Instead of setting `-std` compiler flag directly, set the
`CMAKE_CXX_STANDARD` variable. This lets CMake know the required C++
standard and evaluate the final compiler flag correctly, taking into
account compile features set by `target_compile_features()`.
This change preserves the existing behavior, where compiler extensions
are disabled for other targets than Cygwin.
This replaces a loop over _all_ tasks with one that fetches only pending
tasks, as determined by the working set.
This should be faster for task DB's with large numbers of completed
tasks, although on my medium-sized installation (~5000 total tasks) the
difference is negligible.
The code for parsing epoch timestamps when displaying tasks only
supports values between year 1980 and 9999. Previous to this change, it
was possible to set e.g., the due timestamp to a value outside of these
limits, which would make it impossible to later on show the task.
With this change, we only allow setting values within the same limits
used by the code for displaying tasks.
* libshared: bump for weekstart, epoch defines, eopww fix
mainly those visible changes, and miscellaneous others
see GothenburgBitFactory/taskwarrior#3623 (weekstart)
see GothenburgBitFactory/taskwarrior#3651 (epoch limit defines)
see GothenburgBitFactory/libshared#73 (eopww fix)
* Initialize libshared's weekstart from user's rc.weekstart config
This enables use of newer libshared code that can parse week numbers
according to ISO8601 instead of existing code which is always using
Sunday-based weeks. To get ISO behavior, set rc.weekstart=monday.
Default is still Sunday / old algorithm, as before, since Sunday is in
the hardcoded default rcfile.
Weekstart does not yet fix week-relative shortcuts, which will still
always use Monday.
See #3623 for further details.
* Restore 'load' timer
* [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>
This replicates what the Rust runtime does, and matches what Rust code
expects, for example when writing to a socket which is no longer
connected to the remote end.
There is no easy way to determine the MSRV for TaskChampion, other than
somehow pulling the right version of the source and grepping for it. In
practice, if we update the `taskchampion` dependency to one that has a
higher MSRV, we'll get a build error and find this comment. And if we
get an error building Taskwarrior due to an old MSRV (for example if
something changes on `crates.io`) then we will also find this comment.
This also removes some superfluous dependency versions from the root
workspace. `src/taskchampion-cpp/Cargo.toml` specifies versions
directly.
TC 0.7.0 introduces a new `TaskData` type that maps to Taskwarrior's
`Task` type more cleanly. It also introduces the idea of gathering lists
of operations and "committing" them to a replica.
A consequence of this change is that TaskChampion no longer
automatically maintains dependency information, so Taskwarrior must do
so, with its `TDB2::dependency_sync` method. This method does a very
similar thing to what TaskChampion had been doing, so this is a shift of
responsibility but not a major performance difference.
Cxx is .. not great. It is missing a lot of useful things that make a
general-purpose bridge impractical:
- no support for trait objects
- no support for `Option<T>` (https://github.com/dtolnay/cxx/issues/87)
- no support for `Vec<Box<..>>`
As a result, some creativity is required in writing the bridge, for
example returning a `Vec<OptionTaskData>` from `all_task_data` to allow
individual `TaskData` values to be "taken" from the vector.
That said, Cxx is the current state-of-the-art, and does a good job of
ensuring memory safety, at the cost of some slightly awkward APIs.
Subsequent work can remove the "TDB2" layer and allow commands and other
parts of Taskwarrior to interface directly with the `Replica`.
* Update google-cloud-auth to drop ring@0.16.20ring@0.16.20 doesn't build on ppc and risc-v, and updating
google-cloud-auth pulls in a newer version of jsonwebtoken,
which in turn depends on a newer version of ring that we depend on
already either way.
This necessitated an MSRV bump to 1.73.0
Two of these used EXPFAIL which, because nothing is interpreting the TAP
output, does not work. So, that functionality is removed, and the
expected-to-fail bits are commented out or removed.
There was a conditional on the filename in `bash_tap.sh` which caused it
to not actually do anything and just run the test as a simple shell
script. That, too, has been removed.
