*
* The default set of patterns recognized are ftp://path-of-url,
* http://path-of-url, https://path-of-url but can be expanded upon.
*
* In addition, the following patterns are also recognized.
*
* [url path-of-url]descriptive text[/url] and
* [url=path-of-url]descriptive text[/url].
*
* The [url] allows any path to be defined as link.
*/
public class URLFilter{
private ArrayList schemes = new ArrayList();
// define a preset default set of schemes
public URLFilter() {
schemes.add("http://");
schemes.add("https://");
schemes.add("ftp://");
}
public String applyFilter(String string) {
if (string == null || string.length() == 0) {
return string;
}
int length = string.length();
StringBuffer filtered = new StringBuffer((int) (length * 1.5));
ArrayList urlBlocks = new ArrayList(5);
// search for url's such as [url=..]text[/url] or [url ..]text[/url]
int start = string.indexOf("[url");
while (start != -1 && (start + 5 < length)) {
// check to verify we're not in another block
if (withinAnotherBlock(urlBlocks, start)) {
start = string.indexOf("[url", start + 5);
continue;
}
int end = string.indexOf("[/url]", start + 5);
if (end == -1 || end >= length) {
// went past end of string, skip
break;
}
String u = string.substring(start, end + 6);
int startTagClose = u.indexOf(']');
String url;
String description;
if (startTagClose > 5) {
url = u.substring(5, startTagClose);
description = u.substring(startTagClose + 1, u.length() - 6);
// Check the user entered URL for a "javascript:" or "file:" link. Only
// append the user entered link if it doesn't contain 'javascript:' and 'file:'
String lcURL = url.toLowerCase();
if (lcURL.indexOf("javascript:") == -1 && lcURL.indexOf("file:") == -1) {
URLBlock block = new URLBlock(start, end + 5, url, description);
urlBlocks.add(block);
}
}
else {
url = description = u.substring(startTagClose + 1, u.length() - 6);
// Check the user entered URL for a "javascript:" or "file:" link. Only
// append the user entered link if it doesn't contain 'javascript:' and 'file:'
String lcURL = url.toLowerCase();
if (lcURL.indexOf("javascript:") == -1 && lcURL.indexOf("file:") == -1) {
URLBlock block = new URLBlock(start, end + 5, url);
urlBlocks.add(block);
}
}
start = string.indexOf("[url", end + 6);
}
// now handle all the other urls
Iterator iter = schemes.iterator();
while (iter.hasNext()) {
String scheme = (String) iter.next();
start = string.indexOf(scheme, 0);
while (start != -1) {
int end = start;
// check context, don't handle patterns preceded by any of '"<=
if (start > 0) {
char c = string.charAt(start - 1);
if (c == '\'' || c == '"' || c == '<' || c == '=') {
start = string.indexOf(scheme, start + scheme.length());
continue;
}
}
// check to verify we're not in another block
if (withinAnotherBlock(urlBlocks, start)) {
start = string.indexOf(scheme, start + scheme.length());
continue;
}
// find the end of the url
int cur = start + scheme.length();
while (end == start && cur < length) {
char c = string.charAt(cur);
switch (c) {
case ' ':
end = cur;
break;
case '\t':
end = cur;
break;
case '\'':
end = cur;
break;
case '\"':
end = cur;
break;
case '<':
end = cur;
break;
case '[':
end = cur;
break;
case '\n':
end = cur;
break;
case '\r':
end = cur;
break;
default:
// acceptable character
}
cur++;
}
// if this is true it means the url goes to the end of the string
if (end == start) {
end = length - 1;
}
URLBlock block = new URLBlock(start, end-1, string.substring(start, end));
urlBlocks.add(block);
start = string.indexOf(scheme, end);
}
}
// sort the blocks so that they are in start index order
sortBlocks(urlBlocks);
// now, markup the urls and pass along the filter chain the rest
Iterator blocks = urlBlocks.iterator();
int last = 0;
while (blocks.hasNext()) {
URLBlock block = (URLBlock) blocks.next();
if (block.getStart() > 0) {
filtered.append(string.substring(last, block.getStart()));
}
last = block.getEnd() + 1;
filtered.append("");
if (block.getDescription().length() > 0) {
filtered.append(block.getDescription());
}
else {
filtered.append(block.getUrl());
}
filtered.append("");
}
if (last < string.length() - 1) {
filtered.append(string.substring(last));
}
return filtered.toString();
}
/**
* Returns the current supported uri schemes as a comma seperated string.
*
* @return the current supported uri schemes as a comma seperated string.
*/
public String getSchemes() {
StringBuffer buf = new StringBuffer(50);
for (int i = 0; i < schemes.size(); i++) {
buf.append((String) schemes.get(i)).append(",");
}
buf.deleteCharAt(buf.length() - 1);
return buf.toString();
}
/**
* Sets the current supported uri schemes as a comma seperated string.
*
* @param schemes a comma seperated string of uri schemes.
*/
public void setSchemes(String schemes) {
if (schemes == null) {
return;
}
// enpty the current list
this.schemes.clear();
StringTokenizer st = new StringTokenizer(schemes, ",");
while (st.hasMoreElements()) {
this.schemes.add(st.nextElement());
}
}
private void sortBlocks(ArrayList blocks) {
Collections.sort(blocks, new Comparator() {
public int compare(Object object1, Object object2) {
URLBlock b1 = (URLBlock) object1;
URLBlock b2 = (URLBlock) object2;
return (b1.getStart() > b2.getStart()) ? 1 : -1;
}
});
}
private boolean withinAnotherBlock(List blocks, int start) {
for (int i = 0; i < blocks.size(); i++) {
URLBlock block = (URLBlock) blocks.get(i);
if (start >= block.getStart() && start < block.getEnd()) {
return true;
}
}
return false;
}
class URLBlock {
int start = 0;
int end = 0;
String description = "";
String url = "";
public String getDescription() {
return description;
}
public void setDescription(String description) {
this.description = description;
}
public String getUrl() {
return url;
}
public void setUrl(String url) {
this.url = url;
}
URLBlock(int start, int end, String url) {
this.start = start;
this.end = end;
this.url = url;
}
URLBlock(int start, int end, String url, String description) {
this.start = start;
this.end = end;
this.description = description;
this.url = url;
}
public int getStart() {
return start;
}
public void setStart(int start) {
this.start = start;
}
public int getEnd() {
return end;
}
public void setEnd(int end) {
this.end = end;
}
}
}