On Wed, Apr 20, 2011 at 4:00 PM, Tomek Kott <[email protected]> wrote:
> My development machine is a Win7 x64 machine. The website application I
> develop for I have access to via a webdav connection that is visible as a
> mounted drive in explorer. I use cygwin for access to many linux tasks etc.,
> and I use a terminal through which I do my fossil work.
>>>>
> I essentially want to set up a dev environment in which I have two branches,
> say "trunk" and "dev" so that I can push changes and roll back when
> necessary.
I have done something similar in the past, for publishing data to a webserver.
My process was to create a staging area when I was ready for a
release, fetching the files into the staging area much like I would
into a working copy. Once done testing the proposed release in the
staging area, I would copy the staging area to a new folder on the
server. Once the copy was done, I would switch over by running:
mv data data-prev; mv data-new data
This made for the smoothest transition. However, it was also possible
to just copy from staging to the live website. If a file was still be
written when accessed by the webserver, an incomplete file would be
delivered to the client, so this was not as clean, but it did work.
_______________________________________________
fossil-users mailing list
[email protected]
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
