_char ()) != CHAR_EOF
&& ! IS_TAG(ch) && ! IS_CLOSE (ch))
{
if (eolcomm.length > 0 && MATCH (ch, eolcomm.string))
{
skip_line ();
ch = CHAR_EOF;
break;
}
if (ch == '"')
obstack_1grow (&token_stack, CHAR_QUOTE);
else
obstack_1grow (&token_stack, ch);
}
unget_input(ch);
type = TOKEN_STRING;
}
/* Below we know that expansion != READ_BODY */
else if (IS_ALPHA(ch))
{
obstack_1grow (&token_stack, ch);
while ((ch = next_char ()) != CHAR_EOF && (IS_ALNUM(ch)))
{
obstack_1grow (&token_stack, ch);
}
if (ch == '*')
{
obstack_1grow (&token_stack, ch);
ch = peek_input ();
}
else
unget_input(ch);
type = TOKEN_STRING;
}
else if (IS_RQUOTE(ch))
{
MP4HERROR ((EXIT_FAILURE, 0,
"INTERNAL ERROR: CHAR_RQUOTE found."));
}
else if (IS_LQUOTE(ch)) /* QUOTED STRING */
{
quote_level = 1;
while (1)
{
ch = next_char ();
if (ch == CHAR_EOF)
MP4HERROR ((EXIT_FAILURE, 0,
"INTERNAL ERROR: EOF in string"));
if (IS_BGROUP(ch) || IS_EGROUP(ch))
continue;
else if (IS_RQUOTE(ch))
{
quote_level--;
if (quote_level == 0)
break;
}
else if (IS_LQUOTE(ch))
quote_level++;
else
obstack_1grow (&token_stack, ch);
}
type = TOKEN_QUOTED;
}
else if (IS_BGROUP(ch)) /* BEGIN GROUP */
type = TOKEN_BGROUP;
else if (IS_EGROUP(ch)) /* END GROUP */
type = TOKEN_EGROUP;
else if (ch == '"') /* QUOTED STRING */
{
switch (expansion)
{
case READ_NORMAL:
obstack_1grow (&token_stack, CHAR_QUOTE);
type = TOKEN_SIMPLE;
break;
case READ_ATTRIBUTE:
case READ_ATTR_VERB:
type = TOKEN_QUOTE;
break;
case READ_ATTR_ASIS:
case READ_ATTR_QUOT:
obstack_1grow (&token_stack, '"');
type = TOKEN_QUOTE;
break;
default:
MP4HERROR ((warning_status, 0,
"INTERNAL ERROR: Unknown expansion type"));
exit (1);
}
}
else if (ch == '\\')
{
switch (expansion)
{
case READ_NORMAL:
obstack_1grow (&token_stack, ch);
type = TOKEN_SIMPLE;
break;
case READ_ATTRIBUTE:
case READ_ATTR_QUOT:
ch = next_char();
if (ch == 'n')
obstack_1grow (&token_stack, '\n');
else if (ch == 't')
obstack_1grow (&token_stack, '\t');
else if (ch == 'r')
obstack_1grow (&token_stack, '\r');
else if (ch == '\\')
obstack_1grow (&token_stack, ch);
else if (ch == '"' && in_string)
obstack_1grow (&token_stack, CHAR_QUOTE);
else
{
if (!(exp_flags & EXP_STD_BSLASH))
obstack_1grow (&token_stack, '\\');
obstack_1grow (&token_stack, ch);
}
type = TOKEN_STRING;
break;
case READ_ATTR_VERB:
ch = next_char();
if (ch == '"' && in_string)
obstack_1grow (&token_stack, CHAR_QUOTE);
else
{
obstack_1grow (&token_stack, '\\');
obstack_1grow (&token_stack, ch);
}
type = TOKEN_STRING;
break;
case READ_ATTR_ASIS:
obstack_1grow (&token_stack, ch);
ch = next_char();
obstack_1grow (&token_stack, ch);
type = TOKEN_STRING;
break;
default:
MP4HERROR ((warning_status, 0,
"INTERNAL ERROR: Unknown expansion type"));
exit (1);
}
}
else /* EVERYTHING ELSE */
{
obstack_1grow (&token_stack, ch);
if (IS_OTHER(ch) || IS_NUM(ch))
type = TOKEN_STRING;
else if (IS_SPACE(ch))
{
while ((ch = next_char ()) != CHAR_EOF && IS_SPACE(ch))
obstack_1grow (&token_stack, ch);
unget_input(ch);
type = TOKEN_SPACE;
}
else
type = TOKEN_SIMPLE;
}
}
obstack_1grow (&token_stack, '\0');
TOKEN_DATA_TYPE (td) = TOKEN_TEXT;
TOKEN_DATA_TEXT (td) = obstack_finish (&token_stack);
#ifdef DEBUG_INPUT
print_token("next_token", type, td);
#endif
return type;
}
static int
print_token (const char *s, token_type t, token_data *td)
{
fprintf (stderr, "%s: ", s);
switch (t)
{ /* TOKSW */
case TOKEN_SIMPLE:
