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

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

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

All the plugin interfaces defined in `avocado.core.plugin_interfaces`
ommit the "plugin" term, simply because this is already qualified in
the module name.

Let's also standardize the Varianter interface to "plugin-less" name.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

variants.itertests() changed the format of each variant from a tuple to
a dictionary. After that change, the runner was not updated to provide
the mux_path considering the new format.

This patch adds the mux_path to the factory params in _iter_variants(),
fixing the Test.self.__params resolution.
Signed-off-by: NAmador Pahim <apahim@redhat.com>

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

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

Here the spell fixes for the v47 test plan.
Signed-off-by: NAmador Pahim <apahim@redhat.com>

man is quite outdated. This patch updates the man to reflect the
current options and UI.

Reference: https://trello.com/c/UB6hG2mDSigned-off-by: NAmador Pahim <apahim@redhat.com>

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

There are some old docstrings which are not much helpful, let's improve
them.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Finally after all the refactoring, refining and modifying here is the
up-to-date documentation which follows the current structure and tries
to describe both theoretically and practically how are test's parameters
produced and how they can be used.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Currently the variant format is (variant_id, (variant, mux_path)) which
is not really scalable, nor user friendly. Let's use a dictionary
instead.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

The "ignore_new_data" was used by "replay" to override Varianter and
tell it to stop accepting new data by other plugins (like yaml_to_mux).
With the new changes the plugins are part of the Varianter, so when
"replay" replaces it with the old (parsed) one, it will not parse itself
again, therefor the default_params and variants stays untouched.

To make sure this works well a unittest is included.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

The "mux-skip-defaults" is a generic concept and should be renamed to
"variants-skip-defaults" as the whole generic concept is called
variant/varianter. This change is only internal and not visible to
users.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

This patch replaces the hardcoded mux in Varianter for list of Varianter
plugins, iterating through all of them to produce all specified
variants. The interface is described in `avocado.core.plugin_interfaces`
and we also provide base-implementation for Mux-like plugins in
`avocado.core.mux`.

Mainly this PR is about defining the stevedore entry points, moving the
code around and wrapping the code into a class according the defined
stevedore interface. Doing so allowed us to simplify the `MuxTree`
object which does not need to keep track of the `root` of the tree,
because the root is stored in the pluggin's instance.

There is still one potential issue in Varianter which is probably even
more visible now, which is that we are pickling it in jobdata in order
to be able to reproduce the variants when re-running/diffing the job.
Before this was prone to code-changes but now it can be even worse as
multiple plugins can be enabled and can break loading of the previous
state. To avoid the potential breakage this version simply initializes
only the available extensions which should produce only the supported
variants.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

The idea behind default_params is, that some of the params are
considered the default_params, they can change, but are not part of
variant, nor the variant-id.

This implementation replaces the `Mux.data_inject` with
`Mux.add_default_param` which also accepts `name` to map data to
different plugins/groups in order to allow ignoring only some of them.

During parsing it extends multiplex variants of the defaults (merges
multiplex variants into defaults, therefor the mux overrides the values)
in order to provide the combination while yielding variants. Then when
there are variants, it yields the variant (compound of the variant and
the default_params) and variant-id. If there are no variants, it yields
only the default_params with no variant-id (None).
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Currently the Varianter itself calculates the number of tests, but in
the future we want to allow each variants plugin to report the number of
tests. Let's add support for `__len__` to the Mux plugin and utilize it
in Varianter instead of hardcoded iteration.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Instead of custom handling of human readable output let's create a
varianter method which produces it and use it from the core.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

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

* https://github.com/avocado-framework/avocado:
  yaml_to_mux: fix deprecation message

--filter-only/--filter-out are being deprecated by
--mux-filter-only/--mux-filter-out, but the deprecation
messages are pointing users to --mux-only/--mux-out, which
does not exist.

This patch fixes the deprecation messages, referencing the
correct new options.
Signed-off-by: NAmador Pahim <apahim@redhat.com>

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

With the last pbr update, stevedore 1.8.0 dependencies are broken.

This patch adjusts the requirements for travis and for docs to install
the latest stevedore.
Signed-off-by: NAmador Pahim <apahim@redhat.com>

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

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

* https://github.com/avocado-framework/avocado:
  docs: update the examples according to current UI

This change should decrease the probability of users accidentally
overriding `Test.params` while keeping the usage the same. There is one
caveat and that is Avocado-vt, which modifies `self.params`, which is
not that simple anymore. I have a fix for it, but the problem is it is
not backward compatible and it can only be changed to
forward-compatible. We should consider whether the benefit of having
this "a bit" protected is greater than backward incompatible change with
Avocado-vt (and possibly other "nasty" user's tests)
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

We removed several classes in the history, let's update the get_state to
remove methods, that are not used anymore.

Note that some of the variables were related to Avocado-vt, anyway the
usage of `get_state` from Avocado depends only on it's values. If
Avocado-vt needs to publish it's internal values, it can override this
method and do so, anyway at this point this is not necessary.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Publishing of text_output doesn't makes much sense as what it does is
basically just open `self.logfile` and writes the content into memory.
This was only used by `xunit` plugin and this plugin can simply open the
file and read it.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

The runner queue is essential part of runner->test communication. Let's
decrease the probability of user accidentally replacing the runner queue
by turning it into a property.

For safety reasons our runner also needs to set this property from
outside (if it's not set by the test_loader) so this patch also includes
a method `set_runner_queue` which allows overriding the runner queue
from outside, but we only allow this once to avoid unwanted changes.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

The 'TESTS' line was removed from the UI. Let's update the examples in
our docs accordingly.
Signed-off-by: NAmador Pahim <apahim@redhat.com>

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

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

mkfs function in partition.py is failing to create a btrfs file
system when already a file system exists on the given disk.
adding -f option will resolve the issue. So added the same and
it is working fine now.
Signed-off-by: NNaresh Bannoth <nbannoth@in.ibm.com>

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

It is possible to set relative paths in settings, but once we evaluate
them we should turn them into absolute paths to avoid incorrect
representation on `os.chdir` (or so)
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

This should help beginners to identify the issue when accidentally
overriding test's public variables. Also add a unittest to support this
note.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

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

* https://github.com/avocado-framework/avocado:
  SIGTERM handler
  avocado.utils.process: recursive get_children_pids()

To avoid accidental write into test-traceback-related variables, let's
turn them into properties. This makes it harder to manually fail the
test with custom exceptions, but I consider that ugly and one should
simply raise the exception and the runner should take care of setting
the class, traceback and details.

To prove the direct access is not necessary I tried this with
Avocado-vt which works without any modifications even though it uses
custom exception handling.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

I'm not 100% sure, but I don't see a benefit in allowing people to
override test status from inside test. They are suppose to use
assertions, or `self.fail`-like methods. With this change it'd be harder
to manually override the test status (as can be seen on updated
unittests) but we minimize the accidental write into `self.status`
variable, which can lead to confusing results.

Note we can consider adding `set_status` method to allow setting this
property, but I'd like to avoid allowing direct `self.status = ` usage.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

The `debugdir` and `resultsdir` are not used in Avocado. They were
inherited from Avocado-vt, but there is no point in defining them in
Avocado itself as Avocado-vt creates them anyway.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>