{"thread":{"id":"50553","subject":"[Suggestion] Add Skip for t9020","startedAt":"2019-02-21T20:28:24Z","lastAt":"2019-02-22T15:30:58Z","messageCount":5,"participants":["Randall S. Becker","Todd Zullinger","Jonathan Nieder","Jeff King"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"369848","messageId":"004b01d4ca23$faebcad0$f0c36070$@nexbridge.com","threadId":"50553","inReplyTo":null,"subject":"[Suggestion] Add Skip for t9020","fromName":"Randall S. Becker","fromEmail":"rsbecker@nexbridge.com","sentAt":"2019-02-21T20:28:12Z","receivedAt":"2019-02-21T20:28:24Z","isPatch":false,"sender":{"key":"randall.becker@nexbridge.ca","avatar":"https://avatars.githubusercontent.com/u/28956764?v=4"},"body":"On February 21, 2019 15:00, I wrote:\n> To: 'Junio C Hamano' <gitster@pobox.com>\n> Cc: git@vger.kernel.org; 'Johannes Schindelin'\n> <johannes.schindelin@gmx.de>; 'Duy Nguyen' <pclouds@gmail.com>\n> Subject: RE: [ANNOUNCE] Git v2.21.0-rc2\n> \n> On February 21, 2019 10:55, I wrote:\n> > On February 20, 2019 14:42, Junio C Hamano wrote:\n> > > \"Randall S. Becker\" <rsbecker@nexbridge.com> writes:\n> > >\n> > > > On February 19, 2019 18:29, Junio C Hamano wrote:\n> > > >> A release candidate Git v2.21.0-rc2 is now available for testing\n> > > >> at the usual places.  It is comprised of 474 non-merge commits\n> > > >> since v2.20.0, contributed by 61 people, 16 of which are new faces.\n> > > >\n> > > > Thanks. t5562 works properly on NonStop (3 tests under various\nloads).\n> > > > I am starting a full regression test now - see you in two days or\n> > > > so with a full report.\n> > >\n> > > Thanks.  In the end I ended up doing the following:\n> > >\n> > >  - kept most of your \"feed NULs generated from a helper program via\n> > >    pipe, instead of reading from /dev/zero\" patch,\n> > >\n> > >  - replaced the helper program used in the above with the test-tool\n> > >    one by Dscho,\n> > >\n> > >  - used Max's \"do not reuse the output file---the background writer\n> > >    about to die may still have it open when the next test starts\",\n> > >\n> > >  - used Max's \"when testing the program with overly large input\n> > >    declared with content-length, do not have to feed from /dev/zero;\n> > >    /dev/null suffices as a mistaken program attempting to read would\n> > >    die when seeing EOF anyway\" patch to reduce one mention of\n> > >    /dev/zero (or one pipe from the helper output) further.\n> > >\n> > > Hopefully Dscho's Azure thing would also be happy with one less use\n> > > of Perl script.\n> > >\n> > > As I said earlier, a few places in t/ still mention /dev/zero and I\n> > > think it is a real bug on platforms without /dev/zero in t4153:\n> > >\n> > >     $ git grep -h dev/zero t/\n> > >             test_must_fail test_terminal git am --3way </dev/zero &&\n> > >             test_must_fail test_terminal git am --reject </dev/zero\n> > > &&\n> > >\n> > > Curiously, applying the attached patch and running the test on Linux\n> > > would notice that we do not have such a device file, and the use of\n> > > test_must_fail does *not* make it a-OK [*].  I am not sure why you\n> > > are not getting hit by the same issue.\n> > >\n> > > \tSide note: test_must_fail is to ensure that the command\n> > > \texits in a controlled way without segfault, so at the first\n> > > \tglance, replacing /dev/zero with something nonsensical and\n> > > \tnonexistent ought to make the command pass, but that is not\n> > > \tthe case.  