Currently TAP plugin includes test logs. This changed several times in
the history, but mainly is not needed, so let's disable it by default
and add argument to enable it.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

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

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

The legacy disable value is currently transformed in a loop using the
same variable, which results in TypeError as the transformed output is
not a valid input. Let's fix it by only single transformation in case
`setstate` is not used.

With `setstate` the value can be safely overridden as the value comes
from `setstate`.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

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

* https://github.com/avocado-framework/avocado:
  Removing the deprecated AVOCADO_TEST_DATADIR

AVOCADO_TEST_DATADIR is deprecated.
Removing them from the examples and selftests.
Signed-off-by: NNarasimhan V <sim@linux.vnet.ibm.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>

The check for mock version is only applicable to Python 2 versions,
Python 3 is fine on the versions we support.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Instead of using integers or even strings, let's use a more Pythonic
interface to disable/enable idle states.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The Linux Kernel will not change the content of those files based on
locale.  So, let's stick with honest binary mode when reading those
files.

While at it, also change the regex patterns, as they wi'll be matched
agains /proc/* and /sys/* content, which should really be treated as
binary.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Make it more Pythonic, like return True/False and not needing to
manually compile regexes.

Reference: https://docs.python.org/3/library/re.html#re.compileSigned-off-by: NCleber Rosa <crosa@redhat.com>

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

* https://github.com/avocado-framework/avocado:
  test_utils_service: wrap long lines
  service.py: add mask, unmask and reset-failed to systemd
  service.py: multi-access safe call - bugfix
  service.py: pass runner to service manager

Signed-off-by: NRadek Duda <rduda@redhat.com>

reset-failed, mask, unmask options are added to service management
((majority of which are transerred from
https://github.com/avocado-framework/avocado-vt/blob/master/virttest/staging/service.py)
- see avocado-framework/avocado-vt#1565)
Signed-off-by: NRadek Duda <rduda@redhat.com>

If `service_manager` is called multiple times, its previous instance is
used. The problem arises if different runner is used. Because it leads
to situation when service commands are passed to wrong VM or service is
run as systemd insted of SysVinit or vice versa. Global variables,
where information from previous run were stored were thus removed as well as
then unused functions. Selftest for two different calls was also added.

Steps to invoke:

```
ssn_c = vm_client.vm.wait_for_login()
runner = remote.RemoteRunner(session=ssn_c)
vdagentd = service.specific_service_manager("spice-vdagentd", run=runner.run)
vdagentd.stop() #  send to vm_client
ssn_g = vm_guest.vm.wait_for_login()
runner = remote.RemoteRunner(session=ssn_g)
vdagentd = service.specific_service_manager("spice-vdagentd", run=runner.run)
vdagentd.stop() #  send to vm_client (again!)
```

As a trade-off this may affect speed when invoked in a loop since
`service_manager` won't be cached any more.
Signed-off-by: NRadek Duda <rduda@redhat.com>

If `run` parameter is not passed, `ServiceManager` is always run by
`process.run` while other runner could be used (e.g.
`remote.RemoteRunner().run`).
Signed-off-by: NRadek Duda <rduda@redhat.com>

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

* https://github.com/avocado-framework/avocado:
  Docs: reference Avocado-VT from Avocado's installation instructions

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

* https://github.com/avocado-framework/avocado:
  utils.pci: Treat incorrect value in get_slot_from_sysfs
  utils.pci: Avoid unnecessary match in get_slot_from_sysfs

Moved original whiteboard example test to selftests and then
created a simpler example code.

Reference: https://trello.com/c/sG6J0VpwSigned-off-by: NWainer dos Santos Moschetta <wainersm@redhat.com>

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

Add an Amazon Linux Probe, for both versions 1 and 2
of the distribution.
Signed-off-by: NLucas Meneghel Rodrigues <lookkas@gmail.com>

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

* https://github.com/avocado-framework/avocado:
  avocado.Test: remove once deprecated datadir
  avocado.Test: removal of srcdir

The `avocado.Test.datadir` attribute has been deprecated, and it's now
time to have it removed for good.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

PCI Bridges can be represented in lspci output in different ways.
But they are all of Class 06.
So, handling using class id instead, which is a better way.
Signed-off-by: NNarasimhan V <sim@linux.vnet.ibm.com>

After the deprecation, comes the removal.  Hopefully, "workdir" is now
being used in tests by now.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NLukáš Doktor <ldoktor@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>

The fabric-2.0.0 is a complete rework of fabric using different
approach and will require many changes to our code. Let's keep using the
1.x.y version for now.

The fabric3 is still only based on 1.x.y branch but let's also limit
that one as it is based on the original fabric so it's expected that
they might eventually come-up with 2.x.y py3 fork.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

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

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

* https://github.com/avocado-framework/avocado:
  get_slot_from_sysfs: accomodate more type of slot names in pci utils

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

* https://github.com/avocado-framework/avocado:
  utils.process: Ignore decode-errors when logging the output
  utils.process: Simplify GDBSubProcess docstring
  utils.process: Use the `astring.ENCODING` by default

It might happen that user runs process with encoding that is ritcher
than system one. This is fine by the process as internally it stores
encoded bytes object, but the drainer might fail attempting to log the
output.

Those failures should not be critical as, under described circumstances,
the unavailability of the output is expected and should not prevent the
execution. Let's simply replace those characters using 'replace' (the
'xmlcharrefreplace' might fail with ASCII coder) and keep processing the
data.

To simplify error handling new "astring.to_text" argument "errors" is
added.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Most of the arguments in GDBSubProcess are ignored, let's list all with
a simple generic note to be quickly recognized.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Recently we introduced the "preferred" encoding, which should be better
default than just `sys.getdefaultencoding`. Let's use it in the process
module.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>
Signed-off-by: NCleber Rosa <crosa@redhat.com>

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