Re: anomalies during "make test"

[Chet Ramey]

On 10/26/23 10:50 PM, martin@kurahaupo.gen.nz wrote:

> uname output: Linux treehug.home.kurahaupo.gen.nz 5.15.0-73-generic #80-Ubuntu 
> SMP Mon May 15 15:18:26 UTC 2023 x86_64 x86_64 x86_64 GNU/Linux
> Machine Type: x86_64-pc-linux-gnu

I don't have this exact copy of Ubuntu handy, but I just ran the bash-5.2
test suite on

$ uname -a
Linux ip-172-29-1-186 5.15.0-1026-aws #30-Ubuntu SMP Wed Nov 23 14:15:21 
UTC 2022 x86_64 x86_64 x86_64 GNU/Linux


> The following tests produce unexpected output during "make test".
>
> These definitely appear to indicate bugs:
>
>         run-execscript  (the bashdb directory isn't created during test 
> installation, so debugger can't run)

None of these tests attempt to run the debugger. In fact, the string
`debug' doesn't appear in any of the scripts for this test.

>         run-read        (too few characters read)

A few other people have reported this, but I've never been able to
reproduce it. I changed the timeout duration for bash-5.3.

>         run-redir       (reports "bad N")

The test expects file descriptor 7 to be closed, and it is in my
environment. Maybe it's not in yours.

Maybe I should close it explicitly in the test script.

> This is a bug in the test, rather than in bash:
>
>         run-type        (incorrectly assumes "m" does not exist in $PATH)

Yes, I should use a name less likely to exist in someone's environment.

> These are just because I don't have the relevant locales installed:
>
>         run-glob-test
>         run-intl        (and the number of tests differs)

Yes, that's why the test scripts warn you about this.

--
``The lyf so short, the craft so long to lerne.'' - Chaucer
                 ``Ars longa, vita brevis'' - Hippocrates
Chet Ramey, UTech, CWRU    chet@case.edu    http://tiswww.cwru.edu/~chet/