The shell notices redirection failure and aborts\n> > > \tthe &&-chain.  This behaviour matches what we want, so I\n> > > \twon't complain ;-)\n> > >\n> > > Another mention of /dev/zero appears in t/helper/test-sha1.sh (not\n> > > to be confused with t/helper/test-sha1.c).  This seems to be run\n> > > only with an explicit \"make -C t/helper check-sha1\" request, so\n> > > perhaps nobody on your platform ran it to get hit by it.  I wonder\n> > > if anybody runs this on any platform, to be honest, though.  Duy\n> > > Cc'ed as the last person to touch that particular target---yes, I\n> > > know I was guilty of introducing it in b65bc21e (\"Makefile: add\n> > > framework to verify and\n> > bench sha1 implementations.\", 2006-06-24).\n> > >\n> > >  t/t4153-am-resume-override-opts.sh | 4 ++--\n> > >  1 file changed, 2 insertions(+), 2 deletions(-)\n> > >\n> > > diff --git a/t/t4153-am-resume-override-opts.sh b/t/t4153-am-resume-\n> > > override-opts.sh index 8ea22d1bcb..29ef22c94f 100755\n> > > --- a/t/t4153-am-resume-override-opts.sh\n> > > +++ b/t/t4153-am-resume-override-opts.sh\n> > > @@ -40,7 +40,7 @@ test_expect_success TTY '--3way overrides --no-\n> > 3way' '\n> > >\n> > >  \t# Applying side1 with am --3way will succeed due to the threeway-\n> > > merge.\n> > >  \t# Applying side2 will fail as --3way does not apply to it.\n> > > -\ttest_must_fail test_terminal git am --3way </dev/zero &&\n> > > +\ttest_must_fail test_terminal git am --3way </dev/zarro &&\n> > >  \ttest_path_is_dir .git/rebase-apply &&\n> > >  \ttest side1 = \"$(cat file2)\"\n> > >  '\n> > > @@ -94,7 +94,7 @@ test_expect_success TTY '--reject overrides --no-\n> > reject'\n> > > '\n> > >  \ttest_path_is_dir .git/rebase-apply &&\n> > >  \ttest_path_is_missing file.rej &&\n> > >\n> > > -\ttest_must_fail test_terminal git am --reject </dev/zero &&\n> > > +\ttest_must_fail test_terminal git am --reject </dev/zarro &&\n> > >  \ttest_path_is_dir .git/rebase-apply &&\n> > >  \ttest_path_is_file file.rej\n> > >  '\n> >\n> > Here's what we get with this patch (and without it):\n> >\n> > /home/git/git/t: sh t4153-am-resume-override-opts.sh ok 1 - setup ok 2\n> > # skip --3way overrides --no-3way (missing TTY) ok 3 - --no-quiet\n> > overrides\n> --\n> > quiet ok 4 - --signoff overrides --no-signoff ok 5 # skip --reject\n> overrides --\n> > no-reject (missing TTY) # passed all 5 test(s)\n> > 1..5\n> >\n> > Technically, we have TTY, but not that dev. The TTYs are not under\n> /dev/tty*,\n> > however.\n> \n> t9020 subtests 1,2,5,6 failed - Not new. unsurprising as there is no SVN\nor\n> perl with SVN module on platform. It might be useful to have a detection\nto\n> skip of Perl SVN is not present.\n\nWhile this is a bit of a hack, it might be useful for skipping t9020 in\nenvironments where the svn.remote package is not installed. I can make this\ninto a patch if this style is reasonable - guessing probably not and that\nthe REMOTE_SVN test should go elsewhere if it is called that.