ferro-ta/CHANGELOG.md

Changelog

All notable changes to ferro-ta are documented in this file.

The format follows Keep a Changelog and the project uses Semantic Versioning.

---

Unreleased

[1.2.0] — 2026-06-29

Added

• Runtime CPU-feature dispatch for SIMD hot paths via the multiversion crate (ferro_ta_core/src/simd.rs). One binary now selects baseline / AVX2-FMA / AVX-512 / NEON kernels at load time via CPUID instead of being pinned at build time, so it runs on any CPU of the target architecture with no illegal-instruction crashes on older chips. The simd feature is on by default; --no-default-features gives a pure-scalar build.
• Broader wheel coverage: release builds now also produce Linux aarch64 (manylinux) and musllinux (x86_64 + aarch64) wheels and a Windows arm64 wheel, alongside the existing Linux x86_64, macOS universal2, and Windows x64 wheels.
• abi3 wheels (cp310-abi3): a single stable-ABI wheel per platform now covers CPython 3.10+ (including future 3.14+), replacing the per-version wheel matrix.
• Dynamic Time Warping (DTW, DTW_DISTANCE, BATCH_DTW): Euclidean-cost DTW with optional Sakoe-Chiba band (window= parameter). DTW() returns (distance, path) with the optimal warping path as an (N, 2) index array; DTW_DISTANCE() is the faster distance-only variant; BATCH_DTW() computes distances from each row of a 2-D matrix to a reference series in parallel via rayon. Distance convention matches dtaidistance.dtw.distance().
• Indicator-specific exception types (FerroTaError, InvalidPeriodError, InsufficientDataError, LengthMismatchError, NumericConvergenceError, InvalidInputError): finer-grained errors for catching specific failure modes. All subclass ValueError so existing except ValueError code keeps working (backward compatible).

Changed

• SIMD is now enabled by default and runtime-dispatched. Replaced the compile-time wide crate (which was never actually enabled in published wheels) with multiversion. The wide Cargo feature is removed; use the default simd feature instead.
• Dependency bumps. Rust: log 0.4.32, serde_json 1.0.150, rayon 1.12.0. API service (api/requirements.txt): uvicorn>=0.49.0, pydantic>=2.13.4, ferro-ta>=1.1.4. CI actions: actions/deploy-pages v5, actions/upload-pages-artifact v5, softprops/action-gh-release v3.
• Python coverage threshold raised from 65% to 80% and enforced in CI.

Fixed

• aarch64 Linux containers can now install ferro-ta. Previously no Linux aarch64 wheel was published, so arm64 images (e.g. AWS Graviton) fell back to an sdist build that failed without a Rust toolchain. A manylinux/musllinux aarch64 wheel is now published.
• Published wheels now actually ship SIMD-accelerated kernels; the prior build enabled no SIMD feature at all.

Security

• SLSA build provenance attestations are now generated for every PyPI wheel and sdist via actions/attest-build-provenance. Verify with gh attestation verify <wheel>.
• Sigstore keyless signatures are now published alongside both CycloneDX/SPDX SBOMs on every GitHub Release (.sig + .pem files).
• ferro_ta_core crate now declares #![forbid(unsafe_code)] to prevent regression — the pure-logic layer has no unsafe code and never will.
• Dependabot now covers the WASM npm package in addition to pip, cargo, and GitHub Actions.
• pip-audit fixes: bumped dev lockfile deps idna 3.18, pytest 9.1.1, and urllib3 2.7.0 to clear PYSEC-2026-215, CVE-2025-71176, and PYSEC-2026-141/142.
• pyo3 advisories triaged: RUSTSEC-2026-0176 and RUSTSEC-2026-0177 are ignored in deny.toml with rationale — ferro-ta uses neither affected code path (PyList/PyTuple nth iterators; PyCFunction::new_closure). The upstream fix requires pyo3 >=0.29 (a large API migration), tracked for a follow-up.

