NB: Apologies if I don't get this to thread correctly, I've not got an
email copy of the original post, so can't reply properly.
As an occasional Fossil user, I have previously hosted using
inetd/althttpd[1] and Chisel[2] as well as Dropbox.
Now I have moved to cgi, given ssh access, it seems to be as simple as
just three file copies, right?
scp fossil user@webservers-ssh-domain:/path-to-host's-executables/bin/
## if binary not available for host machine, then instead, do wget of
the sources and ./configure make on the host
scp repos.cgi user@webservers-ssh-domain:/path-to-host's-scripts/cgi-bin/
scp repo1.fossil user@webservers-ssh-domain:/path-to-host's-repositories/repos/
# repeat the above copy for more than one repository on the server
# backup the original copied fossil archives from the client machine,
and 'fossil clone' new copies from the server (fossils can be moved,
but I think should not be duplicated, internals of fossil need state
variables dependent upon the instance)
A further thread I found very valuable is the knowledge that this can
be done on sourceforge for open source projects [3].
I use fossil to share data with people who are normally dropbox users
for their personal use and occasional users of corporate VCS/tracking
tools. I and people I introduce Fossil to feel that it is an
excellent model of elegant design and efficient workflow and would
like to see it adopted more widely, and perhaps bring in extra
valuable contributors.
I find that the idea of hosting [4] is great and that for me; an
additional facility matching dropbox background file syncing could
perhaps widen the user base. A client(Fossil) installed like that for
dropbox[5] on the users local machine would be most handy. As Fossil
already does autosync and autocheckout, some extra client side
automatic checkin and filename splitting on (rare) replication
conflicts may be needed to match dropbox usability.
On top of a Fossil hosting service, this wish I have, could be
characterised as anybody who wants to use fossil like dropbox simply
has to download the client package, open to install it, agree to let
it start on reboot and run in the background. Given an available
hosting solution, the user would see something like: If you're running
this for the first time, you'll be asked to create a new account or
add this to an existing account. Once you do, your Fossil folder will
be created in your home directory.
[1] althttpd, http://www.sqlite.org/docsrc/artifact/14c9965da19?ln=2-62
[2] chisel and flint, http://chiselapp.com/repositories/search/flint
[3] ssh access to sourceforge,
http://thread.gmane.org/gmane.comp.version-control.fossil-scm.user/12163,
really convenient option for self hosting open source
[4] fossil hosting requirements,
http://thread.gmane.org/gmane.comp.version-control.fossil-scm.user/12160/focus=12167,
anybody who wants to host something like chiselapp simply has to
obtain a low-cost internet host, copy the fossil executable into
/usr/bin, create a single CGI script that is less than 10 lines long,
and they are up and running
[5] dropbox client application, https://www.dropbox.com/help/247/en,
seems to have wide adoption
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