fprintf (stderr, "char\t\"%s\"\n", TOKEN_DATA_TEXT (td));
break;
case TOKEN_WORD:
fprintf (stderr, "word\t\"%s\"\n", TOKEN_DATA_TEXT (td));
break;
case TOKEN_ENTITY:
fprintf (stderr, "entity\t\"%s\"\n", TOKEN_DATA_TEXT (td));
break;
case TOKEN_STRING:
fprintf (stderr, "string\t\"%s\"\n", TOKEN_DATA_TEXT (td));
break;
case TOKEN_QUOTED:
fprintf (stderr, "quoted\t\"%s\"\n", TOKEN_DATA_TEXT (td));
break;
case TOKEN_BGROUP:
fprintf (stderr, "bgroup\n");
break;
case TOKEN_EGROUP:
fprintf (stderr, "egroup\n");
break;
case TOKEN_QUOTE:
fprintf (stderr, "quote\n");
break;
case TOKEN_SPACE:
fprintf (stderr, "space\t\"%s\"\n", TOKEN_DATA_TEXT (td));
break;
case TOKEN_MACDEF:
fprintf (stderr, "macro 0x%x\n", (int)TOKEN_DATA_FUNC (td));
break;
case TOKEN_EOF:
fprintf (stderr, "eof\n");
break;
case TOKEN_NONE:
fprintf (stderr, "none\n");
break;
default:
MP4HERROR ((warning_status, 0,
"INTERNAL ERROR: unknown token in print_token"));
break;
}
return 0;
}
token_type
next_token (token_data *td, int *line)
{
int ch;
int quote_level;
token_type type;
#ifdef ENABLE_CHANGEWORD
int startpos;
char *orig_text = NULL;
#endif
const char *file;
int dummy;
obstack_free (&token_stack, token_bottom);
if (!line)
line = &dummy;
/* Can't consume character until after CHAR_MACRO is handled. */
ch = peek_input ();
if (ch == CHAR_EOF)
{
#ifdef DEBUG_INPUT
xfprintf (stderr, "next_token -> EOF\n");
#endif
next_char ();
return TOKEN_EOF;
}
if (ch == CHAR_MACRO)
{
init_macro_token (td);
next_char ();
#ifdef DEBUG_INPUT
xfprintf (stderr, "next_token -> MACDEF (%s)\n",
find_builtin_by_addr (TOKEN_DATA_FUNC (td))->name);
#endif
return TOKEN_MACDEF;
}
next_char (); /* Consume character we already peeked at. */
file = current_file;
*line = current_line;
if (MATCH (ch, bcomm.string, true))
{
obstack_grow (&token_stack, bcomm.string, bcomm.length);
while ((ch = next_char ()) != CHAR_EOF
&& !MATCH (ch, ecomm.string, true))
obstack_1grow (&token_stack, ch);
if (ch != CHAR_EOF)
obstack_grow (&token_stack, ecomm.string, ecomm.length);
else
/* current_file changed to "" if we see CHAR_EOF, use the
previous value we stored earlier. */
M4ERROR_AT_LINE ((EXIT_FAILURE, 0, file, *line,
"ERROR: end of file in comment"));
type = TOKEN_STRING;
}
else if (default_word_regexp && (isalpha (ch) || ch == '_'))
{
obstack_1grow (&token_stack, ch);
while ((ch = peek_input ()) != CHAR_EOF && (isalnum (ch) || ch == '_'))
{
obstack_1grow (&token_stack, ch);
next_char ();
}
type = TOKEN_WORD;
}
#ifdef ENABLE_CHANGEWORD
else if (!default_word_regexp && word_regexp.fastmap[ch])
{
obstack_1grow (&token_stack, ch);
while (1)
{
ch = peek_input ();
if (ch == CHAR_EOF)
break;
obstack_1grow (&token_stack, ch);
startpos = re_search (&word_regexp,
(char *) obstack_base (&token_stack),
obstack_object_size (&token_stack), 0, 0,
®s);
if (startpos ||
regs.end [0] != (regoff_t) obstack_object_size (&token_stack))
{
*(((char *) obstack_base (&token_stack)
+ obstack_object_size (&token_stack)) - 1) = '\0';
break;
}
next_char ();
}
obstack_1grow (&token_stack, '\0');
orig_text = (char *) obstack_finish (&token_stack);
if (regs.start[1] != -1)
obstack_grow (&token_stack,orig_text + regs.start[1],
regs.end[1] - regs.start[1]);
else
obstack_grow (&token_stack, orig_text,regs.end[0]);
type = TOKEN_WORD;
}
#endif /* ENABLE_CHANGEWORD */
else if (!MATCH (ch, lquote.string, true))
{
switch (ch)
{
case '(':
type = TOKEN_OPEN;
break;
case ',':
type = TOKEN_COMMA;
