I got this error message :
java.net.URISyntaxException: Illegal character in query at index 31: http://finance.yahoo.com/q/h?s=^IXIC
A space is encoded to %20 in URLs, and to + in forms submitted data (content type application/x-www-form-urlencoded). You need the former.
Using Guava:
dependencies {
compile 'com.google.guava:guava:28.1-jre'
}
You can use UrlEscapers:
String encodedString = UrlEscapers.urlFragmentEscaper().escape(inputString);
Don't use String.replace, this would only encode the space. Use a library instead.
If you're using RestangularV2
to post to a spring controller in java you can get this exception if you use RestangularV2.one()
instead of RestangularV2.all()
Rather than encoding the URL beforehand you can do the following
String link = "http://example.com";
URL url = null;
URI uri = null;
try {
url = new URL(link);
} catch(MalformedURLException e) {
e.printStackTrace();
}
try{
uri = new URI(url.toString())
} catch(URISyntaxException e {
try {
uri = new URI(url.getProtocol(), url.getUserInfo(), url.getHost(),
url.getPort(), url.getPath(), url.getQuery(),
url.getRef());
} catch(URISyntaxException e1 {
e1.printStackTrace();
}
}
try {
url = uri.toURL()
} catch(MalfomedURLException e) {
e.printStackTrace();
}
String encodedLink = url.toString();
Coudn't imagine nothing better for
http://server.ru:8080/template/get?type=mail&format=html&key=ecm_task_assignment&label=Согласовать с контрагентом&descr=Описание&objectid=2231
that:
public static boolean checkForExternal(String str) {
int length = str.length();
for (int i = 0; i < length; i++) {
if (str.charAt(i) > 0x7F) {
return true;
}
}
return false;
}
private static final Pattern COLON = Pattern.compile("%3A", Pattern.LITERAL);
private static final Pattern SLASH = Pattern.compile("%2F", Pattern.LITERAL);
private static final Pattern QUEST_MARK = Pattern.compile("%3F", Pattern.LITERAL);
private static final Pattern EQUAL = Pattern.compile("%3D", Pattern.LITERAL);
private static final Pattern AMP = Pattern.compile("%26", Pattern.LITERAL);
public static String encodeUrl(String url) {
if (checkForExternal(url)) {
try {
String value = URLEncoder.encode(url, "UTF-8");
value = COLON.matcher(value).replaceAll(":");
value = SLASH.matcher(value).replaceAll("/");
value = QUEST_MARK.matcher(value).replaceAll("?");
value = EQUAL.matcher(value).replaceAll("=");
return AMP.matcher(value).replaceAll("&");
} catch (UnsupportedEncodingException e) {
throw LOGGER.getIllegalStateException(e);
}
} else {
return url;
}
}
A general solution requires parsing the URL into a RFC 2396 compliant URI (note that this is an old version of the URI standard, which java.net.URI uses).
I have written a Java URL parsing library that makes this possible: galimatias. With this library, you can achieve your desired behaviour with this code:
String urlString = //...
URLParsingSettings settings = URLParsingSettings.create()
.withStandard(URLParsingSettings.Standard.RFC_2396);
URL url = URL.parse(settings, urlString);
Note that galimatias is in a very early stage and some features are experimental, but it is already quite solid for this use case.
I had this exception in the case of a test for checking some actual accessed URLs by users.
And the URLs are sometime contains an illegal-character and hang by this error.
So I make a function to encode only the characters in the URL string like this.
String encodeIllegalChar(String uriStr,String enc)
throws URISyntaxException,UnsupportedEncodingException {
String _uriStr = uriStr;
int retryCount = 17;
while(true){
try{
new URI(_uriStr);
break;
}catch(URISyntaxException e){
String reason = e.getReason();
if(reason == null ||
!(
reason.contains("in path") ||
reason.contains("in query") ||
reason.contains("in fragment")
)
){
throw e;
}
if(0 > retryCount--){
throw e;
}
String input = e.getInput();
int idx = e.getIndex();
String illChar = String.valueOf(input.charAt(idx));
_uriStr = input.replace(illChar,URLEncoder.encode(illChar,enc));
}
}
return _uriStr;
}
test:
String q = "\\'|&`^\"<>)(}{][";
String url = "http://test.com/?q=" + q + "#" + q;
String eic = encodeIllegalChar(url,'UTF-8');
System.out.println(String.format(" original:%s",url));
System.out.println(String.format(" encoded:%s",eic));
System.out.println(String.format(" uri-obj:%s",new URI(eic)));
System.out.println(String.format("re-decoded:%s",URLDecoder.decode(eic)));