[1.1.3] — 2026-04-02

Added

• Stock instrument (instrument="stock") in PayoffLeg and StrategyLeg for modelling equity-holding strategies (Covered Call, Protective Put, Collar, Covered Strangle, Stock + Spread). Linear payoff identical to futures. Exposed in all three layers: Rust core, Python, and WASM.
• Extended Greeks (extended_greeks): closed-form vanna (∂Δ/∂σ), volga (∂²V/∂σ²), charm (∂Δ/∂t), speed (∂Γ/∂S), and color (∂Γ/∂t) for BSM. Batch vectorisation supported.
• Digital options (digital_option_price, digital_option_greeks): cash-or-nothing and asset-or-nothing pricing (BSM closed-form) plus numerical delta / gamma / vega. Scalar and batch variants.
• American options (american_option_price, early_exercise_premium): Barone-Adesi-Whaley (1987) quadratic approximation — O(1) per evaluation. Scalar and batch variants.
• Historical volatility estimators (all rolling, annualised): close-to-close, Parkinson, Garman-Klass, Rogers-Satchell, Yang-Zhang. Yang-Zhang is ~14× more efficient than close-to-close and handles overnight gaps.
• Volatility cone (vol_cone): min / p25 / median / p75 / max distribution of realised vol across user-specified window lengths — contextualises current IV against historical norms.
• strategy_value: pre-expiry BSM mid-price value of a multi-leg strategy over a spot grid (time value included), complementing strategy_payoff (expiry intrinsic).
• expected_move: log-normal ±1σ expected price range over N days.
• put_call_parity_deviation: detects stale quotes or data errors by computing C − P − (S·e^{−qT} − K·e^{−rT}).
• All new analytics exposed to WASM (wasm/src/lib.rs): extended_greeks, digital_price, digital_greeks, american_price, early_exercise_premium, close_to_close_vol, parkinson_vol, garman_klass_vol, rogers_satchell_vol, yang_zhang_vol, vol_cone, expected_move, put_call_parity_deviation, strategy_payoff_dense, aggregate_greeks_dense, strategy_value_grid.
• aggregate_greeks_dense added to ferro_ta_core::options::payoff (pure Rust, no PyO3/numpy dependency) enabling WASM reuse.
• Comprehensive docstrings (NumPy style with Parameters / Returns / Notes / Examples) on all new Python functions.
• Accuracy test suite tests/unit/test_derivatives_accuracy.py validates digital options, extended Greeks, American options, and vol estimators against scipy and analytical reference formulas.
• scipy added to dev optional dependencies for reference testing.

Changed

• StrategyLeg.expiry_selector, StrategyLeg.strike_selector, and StrategyLeg.option_type are now Optional (None allowed for stock legs). Existing option legs are unaffected.
• docs/derivatives-analytics.md rewritten to cover all new features with runnable examples and an efficiency comparison table for vol estimators.

[1.1.2] — 2026-04-01

Changed

• WASM npm package now ships both Node.js (CommonJS) and browser/web worker (ESM) builds via conditional exports in package.json.
• Fixed wasm-publish.yml: added job condition, workflow_dispatch inputs, and pre-publish test gate.
• Fixed CI.yml: SBOM job now waits for both PyPI and crates.io publish.
• Aligned Node.js version to 20 across all CI workflows.
• Rewrote wasm/README.md and ferro_ta_core/README.md to reflect full feature parity (200+ WASM exports, 22 core modules).

[1.1.1] — 2026-04-01

Added

• Full feature parity across Rust core, Python, and WASM targets.
• 56 new pure-Rust indicator functions in ferro_ta_core: ROC/ROCP/ROCR/ROCR100, WILLR, AROON/AROONOSC, CCI, BOP, STOCHRSI, APO, PPO, CMO, TRIX, ULTOSC, DEMA, TEMA, TRIMA, KAMA, T3, SAR, SAREXT, MAMA, MIDPOINT, MIDPRICE, MACDFIX, MACDEXT, MA (generic dispatcher), MAVP, VAR, LINEARREG variants, TSF, BETA, CORREL, NATR, and 19 math operators/transforms.
• 120+ new WASM bindings: all 61 candlestick patterns (via macro), 9 streaming API structs, options pricing/greeks/IV/chain/surface, futures basis/roll/curve/ synthetic, backtest engine (close-only + OHLCV), walk-forward analysis, Monte Carlo bootstrap, performance metrics, batch operations, portfolio analytics, and signal utilities.
• workflow_dispatch trigger added to wasm-publish.yml for manual npm publishing.

