{"thread":{"id":"5450","subject":"Joining a fixed archive with its continuation","startedAt":"2006-09-01T11:56:21Z","lastAt":"2006-09-01T14:37:50Z","messageCount":3,"participants":["Sergio Callegari","Jakub Narebski","Johannes Schindelin"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"26183","messageId":"44F81FE5.1040509@arces.unibo.it","threadId":"5450","inReplyTo":null,"subject":"Joining a fixed archive with its continuation","fromName":"Sergio Callegari","fromEmail":"scallegari@arces.unibo.it","sentAt":"2006-09-01T11:56:21Z","receivedAt":"2006-09-01T11:56:21Z","isPatch":false,"sender":{"key":"scallegari@arces.unibo.it","avatar":null},"body":"Hi,\n\nI am trying to follow Johannes (Dscho) suggestions to join a fixed\narchive with a new archive, that is its continuation from the latest\nworking tree.\nI am having some trouble with rebase, and any help would be appreciated.\n>\n> You can \"graft\" the new onto the old branch:\n> \nYes, grafts are fine... however, if I am to use push/pull for\ntransferring data between two computers I am working on, I'd like better\na \"history rewriting\" approach, so that I do not need to manually copy\nthe grafts.\n\n> NOTE! This is the quickest way if you want to have the history _locally_.\n>\n> If you want to be able to distribute it (or synchronize it between your \n> laptop and PC _with git!_), you can rewrite the history by either \n> git-rebase, or by using cg-admin-rewritehist if you are using cogito.\n>\n> \nI have tried using git-rebase (I do not have cg, at least not yet).\nHowever I am encountering some problems...\n\nEverything is fine with the master branch of my continuation archive doing\n\ngit rebase fixed-master master\n\n(where fixed-master is the master head of the recovered archive)\nThis does the trick nicely. I.e. I start with\n\nA---B---C---D E---F---G---H---I\n\n\nwhere D is fixed-master and I is master, and I end up with\n\nA---B---C---D---E'---F'---G'---H'---I'\n\nplus the older tree E---...---I\n\nHowever I am then in trouble with the other branches of the continuation\narchive, which I simply cannot get right...\nIn fact, E...I has actually a branch, say at G, like in\n\nE---F---G---H---I\n \\\n --M---N\n\n\nAnd I cannot make an M'---N' based at G'... the best that I am\nsucceeding in is building an\n\nE''--F''--G''--M'--N'\n\n\nbased on D.\n\n> Ciao,\n> Dscho\n>\n> P.S.: Of course, if you do not insist on a super clean history, you can \n> fake a merge. Just put <40-hex-chars-old> into .git/MERGE_HEAD and commit. \n> This will pretend that your new head and your old head were merged, and \n> the result is the new head. This _should_ even work with git-bisect, but \n> it is slightly ugly.\n\nBefore I try, can you better explain me what shall go on in this case?\nThe man page of commit actually does not say much about commit behavior\nduring a merge (i.e. with MERGE_HEAD set).\n\nThanks,\n\nSergio\n"},{"id":"26185","messageId":"ed9f3e$cvm$1@sea.gmane.org","threadId":"5450","inReplyTo":"44F81FE5.1040509@arces.unibo.it","subject":"Re: Joining a fixed archive with its continuation","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2006-09-01T14:12:04Z","receivedAt":"2006-09-01T14:12:04Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"Sergio Callegari wrote:\n\n>> If you want to be able to distribute it (or synchronize it between your \n>> laptop and PC _with git!_), you can rewrite the history by either \n>> git-rebase, or by using cg-admin-rewritehist if you are using cogito.\n>>\n>>   \n> I have tried using git-rebase (I do not have cg, at least not yet).\n> However I am encountering some problems...\n> \n> Everything is fine with the master branch of my continuation archive doing\n> \n> git rebase fixed-master master\n> \n> (where fixed-master is the master head of the recovered archive)\n> This does the trick nicely. I.e. I start with\n> \n> A---B---C---D    E---F---G---H---I\n> \n> \n> where D is fixed-master and I is master, and I end up with\n> \n> A---B---C---D---E'---F'---G'---H'---I'\n> \n> plus the older tree E---...---I\n> \n> However I am then in trouble with the other branches of the continuation\n> archive, which I simply cannot get right...\n> In fact, E...I has actually a branch, say at G, like in\n> \n> E---F---G---H---I\n>          \\\n>           --M---N\n> \n> \n> And I cannot make an M'---N' based at G'... the best that I am\n> succeeding in is building an\n> \n> E''--F''--G''--M'--N' \n> \n> based on D.\n\nIt is where --onto parameter of git-rebase is used. It means you use\n parameter to identify main branch in the continuation archive,\n parameter being the branch you want to transplant into main\narchive (i.e. .. being the part you want to\ntransplant/rebase), and '--onto ' with being the new\nfork point.\n\nSo in the abovementioned case, try\n\n git rebase --onto G' I N \n\nand tell us if it works.\n-- \nJakub Narebski\nWarsaw, Poland\nShadeHawk on #git\n"},{"id":"26186","messageId":"Pine.LNX.4.63.0609011634370.28360@wbgn013.biozentrum.uni-wuerzburg.de","threadId":"5450","inReplyTo":"44F81FE5.1040509@arces.unibo.it","subject":"Re: Joining a fixed archive with its continuation","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2006-09-01T14:37:50Z","receivedAt":"2006-09-01T14:37:50Z","isPatch":false,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nOn Fri, 1 Sep 2006, Sergio Callegari wrote:\n\n> > P.S.: Of course, if you do not insist on a super clean history, you can fake\n> > a merge. Just put <40-hex-chars-old> into .git/MERGE_HEAD and commit. This\n> > will pretend that your new head and your old head were merged, and the\n> > result is the new head. This _should_ even work with git-bisect, but it is\n> > slightly ugly.\n> \n> Before I try, can you better explain me what shall go on in this case?\n> The man page of commit actually does not say much about commit behavior\n> during a merge (i.e. with MERGE_HEAD set).\n\nA merge commit is almost the same as a regulaar commit; the only \ndifference is that you provide multiple parents to a merge commit. The \nfirst parent is always the current head. And the other parents are in \n.git/MERGE_HEAD (one commit SHA1 per line).\n\nSo, putting the fixed-master SHA1 into .git/MERGE_HEAD pretends that the \nfixed-master merged with the current master is the current master.\n\n>From a view point of correctness, this is _wrong_. _But_ it would work \ncorrectly with pull/push/bisect.\n\nCiao,\nDscho\n"}]}