Commit Graph
2354 Commits
Author SHA1 Message Date
LGUG2Z 2d6423f13c build(nix): relock the nested flakes after the step 1 and crates changes
The root flake.lock pins the content of komorebi-for-kde and komorebi-for-mac (D14 of the flake phase 2 work: the root passes its own self to the nested mkOutputs, and the lock records the nested directories). Their trees changed with the step 1 crate and the crates directory in the source filter, so the pins are updated with nix flake update kfde kfm. Only flake.lock changes.
2026-10-31 01:37:53 -07:00
LGUG2Z 68601ced53 build(nix): give the KDE and macOS builds the shared crates directory
The nested flakes cut the repository down to their own tree and the root crates they depend on by path (rootCrates). Step 1 added crates/komorebi-collections, which both trees now depend on, so crates/ must be in that list or the root flake cannot build kfde (and kfm cannot evaluate its sources). Listing the directory rather than each crate means later shared crates (protocol, config, ...) are picked up without another flake change. Verified: see the following progress entry.
2026-10-31 01:37:53 -07:00
LGUG2Z 8c22a18dce chore(repo): combine the flake phase 2 work with the integration line
Merge revision: wip-main (history merges, fixes, step 0 path dependencies, step 1 komorebi-collections, KDE A3 fixes) and wip-flake (the root flake builds the KDE package with the whole repository as source, the macOS package evaluates the same way, the root flake no longer requires dirtyRev). The root flake.lock pins the content of the nested flakes, so it is relocked in the next revision.
2026-10-31 01:37:53 -07:00
LGUG2Z 412f380ce7 chore(repo): combine the step 1 crate with the KDE A3 fixes
Merge revision on the integration line: wip-step1 (the komorebi-collections crate and its wiring in all three trees, on top of the step 0 path dependencies) and wip-a3-kde (the KDE path-coupling, test-isolation, dirtyRev and lock-order fixes). No file needed manual resolution. Verification of this merge: not run separately; each side passed its own gates (step 1: Windows clippy, KDE 885 tests, macOS type-check; KDE A3: 895 tests three times).
2026-10-31 01:37:53 -07:00
LGUG2Z 74433e30bc build(nix): build the KDE and macOS packages from the whole repository
The root flake no longer re-exports the nested flakes' own outputs. It keeps them as inputs for their pinned nixpkgs, crane and rust-overlay (inputs.<name>.inputs), but calls their lib.mkOutputs with the whole repository (self) as the source, a nested self carrying the root rev, dirtyRev and inputs, and the tree's directory, then re-exports packages, checks (without vm-*), legacyPackages.kfde-vm, nixosModules, overlays and darwinModules from those. flake.lock is updated for the changed nested flakes: the lock pins their content, so it needs 'nix flake update kfde kfm' whenever a file under komorebi-for-kde or komorebi-for-mac changes.

Verified (Linux, x86_64): nix build -L of packages.kfde-komorebic and kfde-komorebi succeed and both binaries run with --help (exit 0); checks kfde-schema-fresh, nix-options-validation, keybindings-fixtures, komorebi-bridge-js, nixos-module-eval, plasma-guests-eval, run-kde-dry-run, run-kde-doctor, komorebi-workspace-fmt, -deny and -nextest build; kfde-komorebi-workspace-toml-fmt FAILS because komorebi-for-kde/Cargo.toml itself is not taplo-formatted (same result running taplo on the file directly; not caused by this change). nix eval of aarch64-darwin kfm-komorebic and kfm-komorebi drvPath print .drv paths (evaluated, not built). Windows clippy and default drvPaths are identical to those before this revision. nix flake check --no-build passes for x86_64-linux. Not verified: any darwin build, kfde clippy/doc, the VM suites.
2026-10-31 01:37:53 -07:00
LGUG2Z 9b8be3ea70 refactor(kde): take ring and lockable_sequence from the komorebi-collections crate
Deletes the KDE copies, adds the komorebi-collections path dependency via the workspace table, re-exports ring and lockable_sequence from lib.rs, #[macro_use] extern crate for impl_ring_elements!, and forwards the schemars feature. Cargo.lock updated.

