path tokens

[Aaron Bentley]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Robert Collins wrote:
> On Thu, 2007-03-15 at 09:57 -0400, Aaron Bentley wrote:
>> "path" is probably not going to work, just because the obvious
>> interpretation is an association of paths (not the contents currently at
>> those paths) with tokens.
> 
> Yes. file is wrong though, because directories are not files (at least,
> not for folk that dont understand how they are implemented down in the
> guts). I'd love a better name.

Perhaps we just want something so vague that you have no idea what it is
until you find out exactly what we mean :-)

>> I think that addressing parallel imports well demands more than
>> addressing the file-id issue.  It would also be able to match up the
>> revisions created by parallel imports.
> 
> One way of doing that is creating new revisions that merge both imports.
> I.e. if you have A and A' which was a parallel import, create A'' which
> has A and A' as parents, and a revision-property to indicate that it
> does this.

Ah, but even identifying A and A' as equivalent will be a pain.  As you
say, not our immediate problem.

>> It's not clear to me that we should use the same primitive to represent
>> both those operations.  The output of a split is two files with no
>> common contents that are both related to the base file.  The output of a
>> copy is two files that have identical contents to the base file.  In the
>> first case, applying a merge from a pre-split tree should apply each
>> change only once.  But in the second case, a merge from a pre-copy tree
>> the changes would be applied twice: once to each file.

> I'd like to take these points to the discussion about what copying
> semantics should be.

Okay, well, at this point I don't know what there is to discuss here.

>> My objection has always been that there are no sane semantics for copy.
>>  Prove me wrong, and I'll be happy.
> 
> Ok, so can I read this email as 'agreed in principle *pending agreement
> on sane copy/combine semantics*' ?

Certainly.

Aaron
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFF+fV50F+nu1YWqI0RAsNWAJ0Ud1ug/TGV0dz5kHb5u8NodtGi9ACfQUZK
BY6rVGAO0l98z+G7YRFwVwo=
=9S5t
-----END PGP SIGNATURE-----