The InstanceData type in the rendering subsystem was previously 17 f32s
plus one u8 which occupied a total of 72 bytes per instance. This meant
that for every character or background cell drawn, 72 bytes were sent to
the GPU. In the case of a 400x100 cell grid, a total of 2.9MB would be
sent.
This patch reduces InstanceData's size to 36 bytes, a 50% improvement!
Using the above example for comparison, a worst case of 1.44MB would be
transferred.
The motivation for this patch comes from macOS. Once the terminal grid
would reach a certain size, performance experienced a sharp and dramatic
drop (render times would go from ~3ms to ~16ms). I don't want to
speculate too much on the underlying issue, but suffice it to say that
this patch alleviates the problem in my testing.
While the performance impact was most significant on macOS, with
rendering times cut by more than 50% in some cases, this also results in
a measurable performance difference on other systems with high density
grids.
Co-authored-by: Christian Duerr <contact@christianduerr.com>
This resolves an issue where negative window positions set in the
configuration file would not place the Alacritty window in the correct
location.
Fixes#4061.
This clarifies the order of subsections in the changelog. All entries
are now defined in the order `Packaging`, `Removed`, `Added`, `Changed`
and `Fixed`.
The sections `Security` and `Deprecated` have been removed. Since
deprecation should always just be a soft deprecation with clear user
warnings, it's not much different from just `Changed`, so that should
make it easier to decide where to put things. The `Security` section is
often not clear enough since a lot of changes might have some form of
security impact depending on interpretation. Since there's unlikely to
be any major security advisories from the Alacritty project, that
section is not particularly useful.
The `Added`, `Changed` and `Fixed` order follows the interest of the
user and the impact on them. This puts the most interesting new features
front and center while allowing to go for a deep dive to see if specific
bugs have been fixed.
The `Removed` section was put last since realistically this should not
affect users. All options should be gradually deprecated and no major
features should just get removed. So most of the time this will just
list deprecated features that are now completely removed, which usually
means there's no impact for users that updated from the last version.
Since Alacritty should not be responsible for packaging and the
maintainers of the snap package have not kept it up to date, this
removes its configuration file.
This should reduce the number of times people with fixed cursor colors
run into troubles when existing text is already colored.
Using just the background color as a metric instead of both background
and foreground color should ensure that the cursor still has a clear
shape, since just changing the foreground color for a cursor might be
difficult to see. Always inverting the entire cursor instead of keeping
the fixed foreground color is important to make sure the contrast isn't
messed up.
Fixes#4016.
This adds support for the secondary DA escape sequence response.
Alacritty's version is formatted allowing for up to 99 minor and patch
versions, which should be sufficient.
The tertiary DA is intentionally not implemented and marked as rejected
in the documentation, since a lot of terminals do not support it, or
report useless data (XTerm/URxvt/Kitty).
Fixes#3100.
Instead of just always falling back to the shell specified in the passwd
file when no config or cli shell was specified, Alacritty will not first
look at the `$SHELL` environment variable. If this is unset, it will
still read the passwd file.
Since macOS is a bit peculiar and does not set the `$SHELL` environment
variable by default, it is set manually to the shell used by Alacritty
while any existing `$SHELL` variables are ignored. This matches the
behavior of iTerm and Terminal.app.
Co-authored-by: Christian Duerr <contact@christianduerr.com>
This resolves a rounding issue when converting to the 26.6 format used
by freetype for character sizes.
This rounding behavior is taken from cairo:
https://gitlab.freedesktop.org/cairo/cairo/-/blob/master/src/cairo-ft-font.c#L900-903
There are various different implementations of the F26Dot6 conversion
online, but the rounding that cairo does seems to be the most common.
Since cairo is very commonly used, it should produce good results
compared with the rest of Linux text rendering.
Fixes#2780.
Co-authored-by: Christian Duerr <contact@christianduerr.com>
Co-authored-by: Kirill Chibisov <contact@kchibisov.com>
This commit removes font dependency from alacritty_terminal,
so it'll simplify the usage of alacritty_terminal as a library,
since you won't link to system's libraries anymore. It also
moves many alacritty related config options from it.
Fixes#3393.
This adds a new regex search which allows searching the entire
scrollback and jumping between matches using the vi mode.
