This PR enables iTerm2-like one button "Set Ghostty as Default Terminal
App" functionality on macOS, making it easier to open a directory in
Ghostty, run shell scripts when mouse clicking, etc.
Related to #8406 (for GTK only)
Spatial split navigation now wraps at the edges.
We first attempt the nearest spatial target using the existing slot
geometry. If there is no candidate in the requested direction, we
synthesize a wrapped target by shifting the current slot by one full
grid in the opposite direction and reuse the same nearest-distance
logic.
This fake target works because the grid is 1x1, so by moving it a full
grid size in the opposite direction, we effectively wrap around to the
other side of the grid.
Fixes#8406
Spatial split navigation now wraps at the edges.
We first attempt the nearest spatial target using the existing slot geometry.
If there is no candidate in the requested direction, we synthesize a wrapped
target by shifting the current slot by one full grid in the opposite
direction and reuse the same nearest-distance logic.
This fake target works because the grid is 1x1, so by moving it a full
grid size in the opposite direction, we effectively wrap around to the
other side of the grid.
Closes#9266.
Big Note: I noticed that this worked properly under `NixOS`, but on my
`Ubuntu` VM it didnt.
The reason is in
[src/build/GhosttyI18n.zig](73a93abf7b/src/build/GhosttyI18n.zig (L24-L31))
because the locale is expected in the `<lang>_<region>` without the
encoding suffix. `<lang>_<region>_<encoding>`
```
// There is no encoding suffix in the LC_MESSAGES path on FreeBSD,
// so we need to remove it from `locale` to have a correct destination string.
// (/usr/local/share/locale/en_AU/LC_MESSAGES)
const target_locale = comptime if (builtin.target.os.tag == .freebsd)
std.mem.trimRight(u8, locale, ".UTF-8")
else
locale;
```
If i force it to always trim the encoding it works, but I am guessing
its there for a reason ,so maybe some of the maintainer can shed some
light in the best way forward, as I am not an expert in how other
systems deal with it. Here you see `Open in Ghostty` -> Abrir con
Ghostty
<img width="353" height="372" alt="image"
src="https://github.com/user-attachments/assets/2c0266f7-cfb3-49e3-aef1-9e98acb16ad8"
/>
- I wanted to format the `py` file with `ruff` but didnt want to drown
the changes, so maybe something that could be worth doing so that also
our `py` files have std formatting.
> [!NOTE]
> Used AI only for helping me debug where the locales could be and why
was it not detected, but no code help whatsoever
This changes the way Ghostty assigns itself and subprocesses to
cgroups and how resource controls are applied.
* Ghostty itself no longer modifies it's own cgroup or moves itself
to a transient scope. To modify the main Ghostty process' resource
controls ensure that you're launching Ghostty with a systemd unit and
use the standard systemd methods for overriding and applying changes
to systemd units.
* If configured (on by default), the process used to run your command
will be moved to a transient systemd scope after it is forked from
Ghostty but before the user's command is executed. Resource controls
will be applied to the transient scope at this time. Changes to
the `linux-cgroup*` configuration entries will not alter existing
commands. If changes are made to the `linux-cgroup*` configuration
entries commands will need to be relaunched. Resource limits can also
be modified after launch outside of Ghostty using systemd tooling. The
transient scope name can be shown by running `systemctl --user whoami`
in a shell running inside Ghostty.
Fixes#2084.
Related to #6669
Example of `systemctl status` showing main Ghostty process and one
surface:
<img width="1132" height="135" alt="Screenshot From 2026-02-07 16-31-14"
src="https://github.com/user-attachments/assets/81dffd0b-8801-4695-adf4-213647cdf0c3"
/>
This PR introduces unit tests and a supporting Mock NSView for testing
the SplitTree implementation in Swift. It includes 51 tests which
achieve approximately 93.13% (949/1019) coverage of SplitTree.swift's
branches.
