Re: [PATCH v3] remotes-hg: bugfix for fetching non local remotes

From: Felipe Contreras <felipe.contreras@gmail.com>
Date: Jul 25, 2013, 20:40 UTC
Message-ID: <CAMP44s2v+CF7x+S6_47CiPb6RMXu+iy06gqWNjus4vff5J8z3g@mail.gmail.com>
In-Reply-To: <CALWbr2wN6k8JBCwLFC=TjTC_sg7Uh8AEsMOBKfH9aBxDEcV4oQ@mail.gmail.com>

On Thu, Jul 25, 2013 at 2:53 PM, Antoine Pelisse <apelisse@gmail.com> wrote:

Show 7 quoted lines

> On Thu, Jul 25, 2013 at 9:12 PM, Felipe Contreras
> <felipe.contreras@gmail.com> wrote:
>> Besides, I don't see
>> the point of having a '.shared/.hg' directory, and nothing else on
>> that '.shared' folder.
>
> Is it not already true about the ".git/hg/$alias/clone/" directory ?

Yeah, but that directory is kind of useful. Somebody might want to clone that, and it's self-explanatory; "Where is the clone of that Mercurial remote? Oh, there".

Show 47 quoted lines

>> So, here's my patch. If only Junio read them.
>>
>> Subject: [PATCH] remote-hg: add shared repo upgrade
>>
>> 6796d49 (remote-hg: use a shared repository store) introduced a bug by
>> making the shared repository '.git/hg', which is already used before
>> that patch, so clones that happened before that patch, fail after that
>> patch, because there's no shared Mercurial repo.
>>
>> It's trivial to upgrade to the new organization by copying the Mercurial
>> repo from one of the remotes (e.g. 'origin'), so let's do so.
>
> I agree with you that we should consider migration. But there's
> another use-case I think can fail.
> What happens with the following:
>
> git clone hg::/my/hg/repo
> cd repo && git remote add newremote hg::http://some/hg/url
>
> Git clone will create .git/hg/origin and with no hg clone (because
> it's a local repository), and then create marks-file in there.
>
>> Reported-by: Joern Hees <dev@joernhees.de>
>> Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
>> ---
>>  contrib/remote-helpers/git-remote-hg.py | 7 +++++++
>>  1 file changed, 7 insertions(+)
>>
>> diff --git a/contrib/remote-helpers/git-remote-hg.py
>> b/contrib/remote-helpers/git-remote-hg.py
>> index 0194c67..57a8ec4 100755
>> --- a/contrib/remote-helpers/git-remote-hg.py
>> +++ b/contrib/remote-helpers/git-remote-hg.py
>> @@ -396,6 +396,13 @@ def get_repo(url, alias):
>>                  hg.clone(myui, {}, url, shared_path, update=False, pull=True)
>>              except:
>>                  die('Repository error')
>> +        else:
>> +            # check and upgrade old organization
>> +            hg_path = os.path.join(shared_path, '.hg')
>> +            if not os.path.exists(hg_path):
>> +                repos = os.listdir(shared_path)
>> +                local_hg = os.path.join(shared_path, repos[0], 'clone', '.hg')
>> +                shutil.copytree(local_hg, hg_path)
>
> With the use-case I described above, I think shutil.copytree() would
> raise an exception because local_hg doesn't exist.

That's true. Maybe something like:

for x in repos:
  local_hg = os.path.join(shared_path, x, 'clone', '.hg')
  if os.path.exists(local_hg):
    shutil.copytree(local_hg, hg_path)
    break

-- 
Felipe Contreras