All visible matches should be highlighted unless their lines are
excessively long. This should help with performance since highlighting
is done during render time.
Fixes#1017.
To make sure that output is consistent even while resizing the window,
the cursor will now reflow with the content whenever the window size is
changed.
Since the saved cursor is more likely to represent a position in the
grid rather than a reference to the content below it and handling of
resize before jumping back to it is more likely than with the primary
cursor, no reflow is performed for the saved cursor
The primary cursor is unfortunately always reflowed automatically by
shells like zsh, which has always caused problems like duplicating parts
of the prompt and stretching it out "infinitely". Since the cursor is
now reflowed appropriately the duplication of the shell prompt should be
reduced, however it is possible that the shell moves the cursor up one
line after it has already been reflowed, which will cause a line of
history to be deleted if there is no duplicated prompt line above the
reflowed prompt. Since this behavior is identical in VTE and Kitty, no
attempt is made to work around it in this patch.
Fixes#3584.
This resolves several problems with handling of the saved cursor when
switching between primary and alternate screen. Additionally ref-tests
are also added for all common interactions to make sure the behavior
does not regress.
The behavior is based on XTerm's behavior except for interaction with
`reset`. XTerm does not reset the alternate screen's saved cursor on
`reset`, but VTE does. Since a `reset` should reset as much as possible,
Alacritty copies VTE here instead of XTerm.
While neither VTE, URxvt nor Kitty handle this, preserving the linewrap
flag across alternate screen switches seems like the correct thing to
do. XTerm also does handle this correctly, which indicates that it is a
bug and not a feature.
This bug was caused by trying to grow the terminal while the cursor line
was wrapped but entirely empty. Resizing the terminal now accounts for
the position of the deleted line and moves the cursor up only when the
line deleted was above it.
The deletion of the line was caused by the shell redrawing itself
whenever the cursor is moved.
Fixes#3583.
This works around a bug where the optimized version of the
`Grid::scroll_down` function would just rotate the entire grid down if
the scrolling region starts at the top of the screen, even if there is
history available.
Since rotations of scrolling regions should not affect the scrollback
history, this optimized version is now only called when the max
scrollback size is 0, making it impossible for the grid to have any
history while it is used.
Since the main usecase of this is the alternate screen buffer, which
never has any history, the performance should not be affected negatively
by this change.
Fixes#3582.
This allows for expanding the selection using the right mouse button.
The new selection type depends on the number of clicks and applies to
both sides of the selection.
Fixes#1554.
This adds a new `Scheduler` which allows for staging events to be
processed at a later time.
If there is a selection active and the mouse is above or below the
window, the viewport will now scroll torwards the direction of the
mouse. The amount of lines scrolled depends on the distance of the mouse
to the boundaries used for selection scrolling.
To make it possible to scroll while in fullscreen, the selection
scrolling area includes the padding of the window and is at least 5
pixels high in case there is not enough padding present.
This reworks the criteria necessary for default bindings to be
overwritten by custom user bindings. It should provide a better
heuristic which accounts for the possibility that notmodes are added
which prevent a conflict between the bindings, so the default isn't
unnecessarily removed.
It is still possible to define a new binding that intersects a default
binding without the default getting removed, if the user explicitly
specifies a mode that doesn't include the default binding's mode. This
is based on the assumption that users explicitly specifying a new mode
are trying to make a mode-specific addition and are capable of removing
the default when desired.
This helps with old modes still being present in user's config files,
since a lot of new config options just have the additional ~Vi mode set.
Fixes#3476.
This commit aims to clear up the separation between Term and Grid to
make way for implementing search.
The `cursor` and `cursor_save` have been moved to the grid, since
they're always bound to their specific grid and this makes updating
easier.
Since the selection is independent of the active grid, it has been moved
to the `Term`.
The env logger has been broken for over a year and is not used by anyone
as far as I know. This removes this option entirely in favor of
Alacritty's built-in logger level selection flags.
This resolves an issue where trying to write a fullwidth character in
the last column would crash Alacritty, if linewrapping was disabled.
Instead of assuming that the linewrap put after the linewrapping spacer
was successful, the character writing is now skipped completely when
trying to put a wide character in the last column.