<details>
<summary>Coverage</summary>
<pre>
./ghostty/macos/Sources/Features/Splits/SplitTree.swift 93.13%
(949/1019)
SplitTree.Path.isEmpty.getter 100.00% (1/1)
SplitTree.isEmpty.getter 100.00% (3/3)
SplitTree.isSplit.getter 100.00% (3/3)
SplitTree.init() 100.00% (3/3)
SplitTree.init(view:) 100.00% (3/3)
SplitTree.contains(_:) 100.00% (4/4)
SplitTree.inserting(view:at:direction:) 100.00% (6/6)
SplitTree.find(id:) 100.00% (4/4)
SplitTree.removing(_:) 93.75% (15/16)
SplitTree.replacing(node:with:) 93.75% (15/16)
SplitTree.focusTarget(for:from:) 82.14% (46/56)
closure #1 in SplitTree.focusTarget(for:from:) 100.00% (1/1)
closure #2 in SplitTree.focusTarget(for:from:) 100.00% (1/1)
closure #3 in SplitTree.focusTarget(for:from:) 100.00% (3/3)
implicit closure #1 in SplitTree.focusTarget(for:from:) 0.00% (0/1)
SplitTree.equalized() 100.00% (5/5)
SplitTree.resizing(node:by:in:with:) 92.00% (69/75)
closure #1 in SplitTree.resizing(node:by:in:with:) 100.00% (1/1)
SplitTree.viewBounds() 100.00% (4/4)
SplitTree.init(from:) 76.00% (19/25)
SplitTree.encode(to:) 100.00% (15/15)
SplitTree.Node.find(id:) 100.00% (13/13)
SplitTree.Node.node(view:) 88.89% (16/18)
SplitTree.Node.path(to:) 100.00% (32/32)
search #1 <A>(_:) in SplitTree.Node.path(to:) 100.00% (27/27)
SplitTree.Node.node(at:) 89.47% (17/19)
SplitTree.Node.inserting(view:at:direction:) 86.84% (33/38)
SplitTree.Node.replacingNode(at:with:) 100.00% (43/43)
replaceInner #1 <A>(current:pathOffset:) in
SplitTree.Node.replacingNode(at:with:) 96.67% (29/30)
SplitTree.Node.remove(_:) 70.27% (26/37)
implicit closure #1 in SplitTree.Node.remove(_:) 100.00% (1/1)
SplitTree.Node.resizing(to:) 100.00% (16/16)
SplitTree.Node.leftmostLeaf() 87.50% (7/8)
SplitTree.Node.rightmostLeaf() 87.50% (7/8)
SplitTree.Node.equalize() 100.00% (4/4)
SplitTree.Node.equalizeWithWeight() 100.00% (30/30)
SplitTree.Node.weightForDirection(_:) 83.33% (10/12)
SplitTree.Node.calculateViewBounds(in:) 100.00% (50/50)
SplitTree.Node.viewBounds() 100.00% (26/26)
SplitTree.Node.spatial(within:) 100.00% (18/18)
SplitTree.Node.dimensions() 80.77% (21/26)
SplitTree.Node.spatialSlots(in:) 100.00% (53/53)
SplitTree.Spatial.slots(in:from:) 100.00% (47/47)
closure #1 in SplitTree.Spatial.slots(in:from:) 100.00% (1/1)
distance #1 <A>(from:to:) in SplitTree.Spatial.slots(in:from:) 100.00%
(6/6)
closure #2 in SplitTree.Spatial.slots(in:from:) 100.00% (3/3)
implicit closure #1 in closure #2 in SplitTree.Spatial.slots(in:from:)
100.00% (1/1)
closure #3 in SplitTree.Spatial.slots(in:from:) 100.00% (3/3)
closure #4 in SplitTree.Spatial.slots(in:from:) 100.00% (3/3)
implicit closure #1 in closure #4 in SplitTree.Spatial.slots(in:from:)
100.00% (1/1)