1.0.6 — 2026-03-24

Added

• Added a repo-managed pre-push gate that mirrors the core local CI and release checks, including version and changelog validation, Rust formatting and clippy, Python linting and type checks, tests, docs, and the WASM smoke suite.
• Added generated API manifest tooling and CI coverage so Python and WASM export drift is detected before release candidates are pushed.
• Expanded the Rust-backed implementation surface for analysis and data-heavy workflows, including backtest signal generation, portfolio loops, payoff and Greeks aggregation, chunked indicator execution, and related helper paths.
• Expanded the WASM package surface with additional indicator exports such as WMA, ADX, and MFI, along with refreshed Node examples and conformance coverage against the Python package.

Changed

• Refreshed benchmark wrappers, perf-contract artifacts, and benchmark comparison helpers so the checked-in performance evidence stays aligned with the current feature set.
• Hardened Python CI and local tooling so they run the same typecheck and test entrypoints, including installing the optional MCP dependency needed by the MCP server tests.
• Updated local pre-commit integration to match the current Ruff configuration and refreshed locked dependencies to pick up the audited PyJWT security fix.

Fixed

• One-off benchmark output files produced in the repository root are now ignored so local benchmarking no longer dirties the repo by default.
• Tightened API typing and MCP helper behavior so the stricter lint and typecheck pipeline passes consistently before release.

1.0.4 — 2026-03-24

Added

• The optional MCP server now exposes the broad public ferro-ta callable surface, including exact top-level exports, non-top-level public analysis and tooling functions, and generic stored-instance tools for stateful classes and returned callables.
• Added a dedicated TA_LIB_COMPATIBILITY.md document so the full TA-Lib coverage matrix remains available without bloating the project homepage.

Changed

• Reworked the root README into a shorter product-first landing page with a compatibility summary and docs map, and refreshed MCP documentation to match the expanded server behavior.
• Updated the MCP implementation to use generated tool registration over the public API while keeping the legacy lowercase aliases (sma, ema, rsi, macd, backtest) available for existing clients.
• Refreshed locked Python dependency resolutions for the latest low-risk direct updates in this release cycle.

Fixed

• The repository no longer tracks the stray .coverage artifact, and coverage outputs are now ignored consistently.
• MCP tests now cover generated tool discovery, stored-instance workflows, and callable-reference execution paths so the broader server surface does not regress silently.

1.0.3 — 2026-03-24

Added

• Public package metadata helpers: ferro_ta.__version__, ferro_ta.about(), and ferro_ta.methods() for quick API discovery across the top-level, indicators, data, and analysis surfaces.
• A standalone derivatives benchmark runner covering selected Black-Scholes-Merton pricing, implied-volatility recovery, Greeks, and Black-76 pricing paths with reproducible machine/runtime metadata, per-run timing samples, variance stats, and Python-tracked allocation snapshots.
• A one-command version bump helper, scripts/bump_version.py, plus make version VERSION=X.Y.Z for aligned Cargo, Python, WASM, Conda, and docs release surfaces.

Changed

• Tightened the homepage and docs product narrative so the core Rust-backed Python TA library leads, while adjacent tooling is called out separately.
• Strengthened benchmark evidence and support documentation with clearer benchmark caveats, support-matrix pages, and more explicit release/version consistency guidance.

Fixed

• Python CI now recognizes the top-level metadata API in type stubs, and the derivatives benchmark smoke test no longer depends on importing the benchmarks package from an installed wheel layout.
• The tag-driven GitHub Release workflow now uses a valid glob trigger and an explicit semantic-version validation step, so pushing v1.0.3-style tags correctly creates the release that fans out into the publish jobs.

1.0.2 — 2026-03-24

Performance

• Optimized rolling statistical kernels (CORREL, BETA, LINEARREG*, TSF) with incremental window math and matching warmup semantics.
• Vectorized Python analysis hotspots in options, backtesting, features, and rank-composition paths, reducing Python-loop overhead on common workflows.
• Added grouped multi-indicator execution for shared-input workloads and refactored batch execution around explicit series-major workspaces.

Added

• Reproducible perf-contract artifacts for single-series, batch, streaming, SIMD, TA-Lib comparison, and WASM benchmark runs.
• Hotspot and TA-Lib regression gates suitable for CI perf smoke coverage.
• Streaming, SIMD, and WASM benchmark scripts plus updated performance docs and benchmark playbook.

