Add named parameters, UI scale setting, OPTIONS/OP dispatch, and help manual

Named parameters (docs/named_parameters_design.md, stages 3-4):
- ParameterTable (src/scene/named_parameters.rs): a document-wide table of
  named formulas, each resolved through a shared expression evaluator with
  cross-parameter references, cycle detection, and division-by-zero/parse
  error reporting. SketchConstraint::driving_param becomes DrivingValue
  (Literal | Named) instead of a bare f64, resolved at solve time by
  sketch_solve::build_constraint via Scene::named_parameters.
- Persistence (src/scene/named_parameters_persist.rs): same lazy XRecord
  model sketch_persist.rs already established, synced around open/save.
  Bumps the sketch-constraint XRecord's own format version (1 -> 2) since
  DrivingValue's shape doesn't line up with a version-1 blob's bytes.
- PARAMETERS command and its editor (src/ui/window/named_parameters.rs):
  add/remove/edit rows, live validation, duplicate-name detection: Apply
  defines the table and re-solves every constraint scope with a
  named-parameter-driven constraint (command_driver.rs). Distance/Angle
  constraint prompts (modules/draw/constrain/value.rs) now accept either a
  literal or an existing parameter's name, mirroring how the constraint's
  own value prompt already worked.
- Undo scope-down carried over from this project's own established pattern
  (constraint-system stage 9): the geometry an Apply moves is one undo
  step, but the parameter table edit itself has no undo entry yet — a
  documented follow-up, not a silent gap.

UI scale setting (src/app/settings.rs, app/mod.rs, ui/window/options.rs):
  a whole-interface scale percentage (50-200%), wired directly into iced's
  own per-window scale_factor hook rather than per-panel font-size
  plumbing, so it applies immediately to every panel, dialog, and the
  ribbon at once. Persisted alongside the existing lineweight-display-scale
  setting.

OPTIONS/OP command dispatch (app/commands/view.rs, app/commands/mod.rs):
  OPTIONS was registered for autocomplete but had no dispatch arm — the
  Start-tab welcome screen's own button was the only way to reach it.
  Found live-testing the UI-scale setting, unrelated to that feature
  itself. "OP" is AutoCAD's own short alias.

help/: a new self-contained static user-help manual (manual.md as the
authored source, tools/build_help.py splits it into the committed static
HTML pages plus search index) — not yet wired into the application.

64 new/updated tests across the named-parameters and constraint-prompt
changes.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
This commit is contained in:
Felix RiestraandClaude Sonnet 5 committed 2026-09-09 13:50:08 +02:00
1 parent e2d39b049d
commit 7cddf31bcb
47 files changed
+4136 -52

