add missing macros

[Siel]

No description provided.

[github-actions]

Bencher Report

Branch	feat/add-missing-macros
Testbed	mhovd-pgx

Click to view all benchmark results

Benchmark	Latency	Benchmark Result nanoseconds (ns) (Result Δ%)	Upper Boundary nanoseconds (ns) (Limit %)
Conditional dose modification	📈 view plot 🚷 view threshold	1,150.30 ns (-5.00%) Baseline: 1,210.88 ns	1,265.76 ns (90.88%)
Create large dataset (100 subjects)	📈 view plot 🚷 view threshold	54,220.00 ns (-1.12%) Baseline: 54,834.46 ns	57,230.64 ns (94.74%)
Data expand complex (1h intervals)	📈 view plot 🚷 view threshold	27,498.00 ns (-1.79%) Baseline: 28,000.24 ns	30,265.88 ns (90.85%)
Data expand simple (1h intervals)	📈 view plot 🚷 view threshold	465.57 ns (-5.23%) Baseline: 491.28 ns	521.47 ns (89.28%)
Data expand with additional time	📈 view plot 🚷 view threshold	38,338.00 ns (-1.37%) Baseline: 38,870.59 ns	42,111.34 ns (91.04%)
Filter exclude subjects	📈 view plot 🚷 view threshold	30,767.00 ns (-0.04%) Baseline: 30,780.84 ns	31,503.83 ns (97.66%)
Filter include subjects	📈 view plot 🚷 view threshold	8,148.20 ns (+2.53%) Baseline: 7,947.35 ns	8,379.02 ns (97.25%)
Modify all bolus doses	📈 view plot 🚷 view threshold	1,121.90 ns (-4.60%) Baseline: 1,175.97 ns	1,223.30 ns (91.71%)
Modify all infusion doses	📈 view plot 🚷 view threshold	1,161.40 ns (-3.84%) Baseline: 1,207.75 ns	1,258.22 ns (92.30%)
SubjectBuilder multi-occasion	📈 view plot 🚷 view threshold	261.08 ns (-1.44%) Baseline: 264.89 ns	275.21 ns (94.87%)
SubjectBuilder simple	📈 view plot 🚷 view threshold	102.41 ns (-1.95%) Baseline: 104.45 ns	109.28 ns (93.72%)
SubjectBuilder with covariates	📈 view plot 🚷 view threshold	283.35 ns (+2.10%) Baseline: 277.53 ns	294.58 ns (96.19%)
nca_auc_cmax_metrics	📈 view plot 🚷 view threshold	600.91 ns (+2.28%) Baseline: 587.50 ns	615.81 ns (97.58%)
nca_population/10	📈 view plot 🚷 view threshold	48,527.00 ns (+3.30%) Baseline: 46,974.94 ns	49,975.42 ns (97.10%)
nca_population/100	📈 view plot 🚷 view threshold	125,590.00 ns (+2.50%) Baseline: 122,525.00 ns	127,813.68 ns (98.26%)
nca_population/500	📈 view plot 🚷 view threshold	365,000.00 ns (-9.36%) Baseline: 402,696.11 ns	432,013.37 ns (84.49%)
nca_single_subject	📈 view plot 🚷 view threshold	1,008.80 ns (-0.29%) Baseline: 1,011.70 ns	1,050.42 ns (96.04%)
one_compartment	📈 view plot 🚷 view threshold	29,515.00 ns (+25.09%) Baseline: 23,595.73 ns	30,159.08 ns (97.86%)
one_compartment_covariates	📈 view plot 🚷 view threshold	43,136.00 ns (+36.53%) Baseline: 31,594.68 ns	46,585.36 ns (92.60%)
readme 20	📈 view plot 🚷 view threshold	340,720.00 ns (+3.19%) Baseline: 330,190.27 ns	351,966.08 ns (96.80%)
two_compartment	📈 view plot 🚷 view threshold	39,449.00 ns (+46.87%) Baseline: 26,859.27 ns	41,887.41 ns (94.18%)

🐰 View full continuous benchmarking report in Bencher

[Copilot]

Pull request overview

This PR expands the “declaration-first” modeling surface by adding missing macros and aligning handwritten model APIs with DSL-style metadata, while also updating DSL authoring keywords and runtime route semantics so route/name lookups stay consistent across ODE/Analytical/SDE paths.

Changes:

• Introduces a new shared handwritten model metadata system and wires it into ODE/Analytical/SDE (including name-based index lookups).
• Adds/exports missing macros (analytical!, sde!) and adds macro-lowering parity tests against handwritten models.
• Updates the DSL authoring surface (model= → name=, kernel= → structure=) and updates runtimes/ABI model info to carry richer route identity (declaration index/kind).