closure #5 in SplitTree.Spatial.slots(in:from:) 100.00% (3/3)
closure #6 in SplitTree.Spatial.slots(in:from:) 100.00% (3/3)
implicit closure #1 in closure #6 in SplitTree.Spatial.slots(in:from:)
100.00% (1/1)
closure #7 in SplitTree.Spatial.slots(in:from:) 100.00% (3/3)
closure #8 in SplitTree.Spatial.slots(in:from:) 100.00% (3/3)
implicit closure #1 in closure #8 in SplitTree.Spatial.slots(in:from:)
100.00% (1/1)
closure #9 in SplitTree.Spatial.slots(in:from:) 100.00% (3/3)
SplitTree.Spatial.doesBorder(side:from:) 100.00% (20/20)
closure #1 in SplitTree.Spatial.doesBorder(side:from:) 100.00% (1/1)
closure #2 in SplitTree.Spatial.doesBorder(side:from:) 100.00% (3/3)
static SplitTree.Node.== infix(_:_:) 100.00% (13/13)
SplitTree.Node.init(from:) 66.67% (12/18)
SplitTree.Node.encode(to:) 100.00% (11/11)
SplitTree.Node.leaves() 100.00% (9/9)
SplitTree.makeIterator() 100.00% (3/3)
implicit closure #1 in SplitTree.makeIterator() 100.00% (1/1)
SplitTree.Node.makeIterator() 0.00% (0/3)
SplitTree.startIndex.getter 100.00% (3/3)
SplitTree.endIndex.getter 100.00% (3/3)
implicit closure #1 in SplitTree.endIndex.getter 100.00% (1/1)
SplitTree.subscript.getter 100.00% (5/5)
implicit closure #1 in SplitTree.subscript.getter 100.00% (1/1)
implicit closure #2 in implicit closure #1 in SplitTree.subscript.getter
100.00% (1/1)
implicit closure #3 in SplitTree.subscript.getter 0.00% (0/1)
implicit closure #4 in SplitTree.subscript.getter 0.00% (0/1)
SplitTree.index(after:) 100.00% (4/4)
implicit closure #1 in SplitTree.index(after:) 100.00% (1/1)
implicit closure #2 in SplitTree.index(after:) 0.00% (0/1)
SplitTree.Node.structuralIdentity.getter 100.00% (3/3)
SplitTree.Node.StructuralIdentity.init(_:) 100.00% (3/3)
static SplitTree.Node.StructuralIdentity.== infix(_:_:) 100.00% (3/3)
SplitTree.Node.StructuralIdentity.hash(into:) 100.00% (3/3)
SplitTree.Node.isStructurallyEqual(to:) 100.00% (18/18)
implicit closure #1 in SplitTree.Node.isStructurallyEqual(to:) 100.00%
(1/1)
implicit closure #2 in SplitTree.Node.isStructurallyEqual(to:) 100.00%
(1/1)
SplitTree.Node.hashStructure(into:) 100.00% (14/14)
SplitTree.structuralIdentity.getter 100.00% (3/3)
SplitTree.StructuralIdentity.init(_:) 100.00% (4/4)
static SplitTree.StructuralIdentity.== infix(_:_:) 100.00% (4/4)
implicit closure #1 in static SplitTree.StructuralIdentity.==
infix(_:_:) 100.00% (1/1)
SplitTree.StructuralIdentity.hash(into:) 80.00% (8/10)
static SplitTree.StructuralIdentity.areNodesStructurallyEqual(_:_:)
90.00% (9/10)
</pre>
</details>
I chose this as a good place to start contributing to Ghostty because I
was curious about the macOS implementation, and there was a specific
request for help with testing (#7879).
