Richard, Thanks for the feedback, that would save me breaking my head more than necessary :)
I don't want either to mess up with fossil concepts so I don't think I'll go by updating the table via sql. Let me explain why I came up to trying such sequence of commands; my plans is to use fossil as follows: 1. I plan to have one fossil repository available via http. 2. I'll sync that http served repository with several developer repositories. 3. One of these dev repositories need though to be able to pull changes from other dev repositories (because some dev repositories won't be able to access the http served repository). 4. The dev repository which will be pulling changes will then propagate the changes back to the http served repository. That was why I was testing the pull from a repository created from scratch. I guess one way to achieve such scenario is to copy or clone the http served repository on each dev environment so that the project id stays the same and afterwards the sync should be possible as required above, right? Or is there another preferred or recommended way to do this as per fossil concepts ? ps: Regarding the rather pointless reply saying that's a big nono to open a repo next ..., my first intent was to test the features in a quick way (and hence a more or less dirty way; if that was causing any issue, I would have expected fossil to complain) before getting to the real things in more structured approach. Thanks for the feedback, Teggy 2011/7/1 Richard Hipp <[email protected]> > > > On Fri, Jul 1, 2011 at 11:31 AM, Teggy V <[email protected]> wrote: > >> >> As reported last week, I am getting a login failure message everytime I >> try to pull from a http url whereas the same http url works fine when used >> with clone command. >> >> As mentioned above, if someone can pinpoint in the above something which >> is not correct; or if someone will be kind enough to try a similar sequence >> of commands and shares these commands with me, that would be great. >> >> > Every fossil repository has a "project id". When you clone a repo, the > project id is copied. But when you create a new repo from scratch, the > project id is different. > > Fossil will not pull (or push or sync) two repositories with different > project ids. That's a feature to prevent you from accidentally pushing or > pulling the entire history of one project into an unrelated project. > > If you create two repos separately and need to sync them, you'll have to > manually set the "project id" to be the same on the two repos. Use "fossil > sql" to do this: > > UPDATE config SET value=? WHERE name='project-code'; > > Substitute the appropriate project id for the ? above, of course. > > Rather than mess with project codes, the best thing to do is simply to > clone or copy the repo rather than trying to recreate it from scratch and > then syncing. > > -- > D. Richard Hipp > [email protected] > > _______________________________________________ > 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
