Local clones and phases, mq (Re: [PATCH 1 of 5 phases] phases: exclude secret when doing a local clone)

[Martin Geisler]

Angel Ezquerra <ezquerra at gmail.com> writes:

> On Jan 15, 2012 1:44 PM, "Martin Geisler" <mg at lazybytes.net> wrote:
>>
>> Matt Mackall <mpm at selenic.com> writes:
>>
>> > On Fri, 2012-01-13 at 02:42 +0100, Pierre-Yves David wrote:
>> >> # HG changeset patch
>> >> # User Pierre-Yves David <pierre-yves.david at logilab.fr>
>> >> # Date 1326299187 -3600
>> >> # Node ID 3f56f39a5135f2253bdd8c14afe1e52a08b96ffc
>> >> # Parent  c47d69ce5208d5b5cfd2fb2f0f1d7a2b4795fbf5
>> >> phases: exclude secret when doing a local clone
>> >>
>> >> This is achieved by denying copy clone when any secret changeset exist.
>>
>> [...]
>>
>> > So enforcing phases here means local clones are a) less useful b)
>> > slower and c) heavier than they could be.
>> >
>> > I'm honestly not sure what the right answer is here, and we can
>> > probably revisit this in the future.
>>
>> I've also thought a bit about this problem and I prefer 'hg clone' to
>> copy everything, including secret changesets. Let people use the
>> --pull flag if they want a clone with what you've get over the wire.
>
> What about clones from smb and cifs?
>
> That's a very common way to share a repo over an intranet, at least in
> my experience, particularly on windows. I think in that case it should
> behave as pull (i.e. ignore the secret changesets). Otherwise that
> method of sharing will not be viable anymore if you want to be able to
> use secret phases.

If you give someone direct file system level access to your repository,
then you've also shown them your secret changesets.

I don't think that's so bad: you can still have a local repository on
your computer for your development and there you will create secret
changesets. When you push to your changeset to your repo on the server,
then the secret changesets wont get pushed.

Also, even if you work directly on the server and have your secret
changesets there, then your co-worker will only get them if he uses
clone instead of pull to get your changesets.

It's definitely something we can debate, but I still think hardlinks are
important and so we should try to link as much as possible.

-- 
Martin Geisler

Mercurial links: http://mercurial.ch/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
URL: <http://selenic.com/pipermail/mercurial-devel/attachments/20120115/fd2d19ec/attachment.pgp>