URL Encode

How to encode a URL

1. Paste the part that needs encoding: a search term, a query value, a path segment, or a complete parameter list.
2. Press Encode URL. The tool applies JavaScript’s encodeURIComponent rule set, which is the same one used by most URL builders and HTTP clients.
3. Copy the result into your link, form action, redirect parameter or API request. The output is plain ASCII, so it survives copy-paste through chat, email and ticketing systems.
4. Need the reverse? Use URL Decode to turn escapes such as %20 back into readable text.

What percent-encoding actually does

Percent-encoding (RFC 3986) escapes bytes, not characters: each byte outside the unreserved set becomes % plus 2 uppercase hex digits, so every escaped byte costs exactly 3 characters. The unreserved set has 66 characters — 26 uppercase + 26 lowercase + 10 digits + - . _ ~ — and 18 characters are reserved (7 generic delimiters plus 11 sub-delimiters), which is why a raw &, =, ? or # changes a URL’s structure.

Non-ASCII text is converted to UTF-8 first, then each byte is escaped. 你 is U+4F60 = E4 BD A0 (3 bytes), so it becomes %E4%BD%A0; 你好 becomes 18 characters. Hello, 世界! is 10 characters and becomes 30, and a b&c=中 grows from 7 to 21. UTF-8 uses 1 to 4 bytes per code point, which is why CJK text and emoji produce so many escapes.

Measured examples (9 November 2026)

This encoder produced five rows on 9 November 2026 using the same encodeURIComponent rules the tool applies, measured in characters: Hello, 世界! (10 into 30 out), 你好 (2 into 18), a b&c=中 (7 into 21), 100% (4 into 6) and hello+world today (17 into 21). The growth comes from UTF-8 byte widths: 10 characters can be 14 bytes once Chinese text is included, and every escaped byte costs 3 characters.

Reserved characters and their escapes

The characters below are the ones that most often break a URL when they are left raw inside a value. The table shows the escape this encoder produces for each of them.

One nuance is worth knowing. encodeURIComponent also leaves ! ' ( ) * unescaped. Those characters are allowed in URLs but can be special in other contexts, so if an API demands a fully conservative escape, replace them after encoding.

Spaces: %20 or +?

Both appear in the wild, and they cost different lengths. In a URL path or query string a space is written %20 — 3 characters for 1 byte. In application/x-www-form-urlencoded data, the body format used by classic HTML forms, a space is a single + (1 character), and a literal plus must be written %2B (3 characters).

This tool always emits %20, the RFC 3986 escape for a URL component. If an API expects a form-encoded body, encode with this tool and then replace %20 with +; do not do that for path segments or for query strings parsed as RFC 3986, or the plus arrives as a literal plus character. The other 2 settings that differ between libraries are the slash: encodeURIComponent escapes it as %2F, while Python’s default quote() keeps it raw.

Encode components, not the whole link

Because encodeURIComponent escapes all 18 reserved characters, pasting a complete https://… address encodes its own separators and produces a string that is no longer a working link. That is sometimes exactly right: when a URL is a value inside another URL, for example a redirect parameter, the inner address must be fully encoded.

When you need a link that works, encode only the values. Keep the scheme, host, the path separators and the parameter names intact, and run this tool on the search term or the value side of each =. A full URL that starts https:// and ends with a path is typically 30–60 characters; encoding only the query value keeps the other 2 structural parts readable.

URL encoding in code

Languages agree on the algorithm and disagree on the defaults. JavaScript's encodeURIComponent(value) escapes all 18 reserved characters and leaves 71 characters unescaped; encodeURI is the whole-URL variant and keeps the separators. Python 3's urllib.parse.quote(value, safe='') matches encodeURIComponent when the safe set is empty, while the default safe='/' keeps 1 character, the slash, unescaped. PHP's rawurlencode($value) uses %20 for a space (1 byte → 3 characters), whereas urlencode() uses + (1 character) and is meant for form bodies.

Java's URLEncoder.encode(value, StandardCharsets.UTF_8) also writes a space as +, so replace it with %20 for a path; PowerShell's [Uri]::EscapeDataString($value) writes %20, and EscapeUriString keeps separators for whole URLs. In a shell, jq -sRr @uri percent-encodes stdin and is handy in pipelines. The 1 setting all of them share: non-ASCII text is converted to UTF-8 bytes before escaping, so 你好 is always 6 bytes and 18 characters, never 2.

Frequently asked questions