This is just a commit to increase the coverage of the test case for jirahook webhook.

There are several pieces of the code that are not being tested.
Some simple checks in existing tests can solve that, but others require separated test cases.

This is not part of #589 sub tasks but it could be.

Signed-off-by: Julio Faracco <jfaracco@redhat.com>

See merge request !1599

Signed-off-by: Scott Weaver <scweaver@redhat.com>

See merge request !1752

This makes it easier to update in future. The main reason is that current owners.yaml provides also list of users, while in old one you'd need to walk over all subsystems.

Signed-off-by: Jan Stancek <jstancek@redhat.com>

See merge request !1757

There's currently no dist-git policy for any rhivos branches, so with the
distgit_ref for automotive source-git branches updated to point to the
appropriate rhivos-* branches in dist-git, we're now failing policy
checks. Add a per-branch flag to permit policy checks to pass w/no policy.

Signed-off-by: Jarod Wilson <jarod@redhat.com>

See merge request !1754

Signed-off-by: Jan Stancek <jstancek@redhat.com>

This makes it easier to update in future (and next commit).

Signed-off-by: Jan Stancek <jstancek@redhat.com>

There's currently no dist-git policy for any rhivos branches, so with the
distgit_ref for automotive source-git branches updated to point to the
appropriate rhivos-* branches in dist-git, we're now failing policy
checks. Add a per-branch flag to permit policy checks to pass w/no policy.

Signed-off-by: Jarod Wilson <jarod@redhat.com>

This branch actually flows into rhivos-1-main in internal dist-git.

Signed-off-by: Jarod Wilson <jarod@redhat.com>

See merge request !1753

This branch actually flows into rhivos-1-main in internal dist-git.

Signed-off-by: Jarod Wilson <jarod@redhat.com>

This reverts commit 7c723699.

Fixes #808 

Signed-off-by: Patrick Talbert <ptalbert@redhat.com>

Closes #808

See merge request !1749

Signed-off-by: Scott Weaver <scweaver@redhat.com>

This is the initial commit of the elnhhook.  

In the time since I opened the original draft of the elnhook, it has changed. Now the hook monitors for new config MRs rather than trigger on the push event. I've kept the original MR description below because the push event code still exists but has been disabled for now.

This webhook now evaluates MR events to determine if a new MR is a kernel-ark config MR and if so it will open an ELN Jira and update the MR description to include a JIRA tag. This hook can be tested using:

```
python3 -m webhook.elnhook --merge-request <gitlab-mr-link>
```

**v1 description:**

_As it is written now, this hook will only open a jira issue when triggered by a config branch push event. That config branch is created by the current functionality running in the kernel-ark scheduled job. That job merges each day with the upstream kernel and adds any new configs that come in from upstream._

_In the future (as described in issue #770), this hook may take over some of the other functionality currently in kernel-ark which is why it triggers on a branch push event._

_I've been testing this using a fake push event that I initially took from [gitlab's webhook log](https://gitlab.com/cki-project/kernel-ark/-/hooks/5222317/hook_logs/10332764157)_


```
$ python3 -m webhook.elnhook --json-message-file fake-push-event.json
```

Fixes: #770 

Signed-off-by: Scott Weaver <scweaver@redhat.com>

Closes #770

See merge request !1712

include RHEL11 and RHIVOS1. Cleanup some UI elements as well. Unify
"strictness" for searching across "ALL" components.

Signed-off-by: Derek Barbosa <debarbos@redhat.com>

See merge request !1750

include RHEL11 and RHIVOS1. Cleanup some UI elements as well. Unify
"strictness" for searching across "ALL" components.

Signed-off-by: Derek Barbosa <debarbos@redhat.com>

This does a few things...

1) Add .direnv to .gitignore, so we don't try to commit and push it
2) Add a bare-bones .envrc file, which has two load-if-exists for an
   /etc/kernel-workflow/envrc and a ~/.config/kernel-workflow/envrc
   additional env config file
3) Add a sample docs/kernel-workflow-envrc file that is more or less
   the snippet that was in the main README.md

With these in place, we can further minimize the amount of copy-pasta
people have to do, they just need to copy kernel-workflow-envrc into
~/.config/kernel-workflow/envrc and update it as needed, and things
should Just Work(tm).

Some of this is further prep work to make this easier to integrate
into the rhel-developer-toolbox as a module, with just about everything
auto-configured by ansible for you (including the quirky repos that
some of the hooks need).

[skip coverage check]

Signed-off-by: Jarod Wilson <jarod@redhat.com>

See merge request !1735

Signed-off-by: Jarod Wilson <jarod@redhat.com>

This lets you invoke update-repo with an empty bucket name and fetch all
the updates just using git, without ever touching awscli, should you so
choose (which is what's currently happening in the RDTx module). It also
adds a setup-remotes option to overlay all the remotes on top of a
freshly cloned linus repo.

Signed-off-by: Jarod Wilson <jarod@redhat.com>

This does a few things...

1) Add .direnv to .gitignore, so we don't try to commit and push it
2) Add a bare-bones .envrc file, which has two load-if-exists for an
   /etc/kernel-workflow/envrc and a ~/.config/kernel-workflow/envrc
   additional env config file
3) Add a sample docs/kernel-workflow-envrc file that is more or less
   the snippet that was in the main README.md

With these in place, we can further minimize the amount of copy-pasta
people have to do, they just need to copy kernel-workflow-envrc into
~/.config/kernel-workflow/envrc and update it as needed, and things
should Just Work(tm).