break;
case ')':
type = TOKEN_CLOSE;
break;
default:
type = TOKEN_SIMPLE;
break;
}
obstack_1grow (&token_stack, ch);
}
else
{
bool fast = lquote.length == 1 && rquote.length == 1;
quote_level = 1;
while (1)
{
/* Try scanning a buffer first. */
const char *buffer = (isp && isp->type == INPUT_STRING
? isp->u.u_s.string : NULL);
if (buffer && *buffer)
{
size_t len = isp->u.u_s.end - buffer;
const char *p = buffer;
do
{
p = (char *) memchr2 (p, *lquote.string, *rquote.string,
buffer + len - p);
}
while (p && fast && (*p++ == *rquote.string
? --quote_level : ++quote_level));
if (p)
{
if (fast)
{
assert (!quote_level);
obstack_grow (&token_stack, buffer, p - buffer - 1);
isp->u.u_s.string += p - buffer;
break;
}
obstack_grow (&token_stack, buffer, p - buffer);
ch = to_uchar (*p);
isp->u.u_s.string += p - buffer + 1;
}
else
{
obstack_grow (&token_stack, buffer, len);
isp->u.u_s.string += len;
continue;
}
}
/* Fall back to a byte. */
else
ch = next_char ();
if (ch == CHAR_EOF)
/* current_file changed to "" if we see CHAR_EOF, use
the previous value we stored earlier. */
M4ERROR_AT_LINE ((EXIT_FAILURE, 0, file, *line,
"ERROR: end of file in string"));
if (MATCH (ch, rquote.string, true))
{
if (--quote_level == 0)
break;
obstack_grow (&token_stack, rquote.string, rquote.length);
}
else if (MATCH (ch, lquote.string, true))
{
quote_level++;
obstack_grow (&token_stack, lquote.string, lquote.length);
}
else
obstack_1grow (&token_stack, ch);
}
type = TOKEN_STRING;
}
obstack_1grow (&token_stack, '\0');
TOKEN_DATA_TYPE (td) = TOKEN_TEXT;
TOKEN_DATA_TEXT (td) = (char *) obstack_finish (&token_stack);
#ifdef ENABLE_CHANGEWORD
if (orig_text == NULL)
orig_text = TOKEN_DATA_TEXT (td);
TOKEN_DATA_ORIG_TEXT (td) = orig_text;
#endif
#ifdef DEBUG_INPUT
xfprintf (stderr, "next_token -> %s (%s)\n",
token_type_string (type), TOKEN_DATA_TEXT (td));
#endif
return type;
}
static bool
expand_argument (struct obstack *obs, token_data *argp)
{
token_type t;
token_data td;
char *text;
int paren_level;
const char *file = current_file;
int line = current_line;
TOKEN_DATA_TYPE (argp) = TOKEN_VOID;
/* Skip leading white space. */
do
{
t = next_token (&td, NULL);
}
while (t == TOKEN_SIMPLE && isspace (to_uchar (*TOKEN_DATA_TEXT (&td))));
paren_level = 0;
while (1)
{
switch (t)
{ /* TOKSW */
case TOKEN_COMMA:
case TOKEN_CLOSE:
if (paren_level == 0)
{
/* The argument MUST be finished, whether we want it or not. */
obstack_1grow (obs, '\0');
text = (char *) obstack_finish (obs);
if (TOKEN_DATA_TYPE (argp) == TOKEN_VOID)
{
TOKEN_DATA_TYPE (argp) = TOKEN_TEXT;
TOKEN_DATA_TEXT (argp) = text;
}
return t == TOKEN_COMMA;
}
/* fallthru */
case TOKEN_OPEN:
case TOKEN_SIMPLE:
text = TOKEN_DATA_TEXT (&td);
if (*text == '(')
paren_level++;
else if (*text == ')')
paren_level--;
expand_token (obs, t, &td, line);
break;
case TOKEN_EOF:
/* current_file changed to "" if we see TOKEN_EOF, use the
previous value we stored earlier. */
M4ERROR_AT_LINE ((EXIT_FAILURE, 0, file, line,
"ERROR: end of file in argument list"));
break;
case TOKEN_WORD:
case TOKEN_STRING:
expand_token (obs, t, &td, line);
break;
case TOKEN_MACDEF:
if (obstack_object_size (obs) == 0)
{
TOKEN_DATA_TYPE (argp) = TOKEN_FUNC;
TOKEN_DATA_FUNC (argp) = TOKEN_DATA_FUNC (&td);
}
break;
default:
M4ERROR ((warning_status, 0,
"INTERNAL ERROR: bad token type in expand_argument ()"));
abort ();
}
t = next_token (&td, NULL);
}
}
