On Wed, May 4, 2022 at 1:00 PM Miro Hrončok <mhroncok@xxxxxxxxxx> wrote:
On 04. 05. 22 15:40, Troy Dawson wrote:
> I just want to make sure that you are querying RHEL 9.0 and not 9.1.
Well, I was querying 9.1, so you have a point. However, with 9.0 it is almost
the same:
$ comm -12 <(repoquery -q
--repo=RHEL9.0-{BaseOS,AppStream,CRB,HighAvailability} -a | pkgname | sort |
uniq ) <(repoquery -q --repo=epel9 -a | pkgname | sort | uniq)
libwmf
libwmf-lite
pybind11-devel
python3-pybind11
tesseract
tesseract-devel
tesseract-langpack-eng
tesseract-tessdata-doc
$ comm -12 <(repoquery -q
--repo=RHEL9.0-{BaseOS,AppStream,CRB,HighAvailability}-source -a | pkgname |
sort | uniq ) <(repoquery -q --repo=epel9-source -a | pkgname | sort | uniq)
libwmf
pybind11
tesseract
tesseract-tessdata
The only difference I can spot is anthy-unicode-devel and
double-conversion-devel, which apparently might be allowed in EPEL 9 for now.
There is now a new function added to the "Will-It" family. "Will-It-Duplicate"
I figured I already had 80% of what was needed already written, so why not.
There are indeed duplicates of upstream in both epel9 and epel8.
It also does epel7, but right now there seems to be a bug in that.
So, now we can easily see the duplicate packages to remove them.
And, we'll get them removed. Removing is much easier than adding.
We are still investigating how this happened for pybind11, tesseract and tesseract-tessdata. Those shouldn't have been able to be branched.
For libwmf it seems it was a race condition.
The new RHEL workflow checks epel9 to see if a package is already there. At the time work started on libwmf, it wasn't in epel9, so it didn't create a bug.
Then libwmf was branched in epel9, because libwmf wasn't in CentOS Stream 9 yet.
They both came out within a couple weeks of each other.
The odds of that happening again are pretty slim, and the effort to fix that sort of race condition is pretty high.
So we'll just let Will-It-Duplicate find those race condition problems.
Troy
_______________________________________________ epel-devel mailing list -- epel-devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to epel-devel-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/epel-devel@xxxxxxxxxxxxxxxxxxxxxxx Do not reply to spam on the list, report it: https://pagure.io/fedora-infrastructure
