{"thread":{"id":"4543","subject":"Re: [PATCH] CVSps fixed; git-cvsimport works, too","startedAt":"2006-06-17T12:21:58Z","lastAt":"2006-06-17T18:51:27Z","messageCount":5,"participants":["Yann Dirson","Chris Shoemaker"],"isPatch":true,"patchVersion":1,"patchTotal":null},"messages":[{"id":"21951","messageId":"20060617122158.GQ7766@nowhere.earth","threadId":"4543","inReplyTo":null,"subject":"Re: [PATCH] CVSps fixed; git-cvsimport works, too","fromName":"Yann Dirson","fromEmail":"ydirson@altern.org","sentAt":"2006-06-17T12:21:58Z","receivedAt":"2006-06-17T12:21:58Z","isPatch":true,"sender":{"key":"ydirson@altern.org","avatar":"https://avatars.githubusercontent.com/u/1190950?v=4"},"body":"[for original thread see\nhttp://www.gelato.unsw.edu.au/archives/git/0511/11812.html]\n\nHi Chris,\n\nI was looking at your patch and test script to decide whether to\ninclude it in my cvsps patch collection, but it seems there are still\nproblems:\n\n* when run with \"-x\" it gives the expected result, even with \"-A\"\n\n* when run without \"-x\" and without \"-A\", the patchset introducing\nfile \"b\" and branch \"stable\" does not appear, without even a warning\n\n* when run with \"-A\" cvsps notices the inconsistency:\n\ninvalid initial_branch for file a, probably from old cache, run with -x.\n\n\nI have included the current version of your patch in branch \"to-check\"\nin http://ydirson.free.fr/soft/git/cvsps.git/.  I did not tak time to\nlook at it closely yet - do you plan to work on this issue ?\n\nBest regards,\n-- \nYann Dirson    <ydirson@altern.org> |\nDebian-related: <dirson@debian.org> |   Support Debian GNU/Linux:\n                                    |  Freedom, Power, Stability, Gratis\n     http://ydirson.free.fr/        | Check <http://www.debian.org/>\n"},{"id":"21953","messageId":"20060617132023.GA29452@pe.Belkin","threadId":"4543","inReplyTo":"20060617122158.GQ7766@nowhere.earth","subject":"Re: [PATCH] CVSps fixed; git-cvsimport works, too","fromName":"Chris Shoemaker","fromEmail":"c.shoemaker@cox.net","sentAt":"2006-06-17T13:20:23Z","receivedAt":"2006-06-17T13:20:23Z","isPatch":true,"sender":{"key":"c.shoemaker@cox.net","avatar":null},"body":"On Sat, Jun 17, 2006 at 02:21:58PM +0200, Yann Dirson wrote:\n> [for original thread see\n> http://www.gelato.unsw.edu.au/archives/git/0511/11812.html]\n> \n> Hi Chris,\n> \n> I was looking at your patch and test script to decide whether to\n> include it in my cvsps patch collection, but it seems there are still\n> problems:\n> \n> * when run with \"-x\" it gives the expected result, even with \"-A\"\n\nThis is expected.\n\n> \n> * when run without \"-x\" and without \"-A\", the patchset introducing\n> file \"b\" and branch \"stable\" does not appear, without even a warning\n\nWith out \"-A\" I don't think the patch is operative.\n\n> \n> * when run with \"-A\" cvsps notices the inconsistency:\n> \n> invalid initial_branch for file a, probably from old cache, run with -x.\n\ngood advice.\n\n> \n> \n> I have included the current version of your patch in branch \"to-check\"\n> in http://ydirson.free.fr/soft/git/cvsps.git/.  I did not tak time to\n> look at it closely yet - do you plan to work on this issue ?\n\nI wondered if you'd find that.  Evidentally, you're being thorough. :)\n\nThe intent of my patch is only to affect the branch ancestry\ncalculation.  IIRC, the patch version is incompatible with cached\nresults from unpatched versions.  Hence I would only expect it to work\nwith '-x -A' (unless there was no pre-existing cache file).