Verified: fast-check KDE cargo check, --lib tests and schemas test: PASS.
2026-10-31 01:37:53 -07:00
LGUG2Z 3a8176875e fix(kde): sync-e2e-core checks against the root e2e/lib, which is the source of truth
The e2e core was vendored twice (komorebi/e2e/lib and the KDE VM lib), synced by a script. The script now checks the KDE copy against ../e2e/lib in the root, treating the root as the single source, as the plan's A3 requires. Verified: running the script reports the copies in sync.
2026-10-31 01:37:53 -07:00
LGUG2Z a57ca82f0c fix(nix): do not require dirtyRev in the root flake
A path flake (a jj workspace without .git) has neither rev nor dirtyRev, so the commit stamp default in mkKomorebiPackages failed to evaluate. It now falls back to "unknown", as the nested flakes do.

Verified: nix eval --raw .#checks.x86_64-linux.komorebi-workspace-clippy.drvPath now prints z9nb5j12qlfp1j8d71ykp9qnafk553ra-komorebi-workspace-clippy-0.1.0.drv (it failed before); .#packages.x86_64-linux.default.drvPath is cq0i40300hgsm0ih2llcdjpkcg9sfqaf-komorebi-workspace-0.1.0.drv. Neither was built.
2026-10-31 01:37:53 -07:00
LGUG2Z 334e56b687 refactor(mac): take ring and lockable_sequence from the komorebi-collections crate
Deletes the macOS copies, adds the komorebi-collections path dependency via the workspace table, re-exports ring and lockable_sequence from lib.rs, #[macro_use] extern crate for impl_ring_elements!, and forwards the schemars feature. Cargo.lock updated.

Verified: fast-check macOS fake-CC cargo check --workspace --all-targets: PASS. Not run: tests.
2026-10-31 01:37:53 -07:00
LGUG2Z ef1d3ef096 fix(kde): take the rule locks in one order, so a rule check and a config load cannot deadlock
window.rs and static_config.rs locked the float, manage and other rule lists in conflicting orders, which deadlocked KDE tests and could in the daemon. Both now take the locks in one order. Found by test hangs during the merge work. The Windows and macOS trees have the same pattern by reading and were not changed. Verified: the KDE test suite no longer hangs; the VM tier was not re-run.
2026-10-31 01:37:53 -07:00
LGUG2Z 4fc3f8e139 build(nix): let the macOS flake take the repository as its source
Same change as for the KDE flake: lib.mkOutputs takes inputs, self, the repository source and the tree's directory; the crane source is the macOS tree plus komorebi-layouts, -scrolling, -themes and the root Cargo.toml; cargoLock, cargoToml and postUnpack point into the tree. The nested flake's own outputs call it with ./. and keep evaluating, but cannot build the Rust workspace on their own.

Verified: nix eval --raw of packages.aarch64-darwin.komorebic.drvPath on the nested flake prints a .drv (evaluated only; nothing darwin was built).
2026-10-31 01:37:53 -07:00
LGUG2Z 4a5733eb1d refactor(collections): move ring.rs and lockable_sequence.rs into komorebi-collections
Moves the identical ring.rs and lockable_sequence.rs out of komorebi/src into the crate (rename). Necessary edits: lib.rs with pub use Lockable and a hidden pastey re-export; #[macro_export] on impl_ring_elements! using $crate::pastey::paste!; the doc test now says komorebi_collections::ring::Ring. The Windows tree gets the path dependency, re-exports (pub use komorebi_collections::{ring, lockable_sequence, Lockable}, #[macro_use] extern crate), forwards its schemars feature, and drops pastey again. Windows is bundled here because it is the tree the files move out of.

Verified: fast-check native, Windows clippy --all-targets -D warnings, KDE and macOS checks: PASS.
2026-10-31 01:37:53 -07:00
LGUG2Z b38a26f289 fix(kde): build without self.dirtyRev when the flake is a nested path input
A path-input flake has no dirtyRev attribute, so evaluation failed. Falls back to "unknown" via self.dirtyRev or. An earlier same-derivation comparison was a false positive because both sides printed the same error; this fix came from retracting it. Verified: the KDE package evaluates and builds from the root flake.
2026-10-31 01:37:53 -07:00
LGUG2Z 8c6714ccaf build(nix): let the KDE flake take the repository as its source
The KDE flake's outputs become a function (lib.mkOutputs) of inputs, self, the repository source and the tree's directory in it. The flake's own outputs still call it with ./. and no directory, so everything evaluates standalone, but the Rust workspace can no longer be built from the nested flake alone: since step 0 it depends on ../komorebi-layouts, -scrolling and -themes, which a nested path input cannot see.

