On Python 3, the calculation of a value for a unit will return
a float.  Let's make things predictable accross Python 2 and 3
and force an integer instead.

Reference: https://github.com/avocado-framework/avocado/pull/2498Signed-off-by: NCleber Rosa <crosa@redhat.com>

Now that we have a second use case for it, let's finally create the
DataSize object.
Signed-off-by: NAmador Pahim <apahim@redhat.com>

In situations where a user of the CmdResult class manually sets the
stdout/stderr attributes to a string, instead of the recommended and
documented content of type "bytes", it won't be possible to "decode"
the bytes into a string of a given encoding, and we'll end up with
a crash.

Since the goal of std{out,err}_text is to return a text version
of stdout, if itself already is of such a type, let's just return
it.

Additionally, if the data cannot be "decode()d", let's raise an
explicit TypeError on this location, rather than later in the code
when its value is attempted to be used.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Looks like the Python 3 module import code behaves differently than
Python 2 with regards to where it looks modules by name:

   File "/home/cleber/src/avocado/avocado/avocado/core/test.py", line 932, in _run_avocado
      raise test_exception
   File "/home/cleber/src/avocado/avocado/avocado/core/test.py", line 819, in _run_avocado
      testMethod()
   File "/home/cleber/src/avocado/avocado/avocado/core/test.py", line 1371, in test
      self.error(self.exception)
   File "/home/cleber/src/avocado/avocado/avocado/core/test.py", line 1041, in error
      raise exceptions.TestError(message)\navocado.core.exceptions.TestError: Traceback (most recent call last):
   File "/usr/lib64/python3.6/imp.py", line 297, in find_module
      raise ImportError(_ERR_MSG.format(name), name=name)
   ImportError: No module named \'whiteboard\'\n\n'

Effectively, this means when running this test, on Python 3, on the
parallel mode, the change of directory implemented there
(on selftests/checkall) will make Python 3 fail to find this module,
and the test will fail.

This makes the reference to the module path absolute, so no matter
where this test is run from, it should be able to find the module.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The avocado.utils.process module has been changed so that the content
of stdout and stderr are now bytes.  The content generated by the
iso9660 backend implementations should also operate on bytes, given
that it makes little sense for the library to assume a given content
type of a file (inside an iso9660 archive).

To make the data returned by the various backend `read()`
implementations always return bytes on both Python 2 and 3, an
explicit conversion is done from strings to bytes (which affects only
Python 2, since Python 3 already operates with bytes on files opened
in binary mode).
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Automatically chosen on Python 3.  This mimmics the (proposed)
behavior in `avocado/core/runner.py`.

Reference: https://github.com/avocado-framework/avocado/pull/2451Signed-off-by: NCleber Rosa <crosa@redhat.com>

And not a directory.  Tests were being always skipped becaused of that.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The directory check is being performed, but without value to the test.
Let's turn it into an assertion.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Whiteboard was (probably accidentally) removed from the test attributes.
This patch recreates the whiteboard.
Signed-off-by: NAmador Pahim <apahim@redhat.com>

In case on of the first cleanups fails the following ones are not
performed, leaving uncleaned files. Lets perform always all steps and
then check for occurred errors.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

The prevailing coding style of Avocado (unittest.TestCase) tests is to
use the specialized assert methods, such as assertNotRegexpMatches.

Unfortunately, the regular expression related assertion methods have
been through a messy renaming.  In theory, assertNotRegexpMatches and
assertRegexpMatches should be available in Python 3.0 and 3.1, but not
in Python >= 3.2, where it got renamed to assertRegex and
assertNotRegex.

In practice, in my system with Python 3.6.3, I see:

>>> unittest.TestCase.assertNotRegexpMatches
<function TestCase._deprecate.<locals>.deprecated_func at 0x7f3b85b709d8>

But in Travis, with Python 3.4, it's not available:

ERROR: test_directives_regex (selftests.unit.test_safeloader.DocstringDirectives)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/travis/build/avocado-framework/avocado/selftests/unit/test_safeloader.py", line 135, in test_directives_regex
    self.assertNotRegexpMatches(directive, safeloader.DOCSTRING_DIRECTIVE_RE)
AttributeError: 'DocstringDirectives' object has no attribute 'assertNotRegexpMatches'

Let's simplify things and just use a regex match and check for its
result.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

make spell is now covering more cases. This patch fixes the actual spell
errors and adds the required exceptions.
Signed-off-by: NAmador Pahim <apahim@redhat.com>

The docstring given on `_greedy_path()` has examples that don't match
the implementation.  To prevent the implementation from deviating
(again?) let's turn those into tests.