\n\nI've had one independent report of success from someone who found and\nused my patch.\n\nI haven't had to use cvsps recently, so I don't know how much time\nI'll have for cvsps hacking, but I'm interested in getting feedback on\nmy current patch.\n\nThere's probably some nicer way to handle the cache incompatibility,\nbut my current approach has been to tell people to delete their old\ncache file or run wih -x.  Suggestions are welcome.\n\n-chris\n"},{"id":"21954","messageId":"20060617134717.GR7766@nowhere.earth","threadId":"4543","inReplyTo":"20060617132023.GA29452@pe.Belkin","subject":"Re: [PATCH] CVSps fixed; git-cvsimport works, too","fromName":"Yann Dirson","fromEmail":"ydirson@altern.org","sentAt":"2006-06-17T13:47:17Z","receivedAt":"2006-06-17T13:47:17Z","isPatch":true,"sender":{"key":"ydirson@altern.org","avatar":"https://avatars.githubusercontent.com/u/1190950?v=4"},"body":"On Sat, Jun 17, 2006 at 09:20:23AM -0400, Chris Shoemaker wrote:\n> > \n> > * when run with \"-A\" cvsps notices the inconsistency:\n> > \n> > invalid initial_branch for file a, probably from old cache, run with -x.\n> \n> good advice.\n> \n> > \n> > \n> > I have included the current version of your patch in branch \"to-check\"\n> > in http://ydirson.free.fr/soft/git/cvsps.git/.  I did not tak time to\n> > look at it closely yet - do you plan to work on this issue ?\n> \n> I wondered if you'd find that.  Evidentally, you're being thorough. :)\n> \n> The intent of my patch is only to affect the branch ancestry\n> calculation.  IIRC, the patch version is incompatible with cached\n> results from unpatched versions.  Hence I would only expect it to work\n> with '-x -A' (unless there was no pre-existing cache file).\n\nAll those tests were done with your patch applied, so the cache was\ngenerated and re-read using the pached version.\n\n> There's probably some nicer way to handle the cache incompatibility,\n> but my current approach has been to tell people to delete their old\n> cache file or run wih -x.  Suggestions are welcome.\n\nIf the cache format is changed, this would be a good candidate for the\n\"v2\" cache format.  I already have 2 (incomplete) patches that well\nneed to bump the cache format version - see branches multitag (allow\nseveral tags per patchset) and semicolon (allow semicolons in\nfilenames).\n\nBest regards,\n-- \nYann Dirson    <ydirson@altern.org> |\nDebian-related: <dirson@debian.org> |   Support Debian GNU/Linux:\n                                    |  Freedom, Power, Stability, Gratis\n     http://ydirson.free.fr/        | Check <http://www.debian.org/>\n"},{"id":"21955","messageId":"20060617143443.GA29602@pe.Belkin","threadId":"4543","inReplyTo":"20060617134717.GR7766@nowhere.earth","subject":"Re: [PATCH] CVSps fixed; git-cvsimport works, too","fromName":"Chris Shoemaker","fromEmail":"c.shoemaker@cox.net","sentAt":"2006-06-17T14:34:43Z","receivedAt":"2006-06-17T14:34:43Z","isPatch":true,"sender":{"key":"c.shoemaker@cox.net","avatar":null},"body":"On Sat, Jun 17, 2006 at 03:47:17PM +0200, Yann Dirson wrote:\n> On Sat, Jun 17, 2006 at 09:20:23AM -0400, Chris Shoemaker wrote:\n> > > \n> > > * when run with \"-A\" cvsps notices the inconsistency:\n> > > \n> > > invalid initial_branch for file a, probably from old cache, run with -x.\n> > \n> > good advice.\n> > \n> > > \n> > > \n> > > I have included the current version of your patch in branch \"to-check\"\n> > > in http://ydirson.free.fr/soft/git/cvsps.git/.  I did not tak time to\n> > > look at it closely yet - do you plan to work on this issue ?\n> > \n> > I wondered if you'd find that.  