\n\ndiff --git a/t/t9020-remote-svn.sh b/t/t9020-remote-svn.sh\nindex 6fca08e5e3..31edf99371 100755\n--- a/t/t9020-remote-svn.sh\n+++ b/t/t9020-remote-svn.sh\n@@ -12,6 +12,12 @@ then\n        test_done\n fi\n\n+python -c \"import svn.remote\" 2>/dev/null >/dev/null\n+if [ $? -eq 0 ]; then\n+       test_set_prereq REMOTE_SVN\n+fi\n+if ! test_have_prereq REMOTE_SVN\n+then\n+       skip_all='skipping remote-svn tests, python svn.remote not\navailable'\n+       test_done\n+fi\n+\n\n# Override svnrdump with our simulator\n PATH=\"$HOME:$PATH\"\n export PATH PYTHON_PATH GIT_BUILD_DIR\n\n"},{"id":"369851","messageId":"20190221205837.GX16125@zaya.teonanacatl.net","threadId":"50553","inReplyTo":"004b01d4ca23$faebcad0$f0c36070$@nexbridge.com","subject":"Re: [Suggestion] Add Skip for t9020","fromName":"Todd Zullinger","fromEmail":"tmz@pobox.com","sentAt":"2019-02-21T20:58:37Z","receivedAt":"2019-02-21T20:58:43Z","isPatch":false,"sender":{"key":"tmz@pobox.com","avatar":"https://avatars.githubusercontent.com/u/806319?v=4"},"body":"Hi,\n\nRandall S. Becker wrote:\n> On February 21, 2019 15:00, I wrote:\n>> t9020 subtests 1,2,5,6 failed - Not new. unsurprising as there is no SVN or\n>> perl with SVN module on platform. It might be useful to have a detection to\n>> skip of Perl SVN is not present.\n> \n> While this is a bit of a hack, it might be useful for skipping t9020 in\n> environments where the svn.remote package is not installed. I can make this\n> into a patch if this style is reasonable - guessing probably not and that\n> the REMOTE_SVN test should go elsewhere if it is called that.\n\nJeff King sent an RFC patch which would remove this test and\nthe rest of the vcs-svn experiment in August[1].  Jonathan\nNieder replied as one user who would rather see it moved to\ncontrib/, so it was held off.\n\nWhether that has any impact on adding a way to skip all the\ntests here, I don't know.  Maybe it's a gentle nudge in\nfavor of moving them to contrib?\n\n[1] https://public-inbox.org/git/20180817190310.GA5360@sigill.intra.peff.net/\n\n-- \nTodd\n"},{"id":"369878","messageId":"20190222004659.GA167507@google.com","threadId":"50553","inReplyTo":"004b01d4ca23$faebcad0$f0c36070$@nexbridge.com","subject":"Re: [Suggestion] Add Skip for t9020","fromName":"Jonathan Nieder","fromEmail":"jrnieder@gmail.com","sentAt":"2019-02-22T00:46:59Z","receivedAt":"2019-02-22T00:47:04Z","isPatch":false,"sender":{"key":"jrnieder@gmail.com","avatar":"https://avatars.githubusercontent.com/u/281595?v=4"},"body":"Hi,\n\nRandall S. Becker wrote:\n\n> While this is a bit of a hack, it might be useful for skipping t9020 in\n> environments where the svn.remote package is not installed. I can make this\n> into a patch if this style is reasonable - guessing probably not and that\n> the REMOTE_SVN test should go elsewhere if it is called that.\n>\n> diff --git a/t/t9020-remote-svn.sh b/t/t9020-remote-svn.sh\n> index 6fca08e5e3..31edf99371 100755\n> --- a/t/t9020-remote-svn.sh\n> +++ b/t/t9020-remote-svn.sh\n> @@ -12,6 +12,12 @@ then\n>         test_done\n>  fi\n>\n> +python -c \"import svn.remote\" 2>/dev/null >/dev/null\n> +if [ $? -eq 0 ]; then\n> +       test_set_prereq REMOTE_SVN\n> +fi\n> +if ! test_have_prereq REMOTE_SVN\n> +then\n> +       skip_all='skipping remote-svn tests, python svn.remote not\n> available'\n> +       test_done\n> +fi\n\nInteresting.  Where do we use the svn.remote package?  I did a quick\ngrep and didn't find any instances.\n\nDo you have output from running \"./t9020-remote-svn.sh -v -i\"?\n\nThanks,\nJonathan\n"},{"id":"369879","messageId":"00ae01d4ca58$d0d8d0f0$728a72d0$@nexbridge.com","threadId":"50553","inReplyTo":"20190222004659.GA167507@google.com","subject":"RE: [Suggestion] Add Skip for t9020","fromName":"Randall S. Becker","fromEmail":"rsbecker@nexbridge.com","sentAt":"2019-02-22T02:46:25Z","receivedAt":"2019-02-22T02:46:45Z","isPatch":false,"sender":{"key":"randall.becker@nexbridge.ca","avatar":"https://avatars.githubusercontent.com/u/28956764?v=4"},"body":"On February 21, 2019 19:47, Jonathan Nieder  wrote:\n> Randall S. Becker wrote:\n> \n> > While this is a bit of a hack, it might be useful for skipping t9020\n> > in environments where the svn.remote package is not installed. I can\n> > make this into a patch if this style is reasonable - guessing probably\n> > not and that the REMOTE_SVN test should go elsewhere if it is called\nthat.