#672: #592-D: register GTK scroll surfaces + hit-test caches on the live path, then delete src/gtk/draw.rs - #687
Merged
JDonaghy merged 2 commits intoAug 28, 2026
Conversation
… the live path, delete src/gtk/draw.rs `src/gtk/draw.rs` (3,733 lines under a file-level `#![allow(dead_code)]`) had zero live callers after the #540 Relm4->ShellApp migration, but it was still the sole writer of two live-read caches: - `engine.scroll_surfaces`: GTK's `dispatch_scroll`/`dispatch_click` hit-test this list every event, but under `ShellApp` nothing ever cleared or registered it for the editor's own scroll paths — the terminal/debug-output registrations #670 already ported were never cleared each frame, so stale entries could accumulate. Now cleared and re-registered at the top of every `render_content` pass, matching TUI's `render_impl.rs`. - `status_segment_map`: `click.rs`'s `pixel_to_click_target` reads this to resolve per-window/separated status bar segment clicks (goto-line, change-language, switch-branch, ...) to a `StatusAction`, but nothing populated it live, so those clicks silently resolved to `ClickTarget::None`. Now populated from `backend.status_bar_layout()`'s hit regions right after each status bar paints. - `sync_ui_font_size`/`UI_FONT()`/`UI_FONT_FAMILY`: re-homed into `mod.rs` (still live — read by the raw-Pango chrome) and wired to run once per frame, fixing a latent bug where the UI font size atomic never updated from settings (its only caller was dead code). Also deleted `src/gtk/quadraui_gtk.rs`: `q_theme()` and its two `RICH_TEXT_POPUP_SB_*` constants had no production caller left once `draw.rs` was gone (only a test used `q_theme`, now switched to the `render::to_quadraui_theme` it wrapped), and the two constants had zero callers anywhere. Added a GtkDriver test that opens the debug-output panel, dispatches a wheel scroll over it, and asserts `debug_output_scroll` advances — this fails red against an empty `scroll_surfaces` list and only passes once the surface is actually registered. No file under src/gtk/ carries a file-level #![allow(dead_code)] anymore. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
…parated status line dispatch Review requested a driver test proving status_segment_map's per-window and separated-status-line click sites (mod.rs) are actually wired live, not just populated. Writing that test for the separated status line surfaced a real bug: unlike the per-window status bar (resolved via ScreenZone::Window -> WindowZone::StatusBar), the separated line paints in its own full-width band outside every window's rect, and pixel_to_click_target had no zone type that ever reached it — every click there silently missed (ClickTarget::None), even after #672's paint-time registration fix. Fixes it with a small shared render.rs helper, status_bar_zone_hit_test, that hit-tests an absolute point against a painted bar rect + its status_segment_map zones (mirrors the per-window arm's contract but doesn't require a window rect). click.rs's pixel_to_click_target now checks the separated status bar first, keyed by active_window_id, before falling through to the existing window/tab-bar zone resolution. Threaded a new separated_status_bar parameter through pixel_to_click_target and its three wrapper functions (handle_mouse_click, handle_mouse_double_click, handle_mouse_drag) and all call sites. Added: - gtk::testing::status_bar_segment_click_opens_go_to_line_picker: clicks the per-window status bar's live Ln/Col segment and asserts the go-to-line picker actually paints (picker_popup_rect), not just an engine flag flip. - gtk::testing::separated_status_line_segment_click_opens_go_to_line_picker: same proof for the separated status line, the test that caught the dead dispatch path above. - Two new Harness accessors (status_segment_center, separated_status_segment_center) that locate a segment's painted pixel target from status_segment_map, and a small separated_status_bar_rect App field (mirroring the existing picker_popup_rect/tab_switcher_popup_rect "painted rect for click+test" pattern) so a test can find the separated bar without re-deriving compute_editor_layout's stacking arithmetic. - Three render::tests unit tests pinning status_bar_zone_hit_test's contract directly (bar-rect bounds check, gap-between-segments miss). cargo build/test/clippy/fmt all clean. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Closes #672
Automated PR opened by coordinator for review of issue #672.