How to remove \u200B (Zero Length Whitespace Unicode Character) from String in Java?

Question

My application is using Spring Integration for email polling from Outlook mailbox.

As, it is receiving the String (email body)from an external system (Outlook), So I have no control over it.

For Example,

String emailBodyStr= "rejected by sundar14-\u200B.";

Now I am trying to remove the unicode character \u200B from this String.

What I tried already.

Try#1:

emailBodyStr = emailBodyStr.replaceAll("\u200B", "");

Try#2:

`emailBodyStr = emailBodyStr.replaceAll("\u200B", "").trim();`

Try#3 (using Apache Commons):

StringEscapeUtils.unescapeJava(emailBodyStr);

Try#4:

StringEscapeUtils.unescapeJava(emailBodyStr).trim();

Nothing worked till now.

When I tried to print this String using below code.

logger.info("Comment BEFORE:{}",emailBodyStr);
logger.info("Comment AFTER :{}",emailBodyStr);

In Eclipse console, it is NOT printing unicode char,

Comment BEFORE:rejected by sundar14-.

But the same code prints the unicode char in Linux console as below.

Comment BEFORE:rejected by sundar14-\u200B.

I read some examples where str.replace() is recommended, but please note that examples uses javascript, PHP and not Java.

Before trying to replace it, I see an unprintable character in the output (shows up as "?"), and the length is 23. After replacing it, the unprintable character is gone and the length is 22. — resueman, Mar 22 '17 at 19:04
Where are you seeing "?", in your IDE console? Which IDE you are using? When I used str.replaceAll(), I was also getting count 23 (before) and 22 (after), but When I stored this string (after str.replaceAll()) in database - I can see '\u200B' in DB. — Sundararaj Govindasamy, Mar 22 '17 at 19:11
I see the question mark when running in Windows command prompt. Also, comparing the `String` with another one created without the \u200B shows that they're equal. — resueman, Mar 22 '17 at 19:16
Big props figuring out your own answer! Fyi though, I believe the reason you originally had this problem is that you used `replaceAll()` instead of `replace()`. `replaceAll()` treats its first argument as a regex string, and since your input of `\u200B` has a \, it parses incorrectly and doesn't replace as it can't find the search string. — rococo, Jun 28 '21 at 18:41

score 24 · Accepted Answer · edited May 23 '17 at 10:31

Finally, I am able to remove 'Zero Width Space' character by using 'Unicode Regex'.

String plainEmailBody = new String();
plainEmailBody = emailBodyStr.replaceAll("[\\p{Cf}]", "");

Reference to find the category of Unicode characters.

Character class from Java lists all of these unicode categories.

Website: http://www.regular-expressions.info/ => Unicode Regular Expressions

Note 1: As I received this string from Outlook Email Body - none of the approaches listed in my question was working.

My application is receiving a String from an external system (Outlook), So I have no control over it.

Note 2: This SO answer helped me to know about Unicode Regular Expressions .

1 Answers1