The crane source is now the repository cut down to the KDE tree, those three root crates and the root Cargo.toml (their inherited workspace dependencies come from it). cargoLock and cargoToml point into the tree and postUnpack enters it.

Verified: nix eval of packages.x86_64-linux.komorebic.drvPath on the nested flake prints a .drv; the real build is verified through the root flake (see the revision that wires it). Reindentation by nixfmt makes the diff large; the logic change is the mkOutputs wrapper, the src filter and three commonArgs attributes.
2026-10-31 01:37:53 -07:00
LGUG2Z 11e38c8c8a build(collections): add the komorebi-collections crate
New crates/komorebi-collections (edition 2024; deps pastey, serde, optional schemars feature, dev serde_json_lenient) added to the root workspace members, with an empty lib.rs. Cargo.lock for this revision alone is not refreshed (it is in the next revision).

Verified: cargo test -p komorebi-collections --offline built (0 tests yet).
2026-10-31 01:37:53 -07:00
LGUG2Z c3dc3a8e49 fix(kde): give each schemas test its own socket
Tests in komorebi/tests/schemas.rs shared one socket path and raced when run in parallel. Each test now gets its own. Verified: cargo test --workspace in the KDE tree.
2026-10-31 01:37:53 -07:00
LGUG2Z 52b8d2df63 chore(repo): combine the step 0 and KDE A3 work
Empty merge revision joining the step 0 line (path dependencies on the root layouts, scrolling and themes crates) with the KDE A3 line (path-coupling fixes). No file changes of its own, so each parent's content is kept as written. Verified: all three trees type-check on the combined line.
2026-10-31 01:37:53 -07:00
LGUG2Z e9e62319e8 refactor(ring): make the macOS ring.rs identical to the documented one
macOS ring.rs becomes the KDE copy (doc comments, unit tests, std path in the macro; pastey was already used). After this the three ring.rs and lockable_sequence.rs files are byte-identical (cmp).

Verified: fast-check macOS fake-CC cargo check --workspace --all-targets: PASS. Not run: tests (macOS cannot run).
2026-10-31 01:37:53 -07:00
LGUG2Z eaf7228a75 fix(kde): default compare-gen to the komorebi and komorebi-for-mac trees around it
compare-gen.py defaulted to sibling checkouts of the old repos. It now defaults to the Windows tree and macOS tree that sit beside komorebi-for-kde/ in this repository. Verified: runs and finds both trees.
2026-10-31 01:37:53 -07:00
LGUG2Z 79ffe8b7a1 refactor(kde): depend on the layouts, scrolling and themes crates of this repository by path
Step 0. KDE depended on komorebi-layouts and komorebi-themes through git revisions (6d862a11 / e8216924) that exist in no komorebi history. It now uses the path crates in the repository root so one copy is built. Cargo.lock is updated. Verified: cargo check --workspace --locked in the KDE tree.
2026-10-31 01:37:53 -07:00
LGUG2Z 1f2f477d2f refactor(lockable): document LockableSequence in the macOS tree as the other trees do
macOS lockable_sequence.rs becomes the KDE copy; the only difference is the module doc header.

