ansible-runner-1.4.7-bp156.2.2

- do not require python-mock: the testsuite is not running anyway
  and newer testsuite (e. g. 2.1.3) will not use mock at all
  https://trello.com/c/S6eADbii/64-remove-python-mock

- Update to version 1.4.7
  * Fixed an issue where byte sequences in the output that are not
    interpretable as UTF-8 would break the job execution.
  * Fixed a bug that broke Ansible playbook execution prior to
    version 2.8 of Ansible
  * Fix an issue with --process_isoloation_*_ paths parsing cli args
  * Switch extravar format so we can support more than just string
    types
  * Make sure job events folder is created earlier to prevent
    errors when using immediately after starting a runner job
  * Annotate all runner_on events with start/end/duration times
- Drop 0001-Use-the-correct-python-executable-for-tests.patch

- update to 1.4.4:
  + Fix some unicode issues when using command line override on python 2.7
  + Fix an issue with file handles not closing on the pexpect pty
  + Fix missing ssh_key parameter from module interface
  + Fix a bug where the ssh agent process would hang around after
    process isolation exit causing a failure to remove temporary directories
  1.4.2 (2019-10-04)
  + Reverted ansible-runner --version to print a semantic version number
  1.4.1 (2019-10-03)
  + Fixed a bug that prevented ANSIBLE_HOST_KEY_CHECKING from being respected
  1.4.0 (2019-09-20)
  + Added changed count to stats data
  + Added initial support for gathering performance statistics using
    the system's cgroup interface
  + Fix command line args override missing from module run kwargs
  + Omit inventory argument entirely if no inventory content is supplied
    this allows ansible to pick up inventory from implicit locations and ansible.cfg
  + Fix an issue where Runner wouldn't properly clean up process
    isolation temporary directories
  + Fix error generated if unsafe parameter is used on vars prompt tasks
  + Fix an issue where additional callback plugins weren't being
    used when defined in the environment
  + Fix an issue where Runner would stop returning events after the
    playbook finished when using run_async
  + Fix an issue where unicode in task data would cause Runner to fail
  + Fix issues using vaulted data that would cause Runner to fail
  + Fix an issue where artifact-dir was only allowed in ad-hoc mode
- run fdupes
- remove upstreamed patch:
  0001-Use-the-correct-python-executable-for-tests.patch
- disable tests again for now

- update to 1.3.4:
  * make class __name__ match Ansible expectation
  * bundle six.ensure_str to avoid pinning to a new six version
  * Updating for 1.3.4 release
  * Adopt modified upstream packaging
  * Add missing lockfile RPM dependency
  * Fix an issue with missing options
  * port old callback tests to integration folder
  * Fix missing toplevel host events docs issue
  * Move integration data folder, add new test
- fix license - ansible-runner also contains code licensed under
  GPL-3.0-or-later

- enable tests in %check
- add 0001-Use-the-correct-python-executable-for-tests.patch to fix
  tests with python3

- Initial packaging (version 1.3.3)