{"thread":{"id":"43357","subject":"Re: Question on rerere","startedAt":"2006-12-19T23:53:38Z","lastAt":"2006-12-20T01:40:46Z","messageCount":4,"participants":["Junio C Hamano","Johannes Schindelin"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"295268","messageId":"Pine.LNX.4.63.0612200045490.19693@wbgn013.biozentrum.uni-wuerzburg.de","threadId":"43357","inReplyTo":null,"subject":"Question on rerere","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2006-12-19T23:53:38Z","receivedAt":"2006-12-19T23:53:38Z","isPatch":false,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nwhile studying the code of rerere I found myself wondering if the code of \ncompute_conflict_name and record_preimage is correct.\n\nIIUC then each conflict hunk is handled _separately_ like this: the \nlexicographically smaller between the two file sections is displayed \nfirst, regardless if a previous hunk had a different order. Not that it \nmatters most of the time, but isn't this dangerous?\n\nCiao,\nDscho\n"},{"id":"294478","messageId":"7vk60no0ua.fsf@assigned-by-dhcp.cox.net","threadId":"43357","inReplyTo":"Pine.LNX.4.63.0612200045490.19693@wbgn013.biozentrum.uni-wuerzburg.de","subject":"Re: Question on rerere","fromName":"Junio C Hamano","fromEmail":"junkio@cox.net","sentAt":"2006-12-20T00:41:33Z","receivedAt":"2006-12-20T00:41:33Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Johannes Schindelin writes:\n\n> IIUC then each conflict hunk is handled _separately_ like this: the \n> lexicographically smaller between the two file sections is displayed \n> first, regardless if a previous hunk had a different order. Not that it \n> matters most of the time, but isn't this dangerous?\n\nYou are probably right. Probably the right thing would be to\nuse the first hunk to determine the flipping order and stick to\nthat for the rest.\n\nNot that I've seen problems with the current behaviour, though.\n"},{"id":"294808","messageId":"7vy7p3mk1i.fsf@assigned-by-dhcp.cox.net","threadId":"43357","inReplyTo":"7vk60no0ua.fsf@assigned-by-dhcp.cox.net","subject":"Re: Question on rerere","fromName":"Junio C Hamano","fromEmail":"junkio@cox.net","sentAt":"2006-12-20T01:29:45Z","receivedAt":"2006-12-20T01:29:45Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Junio C Hamano writes:\n\n> Johannes Schindelin writes:\n>\n>> IIUC then each conflict hunk is handled _separately_ like this: the \n>> lexicographically smaller between the two file sections is displayed \n>> first, regardless if a previous hunk had a different order. Not that it \n>> matters most of the time, but isn't this dangerous?\n>\n> You are probably right. Probably the right thing would be to\n> use the first hunk to determine the flipping order and stick to\n> that for the rest.\n>\n> Not that I've seen problems with the current behaviour, though.\n\nWell, come to think of it, I think the current behaviour makes\nmore sense.\n\nSuppose you start from an original file \"OO\". You have two\nbranches that change it to \"AO\" and \"BO\", and another pair of\nbranches that change it to \"OC\" and \"OD\". Let's call these\nbranches A, B, C, and D.\n\nBy merging A and C, you will get \"AC\"; you can get \"AD\", \"BC\"\nand \"BD\" the same way.\n\nNow suppose you are on \"AC\" and merged \"BD\". You would get\n\"\".\n\nIf you were on \"BD\" and merged \"AC\" you would get \"\".\nIf you were on \"AD\" and merged \"BC\" you would get \"\".\n\nYou got the idea?\n\n\n\n\n\n"},{"id":"296728","messageId":"Pine.LNX.4.63.0612200240360.19693@wbgn013.biozentrum.uni-wuerzburg.de","threadId":"43357","inReplyTo":"7vy7p3mk1i.fsf@assigned-by-dhcp.cox.net","subject":"Re: Question on rerere","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2006-12-20T01:40:46Z","receivedAt":"2006-12-20T01:40:46Z","isPatch":false,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nOn Tue, 19 Dec 2006, Junio C Hamano wrote:\n\n> Junio C Hamano writes:\n> \n> > Johannes Schindelin writes:\n> >\n> >> IIUC then each conflict hunk is handled _separately_ like this: the \n> >> lexicographically smaller between the two file sections is displayed \n> >> first, regardless if a previous hunk had a different order. Not that it \n> >> matters most of the time, but isn't this dangerous?\n> >\n> > You are probably right. Probably the right thing would be to\n> > use the first hunk to determine the flipping order and stick to\n> > that for the rest.\n> >\n> > Not that I've seen problems with the current behaviour, though.\n> \n> Well, come to think of it, I think the current behaviour makes\n> more sense.\n> \n> Suppose you start from an original file \"OO\". You have two\n> branches that change it to \"AO\" and \"BO\", and another pair of\n> branches that change it to \"OC\" and \"OD\". Let's call these\n> branches A, B, C, and D.\n> \n> By merging A and C, you will get \"AC\"; you can get \"AD\", \"BC\"\n> and \"BD\" the same way.\n> \n> Now suppose you are on \"AC\" and merged \"BD\". You would get\n> \"\".\n> \n> If you were on \"BD\" and merged \"AC\" you would get \"\".\n> If you were on \"AD\" and merged \"BC\" you would get \"\".\n> \n> You got the idea?\n\nYes. Thanks!\n\nCiao,\nDscho\n"}]}