I don't know - I knew that my input data wouldn't have to deal with such cases so I didn't have to deal with / check for it. This was part of 'etc' catchall :)
I guess an alternative is _always_ using e notation but don't bother moving the decimal, i.e effectively the same as the encoding but in base 10. That would be both simple and have small maximum string length.
Users probably wouldn't like it thought :P damn users.