\n> >\n> > diff --git a/t/t9020-remote-svn.sh b/t/t9020-remote-svn.sh index\n> > 6fca08e5e3..31edf99371 100755\n> > --- a/t/t9020-remote-svn.sh\n> > +++ b/t/t9020-remote-svn.sh\n> > @@ -12,6 +12,12 @@ then\n> >         test_done\n> >  fi\n> >\n> > +python -c \"import svn.remote\" 2>/dev/null >/dev/null if [ $? -eq 0 ];\n> > +then\n> > +       test_set_prereq REMOTE_SVN\n> > +fi\n> > +if ! test_have_prereq REMOTE_SVN\n> > +then\n> > +       skip_all='skipping remote-svn tests, python svn.remote not\n> > available'\n> > +       test_done\n> > +fi\n> \n> Interesting.  Where do we use the svn.remote package?  I did a quick grep\n> and didn't find any instances.\n\nWell, this was the remote svn test, so I figured I would be specific. We\ndon't have the svn package either so could restrict to that.\n\n> Do you have output from running \"./t9020-remote-svn.sh -v -i\"?\n\nInitialized empty Git repository in /home/git/git/t/trash\ndirectory.t9020-remote-svn/.git/\nexpecting success:\n        init_git &&\n        git fetch svnsim &&\n        test_cmp .git/refs/svn/svnsim/master .git/refs/remotes/svnsim/master\n&&\n        cp .git/refs/remotes/svnsim/master master.good\n\nInitialized empty Git repository in /home/git/git/t/trash\ndirectory.t9020-remote-svn/.git/\nfatal: error reading from fast-import: Resource temporarily unavailable\nprogress Imported commit 1.\nfatal: error while running fast-import\nnot ok 1 - simple fetch\n#\n#               init_git &&\n#               git fetch svnsim &&\n#               test_cmp .git/refs/svn/svnsim/master\n.git/refs/remotes/svnsim/master  &&\n#               cp .git/refs/remotes/svnsim/master master.good\n#\n\n\n"},{"id":"369929","messageId":"20190222153054.GD21524@sigill.intra.peff.net","threadId":"50553","inReplyTo":"20190221205837.GX16125@zaya.teonanacatl.net","subject":"Re: [Suggestion] Add Skip for t9020","fromName":"Jeff King","fromEmail":"peff@peff.net","sentAt":"2019-02-22T15:30:54Z","receivedAt":"2019-02-22T15:30:58Z","isPatch":false,"sender":{"key":"peff@peff.net","avatar":"https://avatars.githubusercontent.com/u/45925?v=4"},"body":"On Thu, Feb 21, 2019 at 03:58:37PM -0500, Todd Zullinger wrote:\n\n> > While this is a bit of a hack, it might be useful for skipping t9020 in\n> > environments where the svn.remote package is not installed. I can make this\n> > into a patch if this style is reasonable - guessing probably not and that\n> > the REMOTE_SVN test should go elsewhere if it is called that.\n> \n> Jeff King sent an RFC patch which would remove this test and\n> the rest of the vcs-svn experiment in August[1].  Jonathan\n> Nieder replied as one user who would rather see it moved to\n> contrib/, so it was held off.\n> \n> Whether that has any impact on adding a way to skip all the\n> tests here, I don't know.  Maybe it's a gentle nudge in\n> favor of moving them to contrib?\n\nI still hoped we could ditch them completely. :) As part of my\n-Wunused-parameter patches, I found that was an easy way to get rid of\nsome warnings. But I'd be OK if they were in contrib/, too.\n\n-Peff\n"}]}