My process for writing the tests was basically reading
[SplitTree.swift](./macos/Sources/Features/Splits/SplitTree.swift) to
understand it, then writing tests for each high-level method and
checking against code coverage to capture all the code paths:
## Running
```bash
rm -rf /tmp/ghostty-test.xcresult
xcodebuild -project macos/Ghostty.xcodeproj \
-scheme GhosttyTest \
-configuration Debug \
test \
-destination 'platform=macOS' \
-enableCodeCoverage YES \
-resultBundlePath /tmp/ghostty-test.xcresult \
-only-testing:GhosttyTests/SplitTreeTests \
2>&1 | xcbeautify
```
## Coverage
```bash
xcrun xccov view --report /tmp/ghostty-test.xcresult | grep 'SplitTree\.'
```
This was originally implemented in [~38
commits](https://github.com/pouwerkerk/ghostty/pull/1/commits), but I
squashed them down to 1 commit for easier review.
## AI Disclosure
The tests were written by me, but I used Opus 4.6 to explain some parts
of the code, and then finally to provide feedback on the tests. It
suggested tests for `nodeStructuralIdentityInSet` and
`nodeStructuralIdentityDistinguishesLeaves` as well as [the
Parameterized
test](6a0bca43f6),
`resizingAdjustsRatio`, which seemed like a clever way to collapse 12
individual tests into 3 parameterized ones that still run 12 cases
total. I didn't know this feature existed, and it seems like a great way
to write tests that are more maintainable. I read this relatively new
feature in the [Swift
Docs](https://developer.apple.com/documentation/testing/parameterizedtesting).
I find this to be a particularly useful feature of Claude/related
agents, where it can suggest better ways of writing something in a more
idiomatic way, and it taught me something new, which is always fun.
I'm more than happy to continue work on tests for #7879 and always
welcome to any feedback you have.
This fixes the issue where our palette generation was changing our
default palette. The default palette is based on some well known values
chosen from various terminals and it was a bit jarring to have it
change.
We now only auto-generate the palette if the user has customized at
least one entry.
This fixes the issue where our palette generation was changing our
default palette. The default palette is based on some well known values
chosen from various terminals and it was a bit jarring to have it
change.
We now only auto-generate the palette if the user has customized at
least one entry.
The HTML page formatter can now track hyperlink state so <a> tags open
and close when the OSC 8 data changes. Also added a new
`writeHtmlEscaped` helper to keep generated markup safe.
Originally written with Copilot, revised by hand.
This changes the way Ghostty assigns itself and subprocesses to
cgroups and how resource controls are applied.
* Ghostty itself no longer modifies it's own cgroup or moves itself
to a transient scope. To modify the main Ghostty process' resource
controls ensure that you're launching Ghostty with a systemd unit and
use the standard systemd methods for overriding and applying changes
to systemd units.
* If configured (on by default), the process used to run your command
will be moved to a transient systemd scope after it is forked from
Ghostty but before the user's command is executed. Resource controls
will be applied to the transient scope at this time. Changes to
the `linux-cgroup*` configuration entries will not alter existing
commands. If changes are made to the `linux-cgroup*` configuration
entries commands will need to be relaunched. Resource limits can also
be modified after launch outside of Ghostty using systemd tooling. The
transient scope name can be shown by running `systemctl --user whoami`
in a shell running inside Ghostty.
Fixes#2084.
Related to #6669
We continue to support bash 3.2 for compatibility with /bin/bash on
macOS. `mapfile` was introduced in bash 4.0, so this change introduces a
`read -r`-based helper function for populating COMPREPLY from a list of
lines.
See: #3042
Hi Ghostty team,
I believe that terminals should generate the 256-color palette based on
the user's base16 theme.
The rationale and approach is written up
[here](https://gist.github.com/jake-stewart/0a8ea46159a7da2c808e5be2177e1783).
I consider it important that terminals support this out of the box so
that such behaviour can become normal and expected, because then
terminal program maintainers will consider the palette a viable choice.
I have created a PR for kitty and the maintainer seems interested. I
plan to offer this to more terminals soon.
