[jcifs] Character Set discussions

[Michael B. Allen]

On Fri, 7 Feb 2003 15:38:57 -0600
"Christopher R. Hertel" <crh at ubiqx.mn.org> wrote:

> On Fri, Feb 07, 2003 at 03:48:00PM -0500, Michael B. Allen wrote:
> :
> > Didn't we(I) conclude previously however that once we start accepting
> > escapes we must also return them? 
> 
> I don't think you do need to return escapes if they are not required by 
> the environment (and that's the sticky point).  For example, if you are 
> given:  smb://f%6Fo/  there's no reason not to return that as: smb://foo/
> 
> > And if we return them then do we escape characters as we drill down
> > doing list() operations? We just can't do that.
> 
> You escape characters for presentation only.  There's no reason to keep
> the URLs in escaped format once they are parsed and stored in an internal
> (parsed) format.

Well the Java URL class (and I believe the URI class) does return the
escapes. It returns what it was given which is a good policy.

However now the problem is if you are given a URL (or URI or IRI) that
contains escapes and then *derive* URLs from it (using list()) you must
ether escape the new part. You cannot partially escape a URL.

[Side note: I'm just talking about the Unicode characters at this
point. Escaping the 7 special characters is separable (I think)]

Being that SMB is inherently Unicode aware you cannot unconditionally
escape Unicode characters. I think we've all accept that.

There is one possibility I have not fully explored. We may conclude that
any derived URL does not escape Unicode characters.

Incedentally here's another problem I just realized. If we accept
both unescaped URLs and URLs that have Unicode characters that have
been converted to UTF-8 and escaped how will we know after unescaping
them that they are really a sequence of UTF-8 bytes encoding a Uniocde
character? Ans: You don't. The only way to know is if you know the URL
will always escape such sequences. It cannot be "optional".

Ugh.

> > Does the URI class remove and discard the escapes? I don't
> > have Java 1.4 on my machines. Is there anyway to get the original escaped
> > URL from you uri object above back? Like from toString()?
> 
> There should be.  Like a computer language, the point here is to translate 
> between something human-readable and computer-usable.  In this case, we 
> also need to go back again because new URL strings can be generated based 
> on combinations (absolute + relative) of other strings.  The new URL 
> strings are then presented back to the user in user-readable form.
> 
> That's what existing browsers, including those that handle SMB URLs, 
> already do.  We're just talking about expanding it to include Unicode.

I seriously doubt these browsers you mention handle Unicode properly. I
doubt they put much thought into it at all really. It would be
irresponsible for us to check however.

Mike

-- 
A  program should be written to model the concepts of the task it
performs rather than the physical world or a process because this
maximizes  the  potential  for it to be applied to tasks that are
conceptually  similar and, more important, to tasks that have not
yet been conceived.