Text to Binary, Morse & More

about this tool

Bytes, not characters

Binary, hex, octal and decimal all work on UTF-8 bytes rather than code points. That is what every other tool means by "the binary of this text", and it is what makes the round trip exact.

The difference shows up the moment the text is not plain English. The letter é is one character but two bytes, so it produces two units, and an emoji produces four. A tool that encodes code points instead would give a different answer and would not decode anywhere else.

Decoding is forgiving about how the input is written: extra spaces, commas between units and a 0x prefix are all accepted. It is strict about the result, though. Bytes that are not valid UTF-8 are reported rather than returned as replacement characters.

Reversing without breaking things

Reversing a string by character index tears apart anything built from more than one code unit. An emoji comes out as two pieces of a broken pair, and a family emoji, which is several emoji joined together, scatters completely.

This reverses by grapheme, meaning by character as a reader sees it, so emoji and accented letters survive intact.

Morse and the NATO alphabet

Morse uses the ITU table, with a space between letters and a slash between words. Case does not exist in Morse, so decoding returns lower case.

The NATO alphabet is for reading something aloud over a bad line: a serial number, a booking reference, a postcode. Each letter and digit has a word chosen to be hard to confuse with any other.

ROT13 and Caesar

Neither is encryption. ROT13 rotates letters by thirteen, which is exactly half the alphabet, so applying it twice returns the original text. It exists to hide a spoiler or a punchline from a casual glance, nothing more. A Caesar shift is the same idea with a number you choose, and is broken by trying all twenty-five.