40.0 Dr Who¶
The Avocado team is proud to present another release: Avocado version 40.0, aka, “Dr Who”, is now available!
Release documentation: Avocado 40.0
The major changes introduced on this version are listed below.
- The introduction of a tool that generated a diff-like report of two jobs. For more information on this feature, please check out its own documentation at Job Diff.
avocado.utils.processlibrary has been enhanced by adding the
avocado.utils.process.SubProcess.get_pid()method, and also by logging the command name, status and execution time when verbose mode is set.
- The introduction of a rr based wrapper.
With such a wrapper, it’s possible to transparently record the
process state (when executed via the
avocado.utils.processAPIs), and deterministically replay them later.
- The coredump generation contrib scripts will check if the user running Avocado is privileged to actually generate those dumps. This means that it won’t give errors in the UI about failures on pre/post scripts, but will record that in the appropriate job log.
- BUGFIX: The
--remote-no-copycommand line option, when added to the
--remote-*options that actually trigger the remote execution of tests, will now skip the local test discovery altogether.
- BUGFIX: The use of the asset fetcher by multiple avocado executions
could result in a race condition. This is now fixed, backed by a
file based utility lock library:
- BUGFIX: The asset fetcher will now properly check the hash on
- BUGFIX: A busy loop in the
avocado.utils.processlibrary that was reported by our users was promptly fixed.
- BUGFIX: Attempts to install Avocado on bare bones environments, such
as virtualenvs, won’t fail anymore due to dependencies required at
setup.pyexecution time. Of course Avocado still requires some external Python libraries, but these will only be required after installation. This should let users to
pip install avocado-frameworksuccessfully.
For more information, please check out the complete Avocado changelog.
The Avocado release meetings are now open to the community via Hangouts on Air. The meetings are recorded and made available on the Avocado Test Framework YouTube channel.
For this release, you can watch the meeting on this link.