Due to the deprecation of the self.skip(), the availability of the skip
decorators and the new CANCEL status, some internal improvements are
important to keep our behaviour sane and our own code sound and clean.

- The test.SkipTest class was renamed to test.MockingTest to be even more
  generic, intending to be overridden by sub-classes that make the test
  to end both with SKIP or CANCEL status. To keep it generic,
  test.MockingTest class will not SKIP the test if used directly anymore.

- The test.TimeOutSkipTest and test.ReplaySkipTest classes now are using
  the skip decorators instead of raising an exception in setUp(), since
  'skipping' the test means 'don't execute anything', not even the
  setUp().

- The test.DryRunTest class, which is expected to log itself in setUp()
  and then abort the test execution, is now using self.cancel() (instead
  of raising a SKIP exception), being now compliant with the concept
  that a SKIP test cannot execute anything.

- Selftests were adjusted accordingly.
Signed-off-by: NAmador Pahim <apahim@redhat.com>

Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

* https://github.com/avocado-framework/avocado:
  utils: Improve error message of pci.get_pci_class_name

When there is no device associated to given pci_address the error
message is a bit misleading. Let's report a better one.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

* https://github.com/avocado-framework/avocado:
  utils: Add "get_pci_class_name" method to pci.py

Signed-off-by: NAmador Pahim <apahim@redhat.com>

Test which are using this utility are failing as they start using
multipath as soon as services are re-satrted. But multipath need
some time to load the services completely. So to load the
multipaths properly we need to give some sleep time which resolve
the issue
Signed-off-by: NNaresh Bannoth <nbannoth@in.ibm.com>

Signed-off-by: NAmador Pahim <apahim@redhat.com>

Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NCleber Rosa <crosa@redhat.com>

Robot plugin was introduced without documentation. This PR adds the
basic information about the Robot plugin.
Signed-off-by: NAmador Pahim <apahim@redhat.com>

Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

* https://github.com/avocado-framework/avocado:
  SPEC file: include resultsdb subpackage
  resultsdb plugin: require avocado-framework

This packages the newly introduced resultsdb optional plugin.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

All optional plugins require the installation (or presence) of
"avocado-framework", so let's do that to the newly introduced
resultsdb plugin too.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

We have a new section "optional_plugins". Let's create a "results"
category, move the "html" plugin documentation there and link to this
place from the "Result formats" section.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

The ResultsDB expects the Testcase, not the Test ID. Even so, the
Testcase can be considered the combination of the test name plus the
variant.

While on this, this patch adds the parameters information to the
ResultsDB 'data' section.
Signed-off-by: NAmador Pahim <apahim@redhat.com>

Some spelling issues, broken link fixes and updates related to recent
changes in Avocado.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

The default for our configuration is currently /var/lib, which was not
reflected in Configuration documentation. This is not really important
as we only share it as "example configuration" but it might create
confussion.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

With the resultdb optional plugin new arguments were added which are
missing in our man pages.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Signed-off-by: NAmador Pahim <apahim@redhat.com>

The avocado-server now requires both, "requirements.txt" and
"requirements-selftests.txt" to run "make check".
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

When --html is used the original in-job-dir results are removed just
before the second render. This was probably due to typo as the original
location was (probably) intended for removal. This patch doesn't do that
because the user-defined dir can be shared dir with other user files so
let's just produce the results and keep the dirs intact (while keeping
the refresh in job-dir as that is guaranteed to be our directory and we
do know it can be removed).
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Given the PCI address, function will return corresponding
PCI Class name
Signed-off-by: NVenkat R B <vrbagal1@linux.vnet.ibm.com>

Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

* https://github.com/avocado-framework/avocado:
  resultsdb plugin

Propagate tests results to resultsdb.

Reference: https://trello.com/c/AcruJdgS/906-result-plugin-for-resultsdbSigned-off-by: NAmador Pahim <apahim@redhat.com>

Signed-off-by: NAmador Pahim <apahim@redhat.com>

As it has already been done to some other tests, two other tests have
been producing false positives.  This has been observed recently in
package builds in build farms.

Let's run them only when `make check-full` is called, which is
usually done at dedicated machines with plenty of resources.

Reference: https://kojipkgs.fedoraproject.org/work/tasks/9189/19059189/build.log
Reference: https://kojipkgs.fedoraproject.org/work/tasks/9445/19069445/build.logSigned-off-by: NCleber Rosa <crosa@redhat.com>

When one extracts a zip file with symlinks multiple times, the symlinks
are not replaced by the `extractall`, which results in our
`_update_zip_extra_attrs` workaround to try to create a symlink to the
content of the file, rather than to the destination. This patch checks
whether the link already exists, removes it and refresh it from the
zipfile.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

* https://github.com/avocado-framework/avocado:
  data_dir: use a directory intended for writable content

Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NAmador Pahim <apahim@redhat.com>

The FileLoader heuristic to find test is quite complicated and there was
one uncovered branch which did not report anything. Let's avoid this by
reporting `MissingTest` at the end of the execution unless we specify
otherwise (by return).
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

The MissingTest is yielded by TestLoaderProxy, but it was implemented
and described in FileLoader, which caused troubles when FileLoader was
not being used. This simple fix allows TestLoaderProxy to define basic
mappings and adds MissingTest to it.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Signed-off-by: NCleber Rosa <crosa@redhat.com>

Avocado's "data_dir" has always been meant for readable **and** writable
content.

Right now, this is broken because the system wide default for
"base_dir" (data_dir's parent) is set to /usr/share/avocado, and
consequently, "data_dir" is set to /usr/share/avocado/data.

Let's use "/var/lib/avocado" for the "base_dir", and
"/var/lib/avocado/data" for the "data_dir".  Content intended to be
read-only, such as example tests and wrappers, are still set to
/usr/share/avocado.

This has been tested with Avocado-VT, making the result of system
wide content, such as test providers and downloadable files be
put under "/var/lib/avocado/data/avocado-vt".

Reference: https://trello.com/c/KU1DvSAzSigned-off-by: NCleber Rosa <crosa@redhat.com>