URL Encoder / Decoder

Pro tips most people learn the hard way

• Component mode for parameter values; Full URI mode for whole URLs. Component (encodeURIComponent) escapes everything including :/?#&=. Full URI (encodeURI) preserves URL syntax. If you encode a complete URL with Component, your https:// becomes https%3A%2F%2F and nothing works. If unsure, Component is the safer default for a single value.
• Spaces encode two different ways and both are correct. Form posts use + for space; modern URL encoding uses %20. URLSearchParams in browsers emits +. The bug is mixing them — always decode with the same rule you encoded with, or your search for "hello world" comes back as "hello+world".
• Accidental double-encoding shows up as %2520 instead of %20. If you see %25 in production URLs, someone encoded an already-encoded string. Run the value through the decoder once — if it still has % sequences, decode again. Each decode strips one layer.
• Modern URL encoding is UTF-8 first. The é in cafés becomes %C3%A9 (two bytes), not %E9. Legacy Latin-1 systems still output the old one-byte form and that's the source of half the "weird character" bugs you'll hit when integrating older APIs.
• Use the query parser when a webhook URL looks chaotic. Pasting a 500-character redirect URL into the decoder and getting a clean table of keys and values is faster than reading it left to right, especially when nested redirects encode their own query strings inside another one.