0 && !isspace(html2[j-1])) {
html2[j++]=' ';
}
visible = 0;
} else {
closing_td_tag = 0;
visible = 1;
}
k = i + 1;
if ((k < len) && (!( (html[k] >= 65 && html[k] <= 90) ||
(html[k] >= 97 && html[k] <= 122) ||
(html[k] == 47) ||
(html[k] == 33) ))) {
/* Not a HTML tag. HTML tags start with a letter, forwardslash or exclamation mark */
visible = 1;
html2[j++]=html[i];
i = k;
const char *w = &(html[k]);
while (j < len && (size_t)(w - html) < len && *w != '<') {
html2[j++]=*w;
w++;
i++;
}
continue;
} else if (html[k]) {
/* find the end of the tag */
while (k < len && html[k] != '<' && html[k] != '>') {k++;}
/* if we've got a tag with a uri, save the address to print later. */
char *url_tag = " ";
int tag_offset = 0, x = 0, y = 0;
for (y = 0; y < num_uri; y++) {
x = strlen(uritag[y].open_tag);
if (strncasecmp(html+i,uritag[y].open_tag,x)==0 && (i+x < len && isspace(html[i+x]))) {
url_tag = uritag[y].uri_tag;
tag_offset = i + x + 1;
break;
}
}
/* tag with uri found */
if (tag_offset > 0) {
size_t url_start; /* start of url tag inclusive [ */
size_t url_tag_len = strlen(url_tag);
char delim = ' ';
/* find start of uri */
for (url_start = tag_offset; url_start <= k; url_start++) {
if (strncasecmp(html + url_start, url_tag, url_tag_len) == 0) {
url_start += url_tag_len;
while (html[url_start] && isspace(html[url_start])) {url_start++;} /* remove spaces before = */
if (html[url_start] == '=') {
url_start++;
while (html[url_start] && isspace(html[url_start])) {url_start++;} /* remove spaces after = */
if (html[url_start] == '"') {
delim = '"';
url_start++;
} else if (html[url_start] == '\'') {
delim = '\'';
url_start++;
} else {
delim = '>';
}
break;
} else {
/* Start of uri tag found but no '=' after the tag.
* Skip the whole tag.
*/
break;
}
} else if ((url_start - tag_offset) >= 50) {
/* The length of the html tag is over 50 characters long without
* finding the start of the url/uri. Skip the whole tag.
*/
break;
}
}
/* find end of uri */
if (delim != ' ') {
if (url_start < len &&
(strncasecmp(html + url_start, "http:", 5) == 0 ||
strncasecmp(html + url_start, "https:", 6) == 0 ||
strncasecmp(html + url_start, "ftp:", 4) == 0)) {
html2[j++]=' ';
const char *w = &(html[url_start]);
/* html2 is a buffer of len + 1, where the +1 is for NULL
* termination. This means we only want to loop to len
* since we will replace html2[j] right after the loop.
*/
while (j < len && (size_t)(w - html) < len && *w != delim) {
html2[j++]=*w;
w++;
}
html2[j++]=' ';
}
}
} else if (strncasecmp(html + i, "", 3) == 0
|| strncasecmp(html + i, "
", 9) == 0
|| strncasecmp(html + i, "", 8) == 0) {
if (j > 0 && html2[j-1] != '\n' && html2[j-1] != '\r') {
html2[j++] = '\n';
}
} else if (strncasecmp(html + i, "= 32)
html2[j++] = x;
i += 4;
} else if (html[i+6]
&& html[i+6] == ';'
&& isdigit(html[i+2])
&& isdigit(html[i+3])
&& isdigit(html[i+4])
&& isdigit(html[i+5])) {
n[0] = html[i+2];
n[1] = html[i+3];
n[2] = html[i+4];
n[3] = html[i+5];
n[4] = 0;
x = atoi(n);
if (x <= 255 && x >= 32)
html2[j++] = x;
i += 6;
} else {
const char *w = &(html[i]);
while (*w != ';' && *w != ' ' && *w != '\t' && *w != '\0') {i++;w++;}
}
visible = 0;
continue;
} else if (html[i] == '&') {
int x = 0, y = 0;
for (y = 0; y < num_chars; y++) {
x = strlen(charset[y].entity);
if (strncasecmp(html+i,charset[y].entity,x)==0) {
if (charset[y].id <= 255)
html2[j++] = charset[y].id;
i += x-1;
visible = 0;
continue;
}
}
}
if (j < len && visible)
html2[j++] = html[i];
if (j >= len)
i = j = len;
}
html2[j] = '\0';
return (char *)html2;
}