{"thread":{"id":"28971","subject":"Re: t/t1304: avoid -d option to setfacl (db82657)","startedAt":"2011-11-18T05:37:06Z","lastAt":"2011-11-29T12:23:33Z","messageCount":3,"participants":["Brandon Casey","Scott Parrish","Jeff King"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"179674","messageId":"CA+sFfMc6BCPvk1FCjZwyY_pWqXBut3D--OfrFEcfK5p97D-KHQ@mail.gmail.com","threadId":"28971","inReplyTo":"gitster/git/commit/db826571e4099066fe44233d95642591016c831b/729354@github.com","subject":"Re: t/t1304: avoid -d option to setfacl (db82657)","fromName":"Brandon Casey","fromEmail":"drafnel@gmail.com","sentAt":"2011-11-18T05:37:06Z","receivedAt":"2011-11-18T05:37:06Z","isPatch":false,"sender":{"key":"drafnel@gmail.com","avatar":"https://avatars.githubusercontent.com/u/921167?v=4"},"body":"[cc git@vger.kernel.org since that is the appropriate place for this discusion]\n\nOn Thu, Nov 17, 2011 at 8:59 PM, wsp\n\nwrote:\n> Could this test case be reviewed again? It fails on FreeBSD where the appropriate way to specify default ACL's is with the \"-d\" option. The \"d[efault]:\" syntax is invalid on FreeBSD.\n\nWell, I'm not sure there is a right answer here.\n\n Linux accepts -d or \"d[efault]\"\n Solaris only accepts \"d[efault]\"\n FreeBSD only accepts -d\n a quick search shows z/OS only accepts \"d[efault]\"\n other?\n\nI think most everything else rolls their own implementation into chmod\nor chacl or something like that.\n\nThe abandoned POSIX draft does actually specify the FreeBSD behavior.\n\nSo I think it's kind of a toss-up. Which option we choose should\nprobably depend on whether we get more test coverage by using the\n\"d[efault]\" notation or by using the -d option. That depends on\nwhether there are more Solaris users compiling git or whether there\nare more FreeBSD users. I don't know the answer to that either. I\ntend to think there are very few of either.\n\n-Brandon\n"},{"id":"179747","messageId":"D95BFEA0-BAC1-4CBD-8B56-27DB4BDF147C@gmail.com","threadId":"28971","inReplyTo":"CA+sFfMc6BCPvk1FCjZwyY_pWqXBut3D--OfrFEcfK5p97D-KHQ@mail.gmail.com","subject":"Re: t/t1304: avoid -d option to setfacl (db82657)","fromName":"Scott Parrish","fromEmail":"wsparrish@gmail.com","sentAt":"2011-11-19T20:28:49Z","receivedAt":"2011-11-19T20:28:49Z","isPatch":false,"sender":{"key":"wsparrish@gmail.com","avatar":null},"body":"Hi Brandon,\n\nThanks for posting my question. I realized shortly after my post to github that I should have posted to the mailing list instead.\n\n\nOn Nov 18, 2011, at 12:37 AM, Brandon Casey wrote:\n\n> [cc git@vger.kernel.org since that is the appropriate place for this discusion]\n> \n> On Thu, Nov 17, 2011 at 8:59 PM, wsp\n> \n> wrote:\n>> Could this test case be reviewed again? It fails on FreeBSD where the appropriate way to specify default ACL's is with the \"-d\" option. The \"d[efault]:\" syntax is invalid on FreeBSD.\n> \n> Well, I'm not sure there is a right answer here.\n> \n> Linux accepts -d or \"d[efault]\"\n> Solaris only accepts \"d[efault]\"\n> FreeBSD only accepts -d\n> a quick search shows z/OS only accepts \"d[efault]\"\n> other?\n> \n> I think most everything else rolls their own implementation into chmod\n> or chacl or something like that.\n\nOS X does indeed roll ACL management into chmod. The test passes there because the prereq isn't met (i.e. the setfacl command does not exist) and thus the tests are skipped.\n> \n> The abandoned POSIX draft does actually specify the FreeBSD behavior.\n> \n> So I think it's kind of a toss-up. Which option we choose should\n> probably depend on whether we get more test coverage by using the\n> \"d[efault]\" notation or by using the -d option. That depends on\n> whether there are more Solaris users compiling git or whether there\n> are more FreeBSD users. I don't know the answer to that either. I\n> tend to think there are very few of either.\n\nIs there a reason conditional logic can't be used (perhaps keying off of `uname -s` or the like) so that we have coverage in all cases?\n\n-S\n\n> \n> -Brandon\n> --\n> To unsubscribe from this list: send the line \"unsubscribe git\" in\n> the body of a message to majordomo@vger.kernel.org\n> More majordomo info at http://vger.kernel.org/majordomo-info.html\n"},{"id":"180085","messageId":"20111129122333.GB30456@sigill.intra.peff.net","threadId":"28971","inReplyTo":"D95BFEA0-BAC1-4CBD-8B56-27DB4BDF147C@gmail.com","subject":"Re: t/t1304: avoid -d option to setfacl (db82657)","fromName":"Jeff King","fromEmail":"peff@peff.net","sentAt":"2011-11-29T12:23:33Z","receivedAt":"2011-11-29T12:23:33Z","isPatch":false,"sender":{"key":"peff@peff.net","avatar":"https://avatars.githubusercontent.com/u/45925?v=4"},"body":"On Sat, Nov 19, 2011 at 03:28:49PM -0500, Scott Parrish wrote:\n\n> > The abandoned POSIX draft does actually specify the FreeBSD behavior.\n> > \n> > So I think it's kind of a toss-up. Which option we choose should\n> > probably depend on whether we get more test coverage by using the\n> > \"d[efault]\" notation or by using the -d option. That depends on\n> > whether there are more Solaris users compiling git or whether there\n> > are more FreeBSD users. I don't know the answer to that either. I\n> > tend to think there are very few of either.\n> \n> Is there a reason conditional logic can't be used (perhaps keying off\n> of `uname -s` or the like) so that we have coverage in all cases?\n\nIt's nice if we can auto-detect such things based on behavior, and then\nthere's no need for maintaining a \"uname -s\" guess. Perhaps we can\ncheck at the beginning of t1304 which one works (or neither), and set a\nflag appropriately? It looks like we already check whether setfacl works\nat all, so this would be a natural extension.\n\nI started to do a \"how about this\" patch, but I realized that in order\nto see whether \"-d\" worked as \"default\" versus \"delete\", we'll have to\ncheck the output of \"getfacl\" to see if it reports the default entry.\nAnd then I realized I have no idea what FreeBSD shows in such a case\n(surely not \"d:*\", right?).\n\n-Peff\n"}]}