EN
Java - normalize URL to canonical form
4 points
In this short article, we would like to show how to normalize URL in Java to common canonical form.
Normalization form:
xxxxxxxxxx1
[INPUT] [OUTPUT]2
3
http://test.com:80 http://test.com4
https://test.com:80 https://test.com:805
http://test.com:443 http://test.com:4436
https://test.com:443 https://test.com7
http://test.com/ http://test.com8
http://test.com? http://test.com9
http://test.com?a=1&&b=2& http://test.com?a=1&b=210
http://test.com# http://test.com11
http://test.com?b=2&a=1 http://test.com?a=1&b=212
http://test.com:80/?# http://test.com13
https://test.com:443/?# https://test.com14
https://test.com/users/john/?p=10 https://test.com/users/john?p=1015
https://test.com/users/john/?p=10# https://test.com/users/john?p=1016
https://test.com/users/john/?p=10#data https://test.com/users/john?p=10#data17
https://test.com/users/john/?#data https://test.com/users/john#data18
https://test.com:8080/users/john/?#data https://test.com:8080/users/john#data19
https://test.com:8080//users//john//?#data https://test.com:8080/users/john#data20
https://test.com:443//?# https://test.com
Program.java file:
xxxxxxxxxx1
package example.utils;2
3
import java.net.URISyntaxException;4
5
public final class UrlUtils {6
7
public static void main(String[] args) throws URISyntaxException {8
9
printNormalized("http://test.com:80"); // http://test.com10
printNormalized("https://test.com:80"); // https://test.com:8011
printNormalized("http://test.com:443"); // http://test.com:44312
printNormalized("https://test.com:443"); // https://test.com13
printNormalized("http://test.com/"); // http://test.com14
printNormalized("http://test.com?"); // http://test.com15
printNormalized("http://test.com?a=1&&b=2&"); // http://test.com?a=1&b=216
printNormalized("http://test.com#"); // http://test.com17
printNormalized("http://test.com?b=2&a=1"); // http://test.com?a=1&b=218
printNormalized("http://test.com:80/?#"); // http://test.com19
printNormalized("https://test.com:443/?#"); // https://test.com20
printNormalized("https://test.com/users/john/?p=10"); // https://test.com/users/john?p=1021
printNormalized("https://test.com/users/john/?p=10#"); // https://test.com/users/john?p=1022
printNormalized("https://test.com/users/john/?p=10#data"); // https://test.com/users/john?p=10#data23
printNormalized("https://test.com/users/john/?#data"); // https://test.com/users/john#data24
printNormalized("https://test.com:8080/users/john/?#data"); // https://test.com:8080/users/john#data25
printNormalized("https://test.com:8080//users//john//?#data"); // https://test.com:8080/users/john#data26
printNormalized("https://test.com:443//?#"); // https://test.com27
}28
29
public static void printNormalized(String url) throws URISyntaxException {30
System.out.println(UrlUtils.normalizeUrl(url));31
}32
}UrlUtils.java file:
xxxxxxxxxx1
package example.utils;2
3
import java.net.URI;4
import java.net.URISyntaxException;5
import java.util.Arrays;6
7
public final class UrlUtils {8
9
private UrlUtils() {10
// nothing here ...11
}12
13
public static String normalizeUrl(String url) throws URISyntaxException {14
URI uri = new URI(url);15
16
String scheme = uri.getScheme();17
18
if (scheme == null) {19
throw new RuntimeException("URL scheme is required.");20
}21
22
String user = uri.getUserInfo();23
String host = uri.getHost();24
25
int port = normalizePort(scheme, uri.getPort());26
String path = normalizePath(uri.getPath());27
String query = normalizeQuery(uri.getQuery());28
String fragment = normalizeFragment(uri.getFragment());29
30
URI result = new URI(scheme, user, host, port, path, query, fragment);31
32
return result.toString();33
}34
35
private static int normalizePort(String scheme, int port) {36
switch (port) {37
case 80:38
if ("http".equals(scheme)) {39
return -1;40
}41
break;42
43
case 443:44
if ("https".equals(scheme)) {45
return -1;46
}47
break;48
}49
return port;50
}51
52
private static String normalizePath(String path) {53
String result = removeDuplicates(path, '/');54
if (result == null || result.isEmpty()) {55
return null;56
}57
int length = result.length();58
char value = result.charAt(length - 1);59
if (value == '/') {60
return result.substring(0, length - 1);61
}62
return result;63
}64
65
private static String normalizeQuery(String query) {66
if (query == null || query.isEmpty()) {67
return null;68
}69
String[] parts = query.split("&");70
if (parts.length > 1) {71
Arrays.sort(parts);72
StringBuilder builder = new StringBuilder();73
for (int i = 0; i < parts.length; ++i) {74
String part = parts[i];75
if (part.isEmpty()) {76
continue;77
}78
int length = builder.length();79
if (length > 0) {80
builder.append("&");81
}82
builder.append(part);83
}84
return builder.toString();85
}86
return query;87
}88
89
private static String normalizeFragment(String fragment) {90
if (fragment == null || fragment.isEmpty()) {91
return null;92
}93
return fragment;94
}95
96
private static String removeDuplicates(String text, char character) {97
if (text == null || text.isEmpty()) {98
return text;99
}100
StringBuilder builder = new StringBuilder();101
int duplicatesCount = 0;102
int textLength = text.length();103
for (int i = 0; i < textLength; ++i) {104
char value = text.charAt(i);105
if (value == character) {106
duplicatesCount += 1;107
if (duplicatesCount > 1) {108
continue;109
}110
} else {111
duplicatesCount = 0;112
}113
builder.append(value);114
}115
return builder.toString();116
}117
}