How does unicode() work?

[Fredrik Lundh]

Robert Latest wrote:

> Here's a test snippet...
> 
> import sys
> for k in sys.stdin:
>     print '%s -> %s' % (k, k.decode('iso-8859-1'))
> 
> ...but it barfs when actually fed with iso8859-1 characters. How is this 
> done right?

it's '%s -> %s' % (byte string, unicode string) that barfs.  try doing

import sys
for k in sys.stdin:
     print '%s -> %s' % (repr(k), k.decode('iso-8859-1'))

instead, to see what's going on.

</F>