URL encoder / decoder

https%3A%2F%2Fexample.com%2Fsearch%3Fq%3Dhello%20world%26lang%3Dru

What this URL encoder does

The tool percent-encodes text so it can be placed safely in a URL, and decodes %XX sequences back to characters. It uses the browser’s own encodeURIComponent / encodeURI and their decode counterparts, so the result is byte-for-byte what your JavaScript would produce — no custom tables. Two modes cover the two real-world jobs: encoding one value (component) and tidying a whole URL (full URI). Output updates as you type; nothing leaves your browser.

How to use it

1. Choose Encode or Decode.
2. Pick a Mode: component (escape all) for a query parameter, path segment or anything that will be pasted into a URL; full URI (keep / : ?) when the input already is a URL and you only want to fix spaces and non-ASCII.
3. Paste the text. The default sample is a URL with a space and a & in a value — try switching modes to see the difference.
4. Copy the output. In Decode mode an invalid sequence shows a red error instead of partial output.

Which characters each function encodes

Character class	encodeURIComponent (component)	encodeURI (full URI)	escape() (legacy)
A–Z a–z 0–9	kept	kept	kept
- _ . ! ~ * ' ( )	kept	kept	- _ . * kept, rest encoded
; , / ? : @ & = + $ #	encoded	kept	+ / @ kept, rest encoded
space	%20	%20	%20
" < > \ ^ [ ] { } | and the backtick	encoded	encoded	encoded
ü (non-ASCII)	%C3%BC (UTF-8)	%C3%BC (UTF-8)	%FC (Latin-1)
€ or emoji	UTF-8 bytes	UTF-8 bytes	%u20AC — not valid in URLs

escape() is deprecated and only listed because you still meet its %uXXXX output in old code; nothing modern will decode it. Note also that encodeURI does not encode #, so a # inside a value will be read as the fragment separator — use component mode for values.

Worked examples

Input: https://example.com/search?q=hello world&lang=ru

Mode	Output
component	https%3A%2F%2Fexample.com%2Fsearch%3Fq%3Dhello%20world%26lang%3Dru
full URI	https://example.com/search?q=hello%20world&lang=ru

The second is a working link. The first is what you want if the whole URL is itself a parameter, for example a redirect_uri or a ?url= value — then the inner ? and & must not be mistaken for the outer query’s separators.

A value that contains a separator, price=1&tax=0.2, becomes price%3D1%26tax%3D0.2 in component mode, so a server reading ?filter=... receives the entire string in one parameter instead of three. In JavaScript the URLSearchParams API does this for you, and additionally writes spaces as +:

const params = new URLSearchParams({ q: 'hello world', lang: 'ru' });
params.toString(); // "q=hello+world&lang=ru"

Decoding gotchas

• + is not a space to decodeURIComponent; it stays +. Only form-encoded query strings use + for space. Replace + with %20 first if the data came from an HTML form, or parse it with URLSearchParams, which does it for you.
• Full-URI decode keeps reserved sequences. decodeURI deliberately leaves %2F, %3F, %23 and the other reserved characters encoded, because turning them back into /, ?, # would change the URL’s structure. Use component mode when you want everything decoded.
• Malformed input throws. %E0%A4%A (truncated), %zz (not hex) or a lone % at the end all produce the “Invalid percent-encoding” error rather than a best-effort string.
• Double encoding shows up as %25 in the input. If decoding once still leaves %20 sequences behind, something upstream encoded the value twice.

Related tools and snippets

• Base64 encoder / decoder — standard Base64 uses + and /, which need encoding before they go in a query string.
• HTML entity encoder / decoder — a URL placed in an href attribute additionally needs & written as &.
• JWT decoder — tokens use Base64URL precisely to avoid percent-encoding.
• Regex tester — build the pattern that extracts a parameter from a URL.
• QR code generator — encode a finished URL into a scannable code.