Some of this is further prep work to make this easier to integrate
into the rhel-developer-toolbox as a module, with just about everything
auto-configured by ansible for you (including the quirky repos that
some of the hooks need).

Signed-off-by: Jarod Wilson <jarod@redhat.com>

private part at: https://gitlab.cee.redhat.com/cki-project/deployment-all/-/merge_requests/3147



Signed-off-by: Jan Stancek <jstancek@redhat.com>

See merge request !1747

This reverts commit 7c723699.

Signed-off-by: Patrick Talbert <ptalbert@redhat.com>

If an item in the list is a string and does not being with `RHEL-` or
`CVE-` then it is not added to any of the buckets and is effectively
lost. Fix this by having the if block which handles strings always put
the item into `issues` or `cves`.

Fixes: #822 

Signed-off-by: Patrick Talbert <ptalbert@redhat.com>

Closes #822

See merge request !1748

If an item in the list is a string and does not being with `RHEL-` or
`CVE-` then it is not added to any of the buckets and is effectively
lost. Fix this by having the if block which handles strings always put
the item into `issues` or `cves`.

Signed-off-by: Patrick Talbert <ptalbert@redhat.com>

Signed-off-by: Jan Stancek <jstancek@redhat.com>

On kernel-ark, we've not been generating end-result diffs for Fedora
kernel config files, since we've only been running make dist-configs,
instead of make dist-configs-check. Switch it up so we report the Fedora
specific changes too.

Signed-off-by: Jarod Wilson <jarod@redhat.com>

See merge request !1744

In dist-git a policy of 'Anything Goes!' of course indicates there is no
policy for this branch. But cpc still tries to transform this into a
list of Token objects which it ends up not knowing how to evaluate.

Save everyone a lot of trouble by looking for this string and if found,
do not try to transform it into tokens. From the user point of view this
is ideal as we'll naturally report there being no policy.

Fixes: #820

Signed-off-by: Patrick Talbert <ptalbert@redhat.com>

Closes #820

See merge request !1746

In dist-git a policy of 'Anything Goes!' of course indicates there is no
policy for this branch. But cpc still tries to transform this into a
list of Token objects which it ends up not knowing how to evaluate.

Save everyone a lot of trouble by looking for this string and if found,
do not try to transform it into tokens. From the user point of view this
is ideal as we'll naturally report there being no policy.

Signed-off-by: Patrick Talbert <ptalbert@redhat.com>

Update the list of webhooks expected by the tests.

Signed-off-by: Scott Weaver <scweaver@redhat.com>

This initial set of unit tests target elnhook's
process_config_merge_request() function to ensure that new Jira issues
will be opened for new config MRs and that the elnhook will ignore MRs
that aren't "config MRs".

Signed-off-by: Scott Weaver <scweaver@redhat.com>

New tests to ensure that we only open issues when we're in
production/staging mode.

Signed-off-by: Scott Weaver <scweaver@redhat.com>

This updates the supporting files needed for the new elnhook.

Signed-off-by: Scott Weaver <scweaver@redhat.com>

This webhook is responsible for creating ELN jira issues for new configs
added to kernel-ark during the merge-window. We would like to open jira
issues in order to bring more awareness to config MR review requests and
allow teams to track their work.

In the future there might be additional functionality added to handle
new upstream commits added to the redhat/configs/pending-rhel directly
(replacing some of that functionality that is run from the kernel-ark
project today)

Signed-off-by: Scott Weaver <scweaver@redhat.com>

On kernel-ark, we've not been generating end-result diffs for Fedora
kernel config files, since we've only been running make dist-configs,
instead of make dist-configs-check. Switch it up so we report the Fedora
specific changes too.

Signed-off-by: Jarod Wilson <jarod@redhat.com>

In the case where a branch name includes forward slashes we won't be
able to use the existing branch name parsing. This removes the
"refs/heads/" prefix, keeping the remaining as the target branch name.
In case the branch ref isn't "refs/heads", the old parsing is kept.

This also adds simple wrappers to return the before and commits fields
along with updating the filter_branch() method to support the
match_to_projects webhook session setting. With push events we may want
to evaluate the source branch name in the hook itself and not bind it to
just the branches defined in rh_metadata.yaml.

Signed-off-by: Scott Weaver <scweaver@redhat.com>

This adds a generic method to open a new Jira issue and a wrapper to
open a new issue in the RHEL project.

Signed-off-by: Scott Weaver <scweaver@redhat.com>

Signed-off-by: Jan Stancek <jstancek@redhat.com>

See merge request !1739

Because we can cache temp_utils loader during tests,
this further improves runtime.

Signed-off-by: Jan Stancek <jstancek@redhat.com>

To save additional time on loading and parsing, cache yamls
based on input path. This doesn't cache loads made with
'contents' parameter, but there's little gain from doing that.

Signed-off-by: Jan Stancek <jstancek@redhat.com>

This brings all tests runtime combined from ~40 minutes down to 4 minutes
when I run it locally on my laptop. In gitlab pipeline it goes down
from usual 15 minutes to ~6.

Couple tests that depend on yaml validation, enable it just for
duration of those tests.

Signed-off-by: Jan Stancek <jstancek@redhat.com>

There's up to 100 discussions in each loop, so 5 repetitions of the
test should be enough - given that we run these tests for each MR
and also locally.

Signed-off-by: Jan Stancek <jstancek@redhat.com>

See merge request !1743

Part of: &123



Signed-off-by: Tales da Aparecida <tales.aparecida@redhat.com>

See merge request !1742