Verified: fast-check macOS fake-CC cargo check --workspace --all-targets: PASS. Not run: tests (macOS cannot run).
2026-10-31 01:37:53 -07:00
LGUG2Z 29457262bb fix(kde): make vm-mac use git+file from a subdirectory of a bigger repository
vm-mac pointed the flake at a repo root that is no longer the KDE directory. It now uses a git+file URL with ?dir=komorebi-for-kde. Verified by evaluation only; the VM was not run for this change.
2026-10-31 01:37:53 -07:00
LGUG2Z ec0c58bc78 refactor(mac): depend on the layouts and themes crates of this repository by path
Step 0 for macOS: replaces the git revision dependencies on komorebi-layouts and komorebi-themes with path dependencies on the root crates, keeping the darwin and schemars features. Verified: macOS type-check with the fake C compiler (check only, nothing linked or run).
2026-10-31 01:37:53 -07:00
LGUG2Z b3519814c8 refactor(ring): make the Windows ring.rs identical to the documented one and use pastey
Windows ring.rs is replaced by the KDE copy (doc comments, tests, pastey::paste!, std::collections::VecDeque path). pastey 0.2 added to the root workspace deps and komorebi/Cargo.toml (paste stays: windows_api.rs and komorebic use it). The VecDeque imports in container.rs and workspace.rs only served the old macro text and are removed.

Verified: fast-check Windows clippy --all-targets -D warnings: PASS (tests compiled, not run: Windows-only).
2026-10-31 01:37:53 -07:00
LGUG2Z 6f6292805d fix(kde): find the KDE tree from run-kde's own location, not git toplevel
git rev-parse --show-toplevel now returns the monorepo root, so scripts/run-kde looked in the wrong place. It now derives the KDE directory from the script path. Verified: the script locates the tree from another working directory.
2026-10-31 01:37:53 -07:00
LGUG2Z e4e196fd76 fix(mac): tolerate a missing self.dirtyRev when the flake is a path input
Same dirtyRev fallback as the KDE flake, for the macOS flake. Verified by evaluation only (darwin outputs cannot be built on Linux).
2026-10-31 01:37:53 -07:00
LGUG2Z 7aee2e53a9 refactor(lockable): define Lockable next to LockableSequence in the Windows tree
Windows lockable_sequence.rs becomes the KDE version (module docs, Lockable defined inline); the trait is removed from lib.rs and re-exported there as pub use lockable_sequence::Lockable, so crate::Lockable paths keep working. Prepares one shared file across the three trees.

