On 4/23/07, Mike Moran <[EMAIL PROTECTED]> wrote:
On 23 Apr 2007, at 00:33, John D. Mitchell wrote:
> Sorry for the slow response, too busy to keep up. :-(
>>
[ ... ]
>> Note that we enforce all the URI spec examples with a set of unit
>> tests.
>> Looking at how tricky those examples are, we feel very confident
>> about the
>> quality of Reference's output :)
>
> Sorry, IMO that's still incorrect...
>
> If 'dir' is a directory then the behavior is as I noted. If 'dir' is
> actually a file *then* the behavior that you mention would be correct.
>
> This behavior is obvious (Principle of Least Surprise) if one mimics
> the behavior by actually trying these out in a shell and move around a
> filesystem. :-)
Just have to butt in here to support Jerome: the Reference is a
representation of a URL and URLs can know nothing about whether a
path represents a directory or a file. It is up to the server to
enforce any semantic breakage that occurs. This *must* be the case,
otherwise URL manipulations would have to round-trip to the server to
discover 'dirness' or 'fileness'.
I'll continue to disagree...
That confusion over the file- or dir-ness is *precisely* why the "add
trailing slash" hack is done everywhere. Jerome's argument that
disagreement is moot because of the "add trailing slash" hack is
plenty of proof. :-)
If directories always have the trailing slash then there's no
(non-intentional) ambiguity. If directories don't always have the
trailing slash then it's always ambiguous and we end up with the
POLS-violating behavior that started this disagreement.
Thanks,
John
