{"thread":{"id":"32744","subject":"[RFC] test-lib.sh: No POSIXPERM for cygwin","startedAt":"2013-01-27T14:57:08Z","lastAt":"2013-02-08T06:08:14Z","messageCount":6,"participants":["Torsten Bögershausen","Erik Faye-Lund","Ramsay Jones","Junio C Hamano"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"207959","messageId":"201301271557.08994.tboegi@web.de","threadId":"32744","inReplyTo":null,"subject":"[RFC] test-lib.sh: No POSIXPERM for cygwin","fromName":"Torsten Bögershausen","fromEmail":"tboegi@web.de","sentAt":"2013-01-27T14:57:08Z","receivedAt":"2013-01-27T14:57:08Z","isPatch":false,"sender":{"key":"tboegi@web.de","avatar":"https://avatars.githubusercontent.com/u/7138363?v=4"},"body":"t0070 and t1301 fail when running the test suite under cygwin.\nSkip the failing tests by unsetting POSIXPERM.\n\nSigned-off-by: Torsten Bögershausen \n---\n t/test-lib.sh | 1 -\n 1 file changed, 1 deletion(-)\n\ndiff --git a/t/test-lib.sh b/t/test-lib.sh\nindex 1a6c4ab..94b097e 100644\n--- a/t/test-lib.sh\n+++ b/t/test-lib.sh\n@@ -669,7 +669,6 @@ case $(uname -s) in\n \ttest_set_prereq SED_STRIPS_CR\n \t;;\n *CYGWIN*)\n-\ttest_set_prereq POSIXPERM\n \ttest_set_prereq EXECKEEPSPID\n \ttest_set_prereq NOT_MINGW\n \ttest_set_prereq SED_STRIPS_CR\n-- \n1.8.1.1\n"},{"id":"208790","messageId":"CABPQNSYmGhaDG0Dzpp2C0ZwKoFwP5EokJW+TC8JT-Bw6rKa79A@mail.gmail.com","threadId":"32744","inReplyTo":"201301271557.08994.tboegi@web.de","subject":"Re: [RFC] test-lib.sh: No POSIXPERM for cygwin","fromName":"Erik Faye-Lund","fromEmail":"kusmabite@gmail.com","sentAt":"2013-02-06T09:34:45Z","receivedAt":"2013-02-06T09:34:45Z","isPatch":false,"sender":{"key":"kusmabite@gmail.com","avatar":"https://avatars.githubusercontent.com/u/47073?v=4"},"body":"On Sun, Jan 27, 2013 at 3:57 PM, Torsten Bögershausen wrote:\n> t0070 and t1301 fail when running the test suite under cygwin.\n> Skip the failing tests by unsetting POSIXPERM.\n>\n\nBut is this the real reason? I thought Cygwin implemented POSIX permissions...?\n"},{"id":"208853","messageId":"5112BA0D.6020200@web.de","threadId":"32744","inReplyTo":"CABPQNSYmGhaDG0Dzpp2C0ZwKoFwP5EokJW+TC8JT-Bw6rKa79A@mail.gmail.com","subject":"Re: [RFC] test-lib.sh: No POSIXPERM for cygwin","fromName":"Torsten Bögershausen","fromEmail":"tboegi@web.de","sentAt":"2013-02-06T20:16:13Z","receivedAt":"2013-02-06T20:16:13Z","isPatch":false,"sender":{"key":"tboegi@web.de","avatar":"https://avatars.githubusercontent.com/u/7138363?v=4"},"body":"Am 2013-02-06 10:34, schrieb Erik Faye-Lund:\n> On Sun, Jan 27, 2013 at 3:57 PM, Torsten Bögershausen wrote:\n>> t0070 and t1301 fail when running the test suite under cygwin.\n>> Skip the failing tests by unsetting POSIXPERM.\n>>\n>\n> But is this the real reason? I thought Cygwin implemented POSIX permissions...?\nt0070:\n 'mktemp to unwritable directory prints filename'\n mkdir cannotwrite &&\n chmod -w cannotwrite &&\n test_when_finished \"chmod +w cannotwrite\" &&\n test_must_fail test-mktemp cannotwrite/testXXXXXX 2>err &&\n grep \"cannotwrite/test\" err\n\nWhen a directory under Linux/*nix has no write permission,\nit is not allowed to create another directory (or file..) here.\nThis is not working under cygwin, a directory/file can be created\neven if the parent directory has chmod 0.\n-------------\ntb@PC /cygdrive/c/temp\n$ mkdir ttt\n\ntb@PC /cygdrive/c/temp\n$ chmod 0 ttt\n\ntb@PC /cygdrive/c/temp\n$ ls -ld ttt\nd---------+ 1 tb None 0 Feb 6 20:33 ttt\n\ntb@PC /cygdrive/c/temp\n$ touch ttt/x\n\ntb@PC /cygdrive/c/temp\n$ ls -ld ttt\nd---------+ 1 tb None 0 Feb 6 20:33 ttt\n\ntb@PC /cygdrive/c/temp\n$ ls -l ttt\ntotal 0\n-rw-r--r--+ 1 tb None 0 Feb 6 20:33 x\n-------------------------------------------\n\nIf this is POSIX compliant? I'm not an expert here.\nOn the other hand:\nThis test case does not test git, but rather the file system,\nso we can probaly remove it?\n\nAbout 1301:\nSome resereach needs to be done, to find out the connection between\numask, cygwin and the mount options.