On 1/3/20 12:35 PM, Neal Gompa wrote:
On Fri, Jan 3, 2020 at 4:01 AM Panu Matilainen
<pmatilai(a)redhat.com> wrote:
>
> On 1/2/20 5:29 PM, Neal Gompa wrote:
>> On Thu, Jan 2, 2020 at 10:14 AM Ben Cotton <bcotton(a)redhat.com> wrote:
>
>> For us, since we don't have the SUSE patches that make PreReq do
>> things, the way we'd declare this with upstream RPM features would be:
>
> Please don't spread misinformation, openSUSE doesn't have patches to
> make PreReq anyhow special. They merely undo the deprecation warning
> that is issued on PreReq. With or without that, a PreReq will be
> translated to Requires(pre,preun) to simulate what the original PreReq did.
>
Huh, for some reason I thought it also did something else...
>>
>> Requires: user(wwwrun)
>> OrderWithRequires: user(wwwrun)
>
> I'm not sure how one is supposed to read the above, but certainly one
> does NOT need both Requires and OrderWithRequires on the same thing in a
> package. Requires(pre) might be in order because for user/group you
> really want them installed first in case of loops, but this is details
> without
OtherWithRequires makes it so that if it's in the same transaction,
it'll get installed before this package. Otherwise it doesn't matter,
yes.
Um, what?
OrderWithRequires behaves exactly like Requires for *ordering*. There's
no additional mystery magic there.
OtherWithRequires can be used to affect ordering in the case no hard
dependency exists. With users, there's always a hard dependency, and
OrderWithRequires achieves nothing at all but obfusctation.
- Panu -