char * striptags(const char *s, unsigned int l)
{
struct template_buffer *buf = buf_init(l);
unsigned char *ptr = (unsigned char *)s;
unsigned char *end = ptr + l;
unsigned char *tag;
unsigned char prev;
char esq[8];
int esl;
for (prev = ' '; ptr < end; ptr++)
{
if ((*ptr == '<') && ((ptr + 2) < end) &&
((*(ptr + 1) == '/') || isalpha(*(ptr + 1))))
{
for (tag = ptr; tag < end; tag++)
{
if (*tag == '>')
{
if (!isspace(prev))
buf_putchar(buf, ' ');
ptr = tag;
prev = ' ';
break;
}
}
}
else if (isspace(*ptr))
{
if (!isspace(prev))
buf_putchar(buf, *ptr);
prev = *ptr;
}
else
{
switch(*ptr)
{
case '"':
case '\'':
case '<':
case '>':
case '&':
esl = snprintf(esq, sizeof(esq), "&#%i;", *ptr);
buf_append(buf, esq, esl);
break;
default:
buf_putchar(buf, *ptr);
break;
}
prev = *ptr;
}
}
return buf_destroy(buf);
}
/* sanitize given string and replace all invalid UTF-8 sequences with "?" */
char * utf8(const char *s, unsigned int l)
{
struct template_buffer *buf = buf_init(l);
unsigned char *ptr = (unsigned char *)s;
unsigned int v, o;
if (!buf)
return NULL;
for (o = 0; o < l; o++)
{
/* ascii char */
if ((*ptr >= 0x01) && (*ptr <= 0x7F))
{
if (!buf_putchar(buf, (char)*ptr++))
break;
}
/* invalid byte or multi byte sequence */
else
{
if (!(v = _validate_utf8(&ptr, l - o, buf)))
break;
o += (v - 1);
}
}
return buf_destroy(buf);
}
/* Sanitize given string and strip all invalid XML bytes
* Validate UTF-8 sequences
* Escape XML control chars */
char * pcdata(const char *s, unsigned int l)
{
struct template_buffer *buf = buf_init(l);
unsigned char *ptr = (unsigned char *)s;
unsigned int o, v;
char esq[8];
int esl;
if (!buf)
return NULL;
for (o = 0; o < l; o++)
{
/* Invalid XML bytes */
if (((*ptr >= 0x00) && (*ptr <= 0x08)) ||
((*ptr >= 0x0B) && (*ptr <= 0x0C)) ||
((*ptr >= 0x0E) && (*ptr <= 0x1F)) ||
(*ptr == 0x7F))
{
ptr++;
}
/* Escapes */
else if ((*ptr == 0x26) ||
(*ptr == 0x27) ||
(*ptr == 0x22) ||
(*ptr == 0x3C) ||
(*ptr == 0x3E))
{
esl = snprintf(esq, sizeof(esq), "&#%i;", *ptr);
if (!buf_append(buf, esq, esl))
break;
ptr++;
}
/* ascii char */
else if (*ptr <= 0x7F)
{
buf_putchar(buf, (char)*ptr++);
}
/* multi byte sequence */
else
{
if (!(v = _validate_utf8(&ptr, l - o, buf)))
break;
o += (v - 1);
}
}
return buf_destroy(buf);
}
/* Replace the current INDEX in the buffer with the given CH value; however, if
CH is a backspace character, revert the current character at INDEX. */
static void replace_char(char ch)
{
bool is_backspace = (ch == 127);
if (is_backspace) {
move_col(-1);
buf_revertchar();
} else {
buf_putchar(ch);
move_col(+1);
}
}
void luastr_escape(struct template_buffer *out, const char *s, unsigned int l,
int escape_xml)
{
int esl;
char esq[8];
char *ptr;
for (ptr = (char *)s; ptr < (s + l); ptr++)
{
switch (*ptr)
{
case '\\':
buf_append(out, "\\\\", 2);
break;
case '"':
if (escape_xml)
buf_append(out, """, 5);
else
buf_append(out, "\\\"", 2);
break;
case '\n':
buf_append(out, "\\n", 2);
break;
case '\'':
case '&':
case '<':
case '>':
if (escape_xml)
{
esl = snprintf(esq, sizeof(esq), "&#%i;", *ptr);
buf_append(out, esq, esl);
break;
}
default:
buf_putchar(out, *ptr);
}
}
}
/* scan given source string, validate UTF-8 sequence and store result
* in given buffer object */
static int _validate_utf8(unsigned char **s, int l, struct template_buffer *buf)
{
unsigned char *ptr = *s;
unsigned int o = 0, v, n;
/* ascii byte without null */
if ((*(ptr+0) >= 0x01) && (*(ptr+0) <= 0x7F))
{
if (!buf_putchar(buf, *ptr++))
return 0;
o = 1;
}
/* multi byte sequence */
else if ((n = mb_num_chars(*ptr)) > 1)
{
/* count valid chars */
for (v = 1; (v <= n) && ((o+v) < l) && mb_is_cont(*(ptr+v)); v++);
switch (n)
{
case 6:
case 5:
/* five and six byte sequences are always invalid */
if (!buf_putchar(buf, '?'))
return 0;
break;
default:
/* if the number of valid continuation bytes matches the
* expected number and if the sequence is legal, copy
* the bytes to the destination buffer */
if ((v == n) && mb_is_shortest(ptr, n) &&
!mb_is_surrogate(ptr, n) && !mb_is_illegal(ptr, n))
{
/* copy sequence */
if (!buf_append(buf, (char *)ptr, n))
return 0;
}
/* the found sequence is illegal, skip it */
else
{
/* invalid sequence */
if (!buf_putchar(buf, '?'))
return 0;
}
break;
}
/* advance beyound the last found valid continuation char */
o = v;
ptr += v;
}
/* invalid byte (0x00) */
else
{
if (!buf_putchar(buf, '?')) /* or 0xEF, 0xBF, 0xBD */
return 0;
o = 1;
ptr++;
}
*s = ptr;
return o;
}
