So, we have a git repository set up on a web server machine 'out there' somewhere. I make my local copy of the repository on my computer and do some editing etc. Then 'git commit' commits the changes to my local repository and a 'git push' sends the changes back to the remote repository.
However I don't see how the files get to 'go live' on the remote web server, my 'git push' has simply put them back into the repository on that system. Presumably someone/somewhere has to extract the files from the git repository on the remote into the working directories from which the web server displays them. Is there a specific git command for doing this or would one use the standard 'git clone' to do it, or what?
On 29 November 2010 17:31, Chris G cl@isbd.net wrote:
However I don't see how the files get to 'go live' on the remote web server, my 'git push' has simply put them back into the repository on that system. Presumably someone/somewhere has to extract the files from the git repository on the remote into the working directories from which the web server displays them. Is there a specific git command for doing this or would one use the standard 'git clone' to do it, or what?
I'd have thought an initial 'git clone' followed by periodic or triggered 'git pull's. There may be tools to do this more automagically - see if Google has any hits on this subject.
But I think there's also a git extract (or whatever equivalent of the svn command is) so you don't also expose the git metadata. I think the manual pages and git books will have more info.
- Srdjan
On Mon, Nov 29, 2010 at 05:46:13PM +0000, Srdjan Todorovic wrote:
On 29 November 2010 17:31, Chris G cl@isbd.net wrote:
However I don't see how the files get to 'go live' on the remote web server, my 'git push' has simply put them back into the repository on that system. Presumably someone/somewhere has to extract the files from the git repository on the remote into the working directories from which the web server displays them. Is there a specific git command for doing this or would one use the standard 'git clone' to do it, or what?
I'd have thought an initial 'git clone' followed by periodic or triggered 'git pull's. There may be tools to do this more automagically - see if Google has any hits on this subject.
But I think there's also a git extract (or whatever equivalent of the svn command is) so you don't also expose the git metadata. I think the manual pages and git books will have more info.
OK, thanks, I'll dig some more (and discuss it with the other git users there).
On 29 Nov 17:46, Srdjan Todorovic wrote:
On 29 November 2010 17:31, Chris G cl@isbd.net wrote:
However I don't see how the files get to 'go live' on the remote web server, my 'git push' has simply put them back into the repository on that system. Presumably someone/somewhere has to extract the files from the git repository on the remote into the working directories from which the web server displays them. Is there a specific git command for doing this or would one use the standard 'git clone' to do it, or what?
I'd have thought an initial 'git clone' followed by periodic or triggered 'git pull's. There may be tools to do this more automagically - see if Google has any hits on this subject.
But I think there's also a git extract (or whatever equivalent of the svn command is) so you don't also expose the git metadata. I think the manual pages and git books will have more info.
It helps if you get the correct svn command... svn export.
But it really depends on how he *wants* to do the deployments, personally I'd just use git pulls from the remote after doing a clone, and stopping the webserver from serving the .git directory.
(The 'equiv' to and svn export is git archive, by the way)
-
Brett Parker -
Chris G -
Srdjan Todorovic