Re: Gate between git and mediawiki : remote-helpers

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Arnaud Lacurie <arnaud.lacurie@xxxxxxxxx> writes:

> This mail referred to that previous one :
> http://article.gmane.org/gmane.comp.version-control.git/173991
>
> 2011/5/22 Claire Fousse <claire.fousse@xxxxxxxxx>:
>> Hi,
>>
>> I'm a member of the team trying to establish a gate between mediawiki
>> powered wiki and git.
>>
>> We've tried several things which seems to work. However, it is
>> something like git-svn and would require some commands such as git-mw
>> to work. Is it recommended to use remote-helpers instead of that ?

I am not Matthew, but my gut feeling is that it largely depends on what
you are interacting with, and how you envision the result to be used in a
larger MediaWiki ecosystem.

In the case of SVN interoperability, there is an established community
that exchanges their work via:

	svn checkout svn://some.where.xz/project
        svn update
	svn commit

and although people on the git side can already participate with:

	git svn init svn://some.where.xz/project
        git svn rebase
        git svn dcommit

it is understandable that they wish to be able to say:

	git clone svn://some.where.xz/project
        git pull -s rebase
        git push

to make it feel more similar to the native git experience. Now, even if
there may be no "svn checkout/update/commit" equivalents in the workflow
of established MediaWiki users, it may be nice to be able to work with:

	git clone --vcs=mediawiki http://some.where.xz/wiki/
        git pull
        git push

>> There is one problem though : nobody wants to git clone the whole
>> Wikipedia for instance.

Then don't let them in your initial version. I do not see any problem in
that. People can gain experience with smaller projects, like so:

	git clone --vcs=mediawiki https://git.wiki.kernel.org/
        git pull
        git push

When we need narrow (in the tree dimention) or shallow (in the history
dimention) in either native or foreign transports, somebody would
eventually add proper support. Currently we do not do "narrow" even for
native transport (I have one cooking privately but it is progressing only
slowly, and I think there may be others who are interested in it).

I would suggest not to be worried too much about narrow/shallow in your
initial round.
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]