No files matched your search

+201
View File
@@ -0,0 +1,201 @@
//! Save/load integration for [`super::named_parameters::ParameterTable`] —
//! design doc §4 stage 2 (the XRecord persistence spike) plus enough of
//! stage 2's "wire it into open/save" to actually be useful, following
//! [`super::sketch_persist`]'s exact precedent (itself the sketch-constraint
//! project's stage 4/§5.2): the "lazy" model, where the table lives only in
//! `Scene::named_parameters` in-memory during editing and is materialized
//! into `document.objects` as one XRecord right before a save, then read
//! back right after a load.
//!
//! The one structural difference from `sketch_persist`: a
//! [`super::named_parameters::ParameterTable`] is document-wide, not
//! per-[`super::sketch_constraints::SketchScope`], so there is exactly one
//! XRecord for the whole document rather than one per scope owner.
//!
//! **Owner handle, and a real round-trip gap found while building this**:
//! the obvious choice was `document.header.named_objects_dict_handle` (the
//! document's own root Named Objects Dictionary) — `extension_dictionary_
//! handle` in the vendored `cadcodec` source explicitly resolves an
//! extension dictionary for a `Dictionary`-type owner (`ObjectType::
//! Dictionary(value) => value.xdictionary_handle`), and that crate's own
//! tests (e.g. `tests/issue51.rs`) use `named_objects_dict_handle` as an
//! XRecord owner directly. In practice this round-tripped through DWG but
//! **silently lost the XRecord through DXF** (caught by this module's own
//! save/load tests, not assumed): `CadDocument::ensure_xrecord` only writes
//! the owner's *own* `xdictionary_handle` field when `get_entity_mut(owner)`
//! succeeds (a real "entity"); a `Dictionary` owner like the root NOD isn't
//! one, so `ensure_xrecord` falls back to a runtime-only side map
//! (`xdic_by_handle`) instead. The DWG writer consults that side map
//! (`io/dwg/dwg_document_builder.rs` et al.); the DXF writer never does (no
//! `xdic_by_handle` reference anywhere under `io/dxf/writer/`) — so a
//! Dictionary-owned extension dictionary is invisible to a DXF save. This
//! is a gap in the vendored crate for this specific owner-type combination,
//! not something introduced here; per this project's own stated preference
//! to avoid needing changes to `acadrust`/`cadkernel`, the fix is to pick a
//! different, already-proven owner rather than patch it.
//!
//! **Owner used instead**: `document.header.model_space_block_handle` — a
//! real `BlockRecord` entity, the exact anchor `sketch_persist`'s own
//! `SketchScope::ModelSpace` case already proved round-trips cleanly
//! through both DWG and DXF. Model space is not a semantically perfect fit
//! for "the whole document" (a parameter table has nothing to do with model
//! space specifically), but it is guaranteed to exist in every
//! `CadDocument` and is a real entity, which is the property that actually
//! matters here.
use super::named_parameters::ParameterTable;
use super::Scene;
const XRECORD_KEY: &str = "OCS_NAMED_PARAMETERS";
/// Prefixed onto the serialized blob so a future schema change can be
/// detected and gracefully skipped rather than silently misreading bytes —
/// same call-out as `sketch_persist::FORMAT_VERSION`, same caveat that the
/// migration path itself is still undesigned.
const FORMAT_VERSION: u8 = 1;
fn encode(table: &ParameterTable) -> Option<Vec<u8>> {
let mut bytes = vec![FORMAT_VERSION];
bytes.extend(bincode::serialize(table).ok()?);
Some(bytes)
}
fn decode(bytes: &[u8]) -> Option<ParameterTable> {
let (&version, body) = bytes.split_first()?;
if version != FORMAT_VERSION {
return None;
}
bincode::deserialize(body).ok()
}
impl Scene {
/// Writes `named_parameters` into the document's `OCS_NAMED_PARAMETERS`
/// XRecord, ready for whatever save call happens next to serialize
/// `self.document` as-is. Called from the same save entry points as
/// `materialize_sketch_constraints_for_save`
/// (`OpenCADStudio::prepare_native_save`, the wasm save path).
///
/// An empty table is skipped rather than writing an empty XRecord — an
/// unconstrained-by-parameters drawing (the common case, at least until
/// this feature is used) should not gain persisted-but-empty
/// bookkeeping on every save.
pub(crate) fn materialize_named_parameters_for_save(&mut self) {
if self.named_parameters.is_empty() {
return;
}
