/**
* Encode a Human Readable Resource Identifier to a URI. Leading and trailing spaces are removed
* first.
*
* <p>NOTE: See more recent W3C note: http://www.w3.org/TR/2008/NOTE-leiri-20081103/
*
* @param uriString URI to encode
* @param processSpace whether to process the space character or leave it unchanged
* @return encoded URI, or null if uriString was null
*/
public static String encodeHRRI(String uriString, boolean processSpace) {
if (uriString == null) return null;
// Note that the XML Schema spec says "Spaces are, in principle, allowed in the ·lexical space·
// of anyURI,
// however, their use is highly discouraged (unless they are encoded by %20).".
// We assume that we never want leading or trailing spaces. You can use %20 if you really want
// this.
uriString = uriString.trim();
// We try below to follow the "Human Readable Resource Identifiers" RFC, in draft as of
// 2007-06-06.
// * the control characters #x0 to #x1F and #x7F to #x9F
// * space #x20
// * the delimiters "<" #x3C, ">" #x3E, and """ #x22
// * the unwise characters "{" #x7B, "}" #x7D, "|" #x7C, "\" #x5C, "^" #x5E, and "`" #x60
final StringBuilder sb = new StringBuilder(uriString.length() * 2);
for (int i = 0; i < uriString.length(); i++) {
final char currentChar = uriString.charAt(i);
if (currentChar >= 0
&& (currentChar <= 0x1f
|| (processSpace && currentChar == 0x20)
|| currentChar == 0x22
|| currentChar == 0x3c
|| currentChar == 0x3e
|| currentChar == 0x5c
|| currentChar == 0x5e
|| currentChar == 0x60
|| (currentChar >= 0x7b && currentChar <= 0x7d)
|| (currentChar >= 0x7f && currentChar <= 0x9f))) {
sb.append('%');
sb.append(NumberUtils.toHexString((byte) currentChar).toUpperCase());
} else {
sb.append(currentChar);
}
}
return sb.toString();
}