1.0.1 — 2026-03-24

Added

• crates/ferro_ta_core/README.md is now shipped with the published Rust crate, and ferro_ta_core metadata now points documentation to docs.rs.

Fixed

• CI has been modularized into focused workflow files (ci-rust.yml, ci-python.yml, ci-wasm.yml, ci-docs.yml) while keeping the release publishing jobs in CI.yml for PyPI and crates.io trusted-publisher compatibility.
• The ci-complete gate no longer fails successful runs because of an escaped shell variable, and the release SBOM job now uses a valid anchore/sbom-action version.
• The npm publish workflow now uses GitHub OIDC trusted publishing, installs the wasm32-unknown-unknown target, and no longer depends on an NPM_TOKEN secret.
• The WASM npm package now removes the generated pkg/.gitignore during prepack, so the published tarball includes the built pkg/ artifacts.

1.0.0 — 2026-03-23 (initial stable release)

Performance

• SMA/EMA (src/overlap/sma.rs, src/overlap/ema.rs): Replaced per-bar ta::SimpleMovingAverage / ta::ExponentialMovingAverage state-machine objects with ferro_ta_core::overlap::sma (O(n) sliding-window sum) and ferro_ta_core::overlap::ema (O(n) recurrence). SMA/EMA now run at 200–600 M bars/s on 1 M input.
• WMA (crates/ferro_ta_core/src/overlap.rs, src/overlap/wma.rs): Replaced O(n × period) double-loop with an O(n) incremental algorithm using running weighted sum T[i] = T[i-1] + n·close[i] - S[i-1] and sliding sum S. ~10× speedup vs previous implementation for large periods.
• BBANDS (crates/ferro_ta_core/src/overlap.rs, src/overlap/bbands.rs): Replaced O(n × period) per-window variance with O(n) sliding sum and sum_sq accumulators (var = sum_sq/n - mean²). ~10× speedup.
• MACD (crates/ferro_ta_core/src/overlap.rs, src/overlap/macd.rs): Replaced ta::MovingAverageConvergenceDivergence per-bar object with a pure-Rust implementation. Fast and slow EMAs now advance in a single combined loop to minimise allocation and memory round-trips.
• MFI (src/momentum/mfi.rs): Removed per-bar ta::DataItem::builder().build() allocation. Replaced ta::MoneyFlowIndex with ferro_ta_core::volume::mfi — a direct O(n) sliding-window implementation on raw high/low/close/volume slices. ~5× speedup.
• batch_sma / batch_ema (src/batch/mod.rs): Batch functions now delegate to ferro_ta_core O(n) implementations instead of constructing per-bar ta indicator objects.

Fixed

• Rust clippy: Removed dead code compute_ema function from src/extended/mod.rs.
• fuzz/Cargo.toml: Added [workspace] table to prevent cargo workspace detection error (same fix as wasm/Cargo.toml).
• Python lint: Replaced deprecated typing.Dict/List/Tuple/Type with built-in equivalents across 21 Python files (ruff UP035).
• Type checking (mypy): Fixed _normalize_rust_error return type to NoReturn; fixed type errors in _utils.py, crypto.py, chunked.py, regime.py, features.py, dsl.py, mcp/__init__.py.
• Type checking (pyright): Set reportMissingImports = false to handle Rust extension and optional deps; fixed gpu.py cupy handling with Any type annotation.
• Sphinx docs: Fixed RST title underline lengths; fixed unexpected indentation in plugins.rst; fixed invalid :doc: references in index.rst and contributing.rst.
• Sphinx autodoc: Fixed conf.py to not override sys.path when the wheel is installed; added suppress_warnings for autodoc import failures.
• CI test coverage: Added pandas, polars, hypothesis, pyyaml to CI test dependencies; coverage threshold adjusted from 80% to 65% (up from failing 59%).
• Exception hierarchy: All FerroTAError subclasses now accept code and suggestion keyword arguments; validation helpers (check_timeperiod, check_equal_length, check_finite, check_min_length) populate error codes and actionable suggestion hints.

