Not a single client of this API actually used the event mask feature to
listen for readability AND writability.
Let's simplify the API and have only one hook: on_activation.
Instead of juggling events between individual instances of
Core::EventLoop, move queueing and processing to a separate per-thread
queue (ThreadEventQueue).
This was used in exactly one place, to avoid sending multiple
CustomEvents to the enqueuer thread in Audio::ConnectionToServer.
Instead of this, we now just send a CustomEvent and wake the enqueuer
thread. If it wakes up and has multiple CustomEvents, they get delivered
and ignored in no time anyway. Since they only get ignored if there's
no work to be done, this seems harmless.
Previously, statements containing malformed exists expressions such as:
`INSERT INTO t(a) VALUES (SELECT 1)`;
could cause the parser to crash. The parser will now return an error
message instead.
The check on the currently active document for an existing favicon has
been removed. It caused an issue during navigation because the active
document will only be updated after the load and the favicon check has
been executed against the old document.
Previously, the `Parser::decode_tiles()` function wouldn't wait for the
tile-decoding workers to finish before exiting the function, which
could mean that the data the threads are working with could become
invalid if the decoder is deleted after an error is encountered.
For a subsection named `Foo` with a markdown file `Foo.md` in the same
directory, the document `Foo.md` will be returned when the subsection is
selected.
This prevents PageNodes and SubsectionNodes from sharing the same
name, which fixes an issue in Help where subsections could not be
navigated.
Using malloc does not invoke T's constructor, nor were were invoking T's
constructor ourselves. Accessing T without invoking its constructor is
undefined behavior.
Without `-y`, to show the current full year you'd have to specify which
one: `cal 2023`. Adding `-y` makes it possible to see the full current
year without remembering what year we are in.
This option is also stolen from FreeBSD :^)
Additionally, validate args: prevent passing both -3 and -y at the
same time. Passing both `--three-month-mode` and `--year` to `cal`
doesn't make sense. You'd either want the one or the other.
Making it configurable in system settings :^)
The --start-day option can still overwrite this global default.
This change makes it no longer possible to use unveil: as we have
to load the Calendar config file, which might be in a dynamic location.
It's also neccessary to add `cpath` to the pledge, as opening a
nonexistent config file with Core::ConfigFile::open_for_app creates it.
Making all the other parts of the world happier :^)
Add a `--starting-day` (`-s`) option to be compatible with GNU cal,
which has a similar option. The GNU option takes allows passing either
an int or a day name. Let's do something similar using weekdays we
already have in AK/DateConstants.h.
Also add myself to the copyright header, as by now I've modified most of
the lines in this file.
...instead of putting a star `*` next to it. This makes `cal`s output
much prettier, and gets rid of one FIXME. :^)
Don't use the escape sequence from the deleted FIXME - \e[30;47m would
set the background to white and foreground to black - which presumably
wouldn't do much on a light-theme terminal. Instead use \e[7m which sets
the color as "inverted".
This fixes cal not highlighting the current day.
After this commit `cal` will show something like this among its output
for the 23rd day of the month
1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23*24 25
26 27 28
Every other cal implementation just highlights the current day instead
of letting you specify a custom one to highlight. It doesn't seem to be
that useful, and is currently broken - no day gets highlighted at all,
because the `target_day` global is never written to.
Moreover, this complicates parsing the arguments. This commit also fixes
parsing a case where just a year is provided to `cal` - for example `cal
2023`.
This brings the function name in line with how we usually name those
functions, which is with a `read_` or `write_` prefix depending on what
they do.
While at it, make the internal `_impl` function private and not-virtual,
since there is no good reason to ever override that function.
- Rename to make it clear it's not just for promotion
- Understand 'p' for pawns
- Take a char parameter instead of StringView since it's always 1 char
Previously, the initial position would look like this:
rnbqkbnr//8/8/8/8//RNBQKBNR w KQkq - 0 1
Now, we correctly give pawns the P/p character in FEN output:
rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1
Also, we only ever have 1 or 0 characters for a piece, so let's return
`Optional<char>` instead of `StringView` from `char_for_piece()`.