Also, to make the purpose of that function more clear, this
changes its name to `_greedy_path_to_re()`.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The order in which test classes and methods are found is relevant
to the loader result.  Currently, this ordering is not dependable
as the loader findings are stored on a regular dictionary.

This change makes the ordering dependable, and so happens to also
make the test pass on a Python 3 environment.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Which, different than standard Python xml.etree library, can do
schema validation.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

That's the only sane way to handle data, as bytes.  Let's not assume
any enconding on them.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Move tests to the plugin's own directory.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Move tests to the plugin's own directory.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

And drop the notion of a "multiplex capable" Avocado.  If the plugin is
installed, then we assume the dependencies have been fulfilled.  The plugin
presence is used then for the creation of the test suite, including the
plugin's own tests.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

This drops the notion of a "multiplex capable" Avocado.  If the plugin
is installed, then we assume the dependencies have been fulfilled.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Some versions of the unittest mock library cannot easily deal with
(builtin) open().  Let's skip these tests on those platforms, instead
of coming up with cumbersome code to support them on all (ancient)
versions.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

This simple test, after the s390x version, checks the proper parsing
of and x86_64 /proc/cpuinfo.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The output in /proc/cpuinfo is different between s390x and other archs.
This is to support to parse the output on s390x.
Signed-off-by: NDan Zheng <dzheng@redhat.com>
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The `avocado.core.tree.TreeNode` is the foundation of the current
Parameter System (`avocado.core.parameters.AvocadoParams`), but it
lacks unittests that only deal with itself.

Some tests that exercise the Multiplexer version of the `TreeNodes`
(`MuxTreeNodes`) exist, but again, that's not what the core Parameter
System is based on.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The AvocadoParams, as a database of test parameters, has no need
to know information about the test.  In fact, the test id parameter
has never had any real use so far.

Let's remove it.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

A lot of the core Avocado code still refers to "mux" based terminology.
For instance, what was previously "mux_path" is now "parameter paths",
and what was previously "mux" is now really variants.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

This is an attempt to catch most (if not all) usages of open
that do not follow the context manager pattern.

Sometimes, for better readability, our own `genio` library is
used.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Despite the reviews we have some W0102s that potentially corrupts the
arguments. Let's get rid of them.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

The default parameters mechanism proved to be confusing, overlapping
and thus was scheduled to be removed.  This implements the removal
of the default_params API.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Our goal to completely separate the basic parameter interface and the
varianter pass through here: splitting these modules into their
reespective ones.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The `_reference_check()` utility method now provides more meaningful
results.  This allows one to check wether a check was actually
performed.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The code is basically the same for all different files.  Let's
have a single method instead.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

There was once an idea that the "mux" would be an interface that would
be reused by other implementations (say, json_to_mux, foo_to_mux,
etc).  Now, it seems much more important to make the Avocado varianter
design and interfaces better, and advocate its use instead.

Additionally, there has always been a gray line between the
Multiplexer and the Avocado params.  Moving the mux outside of the
Avocado core will help to draw clearer lines.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

And include jobdata results and a test for 52.0 (LTS).

The 52.0 release, an LTS release, was meant to allow transitions from
the previous LTS version (36.0lts) and it did so including the ability
to replay jobs created with the older version.

Now, we're 6+ months into the 52.0 release, and as promised, we'll
stop maintaining 36.0lts.  To keep to code clean, let's remove all
this compatibility with versions before 52.0.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

The FDDrainer reads from the PIPEs and writes to both the internal
StringIO instances and to stream logging handlers.  There are no
guarantees that the logging handlers will flush the content before the
FDDrainer finishes.

Let's explicitly close all the handlers (which should really be
FileHandlers) associated with the stream loggers, which should flush
and sync the content on the files themselves.

Additionally, the combined drainer was being missed from the flush
process, and was causing some output in that mode to be unavailable on
generated output.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

This is refactor of the current locks/file/threads created in the
SubProcess class to a more contained an isolated (and tested)
block of functionality.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

And raise an exception early if an invalid value is given.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

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

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

Instead of the iterator next method.
Signed-off-by: NCleber Rosa <crosa@redhat.com>