Added

• Dependabot: Added .github/dependabot.yml for weekly automated dependency updates (Python, Rust, GitHub Actions).
• Error codes: Every FerroTAError exception now carries a short code (e.g. FTERR001–FTERR006) for programmatic handling; see ferro_ta.exceptions.ERROR_CODES.
• Observability / Logging (ferro_ta.logging_utils): New module with enable_debug(), disable_debug(), debug_mode() context manager, log_call(), benchmark(), and traced() decorator. Re-exported from the ferro_ta namespace.
• API discovery (ferro_ta.api_info): New ferro_ta.indicators(category=None) function listing all 160+ indicators with metadata; ferro_ta.info(func) returning signature, docstring and parameter info. Re-exported from the ferro_ta namespace.
• Developer experience: Added Makefile with make dev/build/test/lint/fmt/typecheck/docs/bench/audit/clean targets; added .devcontainer/devcontainer.json for zero-friction VS Code/Codespaces onboarding; added TROUBLESHOOTING.md for common build issues.
• Security: Added deny.toml for cargo-deny license and advisory checking.
• Test fixtures: Added tests/fixtures/ohlcv_daily.csv (252-bar synthetic OHLCV dataset); added tests/test_integration.py with end-to-end indicator tests on the fixture.

Changed

• Python 3.10 minimum: Dropped support for Python 3.8 and 3.9. requires-python is now >=3.10 so optional dependencies (e.g. mcp) resolve correctly with uv/pip. CI, docs, PLATFORMS.md, VERSIONING.md, CONTRIBUTING.md, and conda recipe updated accordingly.

Added — Rust-first migration: streaming, extended indicators, math operators

• Rust streaming classes (src/streaming/mod.rs): All 9 streaming classes (StreamingSMA, StreamingEMA, StreamingRSI, StreamingATR, StreamingBBands, StreamingMACD, StreamingStoch, StreamingVWAP, StreamingSupertrend) are now PyO3 #[pyclass] types compiled into _ferro_ta. Zero Python overhead for bar-by-bar updates in live-trading use. Python streaming.py re-exports the Rust classes from the _ferro_ta extension; there is no Python fallback (the extension must be built).
• Rust extended indicators (src/extended/mod.rs): All 10 extended indicators (VWAP, SUPERTREND, DONCHIAN, ICHIMOKU, PIVOT_POINTS, KELTNER_CHANNELS, HULL_MA, CHANDELIER_EXIT, VWMA, CHOPPINESS_INDEX) now compute entirely in Rust. The SUPERTREND sequential band-adjustment loop, DONCHIAN/CHANDELIER rolling max/min, and CHOPPINESS_INDEX rolling window are now O(n) monotonic deque operations in Rust — no Python loops remain.
• Rust rolling math operators (src/math_ops/mod.rs): rolling_sum, rolling_max, rolling_min, rolling_maxindex, rolling_minindex — all using O(n) prefix-sum or monotonic deque algorithms. Python SUM, MAX, MIN, MAXINDEX, MININDEX in math_ops.py now delegate to Rust.
• docs/rust_first.md: New Rust-first architecture policy document. Defines the Python/Rust boundary, porting rules, forbidden patterns, a checklist for new indicator PRs, and a status table of all modules.
• ferro_ta.raw expanded: Added streaming classes (StreamingSMA, …), extended indicator functions (supertrend, donchian, vwap, …), and rolling math operators (rolling_sum, rolling_max, …) to raw.py.
• docs/index.rst: Added link to docs/rust_first.md.

Added — Rust batch API, raw submodule, stability docs, and production polish

• Rust batch API: Added src/batch/mod.rs with batch_sma, batch_ema, batch_rsi Rust functions that accept 2-D numpy arrays and process all columns in a single Rust call (one GIL release for all columns). Eliminates the per-column Python round-trip in the previous implementation.
• Python batch fast path: ferro_ta.batch.batch_sma/ema/rsi call the Rust batch functions for 2-D input (no Python fallback; extension required). The generic batch_apply remains for arbitrary indicators that do not have a Rust batch implementation.
• ferro_ta.raw submodule: New python/ferro_ta/raw.py that re-exports all compiled Rust functions without pandas/polars wrapping, validation, or _to_f64 conversion. Use when you have pre-converted float64 arrays and need minimal overhead. Includes the new batch_sma/ema/rsi Rust functions.
• docs/stability.md: New API stability policy document: stable vs experimental tiers, versioning table, deprecation policy (keep deprecated name for ≥1 minor release with DeprecationWarning).
• docs/plans/2026-03-08-production-grade.md: Implementation plan tracking all parts of the production-grade plan with status and commit references.
• ndarray dependency: Added ndarray = "0.16" to Cargo.toml to support 2-D array operations in the batch Rust module.
• docs/index.rst: Added link to docs/stability.md.
• CONTRIBUTING.md: Added uv-based development workflow as the recommended setup path; pip-based alternative preserved for users who prefer it.
• RELEASE.md: Added security audit step (cargo audit + pip-audit) to pre-release checklist; added CHANGELOG completeness requirement.

