* Test for unusual task data
A task might have any combination of keys and values, but Taskwarrior
often assumes that only valid values can occur, and crashes otherwise.
This is highly inconvenient, as it's often impossible to do anything
with the invalid task -- Taskwarrior just fails without modifying it.
So, this is the beginning of some testing for such invalid tasks, with
the goal of making Taskwarrior due something reasonable. In general, an
invalid attribute value is treated as if it was not set. This is not
exhaustive, and there are likely still bugs of this sort, but as we find
them we can fix and add regression tests to this script.
This introduces a new test-only binary that creates a "bare" task using
TaskChampion, avoiding Taskwarrior's efforts to not create "unusual"
tasks.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* fix path to task executable in pyton tests
The current approach would copy the current files into the `build/test`
directory. Updating the paths according to the custom user setup.
By the copy I appended `.py` to have a clear visible distingtion which
ones are the python tests.
As soon as a source file in the normal directory is changed, it is
copied over and the corresponding file is updated.
From now on the python tests would need to get run in the according
build directory.
* reflect the current build instruction in PR template
* update paths and globing in run_all
* add line break for every cpp test
* remove .gitignore in test folder
As now all the auxillary files such as `all.log` as well as the
executables are present in the `build` directory there is no longer a
need to ignore them.
* update paths in python test scripts and enable deactivated
* remove .py extension when copy to build
Further remove glob pattern for `*.t.py` tests.
* remove accidentally added template.t from test files
This removes use of gnutls and the TLS implementation, which is no
longer needed (task synchronization is handled via Taskchampion, which
uses `reqwest`, which handles TLS via other Rust dependencies). This
incidentally removes the following config options:
* `debug.tls`
* `taskd.ca`
* `taskd.certificate`
* `taskd.ciphers`
* `taskd.credentials`
* `taskd.key`
* `taskd.server`
* `taskd.trust`
These use tempfile.NamedTemporaryFile but allow execution without
raising "Text file busy".
The file is removed at exit (of the process) so no cleanup is necessary.
* It is now possible to test:
* Hook Input/Output on STDIN/STDOUT channels
* Exit code of hook script
* Execution count (how many times the hook was executed)
* Timestamp execution (when was the hook executed - milisec resolution)
* When testing multiple clients vs server versions, repeating client
tests is unnecessary. By setting the env variables TASKW_SKIP and
TASKD_SKIP it will now be possible to skip all task (client only) and
taskd (client + server) tests, respectively.
* The new unblocking strategy for subprocesses involves timeouts and
waiting periods. If the timeout is too short (or absent) tests may show
intermittent failure.
* Increasing the timeout may make tests a bit slower but more reliable.
* Processes that blocked waiting for stdin data will now be aborted
after a 1 second timeout.
* As a side-effect any process that takes longer than 1 second to
finish will also be aborted.
* Taskd and Taskw classes for testing are now available
* Testing of server and client can now be performed.
* The newer test wrappers will eventually replace the BaseTest class
