[Bug 1288731] Review Request: os-autoinst - OS-level test automation

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



https://bugzilla.redhat.com/show_bug.cgi?id=1288731



--- Comment #14 from awilliam@xxxxxxxxxx <awilliam@xxxxxxxxxx> ---
"- /usr/libexec/os-autoinst/dmidata calls /usr/bin/dmidecode, do you need
R:dmidecode?"

no, I don't think so; the tool seems to be there only for admins to use
manually in a fairly specific situation (you want to have the VM pretend to be
a laptop) - see
https://github.com/os-autoinst/os-autoinst/commit/0b8f9bf75558add140a4d455a9f857fde6644127
. A requires doesn't seem reasonable.

os-autoinst 4.3 came out recently. I've rebased the spec to that, and addressed
a lot of the issues raised. However, I noticed a significant problem with the
'perl bits in libexecdir stuff' which I hadn't noticed before. The perl
auto-provides / auto-requires filters produce provides and requires for all
those perl bits, so os-autoinst has a bunch of bogus provides: and requires:
like this:

    perl(autotest)
    perl(backend::baseclass)
    perl(backend::console_proxy)
    perl(backend::driver)
    perl(backend::ipmi)
    perl(backend::qemu)
    perl(backend::s390x)
    perl(backend::svirt)

etc etc. They're bogus because they're not in the standard location, obviously
- any other package which actually wanted a perl module called backend::qemu or
whatever wouldn't actually be able to load it in the standard fashion
(iso2video - the main executable in os-autoinst - futzes about with the perl
include path to load the modules).

So I'm inclined to agree with you that the installation to libexecdir should
just be considered wrong, or at the very least, a bad idea (I'm guessing SUSE
doesn't use perl auto-provides / auto-requires). I've opened a ticket upstream
to discuss this:

https://github.com/os-autoinst/os-autoinst/issues/387

Here's the latest spec and .src.rpm. For now I'm just leaving the bogus
provides/requires in, so at least the package *works*, but I agree it probably
shouldn't pass review until this is resolved.

https://www.happyassassin.net/reviews/os-autoinst/os-autoinst.spec
https://www.happyassassin.net/reviews/os-autoinst/os-autoinst-4.3-1.fc24.src.rpm

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are always notified about changes to this product and component
_______________________________________________
package-review mailing list
package-review@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/package-review




[Index of Archives]     [Fedora Legacy]     [Fedora Desktop]     [Fedora SELinux]     [Yosemite News]     [KDE Users]     [Fedora Tools]