OpenCADStudio/plugins#901 (comment)
- Add BuiltinPlugin::on_load(&mut self, host: &mut dyn HostApi) with a default
no-op implementation so existing plugins remain source-compatible.
- Invoke on_load once for V4 plugins immediately after the runner connects
and before entering the request loop, giving plugins a HostApi surface to
cache plugin_request_sender() or set up worker threads before any dispatch.
- V2/V3 plugins are unchanged: HostApi is only available during dispatch.
The 9 themed* functions that take &'static [u8] artwork each built a
fresh svg::Handle::from_memory(bytes) on every call, with 113
icons::themed call sites per frame across ribbon/statusbar/popups.
Add a thread_local THEMED_CACHE keyed by the bytes' (address, length)
and a single themed_handle(bytes) helper; the 9 themed* functions all
go through it. The 13 wrapper functions (themed_arrow_*, themed_home,
themed_undo, themed_redo, themed_check_cell, ...) delegate to those 9
and pick up the cache transparently, so all 22 themed* functions (and
the 113 call sites) are deduped.
The cache key is (address, length) -- not pointer alone -- to match the
existing SEMANTIC_CACHE belt-and-braces pattern and guard against any
future call site that passes a sub-slice of a static (e.g. &BYTES[1..]),
which would otherwise alias the wrong handle. Every include_bytes!
produces a unique static slice today, so the length is defensive.
Construction is amortised to once per artwork per process. No semantic
change; visual output is byte-identical.
Measured performance gains (integrated
themed_cache_speedup_over_uncached_handle_creation bench, 20_000
iterations each, same machine):
| | dev profile | release profile |
|----------------|---------------|-----------------|
| Before | 235 ns/iter | 26 ns/iter |
| After (cached) | 78 ns/iter | 9 ns/iter |
| Cold miss | 900 ns | 100 ns |
| Speedup | 3.0x | 2.9x |
Release-mode warm path: 9 ns/iter. At 60 fps with up to 113 themed*
calls per frame, that is ~1 us/frame of handle work -- a rounding
error against a 16.6 ms frame budget. The cold-miss cost (one parse
+ one insert, 100 ns release) is the one-time price per new artwork
and is amortised to negligible per-frame overhead (it happens at
most once per artwork per process). The speedup is stable at ~3x
across dev and release because the ratio of work is the same:
hashmap-get + Handle clone vs. a full SVG parse + allocation.
test ui::icons::themed_cache_tests::themed_cache_returns_same_handle_for_same_bytes
test ui::icons::themed_cache_tests::themed_cache_caches_distinct_bytes_separately
test ui::icons::themed_cache_tests::themed_cache_speedup_over_uncached_handle_creation
Add a new #[cfg(test)] mod chprop_integration_tests to src/app/properties.rs
with four end-to-end tests that exercise the converted CHPROP / MATCHPROP
handlers through the real Message dispatcher and the CLI command driver:
- prop_color_change_multiple_entities_undo_redo: a colour change on two
selected lines flows through Message::PropColorChanged -> the property-op
handler -> apply_property_op, and is reversible via undo / redo. Seeds
the panel's source_handles (selected_handles() is empty in this test
setup) so the converted non-empty branch runs.
- prop_change_on_locked_layer_is_ignored: when a selected entity's layer
is locked, property_target_handles filters it out and the handler's
empty-handles branch leaves the entity itself untouched. Asserts the
entity-level invariant only; the empty branch updates the document
header (CECOLOR) and sets dirty as a separate, pre-existing behaviour.
- ribbon_lineweight_updates_after_change: the Home-ribbon Lineweight chip
is updated by the converted RibbonLineweightChanged handler. Uses
acadrust::types::LineWeight::from_value(50) (the real constructor;
Lineweight::from_mm does not exist).
- cli_chprop_undo_restores_previous_state: regression test for the CLI
CHPROP command. Before Mission #15 it recorded the undo snapshot after
mutating the entities, so undo restored the post-edit state (a no-op
undo); converting the site to apply_property_op moved the snapshot
before the mutation. #[ignore]d: CLI CHPROP dispatch via
run_command_line / automation-op "run" does not reach the handler in
this test setup. The helper-level invariant is already covered by
apply_property_op_tests::applies_to_each_handle_sets_dirty_and_mutates
and the strengthened missing_entity_does_not_abort_loop.
The fifth proposed test (matchprop_empty_dest_no_dirty) was not added:
MATCHPROP is the interactive command driver (source pick -> dest pick)
and cannot be exercised from a unit test in isolation. The empty-handles
early-return is already covered at the helper level by
empty_handles_does_not_invoke_or_dirty.
Setup lessons that shaped these tests (panel vs scene selection,
LineWeight::from_value, the layer-locking pattern, undo not clearing
dirty) are documented in MISSION_15_REVIEW.md section 6.2.
Full suite: 478 passed / 1 pre-existing fail (diameter_dim_bakes_through_center,
unrelated) / 5 ignored. No regression. No production code changed -- the
new module is #[cfg(test)]-only.
test app::properties::chprop_integration_tests::prop_color_change_multiple_entities_undo_redo
test app::properties::chprop_integration_tests::cli_chprop_undo_restores_previous_state
test app::properties::chprop_integration_tests::prop_change_on_locked_layer_is_ignored
test app::properties::chprop_integration_tests::ribbon_lineweight_updates_after_change
Extract the push-undo → per-handle → invalidate → dirty → refresh
contract into one helper on OpenCADStudio. The helper is the single
source of truth for property-op side effects, so future property
handlers can't forget the cache invalidation or the panel refresh.
Note: MATCHPROP now no-ops when the destination selection is empty
(previously it marked the tab dirty and refreshed the panel). This is
a minor correctness improvement; restore the old behaviour with a
2-line guard at the MATCHPROP call site if desired.
test app::properties::apply_property_op_tests::empty_handles_does_not_invoke_or_dirty
test app::properties::apply_property_op_tests::applies_to_each_handle_sets_dirty_and_mutates
test app::properties::apply_property_op_tests::missing_entity_does_not_abort_loop
Sebastian