UNIX-to-UNIX Encode

(U-U-encode)

Since early UNIX E-mail programs could only transmit data as 7-bit ASCII characters the transmission of images and other binary data was a problem. The solution to that problem was to encode the binary data using the ASCII character set. While this increased the size of the transmitted message (the ASCII encoding of the binary bytes result in a message about 42% larger) it solved the problem of not being able to transmit the data at all. Thus was born UUENCODE and UUDECODE.

In a UUENCODED message the first line starts with "begin". If there is a number following "begin" it represents the UNIX file permissions. The file name follows this. The encoded data starts the next line. Encoded lines are generally all the same length and start with an encoded length character followed by the encoded data. The input data stream is taken in blocks of 24-bits each (three 8-bit bytes). The block is then broken into four six-bit characters which are converted to ASCII strings using an encoding character added to the data. This results in four ASCII characters for each three bytes of data. The end of the encoding is signaled by a line with "`" alone on it followed by a line with the word "end" by itself.

While a standard encoding scheme there are potential problems. If a file is encoded on a system with nine bits/byte and sent to a system with eight bits/byte there could be problems decoding the file. Other problems can occur when the encoding character differs and/or spaces fall at the end of a line.

More Information

• NetBSD Manual Page
• Single UNIX Specification Page