Reviewed changes

Copilot reviewed 45 out of 46 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
tests/support/runtime_corpus.rs	Updates embedded DSL fixtures to use name = ... / structure = ....
tests/support/bimodal_ke.rs	Updates DSL fixture to name = ....
tests/sde_macro_lowering.rs	Adds parity tests for sde! lowering vs handwritten SDE (metadata + predictions).
tests/ode_macro_lowering.rs	Adds parity tests for ode! lowering vs handwritten ODE (metadata + predictions).
tests/browser-e2e/site/app.mjs	Updates browser E2E DSL sources from model = to name =.
tests/analytical_macro_lowering.rs	Adds parity tests for analytical! lowering vs handwritten analytical models.
src/test_fixtures.rs	Updates fixture DSL kernel → structure.
src/simulator/equation/sde/mod.rs	Adds metadata attachment + bolus injection mapping hook + SDE event handling changes.
src/simulator/equation/ode/mod.rs	Adds metadata attachment and dimension validation for handwritten ODE models.
src/simulator/equation/mod.rs	Renames meta module to metadata and re-exports DSL enums.
src/simulator/equation/metadata.rs	New shared handwritten metadata builder + validation + lookup helpers.
src/simulator/equation/meta.rs	Removes old Meta container.
src/simulator/equation/analytical/mod.rs	Adds metadata attachment and validation for handwritten analytical models.
src/lib.rs	Re-exports new macros and metadata types through crate root and prelude.
src/dsl/wasm_direct_emitter.rs	Updates route-input load representation to the new ExecutionLoad::RouteInput { .. } form.
src/dsl/wasm_compile.rs	Updates DSL test sources to use name = ....
src/dsl/wasm.rs	Updates WASM tests and route metadata shape expectations (but currently introduces a compile error).
src/dsl/rust_backend.rs	Updates route-input load representation to the new ExecutionLoad::RouteInput { .. } form.
src/dsl/native.rs	Adds route-kind-aware input validation and refines injection semantics for compiled/native runtime.
src/dsl/model_info.rs	Extends native model info route metadata (declaration index/kind) and improves explicit route usage detection.
src/dsl/jit.rs	Updates JIT lowering for new ExecutionLoad::RouteInput { .. } and adds shared-channel test.
src/dsl/compiled_backend_abi.rs	Updates ABI test fixtures for expanded route info fields.
pharmsol-macros/Cargo.toml	Enables syn’s visit feature to support macro parsing/analysis.
pharmsol-dsl/tests/dsl_authoring_edge_cases.rs	Updates DSL authoring fixtures to name = ....
pharmsol-dsl/src/test_fixtures.rs	Updates fixtures to structure = ... and name = ....
pharmsol-dsl/src/semantic.rs	Renames analytical kernel to structure and preserves route kind in typed IR.
pharmsol-dsl/src/parser.rs	Renames analytical block key to structure and updates authoring tests to name = ....
pharmsol-dsl/src/ir.rs	Extends typed IR to store optional RouteKind and renames analytical kernel to structure.
pharmsol-dsl/src/execution.rs	Introduces declaration-index-aware route metadata, route-kind-aware channel assignment, and route-input load encoding.
pharmsol-dsl/src/authoring.rs	Renames model = ... → name = ..., kernel = ... → structure = ..., and preserves route order/kind.
pharmsol-dsl/src/ast.rs	Adds RouteKind and renames analytical kernel to structure in AST + formatter.
examples/two_compartment.rs	Migrates example to declaration-first ode! with metadata-backed indices.
examples/sde_readme.rs	New SDE macro example showing name-based indices + prediction calls.
examples/one_compartment.rs	Migrates example to analytical!/ode! but currently introduces a compile error.
examples/ode_readme.rs	Updates README-style ODE example to ode! and name-based indices.
examples/macro_vs_handwritten_two_cpt.rs	New parity example comparing macro vs handwritten ODE for shared-channel dosing.
examples/macro_vs_handwritten_one_cpt.rs	New parity example comparing macro vs handwritten ODE for one-compartment IV.
examples/dsl_runtime_jit.rs	Updates embedded authoring DSL to name = ....
examples/covariates.rs	Migrates covariates example to ode! with covariate declarations and name-based indices.
examples/compare_solvers.rs	Migrates solver comparison example to declaration-first ode! and shared-channel route semantics.
examples/analytical_vs_ode.rs	Migrates comparison example to macro surfaces and name-based indices.
examples/analytical_readme.rs	New analytical macro example for README-style docs.
README.md	Rewrites quick start to prefer macro surfaces and links to new examples/migration docs.
.gitignore	Adds docs/ to ignored paths.

Comments suppressed due to low confidence (3)

src/dsl/wasm.rs:932

• CompiledModelInfoEnvelope doesn’t define a name field; it requires model: NativeModelInfo. As written this won’t compile and it ignores the model_info variable defined just above. Use the model field here as in other call sites.

    fn rejects_compiled_metadata_abi_version_mismatch() {
        let model_info = loader_test_model_info("metadata_api_version_mismatch");
        let metadata = serde_json::to_vec(&CompiledModelInfoEnvelope {
            abi_version: WASM_API_VERSION + 1,
            name: "model_info",
            kernels: CompiledKernelAvailability {
                outputs: true,
                ..CompiledKernelAvailability::default()
            },
        })

src/dsl/wasm.rs:958

• CompiledModelInfoEnvelope is being constructed with a nonexistent name field and without the required model field. This will fail to compile; pass the NativeModelInfo in the model field instead.

    fn rejects_kernel_metadata_mismatch_from_compiled_envelope() {
        let model_info = loader_test_model_info("kernel_metadata_mismatch");
        let metadata = serde_json::to_vec(&CompiledModelInfoEnvelope {
            abi_version: WASM_API_VERSION,
            name: "model_info",
            kernels: CompiledKernelAvailability {
                outputs: true,
                ..CompiledKernelAvailability::default()
            },
        })

src/dsl/wasm.rs:903

• CompiledModelInfoEnvelope does not have a name field (it has model: NativeModelInfo). This test code will not compile and also drops the model_info value it just computed. Populate the model field instead (and remove the now-unused model_info binding if appropriate).

        let model_info = loader_test_model_info("api_version_export_mismatch");
        let metadata = serde_json::to_vec(&CompiledModelInfoEnvelope {
            abi_version: WASM_API_VERSION,
            name: "model_info",
            kernels: CompiledKernelAvailability {
                outputs: true,
                ..CompiledKernelAvailability::default()
            },
        })

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[mhovd]

It is not apparent to me what an "authoring model" is