Encoding and escaping
for components, URIs and form values
Conversion stays in this browser. Input is not uploaded and URLs are not visited.
Uses encodeURIComponent character rules. Spaces become %20; + becomes %2B. Decoding keeps a literal +.
UTF-8 only, one decode pass. Decoding rejects malformed percent escapes and invalid UTF-8. Both directions reject isolated UTF-16 surrogates. Form decoding is strict, unlike permissive URLSearchParams parsing.
Input: 1,048,576 UTF-16 units; output: 4,194,304 units. Empty input is valid. Pasting may normalize line endings.
Enter text or load an example. Empty input converts to empty output.
Reverse uses the fresh result as input and changes direction after confirmation. It is unavailable when the result exceeds the input limit.
Use this URL encoder to prepare a query value, inspect a copied address or decode a form field. Choose the context before converting: the same ampersand, slash or plus sign can have a different role in each context.
The URL decoder reads UTF-8 percent encoding once. It leaves every result as plain text, so an address, fragment or script-shaped string can be inspected without visiting or executing it.
The URL encoder distinguishes a single component, a complete URI and one form field value. Component mode escapes delimiters; URI mode preserves structure; form encoding uses + for spaces. Choosing deliberately avoids turning an ampersand in a value into a separator.
Malformed % escapes and invalid UTF-8 produce a source line and column, with the error line highlighted. The URL encoder does not silently substitute replacement characters or export a partially decoded value.
The URL encoder starts in component encoding mode with an empty editor. Paste one value, or select URI for whole-address character rules. Form means a single field value: a=1&b=2 is treated as data, not parsed into fields.
Automatic conversion waits 350 ms after an edit. Turn it off to show Convert and use Ctrl/Cmd+Enter for a manual run. Editing input or options immediately disables export until a new result succeeds. Empty input is valid.
Example, Clear and Reverse show a preview before Apply. Reverse uses the result as input and switches direction. Undo restores input, options and output together. Copy and TXT download include the complete current result.
Component encoding follows encodeURIComponent: ASCII letters, digits and -_.!~*'() remain unchanged. Spaces become %20, ampersands %26 and slashes %2F. Percent signs are encoded again as %25.
Full URI encoding also preserves ;/?:@&=+$,#. URI decoding leaves their escaped forms intact, including hex letter case: %2f stays %2f. It does not parse, validate or repair the address.
For one form value, ASCII letters, digits and *-._ remain unchanged. A space becomes + and a plus sign becomes %2B. Decoding %2B produces a literal plus, without converting that result into a space.
Non-ASCII characters become their UTF-8 bytes, including accented letters and emoji. Error positions use one-based lines and UTF-16 columns. Invalid multi-byte sequences are reported at the starting percent sign.
Both can represent a space in the appropriate context. Component and URI encoding use %20. Form encoding uses +, and only form decoding interprets an original + as a space. A literal plus is encoded as %2B in a component or form value.
A malformed %HH escape or invalid UTF-8 sequence fails strict decoding. Check the indicated source line and column. Isolated UTF-16 surrogates also fail under this tool’s source-text policy; no partially decoded result can be exported.
Choose component mode for one query value or path segment. Choose URI mode when structural separators should remain. URI mode cannot infer whether an ampersand is a separator or part of a value, so encode individual values before assembling an address.
It commonly indicates an extra encoding layer: % became %25, leaving 20 after it. One decode turns %2520 into %20. Decode again only when the receiving format requires that second step.
Conversion runs in this browser and does not upload the input. Pasted addresses are handled as text and are not visited. You can copy the current result or save it as a TXT file.
Found a percent-encoding edge case?
Select a context, compare the percent encoding and copy the full result after conversion succeeds.