Verified: fast-check Windows clippy --all-targets -D warnings: PASS.
2026-10-31 01:37:53 -07:00
LGUG2Z 2435087e2c chore(repo): merge komorebi-for-mac history
Merges the macOS old repo's history, with its original commit hashes, under komorebi-for-mac/. The subtree is byte-identical to the old tip 9a58e511 (306/306 renames in the move commit). Verified: tree equality, git blame, git log --follow, cargo metadata, and that no macOS workflow sits outside komorebi-for-mac/.
2026-10-31 01:37:53 -07:00
LGUG2Z fc02fb9c96 fix(mac): resolve check_schema_docs.py root from the script location
The script assumed it ran from the repository root, which is now komorebi-for-mac/ inside a larger repo. It now finds the root from its own path. Verified by running it from two working directories.
2026-10-31 01:37:53 -07:00
LGUG2Z a7dd44c75b chore(repo): merge komorebi-for-kde history
Merges the KDE old repo's history, with its original commit hashes, under komorebi-for-kde/. The subtree is byte-identical to the old tip 7d19fd40 (654/654 renames in the move commit). Verified: tree equality, git blame, git log --follow, cargo metadata, and that no KDE workflow sits outside komorebi-for-kde/.
2026-10-31 01:37:53 -07:00
LGUG2Z c155917d71 build(nix): re-export the KDE and macOS flakes from the root flake
Phase 1 of A3b. The root flake imports the nested flakes' lib.mkOutputs with the whole repository as source and re-exports packages, checks (without vm-*), nixosModules, overlays and darwinModules with kfde-/kfm- prefixes. Windows outputs are untouched. Verified: the KDE package builds; the macOS package evaluates (not built).
2026-10-31 01:37:53 -07:00
LGUG2Z 2e78aa69df build(nix): keep the other platforms' trees out of the Windows build source
Excludes /komorebi-for-kde and /komorebi-for-mac from the Windows crane source filter so their edits do not invalidate Windows builds and their Cargo.lock files are not swept in. Verified: the Windows clippy check still passes.
2026-10-31 01:37:53 -07:00
LGUG2Z 4712ea3e96 fix(nix): include the pan transcript fixtures in the build source
komorebi/src/pan.rs and komorebic/src/pan.rs include_str! the .txt files in komorebic/fixtures/, which the crane source filter dropped, so the nix build failed. Adds the fixtures to the filter. Verified: nix komorebi-workspace-clippy passes.
2026-10-31 01:37:53 -07:00
LGUG2Z 1f1805f66e fix(e2e): drop the unsafe block nested inside the testwin unsafe block
e2e/guest/testwin/src/main.rs had an unsafe block inside an unsafe block, which clippy -D warnings rejects (unused_unsafe). Removes the inner one without changing behaviour. Verified: Windows clippy passes.
2026-10-31 01:37:53 -07:00
LGUG2Z 524074edfd chore(repo): move into komorebi-for-mac/ 2026-10-30 19:28:07 -07:00
LGUG2Z 8b463317b3 chore(repo): move into komorebi-for-kde/ 2026-10-30 19:28:06 -07:00
LGUG2Z ff3711dcb5 docs(e2e): document the Windows guest and what it found
README.md: building the image, running scenarios, the variables, the known
failures. W1.md: the measurements and findings behind them.
2026-10-30 18:59:32 -07:00
LGUG2Z 0aec364345 feat(e2e): no known failure of maximizing.user once kfw has the fix 2026-10-30 18:59:32 -07:00
LGUG2Z 7f6e530b35 feat(e2e): sync the neutral core from kfde 7d19fd40 2026-10-30 18:54:37 -07:00
LGUG2Z 7d19fd4045 test(nix): maximizing.user has toggle-monocle as both ports do
toggle-monocle leaves IGNORED: it turns the monocle of the tiles that
are left on and off while the window stays maximized, as probed on
komorebi for Windows.
2026-10-30 18:54:19 -07:00
LGUG2Z 3e7451f132 feat(process_command): toggle-monocle acts while a window is maximized
komorebi for Windows does not ignore it while a window is maximized: the
monocle of the tiles that are left comes on and goes off, and the window
stays maximized. Match it, and drop toggle-monocle from the commands
that are about the tiles.
2026-10-30 18:54:19 -07:00
LGUG2Z 503ccbc9c6 feat(e2e): add a probe of toggle-monocle under a user's maximize 2026-10-30 18:49:48 -07:00
LGUG2Z 8cf77519a2 fix(e2e): name toggle-monocle as what maximizing.user still fails 2026-10-30 18:45:15 -07:00
LGUG2Z a2ab755352 feat(e2e): sync the neutral core from kfde b3dc2ed9 2026-10-30 18:25:26 -07:00
LGUG2Z e5c1e2fa75 feat(e2e): run the cargo tests of the workspace in the Windows guest
The tests are cross-compiled for Windows, so they run there: cargo_tests.py
uploads each test executable, lists its tests and runs every one in a
process of its own, as nextest does. The nix check never ran them
(doCheck is false).
2026-10-30 18:22:45 -07:00
LGUG2Z 725c95ffc2 feat(nix): OpenSSL for the host build script in the devshell
komorebic's build script (reqwest) runs on the build host and needs
pkg-config and OpenSSL, as in the nix build, so cargo test --workspace works
from nix develop.
2026-10-30 18:22:45 -07:00
LGUG2Z 345ddf30ea feat(e2e): the known difference of maximizing.user after kfw tracks a user's maximize
On a komorebi with 3fa3d429 the step passes its tracking checks; what is
left is that toggle-float and stack left/right are not ignored while a
window is maximized. The marker says so for such a build.
2026-10-30 18:22:45 -07:00
LGUG2Z b1d053b358 perf(nix): build the test window with the constant stamp too
komorebi-testwin-e2e is the e2e build's test window, so a change to it
rebuilds the crate and not its dependencies (86 s instead of minutes);
build-ref.sh uses it.
2026-10-30 18:22:45 -07:00
LGUG2Z 0c21138e25 fix(e2e): make the test window gray, not blue
The pixel checks of the shared scenarios (the ring of a border by hue) take
the window's own color to be neutral. The default was (48, 80, 128), which
is cool by their measure, so a warm ring beside it read as mixed.
2026-10-30 18:22:45 -07:00
LGUG2Z 7e5f956415 feat(e2e): sync the neutral core from kfde 72631512 2026-10-30 18:22:45 -07:00