Cross-platform window manager written in Rust. https://srdusr.com/git/srdwm/atom?h=main 2025-02-15T12:56:00+00:00 2025-02-15T12:56:00+00:00 srdusr 99972264+srdusr@users.noreply.github.com 2025-02-15T12:56:00+00:00 urn:sha1:0a4c4b5941fe982ccb3d3175e26d9d83f0025ffd Safety commit before reconciling this worktree with main, which has diverged with its own separate fixes today. Nothing here is reviewed or curated yet - this exists purely so none of this work can be lost to a git operation, disk issue, or worktree cleanup while that reconciliation happens. 2025-02-14T19:36:00+00:00 srdusr 99972264+srdusr@users.noreply.github.com 2025-02-14T19:36:00+00:00 urn:sha1:413daa7ba2ea0ebd1424c024fd0566423aaea3f8 sync()'s per-tick platform.focus(id) re-assertion (added to keep real Wayland/X11 keyboard focus following core's own bookkeeping) ran unconditionally on every dirty tick, including when nothing about focus had actually changed. focus_window (core) has its own, separate side effect of switching to the focused window's workspace when it differs from the current one - correct when focus genuinely moves to a window on another workspace, but this call was never gated on focus having changed at all: switching workspace via activate_workspace left the still-focused window's own workspace field untouched, so the very next dirty tick's blind re-assertion of that same focus saw a mismatch against the just-changed current_workspace and switched straight back. Confirmed live via temporary core-side logging: two switch_workspace calls a few milliseconds apart, the second one undoing the first every single time, for every workspace switch that didn't also change which window was focused. Gated the re-assertion on the focused id actually changing since the last sync() call. Real focus-follows-real-platform-focus still happens on every genuine change, which is all the original fix needed. 2025-02-11T12:00:00+00:00 srdusr 99972264+srdusr@users.noreply.github.com 2025-02-11T12:00:00+00:00 urn:sha1:b1d69a552cdcc7370c4befd42ef2111c8b367c46 1. A popup's xdg_popup.grab (Firefox's own right-click menu, concretely) receiving zero pointer input at all - no hover highlight, no click effect, not even dismiss-on-miss - logs whether grab_popup actually succeeds, since a silent failure there would explain exactly this. 2. srd dispatch activate_workspace returning {"ok":true} without ever changing the current workspace, confirmed via a raw socket request bypassing the CLI entirely. switch_workspace's own logic reads correct; logs its actual inputs/state to find out why the real process disagrees with it. Remove once both are resolved. 2024-08-09T07:36:00+00:00 srdusr 99972264+srdusr@users.noreply.github.com 2024-08-09T07:36:00+00:00 urn:sha1:31de54096243fb000fa6e7008a914861293482a5 visible_windows' doc comment claimed windows show "on the active workspace of whichever monitor they're assigned to" - the code never reads w.monitor at all; current_workspace is one flat value shared by every monitor, not per-output. Documented that explicitly on both the field and the method, since this is a real behavioral difference from Hyprland worth a reader actually seeing, not just an inaccurate comment to fix quietly. monitor.primary_workspace/monitor.workspace_count describe a per- monitor-workspace design that doesn't exist; workspace.auto_switch/ workspace.persistent were never wired to any behavior. All four were seeded into default_config() and documented in DEFAULTS.md, so srd.set()/srd.get() on them silently succeeded while doing nothing -- removed from both, matching the precedent already set by general. rounded_corners' deliberate absence from default_config for a different reason (backend-dependent default rather than unbuilt). 2024-07-11T12:17:00+00:00 srdusr 99972264+srdusr@users.noreply.github.com 2024-07-11T12:17:00+00:00 urn:sha1:86b557a6c0baa4146b30712d8b7633f1b9b46d43 Pure reorganization, no behavior change - verified by diffing the function-name set before/after (identical 130 functions) plus a full cargo test pass. WindowManager's struct/field definitions, Default, new(), and the three trivial constructors (add_rule/register_layout/ available_layouts) stay in mod.rs; the rest of the single ~950-line impl block is split into one file per the section comments the file already had (monitors, windows, focus, winops, hittest, dragresize, workspaces, layout). Three methods called across section boundaries (monitor_for, windows_on_workspace, cycle_focus) went from private to pub(super) - Rust's privacy model doesn't let sibling submodules see each other's private items, only a defining module's own descendants. The ~1000-line test module moves to manager/tests.rs unsplit: its helpers (wm_with_monitor, two_monitors, monitor_with_dock) are shared across tests for every section, so splitting further would mean duplicating them or adding another shared-support file for little benefit.