emded revision control in Python application?

[duncan smith]

On 23/06/12 06:45, rusi wrote:
> On Jun 22, 8:58 pm, duncan smith<buzz... at urubu.freeserve.co.uk>
> wrote:
>> Hello,
>>         I have an application that would benefit from collaborative
>> working. Over time users construct a "data environment" which is a
>> number of files in JSON format contained in a few directories (in the
>> future I'll probably place these in a zip so the environment is
>> contained within a single file). At the moment there is one individual
>> constructing the data environment, and me occasionally applying
>> corrections after being e-mailed the files. But in the future there
>> might be several individuals in various locations.
>>
>> As a minimum requirement I need to embed some sort of version control,
>> so that changes committed by one individual will be seen in the local
>> environments of the others. Some of the work involves editing graphs
>> which have restrictions on their structure. In this case it would be
>> useful for edits to be committed / seen in real time. The users will not
>> be particularly technical, so the version control will have to happen
>> relatively quietly in the background.
>>
>> My immediate thoughts are to (somehow) embed Mercurial or Subversion. It
>> would certainly be useful to be able to revert to a previous version of
>> the data environment if an individual does something silly. But I'm not
>> actually convinced that this is the whole solution for collaborative
>> working. Any advice regarding the embedding of a version control system
>> or alternative approaches would be appreciated. I haven't tried anything
>> like this before. The desktop application is written in Python (2.6)
>> with a wxPython (2.8) GUI. Given the nature of the application / data
>> the machines involved might be locally networked but without web access
>> (if this makes a difference). TIA.
>>
>> Duncan
>
> If you are looking at mercurial and subversion you may want to look at
> git also.
>
>  From http://en.wikipedia.org/wiki/Git_%28software%29#Implementation
> (quoting Linus Torvalds)
> ---------------------------
> In many ways you can just see git as a filesystem — it's content-
> addressable, and it has a notion of versioning, but I really really
> designed it coming at the problem from the viewpoint of a filesystem
> person (hey, kernels is what I do), and I actually have absolutely
> zero interest in creating a traditional SCM system.
>
> More details https://git.wiki.kernel.org/index.php/Git#Design
> -------------------------
> Of course its good to say upfront that git is mostly C+shell ie its
> not python
> There is gitpython http://packages.python.org/GitPython/0.1/tutorial.html
> but I know nothing about it

Thanks. I'm trying to figure out whether I'm better of with a version 
control system, a virtual filesystem (e.g. 
http://code.google.com/p/pyfilesystem/), remote procedure calls or some 
combination of these.

What I really need is a flexible framework that I can experiment with, 
as it's not clear what the best strategy for collaborative working might 
be. e.g. It might be best to restrict working on certain elements of the 
data environment to a single individual. Cheers.

Duncan