CI: Skip permissions test when running as sudo

[TomAugspurger]

Closes #33210

[TomAugspurger]

Is this a known failure?

=================================== FAILURES ===================================

__________________ TestHDFStore.test_append_with_data_columns __________________

[gw0] linux -- Python 3.6.6 /home/travis/miniconda3/envs/pandas-dev/bin/python

self = <pandas.tests.io.pytables.test_store.TestHDFStore object at 0x7fbb644fd588>

setup_path = 'tmp.__PzdfGEaMNt__.h5'

    def test_append_with_data_columns(self, setup_path):
...
            result = store.select(

                "df", "string='foo' and string2='foo' and A>0 and B<0"

            )

            expected = df_new[

                (df_new.string == "foo")

                & (df_new.string2 == "foo")

                & (df_new.A > 0)

                & (df_new.B < 0)

            ]

>           tm.assert_frame_equal(result, expected, check_index_type=False)

E           AssertionError: (None, <17 * BusinessDays>)

Restarting CI to see.

[jbrockmendel]

Looks like the Travis failure is from one of the newly-implemented freq checks. Should be unrelated. I'll try to track that down.

[WillAyd]

Is chmod(0) even valid? Wonder if we are just using the wrong bit setting and instead shouldn't do something like chmod("u-r")

[TomAugspurger]

On my system, that results in a file with these permissions.

----------  1 taugspurger  wheel     0B Apr 28 08:22 a

[WillAyd]

Yea not sure that is specified by POSIX though; at least according to Wikipedia I don’t see anything about that, so perhaps the lack of a standard could explain why it doesn’t appear to work on all systems: https://en.wikipedia.org/wiki/Chmod <https://en.wikipedia.org/wiki/Chmod>

On Apr 28, 2020, at 8:57 AM, Tom Augspurger ***@***.***> wrote: On my system, that results in a file with these permissions. ---------- 1 taugspurger wheel 0B Apr 28 08:22 a — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#33847 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAEU4UL2G4HBCRU66QU6ZSDRO34FBANCNFSM4MS2RIUA>.

William Ayd [email protected]

[jbrockmendel]

could do an os.geteuid() == 0 check?

[TomAugspurger]

I'm happy with the skip as written.

…

On Tue, Apr 28, 2020 at 11:03 AM jbrockmendel ***@***.***> wrote: could do an os.geteuid() == 0 check? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#33847 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAKAOIWGFTKCXOGQP43TAQDRO3433ANCNFSM4MS2RIUA> .

[jreback]

this lgtm.

[WillAyd]

Thanks @TomAugspurger

[simonjayhawkins]

@meeseeksdev backport to 1.0.x

[lumberbot-app]

Owee, I'm MrMeeseeks, Look at me.

There seem to be a conflict, please backport manually. Here are approximate instructions:

1. Checkout backport branch and update it.

$ git checkout 1.0.x
$ git pull

2. Cherry pick the first parent branch of the this PR on top of the older branch:

$ git cherry-pick -m1 12c9c626dc30a8d0f1c2305c3fdf8cc3e9acbcde

3. You will likely have some merge/cherry-pick conflict here, fix them and commit:

$ git commit -am 'Backport PR #33847: CI: Skip permissions test when running as sudo'

4. Push to a named branch :

git push YOURFORK 1.0.x:auto-backport-of-pr-33847-on-1.0.x

5. Create a PR against branch 1.0.x, I would have named this PR:

"Backport PR #33847 on branch 1.0.x"

And apply the correct labels and milestones.

Congratulation you did some good work ! Hopefully your backport PR will be tested by the continuous integration and merged soon!

If these instruction are inaccurate, feel free to suggest an improvement.

[simonjayhawkins]

xref #33300 (comment)