Octstr *html_to_sms(Octstr *html)
{
long i, len;
int c;
Octstr *sms;
sms = octstr_create("");
len = octstr_len(html);
i = 0;
while (i < len) {
c = octstr_get_char(html, i);
switch (c) {
case '<':
if (html_comment_begins(html, i))
skip_html_comment(html, &i);
else
skip_html_tag(html, &i);
break;
case '&':
convert_html_entity(sms, html, &i);
break;
default:
octstr_append_char(sms, c);
++i;
break;
}
}
octstr_shrink_blanks(sms);
octstr_strip_blanks(sms);
return sms;
}
static void
parse_html_decode(
unsigned char *buf,
const unsigned char *start,
const unsigned char *end)
{
const unsigned char *p = start;
unsigned char *out = buf;
while (*p && p != end) {
if (*p != '&') {
*out++ = *p++;
continue;
}
++p;
if (!*p || p == end) {
*out++ = '&';
continue;
}
if (*p == '#' && p[1] == 'x') {
const unsigned char *q = p;
p += 2;
while (isxdigit(*p)) ++p;
int e = convert_html_entity(q + 2, p, 16);
if (e >= 0) {
out = ucs4_to_utf8_char(out, e);
if (*p == ';') ++p;
} else {
*out++ = '&';
p = q;
}
} else if (*p == '#') {
const unsigned char *q = p;
++p;
while (isdigit(*p)) ++p;
int e = convert_html_entity(q + 1, p, 10);
if (e >= 0) {
out = ucs4_to_utf8_char(out, e);
if (*p == ';') ++p;
} else {
*out++ = '&';
p = q;
}
} else if (isalpha(*p)) {
// entity expansion is not performed...
const unsigned char *q = p;
while (isalnum(*p)) ++p;
int e = find_html_entity(q, p);
if (e >= 0) {
out = ucs4_to_utf8_char(out, e);
if (*p == ';') ++p;
} else {
*out++ = '&';
p = q;
}
} else {
*out++ = '&';
continue;
}
}
*out = 0;
}
