You don't really need a local branch at all.<br><br>A way to branching project a into repository b is<br><br>svk mirror PROJECT-A-URL //mirrors/a<br>svk mirror PROJECT-B-URL //mirrors/b<br>svk sync -a<br>svk mkdir -p //mirrors/b/branches -m "Made branches dir."
<br>svk sm -B //mirrors/a/trunk //mirrors/b/branches/bbranchofa<br>svk co //mirrors/b/branches/bbranchofa<br>cd bbranchofa<br>[...hack away...commit changes...rinse..repeat...]<br># To get upstream changes from PROJECT-A-URL use either
<br>svk pull<br># which gets you a commit in your branch for each upstream commit, or run<br>svk sync -a<br>svk sm //mirrors/a/trunk .<br># resolve conflicts, test etc.<br>svk ci<br># To make a single local commit for a set of upstream changes and allow for local testing first.
<br><br>Now any user of the PROJECT-B-URL can use either svk or just plain svn to hack on the local branch of project-a. Only those people actually pulling in upstream changes need a mirror of both repositories. In addition since svk stores merge tracking info as properties in the branch on the PROJECT-B-URL repository, the entire branched history is preserved in that repository and in ever developers mirror of it.
<br><br>Note that even if you have a local branch as an intermediate between the 2 mirrors such as suggested earlier in this thread, the merge history of the project still travels with the project to whatever svn repository you sm -B it to.
<br><br>Michael<br><br><br><div><span class="gmail_quote">On 10/24/06, <b class="gmail_sendername">Peter Valdemar Morch</b> <<a href="mailto:swp5jhu02@sneakemail.com">swp5jhu02@sneakemail.com</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
In my opinion Victor brings up a great question:<br>> What I need to do though is have those changes saved back to<br>> server-b's SVN repository in case my local computer blows up.<br><br>At least I have this question:
<br><br>What happens if the workstation running svk blows up. Is there any way<br>to recreate an svk installation from scratch so that the next svk smerge<br>from there works "properly" using only the information on
<a href="http://server-a.com">server-a.com</a><br>and <a href="http://server-b.com">server-b.com</a>?<br><br>Or: Would others be able to maintain <a href="http://server-b.com">server-b.com</a> and merge "upstream"
<br>changes on <a href="http://server-a.com">server-a.com</a> to <a href="http://server-b.com">server-b.com</a>?<br><br>Peter<br><br>--<br>Peter Valdemar Mørch<br><a href="http://www........">http://www........</a>..<br>_______________________________________________
<br>svk-devel mailing list<br><a href="mailto:svk-devel@bestpractical.com">svk-devel@bestpractical.com</a><br><a href="http://lists.bestpractical.com/cgi-bin/mailman/listinfo/svk-devel">http://lists.bestpractical.com/cgi-bin/mailman/listinfo/svk-devel
</a><br></blockquote></div><br>