Use the new `Index.detect_runner` method that will give us the best
available runner for a requested one. To do so a new `pipeline.Runner`
class is introduced that stores the `meta.RunnerInfo` class for the
specific runner and the original name that was requested.
In the manifest loading and describing functions of the formats, use
`Index.detect_runner` to get the `RunnerInfo` for a requested runner
and then wrap it in a `pipeline.Runner` object, which is then passed
to the `Manifest.add_pipeline` method.
See also commit "meta: ability to auto-detect runner".
Adjust all test.
Instead of using a non-existing runner `org.osbuild.test` use an
existing one `org.osbuild.linux`. This prepares the switch to
using runner auto-detection, which will rely on existing runners.
Improve the file-enumeration to first check for all files ending in
`*.py`, and for everything else run `/bin/file` on it, and check for
python mime-types.
Note that we expect the caller to open all those files anyway, and thus
this should not any significant overhead, unless we end up with lots of
non-python files in the repository (which is highly unlikely). But even
then, `/bin/file` only reads the first few bytes of a file, and will
default if those are not sufficient to detect the file type. Hence, not
much overhead is expected by this.
This change will now start adding scripts in ./tools/ to the linter,
including osbuild-mpp.
Signed-off-by: David Rheinsberg <david.rheinsberg@gmail.com>
Update the osbuild-ci container and privdocker action to the most recent
builds.
This changes the ostree-image-tests since the Fedora update pulled in
util-linux/fdisk changes that align partition sizes. Hence, the ostree
tests need to be changed to have aligned partition sizes as well. For
more information, see:
commit 921c7da55ec78350e4067b3fd6b7de6f299106ee
Author: Karel Zak <kzak@redhat.com>
Date: Thu Jan 27 10:50:45 2022 +0100
libfdisk: (gpt) align size of partition by default
Signed-off-by: David Rheinsberg <david.rheinsberg@gmail.com>
Make isort print the diff so we can actually fix the problems. This
might be overly verbose when introducing isort, but for new changes it
is very nice to see the actual code it complains about.
Signed-off-by: David Rheinsberg <david.rheinsberg@gmail.com>
For some reasons I forgot to fix those in the previous runs. Fix a
linter and pep8 warning.
Signed-off-by: David Rheinsberg <david.rheinsberg@gmail.com>
Newer warning from pylint, also consistent with how we do things
elsewhere. Note that this only applies to one file in the tests but
disabling it would be very weird for such a small fix.
Fedora 34 is end of life and we ought to be testing with newer
releases so that we catch issues like the authselect one, that
is detailed in "stages/rpm: support marking install as ostree".
Convert the manifest to use version 2 format. Version 1 is really
not used much anymore because composer was completely ported to
using v2. Welcome to the future, ostree commit.
Include cryptsetup and lvm2 so that commit can be deployed to
images that are encrypted and have lvm2 layout. Additionally,
include greenboot. Remove the explicit dependency on ostree
and systemd-udev, which are automatically pulled in.
Instead of using `subprocess` with `check=True`, which will
echo the command used, including all the files it checked,
check the error code and if non-zero, fail the test with a
nice small error message.
This stage allows to manifests to define information to be put into a
built image. The information can then later be used by RHSM/Insights.
The facts are stored in `/usr/share/osbuild/rhsm.facts` and then
symlinked into `/etc/rhsm/facts/osbuild.facts`.
From IEEE Std 1003.1-2017:
3.437 User Name:
A string that is used to identify a user; see also User Database. To be
portable across systems conforming to POSIX.1-2017, the value is composed
of characters from the portable filename character set. The <hyphen-minus>
character should not be used as the first character of a portable user name.
3.282 Portable Filename Character Set
The set of characters from which portable filenames are constructed.
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
a b c d e f g h i j k l m n o p q r s t u v w x y z
0 1 2 3 4 5 6 7 8 9 . _ -
The last three characters are the <period>, <underscore>, and <hyphen-minus>
characters, respectively.
---
The regex checking the username in org.osbuild.users stage omitted the
<period> character before. This commit adds it and modifies the stage
test in order to cover this change.
Fixes: rhbz#2104464
New stage to create or modify `containers-storage.conf` config
files. The currently implement strategy is to merge the given
and existing configuration options.
The oscap.remediation stage now runs the
remediation during build time rather than
creating the necessary files to run the
remediation on first-boot. The tests have
been updated to reflect these changes.
Add a new stage to import public keys into the RPM database via the
`rpmkeys` command. This is similar as to what the `org.osbuild.rpm`
stage already does but it uses inputs instead of inline data.
This commit introduces a new vagrant stage and initially only supports
the libvirt backend. The vagrant stage takes a disk image as input
generates a Vagrantfile and metadata.json.
The idea of this test case was to check that two identical trees are
only stored once, via their treesum in the object store; but this
functionality was removed in commit e97f6ef34 and instead of treesums
random uuids are now used. As a result there is no de-duplication
anymore -- the subject of the test. So remove the test.
Currently we hard code the vpc options `subformat=fixed` and
`force_size`, which are needed to generate valid azure images
with newer versions of qemu. But for other use cases or other
versions of qemu these options might not be wanted or valid.
Expose all the options but with defaults corresponding to the
old behavior.
Add a unit test for the `force_size` option to check its
effect. Also add a check for the correct size to the existing,
default value (i.e. `force_size` being `true`).
Can be used to create partition tables via GPT laypout via `sgdisk(8)`.
The schema of `partitions` is intentionally kept identical to the one
in `org.osbuild.sfdisk`.
Add corresponding tests.