Evidentally, you're being thorough. :)\n> > \n> > The intent of my patch is only to affect the branch ancestry\n> > calculation.  IIRC, the patch version is incompatible with cached\n> > results from unpatched versions.  Hence I would only expect it to work\n> > with '-x -A' (unless there was no pre-existing cache file).\n> \n> All those tests were done with your patch applied, so the cache was\n> generated and re-read using the pached version.\n\nAre you saying there was no pre-existing cache file when the patched\nversion was first run?  Did you delete it?  If so, then there's a bug\nsomewhere.\n\n> \n> > There's probably some nicer way to handle the cache incompatibility,\n> > but my current approach has been to tell people to delete their old\n> > cache file or run wih -x.  Suggestions are welcome.\n> \n> If the cache format is changed, this would be a good candidate for the\n> \"v2\" cache format.  I already have 2 (incomplete) patches that well\n> need to bump the cache format version - see branches multitag (allow\n> several tags per patchset) and semicolon (allow semicolons in\n> filenames).\n\nIIRC, it's not syntactically incompatible - it's just that the patched\nversion can tell that it doesn't trust the initial branch\ndeterminations that are stored in the cache file by an unpatched\nversion.\n\nThe complaint is basically, Hey, there's an initial branch\ndetermination already made for this file, but it's not the one I would\nhave made, so it must have been restored from a cache file that I\ndidn't make.\n\n-chris\n\n> \n> Best regards,\n> -- \n> Yann Dirson    <ydirson@altern.org> |\n> Debian-related: <dirson@debian.org> |   Support Debian GNU/Linux:\n>                                     |  Freedom, Power, Stability, Gratis\n>      http://ydirson.free.fr/        | Check <http://www.debian.org/>\n"},{"id":"21963","messageId":"20060617185126.GS7766@nowhere.earth","threadId":"4543","inReplyTo":"20060617143443.GA29602@pe.Belkin","subject":"Re: [PATCH] CVSps fixed; git-cvsimport works, too","fromName":"Yann Dirson","fromEmail":"ydirson@altern.org","sentAt":"2006-06-17T18:51:27Z","receivedAt":"2006-06-17T18:51:27Z","isPatch":true,"sender":{"key":"ydirson@altern.org","avatar":"https://avatars.githubusercontent.com/u/1190950?v=4"},"body":"On Sat, Jun 17, 2006 at 10:34:43AM -0400, Chris Shoemaker wrote:\n> Are you saying there was no pre-existing cache file when the patched\n> version was first run?\n\nIt was the sample script you provided to demonstrate the problem which\nshowed the new issue: it does a \"cvsps -x\" first, then runs\ngit-cvsimport, which runs \"cvsps -u\".  That one does fails.  I'm\npretty sure the cache does not exist beforehand, since your script\ncreates the repo in a temporary directory.\n\nBTW, I'll see about adding a testsuite to cvsps, based on the\nframework used by git.  All these scriptlets used to exhibit the\nvarious problems have to be kept somewhere.\n\n\n>  Did you delete it?  If so, then there's a bug somewhere.\n\nI do think there is a bug somewhere :)\n\n\n> The complaint is basically, Hey, there's an initial branch\n> determination already made for this file, but it's not the one I would\n> have made, so it must have been restored from a cache file that I\n> didn't make.\n\nOK, thanks for the clarification.\n\nBest regards,\n-- \nYann Dirson    <ydirson@altern.org> |\nDebian-related: <dirson@debian.org> |   Support Debian GNU/Linux:\n                                    |  Freedom, Power, Stability, Gratis\n     http://ydirson.free.fr/        | Check <http://www.debian.org/>\n"}]}