let Some(bytes) = encode(&self.named_parameters) else { return };
let owner = self.document.header.model_space_block_handle;
if owner.is_null() {
return;
}
self.document.ensure_xrecord(owner, XRECORD_KEY);
if let Some(record) = self.document.xrecord_mut(owner, XRECORD_KEY) {
// Overwrite, not append: a resave must replace the prior blob,
// not accumulate one more Chunk entry every time.
record.entries.clear();
record
.entries
.push(acadrust::objects::XRecordEntry::new(310, acadrust::objects::XRecordValue::Chunk(bytes)));
}
}
/// Populates `named_parameters` from the just-loaded `self.document`'s
/// `OCS_NAMED_PARAMETERS` XRecord, if any — called right after a
/// document open installs its `CadDocument` into this `Scene`, the same
/// call sites `load_sketch_constraints_from_document` already has
/// (`OpenCADStudio::on_file_opened`, the automation `"open"`/`"new"`
/// ops). Replaces whatever was already in `named_parameters` (a fresh
/// table for a real open; `"new"`'s blank document has nothing to find
/// anyway).
pub(crate) fn load_named_parameters_from_document(&mut self) {
self.named_parameters = ParameterTable::new();
let owner = self.document.header.model_space_block_handle;
if owner.is_null() {
return;
}
let Some(record) = self.document.xrecord(owner, XRECORD_KEY) else { return };
let Some(bytes) = record.entries.iter().find_map(|entry| match &entry.value {
acadrust::objects::XRecordValue::Chunk(bytes) => Some(bytes.clone()),
_ => None,
}) else {
return;
};
if let Some(table) = decode(&bytes) {
self.named_parameters = table;
}
}
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn a_parameter_table_survives_materialize_and_reload() {
let mut scene = Scene::new();
scene.named_parameters_mut().set("hole_dia", "5").unwrap();
scene.named_parameters_mut().set("hole_spacing", "2 * hole_dia + 1.5").unwrap();
scene.materialize_named_parameters_for_save();
// Simulate the load-time side of a save/reload round trip: drop the
// in-memory table (a fresh open starts empty) and repopulate purely
// from what materialize just wrote into `document.objects`.
scene.named_parameters = ParameterTable::new();
scene.load_named_parameters_from_document();
assert_eq!(scene.named_parameters().len(), 2);
assert_eq!(scene.named_parameters().resolve("hole_spacing"), Ok(11.5));
assert_eq!(scene.named_parameters().get("hole_dia").unwrap().source, "5");
}
/// The above test only proves `materialize`/`load` agree with each
/// other on the same in-memory `CadDocument` — this drives a real
/// `save_to_bytes`/`load_bytes` round trip (the same primitives
/// `OpenCADStudio::on_file_opened`'s native save/open path uses),
/// through both supported formats, mirroring
/// `sketch_persist`'s own real-bytes tests.
fn full_bytes_roundtrip(ext: &str) -> ParameterTable {
let mut scene = Scene::new();
scene.named_parameters_mut().set("plate_len", "42.5").unwrap();
scene.named_parameters_mut().set("plate_width", "plate_len / 2").unwrap();
scene.materialize_named_parameters_for_save();
let bytes = crate::io::save_to_bytes(&scene.document, ext, scene.document.version)
.unwrap_or_else(|e| panic!("save to {ext} bytes: {e}"));
let mut reloaded_scene = Scene::new();
reloaded_scene.document = crate::io::load_bytes(&format!("named_params_roundtrip.{ext}"), bytes)
.unwrap_or_else(|e| panic!("reload {ext} bytes: {e}"));
reloaded_scene.load_named_parameters_from_document();
reloaded_scene.named_parameters
}
#[test]
fn a_parameter_table_survives_a_real_dxf_save_and_load_round_trip() {
let restored = full_bytes_roundtrip("dxf");
assert_eq!(restored.len(), 2);
assert_eq!(restored.resolve("plate_width"), Ok(21.25));
}
#[test]
fn a_parameter_table_survives_a_real_dwg_save_and_load_round_trip() {
let restored = full_bytes_roundtrip("dwg");
assert_eq!(restored.len(), 2);
assert_eq!(restored.resolve("plate_width"), Ok(21.25));
}
#[test]
fn an_empty_table_is_not_materialized() {
let mut scene = Scene::new();
scene.materialize_named_parameters_for_save();
let owner = scene.document.header.model_space_block_handle;
assert!(scene.document.xrecord(owner, XRECORD_KEY).is_none(), "an empty table should not create an XRecord");
}
#[test]
fn decode_rejects_a_mismatched_format_version() {
let mut bytes = vec![FORMAT_VERSION.wrapping_add(1)];
bytes.extend(bincode::serialize(&ParameterTable::new()).unwrap());
assert!(decode(&bytes).is_none(), "a future/unknown format version must be rejected, not misread");
}
}