Re: URIs vs. half-baked URIs (was Filesel drag and drop)

[Derek Simkowiak <dereks realloc net>]

-> AFAIK uri encoding/decding is strictly one-pass process and no unencoded
-> %-s are allowed inside it. OTOH, it's been a while I looked at the spec.

	This is getting offtopic now, but I'm not sure what you mean by
"one-pass process".

	URL encoding can be iteratively applied to any string.  So can
decoding:

Here are the results of several encodings (by hand, I think I got the
hex codes right :)

http://server/foo bar   (NOTE: malformed URL)
http://server/foo%20bar    (' ' encoded into %20)
http://server/foo%2520bar   (NOTE: The '%' was turned to '%25')
http://server/foo%252520bar   (...and again...)
http://server/foo%25252520bar   (...and again, etc.)


And decoding:

http://server/foo%252520bar
http://server/foo%2520bar
http://server/foo%20bar
http://server/foo bar
http://server/foo bar
http://server/foo bar
...etc...

	The same process would apply to any of the reserved characters,
such as a '/' or '?' appearing in a non-URI context.