URL Batch Encoder/Decoder

Encoding a whole URL is not encoding a parameter

Two different jobs. Encoding an entire URL must preserve the structural characters — the colon, slashes, question mark and ampersands — or the result stops being a URL. Encoding a value that goes inside a parameter must escape those same characters, because there they are data rather than structure. In JavaScript this is the difference between encodeURI and encodeURIComponent, and choosing wrong breaks either the URL or the value.

Double encoding is the classic bulk failure

Running an already-encoded list through an encoder turns %20 into %2520, because the percent sign itself gets escaped. In a batch this happens silently to every row. Checking whether input already contains percent sequences before encoding is worth doing, since the damage is invisible until a link is clicked.

Spaces have two encodings

A space is %20 in a path and may be + in a query string under the form-encoded convention. Decoders differ on whether they translate a literal plus back into a space, which is why a search term containing a plus sign sometimes arrives as a space and sometimes intact.

Non-ASCII must be UTF-8 first

Percent-encoding operates on bytes, so a character is converted to UTF-8 and each byte escaped. An é becomes %C3%A9, two bytes and two sequences. Encoding from a different character set produces sequences that decode into mojibake rather than failing outright.

Reserved characters and consistency

Hyphen, underscore, full stop and tilde are unreserved and never need encoding. Encoders differ on the tilde, and while both forms are valid, inconsistency matters when URLs are compared, deduplicated or signed — two spellings of the same address will not match.