/*
* ALNUMBLOCK = *(ALPHA / DIGIT)
*/
int
XSkip_alnumBlock(const char *head, const char *tail, const char **nextp)
{
const char *p;
for (p = head; p < tail && IS_ALNUM(*p); ++p);
*nextp = p;
return *nextp - head;
} // end function : XSkip_alnumBlock
int
gethtmlcmd(char **s)
{
extern Hash_si tagtable;
char cmdstr[MAX_CMD_LEN];
char *p = cmdstr;
char *save = *s;
int cmd;
(*s)++;
/* first character */
if (IS_ALNUM(**s) || **s == '_' || **s == '/') {
*(p++) = TOLOWER(**s);
(*s)++;
}
else
return HTML_UNKNOWN;
if (p[-1] == '/')
SKIP_BLANKS(*s);
while ((IS_ALNUM(**s) || **s == '_') && p - cmdstr < MAX_CMD_LEN) {
*(p++) = TOLOWER(**s);
(*s)++;
}
if (p - cmdstr == MAX_CMD_LEN) {
/* buffer overflow: perhaps caused by bad HTML source */
*s = save + 1;
return HTML_UNKNOWN;
}
*p = '\0';
/* hash search */
cmd = getHash_si(&tagtable, cmdstr, HTML_UNKNOWN);
while (**s && **s != '>')
(*s)++;
if (**s == '>')
(*s)++;
return cmd;
}
/*
* [RFC4871]
* x-sig-a-tag-k = ALPHA *(ALPHA / DIGIT) ; for later extension
* x-sig-a-tag-h = ALPHA *(ALPHA / DIGIT) ; for later extension
*/
int
XSkip_alphaAlnum(const char *head, const char *tail, const char **nextp)
{
const char *p = head;
// 先頭の文字は ALPHA
if (tail <= p || !IS_ALPHA(*p)) {
*nextp = p;
return *nextp - head;
} // end if
for (++p; p < tail && IS_ALNUM(*p); ++p);
*nextp = p;
return *nextp - head;
} // end function : XSkip_alphaAlnum
// lpPosから次のトークンまでコピー
void ScannerSub::CopyLabel(LPSTR& lpPos){
unsigned char c;
LPSTR lp = labelbuf;
while(true){
c = *lpPos;
if(IS_KANJI1(c)){
*(lp++) = c;
*(lp++) = c; //ほんとはここにsjisの2バイト目かの判別を入れる
lpPos++;
}else if(IS_ALNUM(c) || c == '#' || c == '@' || c == '$'){
*(lp++) = c;
}else{
break;
}
lpPos++;
}
*lp = '\0';
}
Str
quote_mailcap(char *s, int flag)
{
Str d;
d = Strnew();
for (;; ++s)
switch (*s) {
case '\0':
goto end;
case '$':
case '`':
case '"':
case '\\':
if (!(flag & MCF_SQUOTED))
Strcat_char(d, '\\');
Strcat_char(d, *s);
break;
case '\'':
if (flag & MCF_SQUOTED) {
Strcat_charp(d, "'\\''");
break;
}
default:
if (!flag && !IS_ALNUM(*s))
Strcat_char(d, '\\');
case '_':
case '.':
case ':':
case '/':
Strcat_char(d, *s);
break;
}
end:
return d;
}
/** 関数名の取得
@param[in] buf 行(先頭から)
@param[in] end buf末尾
@param[in] p 解析の現在位置
関数名はatom．atomは 小文字アルファベット，_, @ のいずれかから始まる
英数文字列か，あるいはシングルクォーテーションで囲まれた文字列．
*/
const wchar_t* COutlineErlang::ScanFuncName( const wchar_t* buf, const wchar_t* end, const wchar_t* p )
{
assert( m_state == STATE_NORMAL );
if( p > buf || ! ( IS_ATOM_HEAD( *p ) || *p == L'\'' )) {
return end;
}
if( *p == L'\'' ){
do {
++p;
} while( *p != L'\'' && p < end );
if( p >= end ){
// invalid atom
return p;
}
++p;
}
else {
do {
++p;
} while( IS_ALNUM( *p ) && p < end );
}
int buf_len = sizeof( m_func ) / sizeof( m_func[0]);
int len = p - buf;
if( buf[0] == L'\'' ){
++buf;
len -= 2;
--buf_len;
}
len = len < buf_len - 1 ? len : buf_len - 1;
