Convert characters to Unicode escapes in seven notations, or resolve a mixture of them back to text. Percent escapes are decoded as UTF-8 byte runs, correctly.
Waiting for text
Writes each character as an escape in the notation you choose.
The result appears here
Each character written as an escape, ready to paste into source that must stay ASCII.
Unicode Escape Converter runs entirely on your device — nothing is uploaded.
You’re on the Free plan. Compare plans
Seven notations for the same thing, because seven places need it written differently. A backslash-u form in JavaScript, a braced form for anything past the first plane, a backslash-hex form with its significant trailing space in CSS, hexadecimal and decimal entities in HTML, the U-plus form used in prose and bug reports, and percent escapes in a URL.
This is a different question from escaping for a language. The string escape tool answers how do I put this inside a quoted literal — backslashes, quotes, newlines. This answers how do I write this character without typing the character, which is what you need when a file must stay ASCII, when a build tool mangles non-ASCII source, or when a character has to be named precisely in a code review.
Unescaping reads all seven at once, in a single pass, because pasted text frequently carries a mixture. Anything that looks like an escape but is not well-formed is left exactly as written and counted, rather than guessed at — a tool that silently half-decodes a sequence corrupts the string in a way that is invisible until much later.
Percent escapes get particular care. They encode UTF-8 bytes rather than code points, so a run of them has to be decoded together rather than one byte at a time. That is the detail most hand-written converters get wrong, and it is why a percent-escaped tick mark comes back here as one character rather than three broken ones.
The CSS notation ends with a space, which is part of the escape and not decoration. Without it the next character would be read as further hex digits, which is how a CSS escape followed by a letter silently becomes a different character entirely. Nothing is uploaded.
That one makes text safe inside a quoted literal — backslashes, quotes, newlines. This one writes a character as a numeric escape so it can be typed in ASCII. Different questions, which is why they are different tools.
Because the space terminates it. Without it the following character is read as more hex digits, so a CSS escape followed by a letter silently resolves to a different character. The trailing space is part of the escape.
Because they encode UTF-8 bytes rather than code points, and anything outside ASCII is two to four bytes. Decoding them has to treat a run as one sequence, which is what this does and what most hand-rolled decoders do not.
Yes. Unescaping resolves all seven in one pass, which is what pasted text usually needs. Anything malformed is left exactly as written and reported in the count.
The four-digit form can only address the first plane, so a character above it needs two of them as a surrogate pair. The braced form takes the code point directly and is the clearer choice for emoji and anything else above the first plane.
No. An escape is another way of writing the same character, so a parser that understands the notation produces exactly the original string. The point is to make the text typable in ASCII, not to change it.