Added — Performance, uv, CI improvements, and architecture docs

• _to_f64 fast path: 1-D C-contiguous float64 NumPy arrays are returned as-is (zero copy/allocation) instead of always calling np.ascontiguousarray.
• polars zero-copy result: polars_wrap now builds pl.Series from the NumPy buffer via pl.Series(name, np.asarray(result)) instead of the O(n) .tolist() path, improving polars throughput for all indicators.
• uv project manager support: Added [tool.uv] section to pyproject.toml with dev-dependencies; added a dev extra to [project.optional-dependencies]. Development workflow: uv sync --extra dev then uv run pytest tests/.
• CI — separate optional jobs: Rust tarpaulin coverage moved to a dedicated rust-coverage job (marked continue-on-error: true at job level, not step level); fuzz job similarly isolated. All required CI steps are in blocking jobs. The continue-on-error flag is no longer scattered across individual steps, making failures visible in the CI summary.
• CI — uv in lint/typecheck/audit: lint, typecheck, and audit jobs install uv and run tools via uv run --with <tool>.
• Docs — docs/architecture.md: New document describing the two-crate Rust layout, Python binding flow, module table, packaging details, and where validation lives.
• Docs — docs/performance.md: New guide covering the fast path for contiguous arrays, raw _ferro_ta API, pandas/polars overhead, batch limitations, streaming characteristics, and practical tips.
• Docs — docs/index.rst: Added links to architecture and performance docs.

Added — Production-grade hardening (validation, CI, docs)

• Validation: All Python indicator wrappers now call check_timeperiod() and check_equal_length() where applicable and re-raise Rust ValueError as FerroTAValueError/FerroTAInputError via _normalize_rust_error(). New helper check_min_length() in ferro_ta.exceptions.
• CI: Coverage gate (pytest --cov-fail-under=80), lint job (ruff check + format), pyright in typecheck job, CHANGELOG check for PRs, audit and fuzz no longer use continue-on-error.
• Docs: docs/error_handling.rst, docs/api/exceptions.rst, CONTRIBUTING updated for modular Rust layout (src/pattern/mod.rs + per-pattern files), Sphinx release from FERRO_TA_VERSION env.
• Tests: tests/test_validation.py (invalid timeperiod, mismatched lengths, empty/short arrays, exception inheritance), tests/test_property_based.py (Hypothesis), hypothesis optional dependency.
• Tooling: Ruff and pre-commit config (.pre-commit-config.yaml), mypy warn_return_any = true, pyright in CI, RELEASE.md and SECURITY.md updated.

Added — TA-Lib numerical parity documentation

• Added MAMA, SAR/SAREXT, and all HT_* tests to tests/test_vs_talib.py with documented justification for each remaining "Corr/Shape" difference.
• issues/Stages1-10.md created with known-difference table for MAMA, SAR, SAREXT, HT_DCPERIOD, HT_DCPHASE, HT_PHASOR, HT_SINE, HT_TRENDLINE, HT_TRENDMODE.

Added — Pure Rust core library

• New Cargo workspace: root Cargo.toml declares workspace members [".","crates/ferro_ta_core"].
• crates/ferro_ta_core — pure Rust library crate with no PyO3/numpy dependency.
• Core modules: overlap (SMA/EMA/WMA/BBANDS), momentum (RSI/MOM), volatility (ATR/TRANGE), volume (OBV), statistic (STDDEV), math (SUM/MAX/MIN).
• cargo test -p ferro_ta_core passes (12 tests).
• CI rust-core job: cargo build -p ferro_ta_core && cargo test -p ferro_ta_core.
• README and CONTRIBUTING describe the two-layer architecture.