wcsncpy( m_func, buf, len );
m_func[len] = L'\0';
m_state = STATE_FUNC_CANDIDATE_FIN;
return p;
}
static int str_copy(CONF *conf, char *section, char **pto, char *from)
{
int q, r, rr = 0, to = 0, len = 0;
char *s, *e, *rp, *p, *rrp, *np, *cp, v;
BUF_MEM *buf;
if ((buf = BUF_MEM_new()) == NULL)
return 0;
len = strlen(from) + 1;
if (!BUF_MEM_grow(buf, len))
goto err;
for (;;) {
if (IS_QUOTE(conf, *from)) {
q = *from;
from++;
while (!IS_EOF(conf, *from) && (*from != q)) {
if (IS_ESC(conf, *from)) {
from++;
if (IS_EOF(conf, *from))
break;
}
buf->data[to++] = *(from++);
}
if (*from == q)
from++;
} else if (IS_DQUOTE(conf, *from)) {
q = *from;
from++;
while (!IS_EOF(conf, *from)) {
if (*from == q) {
if (*(from + 1) == q) {
from++;
} else {
break;
}
}
buf->data[to++] = *(from++);
}
if (*from == q)
from++;
} else if (IS_ESC(conf, *from)) {
from++;
v = *(from++);
if (IS_EOF(conf, v))
break;
else if (v == 'r')
v = '\r';
else if (v == 'n')
v = '\n';
else if (v == 'b')
v = '\b';
else if (v == 't')
v = '\t';
buf->data[to++] = v;
} else if (IS_EOF(conf, *from))
break;
else if (*from == '$') {
size_t newsize;
/* try to expand it */
rrp = NULL;
s = &(from[1]);
if (*s == '{')
q = '}';
else if (*s == '(')
q = ')';
else
q = 0;
if (q)
s++;
cp = section;
e = np = s;
while (IS_ALNUM(conf, *e))
e++;
if ((e[0] == ':') && (e[1] == ':')) {
cp = np;
rrp = e;
rr = *e;
*rrp = '\0';
e += 2;
np = e;
while (IS_ALNUM(conf, *e))
e++;
}
r = *e;
*e = '\0';
rp = e;
if (q) {
if (r != q) {
CONFerr(CONF_F_STR_COPY, CONF_R_NO_CLOSE_BRACE);
goto err;
}
e++;
}
/*-
* So at this point we have
* np which is the start of the name string which is
* '\0' terminated.
* cp which is the start of the section string which is
* '\0' terminated.
* e is the 'next point after'.
* r and rr are the chars replaced by the '\0'
* rp and rrp is where 'r' and 'rr' came from.
*/
p = _CONF_get_string(conf, cp, np);
if (rrp != NULL)
*rrp = rr;
*rp = r;
if (p == NULL) {
CONFerr(CONF_F_STR_COPY, CONF_R_VARIABLE_HAS_NO_VALUE);
goto err;
}
newsize = strlen(p) + buf->length - (e - from);
if (newsize > MAX_CONF_VALUE_LENGTH) {
CONFerr(CONF_F_STR_COPY, CONF_R_VARIABLE_EXPANSION_TOO_LONG);
goto err;
}
if (!BUF_MEM_grow_clean(buf, newsize)) {
CONFerr(CONF_F_STR_COPY, ERR_R_MALLOC_FAILURE);
goto err;
}
while (*p)
buf->data[to++] = *(p++);
/*
* Since we change the pointer 'from', we also have to change the
* perceived length of the string it points at. /RL
*/
len -= e - from;
from = e;
/*
* In case there were no braces or parenthesis around the
* variable reference, we have to put back the character that was
* replaced with a '\0'. /RL
*/
*rp = r;
} else
buf->data[to++] = *(from++);
}
buf->data[to] = '\0';
OPENSSL_free(*pto);
*pto = buf->data;
OPENSSL_free(buf);
return 1;
err:
BUF_MEM_free(buf);
return 0;
}
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;
}
