I think Andy's suggestion(s) are your best bet but I want to mention what
I've done in the past.
Is your intent to co-mingle same-named branches? I.e. if there is a trunk
in both source repos do you wish to merge both dev streams to a single
trunk? I have done this via brute force in the past and it can work well
*if* there are no overlapps in the files (or if you can ignore the
consequences of overlaps). Here is how I did it (roughly and from memory):
get all the commits from both repos for the branch to be intertwined and
sort them by time.
create a new bare repo
open each of the source repos in separate areas
for each commit
update to that commit in the respective repo
copy the files into the new repo using rsync -az (to not disturb
timestamps)
Add/remove with fossil addrem
commit the changes using the commit message from the source repo and
force the timestamp from the original commit
It would be very difficult to capture any branching with this method and it
can be *very* time consuming but it can be useful. For non-overlapping
trees you can capture deletions by adding --delete to the rsync.
On Thu, Jul 16, 2015 at 7:56 PM, Andy Bradford <[email protected]>
wrote:
> Thus said =?UTF-8?B?Wm9sdMOhbiBLw7Njc2k=?= on Fri, 17 Jul 2015 09:27:48
> +1000:
>
> > My question is, is it possible to move artefacts from one Fossil
> > database to an other?
>
> This is all hypothetical...
>
> It's certainly not in the realm of impossibility. But Fossil wasn't
> really designed to do such a thing. You could change the project-code in
> one clone to that of another and do a pull. That would bring in a lot of
> artifacts, but they would have no common heritage. So the problem is,
> how do you connect the trees so you don't just end up with 1 repository
> with 2 separate trees of development?
>
> Maybe rather than doing a pull, you could deconstruct both repositories,
> put the extracted artifacts into the same place, fabricate a manifest
> that somehow merges the two timelines with files from both trees?
>
> In other words, I'm not aware of an easy way to do this. :-)
>
> Maybe someone else?
>
> Andy
> --
> TAI64 timestamp: 4000000055a86f1d
>
>
> _______________________________________________
> fossil-users mailing list
> [email protected]
> http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
>
_______________________________________________
fossil-users mailing list
[email protected]
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
