nordisk/facett

facett-timeline — crate card

Reverse-engineered (from source — verify, don't trust the name)

• One-line pitch: A swimlane Gantt — a wall-clock time axis with stacked named lanes of segmented phase bars (the Spark jobs/stage/task event timeline).
• Does: From one engine-agnostic TimelineModel (lanes / bars / segments + a [t0,t1] window), TimelineView paints a time axis, a lane gutter, and status-coloured, phase-segmented bars with click-select and scroll-zoom. Covers the Spark Jobs event timeline, per-stage task timeline, and Lakeflow run Gantt (a host maps Spark's event log / REST API onto it; the crate never learns what a "stage" is). Two halves: pure egui-free serializable model, and the Facet. A Msg+update surface is robot/CLI-addressable (select bar by stable id, zoom window, toggle legend); painter, headless layout helper, and state_json all read bar offsets from one formula (TimelineModel::bar_fraction_in), so asserted DATA == pixels. FC: portable_state/load_state.
• Key deps: facett-core (Facet), egui, serde_json. Dev: egui_kittest.
• Shape: lib, ~944 lines: lib.rs (view + BarLayout/layout, 617), model.rs (327). Key types: BarLayout, Msg, TimelineView, model::{TimelineModel,Bar}.
• Observed role in the workspace: A Spark/Databricks-parity timeline pane; a host feeds a mapped TimelineModel and mounts it in a FacetDeck.

Fill in (one line each — only the human knows these)

• Primary consumer(s):
• Why it exists (vs the obvious alternative):
• Hard constraint (airgap / no-C-deps / no-FFI / pure-Rust / etc.):
• Release cadence & independence (rides a parent? own repo someday?):
• Status (production / prototype / experimental / stubbed):
• Must-NOT-change invariant:
• Biggest known gap or TODO:
• Anything a newcomer would get wrong about it: