Re: How to fix issue #4023 (on Windows, 'svn rm missingItem' deletes the unversioned 'MissingItem' from disk)?

On Tue, Jan 24, 2012 at 3:03 PM, Daniel Shahaf <danielsh_at_elego.de> wrote:
> Hyrum K Wright wrote on Tue, Jan 24, 2012 at 13:57:28 +0000:
>> On Tue, Jan 24, 2012 at 1:54 PM, Daniel Shahaf <danielsh_at_elego.de> wrote:
>> > Philip Martin wrote on Tue, Jan 24, 2012 at 13:45:49 +0000:
>> >> Johan Corveleyn <jcorvel_at_gmail.com> writes:
>> >>
>> >> > On Mon, Jan 23, 2012 at 4:28 PM, Philip Martin
>> >> > <philip.martin_at_wandisco.com> wrote:
>> >> >> Is that sufficient?  Given three files "foo", "FOO" and "Foo", one in
>> >> >> wc.db, one on disk and one on the command line, is that the same file?
>> >> >> Add another "fOO" to the database.  It's now ambiguous?
>> >> >
>> >> > No, it's not:
>> >>
>> >> Is that answering the first question "same file?" or the second question
>> >> "ambiguous?" or both?
>> >>
>> >> >
>> >> > - Both fOO and foo exist in wc.db.
>> >> > - FOO is on disk.
>> >> > - I invoke 'svn <subcommand> Foo':
>> >> >   1. Is there a case-exact match in wc.db? No
>> >> >   2. Ok, then apply truepath-ing: so the user is meaning FOO.
>> >> >
>> >> > But the user can still do useful things to fOO and foo, if he gives
>> >> > the exact correct casing.
>> >>
>> >> What about the first bit.
>> >>
>> >>   Given three files "foo", "FOO" and "Foo", one in
>> >>   wc.db, one on disk and one on the command line, is that the same file?
>> >>
>> >> My understanding is that a Windows user expects those to be the same
>> >> file.  So "svn st FOO", "svn st foo" and "svn st Foo" all refer to the
>> >> file in wc.db and the file on disk even though the cases don't match.
>> >
>> > But normal windows programs don't have the concept of "one on disk and
>> > one in wc.db". And I imagine we want to have some syntax to refer to the
>> > latter.
>>
>> @WORKING ?
>
> I'm not sure this addresses the question of 'what object is identified
> by the the stem (ie, path sans peg)'.  Perhaps wc.db:/FOO? (using the
> pseudo-drive syntax, so this isn't currently in use for anythin)

I'm sorry, but I don't understand the point of this discussion
anymore. There is already a solution for this problem: it's the fix
for issue #3865. This implements the following for the command line
client:

   If there is no case-exact match on disk, but there is a case-exact match
   in wc.db, assume the user refers to the latter. Otherwise case-normalize
   like any other Windows app.

IMHO this is the most natural and simple way to do this. I don't see
the need for more complicated schemes.

-- 
Johan