yEnc Decoder

How it is calculated

The decoder first drops any line that starts with =ybegin, =ypart or =yend (they are metadata), then joins the remaining lines, because line breaks are not data. It reads the text one character at a time. A normal character with code c becomes the byte (c − 42) mod 256. The escape character = takes the next character with code c and turns it into (c − 64 − 42) mod 256; this is how bytes that would be troublesome in text (such as NUL, LF, CR and = itself) are carried. A lone = at the very end is ignored. If the resulting bytes are valid printable text, it shows that text; if they contain control characters or invalid UTF-8, it shows the bytes as space-separated hex pairs instead. The tool also reports the byte count. It only skips the header lines: it does not read the filename or part range and does not check the size or CRC.

Common mistakes

• Expecting a downloadable file. The tool is for a short fragment that you read as text or hex; a binary file such as an image comes out as a hex list, not as a file.
• Putting a space or other character before the header word. Only lines that begin exactly with =ybegin, =ypart or =yend are skipped, so " =ybegin" with a leading space is decoded as data and adds meaningless bytes.
• Using text that was copied through a program that changed the characters, for instance a mail client replacing non-ASCII bytes or trimming spaces. A single changed or missing character shifts or corrupts the output.
• Dropping the = escape in front of a character. A lone = consumes the next character, so deleting or adding one shifts the meaning of the character after it.
• Confusing yEnc with Base64. They are different formats; Base64 text will decode into unrelated bytes here, though the tool will not warn you. Also, since the CRC is not checked, a corrupted fragment is decoded without any error message.