{"thread":{"id":"19723","subject":"mkstemps compat issue","startedAt":"2009-06-07T12:32:50Z","lastAt":"2009-06-08T01:13:35Z","messageCount":3,"participants":["Andreas Färber","David Aguilar"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"115699","messageId":"ED882BBA-E0E7-4CCC-9C6B-72FEBED1C9E0@web.de","threadId":"19723","inReplyTo":null,"subject":"mkstemps compat issue","fromName":"Andreas Färber","fromEmail":"andreas.faerber@web.de","sentAt":"2009-06-07T12:32:50Z","receivedAt":"2009-06-07T12:32:50Z","isPatch":false,"sender":{"key":"andreas.faerber@web.de","avatar":"https://gravatar.com/avatar/ac719235a76c4a5e248080fee1f7b6f507bdb1f3d71f7ae48457f3ead91ed1c0?d=mp&s=160"},"body":"Hello,\n\nThe merge commit b11cf09043f18b368ec0d988f064ea21247c843d introduced a  \ngit_mkstemps function in path.c, using mkstemps for the  \nimplementation. Unlike mkstemp that function is not part of POSIX, so  \nwe seem to need a NO_MKSTEMPS option and corresponding alternative  \ncode path.\n\nAny suggestions how to best replace mkstemps?\n\nThanks,\nAndreas\n\n[Please keep me cc'ed.]\n"},{"id":"115700","messageId":"DAD17464-336A-451A-AEFE-26481D6B39B2@web.de","threadId":"19723","inReplyTo":"ED882BBA-E0E7-4CCC-9C6B-72FEBED1C9E0@web.de","subject":"Re: mkstemps compat issue","fromName":"Andreas Färber","fromEmail":"andreas.faerber@web.de","sentAt":"2009-06-07T12:49:16Z","receivedAt":"2009-06-07T12:49:16Z","isPatch":false,"sender":{"key":"andreas.faerber@web.de","avatar":"https://gravatar.com/avatar/ac719235a76c4a5e248080fee1f7b6f507bdb1f3d71f7ae48457f3ead91ed1c0?d=mp&s=160"},"body":"\nAm 07.06.2009 um 14:32 schrieb Andreas Färber:\n\n> The merge commit b11cf09043f18b368ec0d988f064ea21247c843d introduced  \n> a git_mkstemps function in path.c, using mkstemps for the  \n> implementation. Unlike mkstemp that function is not part of POSIX,  \n> so we seem to need a NO_MKSTEMPS option and corresponding  \n> alternative code path.\n>\n> Any suggestions how to best replace mkstemps?\n\nSorry for the noise... NO_MKSTEMPS=YesPlease does work correctly if  \ntyped correctly. (Thanks for the hint!)\n\nIt was a little confusing seeing mkstemps being used unconditionally  \nin path.c, but it's being #defined in git-compat-util.h, so everything  \nis fine.\n\nRegards,\n\nAndreas\n"},{"id":"115759","messageId":"20090608011334.GA5974@gmail.com","threadId":"19723","inReplyTo":"DAD17464-336A-451A-AEFE-26481D6B39B2@web.de","subject":"Re: mkstemps compat issue","fromName":"David Aguilar","fromEmail":"davvid@gmail.com","sentAt":"2009-06-08T01:13:35Z","receivedAt":"2009-06-08T01:13:35Z","isPatch":false,"sender":{"key":"davvid@gmail.com","avatar":"https://avatars.githubusercontent.com/u/13196?v=4"},"body":"On Sun, Jun 07, 2009 at 02:49:16PM +0200, Andreas Färber wrote:\n>\n> Am 07.06.2009 um 14:32 schrieb Andreas Färber:\n>\n>> The merge commit b11cf09043f18b368ec0d988f064ea21247c843d introduced a \n>> git_mkstemps function in path.c, using mkstemps for the  \n>> implementation. Unlike mkstemp that function is not part of POSIX, so \n>> we seem to need a NO_MKSTEMPS option and corresponding alternative code \n>> path.\n>>\n>> Any suggestions how to best replace mkstemps?\n>\n> Sorry for the noise... NO_MKSTEMPS=YesPlease does work correctly if  \n> typed correctly. (Thanks for the hint!)\n\nThanks for keeping an eye on it.\n\nI'm still keeping my eye on the patch that introduced the\nbasename() compat function to see if we missed any Makefile\nsettings for the non-linux platforms.  We support a lot of\nplatforms! =)\n\n\n-- \n\n\tDavid\n"}]}