Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

Similar to the approach taken on the previous commits, let's
unwind apply_using and improve its docstrings.

This also new tests for this specific unit of code, and a functional
test.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The function where this utility method lives is really big and
complex.  Let's break it down for the sake of code readability,
while adding better docstrings.

I've timed the performance changes caused by this, and on most
trials they presented similar performance.

This also adds new tests for this specific unit of code.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The function where this utility method lives is really big and
complex.  Let's break it down for the sake of code readability,
while adding better docstrings.

I've timed the performance changes caused by this, and on most
trials they presented similar performance.

This also adds new tests for this specific unit of code.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The action is the same, only the condition changes, so let's use a
membership test for the type of the control tag.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The function where this utility method lives is really big and
complex.  Let's break it down for the sake of code readability,
while adding better docstrings.  At the same time, this allows
us to write tests for those smaller units.

I've timed the performance changes caused by this, and on all
trials they presented slightly better (but overall insignifcant)
improvements.  This is just mentioned to assert that there are
no performance downsides to this change.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

The changes in this commit are causing intermittent failures in our CI
environment under Python 3.x.

This reverts commit 55f420f7.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

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

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

The HTMLParser class on Python 3 expects to receive text, and urlopen
returns bytes.  To convert bytes into text, an encoding is needed.

Most providers utf-8 encoding on the responses, but Fedora's servers
use ISO-8859-1:

   HTTP/1.1 200 OK
   Date: Wed, 19 Sep 2018 00:46:43 GMT
   Server: Apache/2.4.6 (Red Hat Enterprise Linux) OpenSSL/1.0.2k-fips
   ...
   Content-Type: text/html;charset=ISO-8859-1
   ...

We could get away with utf-8, but let's be thorough here.  This fixes
a "must be str, not bytes" error on Python 3.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

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

Which doesn't depend on running `ps`, but is still Linux specific.  In
the future, it'd be nice to have a multi-platform abstraction.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

That returns the parent PID for the given process.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

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

Let's provide the "__str__" method for "CmdError" to be able to see the
actual content on print(error).
Signed-off-by: NJunxiang Li <junli@redhat.com>
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

The use of "deepcopy" should not be necessary as we only use it here to
override certain keys with our values, not to modify their value.

Even if (theoretically) modified those mutable values later in the
code, it'd have been done even without "deepcopy" for cases where we
didn't forced the "args".
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

This commit makes sure all inputs in yaml2mux plugin are decoded and
passed as unicodes to variants. As a consequence this allows to actually
use this plugin on python3 environment.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

The fingerprint might contain unicode characters, let's use
"astring.ENCODING" to avoid EncodeErrors.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

For computers this does not really matters, but for people the use of
'\n' make debugging slightly easier.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

This replaces the naive "str()" approach of turning objects into strings
with proper "astring.to_text" alternative that should take care of the
proper encoding and works well on py2 as well as py3.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Explicitly close the files we are using.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

We use "iteritems()" on this object which on py2 results int
"dict.iteritems", but on py3 it uses "dict.items". Because we want to
modify the outcome we need to replace the "dict.items" with the slow
implementation using "__getitem__" internally.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

We use the __str__ to generate fingerprint and re-ordering the set leads
to different fingerprints.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

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

Just a simple style fix and adding a missing docstring.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Instead of executing `lsmod`, the same information is available
in /proc/modules.  Also, add test to that function.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The same checks are performed repeated times.  Let's remove those.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Since there are three possible states for a module, this fits well in
the Enum type.  Let's use that, and add compatibility definitions for
the time being.

Reference: https://trello.com/c/BMC7yFbXSigned-off-by: NCleber Rosa <crosa@redhat.com>

This simply moves the recent_mock() function to a common location so
that it can be used by other test modules.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

I tend to favor more predictable and standard behavior, and the
currently used approach to "protect" the import of Python standard
libraries seems to me to be far from standard.

With this change, if the system's Python interpreter is broken, or if
Python standard libraries can not be imported/loaded, the default
behavior will be executed.  If it seems to be an Avocado problem, then
we try to print a line describing that.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

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

Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

This change updates the handle_exception function so it doesn't need to
use a bare-except anymore. The code except responsible for get the crash
directory was kept in a separeted function. This specific code snippet
can't raise any exception. Internally it uses the
avocado.utils.path.usable_rw_dir that already deals with possible
exceptions.

Another possible point of exception could be the import of
avocado.core.data_dir module before avocado be installed itself. However
this import was moved to the initial block of import and dealed proper
as ImportError.

This way seems all the code insisde try block keep being processed by
try-except blocks without bare-except like before.
Signed-off-by: NCaio Carrara <ccarrara@redhat.com>

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