\n\nOn my system I have:\n$mount\nC: on /cygdrive/c type ntfs (binary,posix=0,user,noumount,auto)\n\n/Torsten\n"},{"id":"208929","messageId":"5113F1B1.3010102@ramsay1.demon.co.uk","threadId":"32744","inReplyTo":"201301271557.08994.tboegi@web.de","subject":"Re: [RFC] test-lib.sh: No POSIXPERM for cygwin","fromName":"Ramsay Jones","fromEmail":"ramsay@ramsay1.demon.co.uk","sentAt":"2013-02-07T18:25:53Z","receivedAt":"2013-02-07T18:25:53Z","isPatch":false,"sender":{"key":"ramsay@ramsayjones.plus.com","avatar":"https://avatars.githubusercontent.com/u/33702710?v=4"},"body":"Torsten Bögershausen wrote:\n> t0070 and t1301 fail when running the test suite under cygwin.\n> Skip the failing tests by unsetting POSIXPERM.\n\nt1301 does not fail for me. (WIN XP (SP3) on NTFS)\n[It's so long since I looked, but I'm pretty sure that the failure\nin t0070 is caused by *git*, not by cygwin not supporting POSIXPERM]\n\n> Signed-off-by: Torsten Bögershausen \n> ---\n> t/test-lib.sh | 1 -\n> 1 file changed, 1 deletion(-)\n> \n> diff --git a/t/test-lib.sh b/t/test-lib.sh\n> index 1a6c4ab..94b097e 100644\n> --- a/t/test-lib.sh\n> +++ b/t/test-lib.sh\n> @@ -669,7 +669,6 @@ case $(uname -s) in\n> \ttest_set_prereq SED_STRIPS_CR\n> \t;;\n> *CYGWIN*)\n> -\ttest_set_prereq POSIXPERM\n> \ttest_set_prereq EXECKEEPSPID\n> \ttest_set_prereq NOT_MINGW\n> \ttest_set_prereq SED_STRIPS_CR\n> \n\nSo, I'm not in favour of this, FWIW.\n\nATB,\nRamsay Jones\n"},{"id":"208937","messageId":"7vd2wb7w96.fsf@alter.siamese.dyndns.org","threadId":"32744","inReplyTo":"5113F1B1.3010102@ramsay1.demon.co.uk","subject":"Re: [RFC] test-lib.sh: No POSIXPERM for cygwin","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2013-02-07T19:35:17Z","receivedAt":"2013-02-07T19:35:17Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Ramsay Jones writes:\n\n> Torsten Bögershausen wrote:\n>> t0070 and t1301 fail when running the test suite under cygwin.\n>> Skip the failing tests by unsetting POSIXPERM.\n>\n> t1301 does not fail for me. (WIN XP (SP3) on NTFS)\n\nOthers run Cygwin with vfat or some other filesystem, and some of\nthem do not cope will with POSIXPERM, perhaps?\n\nNot having POSIXPERM by default for Cygwin may be a saner default\nthan having one, if we have to pick one.\n\nIt may be debatable to have this default as platform attribute,\nthough.\n\n>> Signed-off-by: Torsten Bögershausen \n>> ---\n>> t/test-lib.sh | 1 -\n>> 1 file changed, 1 deletion(-)\n>> \n>> diff --git a/t/test-lib.sh b/t/test-lib.sh\n>> index 1a6c4ab..94b097e 100644\n>> --- a/t/test-lib.sh\n>> +++ b/t/test-lib.sh\n>> @@ -669,7 +669,6 @@ case $(uname -s) in\n>> \ttest_set_prereq SED_STRIPS_CR\n>> \t;;\n>> *CYGWIN*)\n>> -\ttest_set_prereq POSIXPERM\n>> \ttest_set_prereq EXECKEEPSPID\n>> \ttest_set_prereq NOT_MINGW\n>> \ttest_set_prereq SED_STRIPS_CR\n>> \n>\n> So, I'm not in favour of this, FWIW.\n>\n> ATB,\n> Ramsay Jones\n"},{"id":"208975","messageId":"5114964E.9010609@web.de","threadId":"32744","inReplyTo":"7vd2wb7w96.fsf@alter.siamese.dyndns.org","subject":"Re: [RFC] test-lib.sh: No POSIXPERM for cygwin","fromName":"Torsten Bögershausen","fromEmail":"tboegi@web.de","sentAt":"2013-02-08T06:08:14Z","receivedAt":"2013-02-08T06:08:14Z","isPatch":false,"sender":{"key":"tboegi@web.de","avatar":"https://avatars.githubusercontent.com/u/7138363?v=4"},"body":"On 07.02.13 20:35, Junio C Hamano wrote:\n> Ramsay Jones writes:\n>\n>> Torsten Bögershausen wrote:\n>>> t0070 and t1301 fail when running the test suite under cygwin.\n>>> Skip the failing tests by unsetting POSIXPERM.\n>> t1301 does not fail for me. (WIN XP (SP3) on NTFS)\n> Others run Cygwin with vfat or some other filesystem, and some of\n> them do not cope will with POSIXPERM, perhaps?\n>\n> Not having POSIXPERM by default for Cygwin may be a saner default\n> than having one, if we have to pick one.\n>\n> It may be debatable to have this default as platform attribute,\n> though.\n>\nYes, 1301 passes on cygwin 1.5, but not on 1.7.\nAnd it breaks on VFAT, for all kind of OS.\n\nThanks for comments, I'll put more investigations on my todo stack.\n/Torsten\n"}]}