Added — Batch execution API

• New ferro_ta.batch module: batch_sma, batch_ema, batch_rsi, batch_apply.
• Accepts 2-D (n_samples × n_series) arrays; returns same shape.
• 1-D input falls back to single-series behaviour (backward compatible).
• Exported from ferro_ta.__init__; documented in docs/batch.rst.

Added — Documentation CI

• New CI job docs: installs Sphinx + ferro_ta, runs sphinx-build -b html docs docs/_build -W.
• docs/batch.rst and docs/api/batch.rst added; linked from docs/index.rst.
• Feature list in docs/index.rst updated to mention batch API and Rust core.

Added — Rust coverage

• CI rust job installs cargo-tarpaulin and collects XML coverage for ferro_ta_core.
• Coverage artifact rust-coverage uploaded per-run.
• CONTRIBUTING updated with cargo tarpaulin instructions.

Added — Community governance (issues/ directory)

• issues/Stages1-10.md — full issue text for stages 1–10 (linked from ROADMAP.md).
• issues/Stages11-20.md — stage overview for stages 11–20.

Added — Release and versioning playbook

• RELEASE.md — step-by-step release playbook (version bump → CHANGELOG → tag → PyPI verify).
• CI version-check job: fails if Cargo.toml and pyproject.toml versions diverge.
• CONTRIBUTING.md updated with release process, changelog policy, and fuzzing instructions.

Added — Optional GPU backend (PyTorch)

• ferro_ta.gpu module: sma, ema, rsi — GPU-accelerated when PyTorch is available.
• ferro_ta[gpu] optional extra in pyproject.toml.
• docs/gpu-backend.md — design doc with scope, limitations, and benchmark table.
• benchmarks/bench_gpu.py — CPU vs GPU benchmark script.

Added — WASM binding expansion

• WASM macd() added to wasm/src/lib.rs (7 indicators total).
• CI WASM job builds package and uploads wasm-pkg artifact.
• wasm/README.md updated with Node.js + browser examples and CI artifact docs.

Added — Fuzzing and robustness

• fuzz/ directory with cargo-fuzz targets for SMA and RSI.
• CI fuzz job: nightly Rust, 1000 iterations per target, uploads crash artifacts.
• Fuzzing instructions added to CONTRIBUTING.md.

Added — Indicator pipeline / composition API

• ferro_ta.pipeline module: Pipeline class, make_pipeline factory.
• Chain multiple indicators; results returned as a named dictionary.
• Supports multi-output indicators (BBANDS, MACD) via output_keys.

Added — Polars integration

• Transparent polars.Series support via polars_wrap decorator in _utils.py.
• ferro_ta[polars] optional extra in pyproject.toml.
• Polars Series in → Polars Series out; NumPy path unchanged.

Added — Configuration and defaults management

• ferro_ta.config module: set_default, get_default, get_defaults_for, reset, list_defaults.
• Config context manager for temporary parameter overrides.
• Thread-local storage — safe for concurrent tests.

Added — Additional WASM indicators

• WASM mom() (Momentum) and stochf() (Fast Stochastic) added (9 indicators total).
• Tests for both new indicators in wasm/src/lib.rs.
• wasm/README.md updated with expanded indicator table.

Added — Jupyter notebook examples

• examples/quickstart.ipynb — core API tour (SMA, RSI, MACD, BBANDS, batch, pipeline, pandas).
• examples/streaming.ipynb — streaming bar-by-bar API demonstration.
• examples/backtesting.ipynb — backtesting harness, pipeline feature engineering, config defaults.
• examples/README.md — index of all notebooks with run instructions.

Added — v1.0 preparation and API stability

• VERSIONING.md updated with API stability guarantees and compatibility matrix.
• ROADMAP.md updated: stages 15–20 marked Done.
• README updated with Pipeline, Polars, and Config API sections.

Added — Alternative language bindings (WASM)

• New wasm/ directory: WebAssembly bindings via wasm-bindgen / wasm-pack.
• Exposes sma, ema, bbands, rsi, atr, obv for Node.js and browsers.
• wasm/README.md — build & usage instructions; wasm/package.json.
• CI job wasm builds and tests the WASM crate with wasm-pack test --node.

Added — Distribution & packaging maturity

• Python 3.13 added to CI test matrix.
• conda/meta.yaml — Conda recipe for conda-forge / local channel builds.
• Supported platforms documented in PLATFORMS.md.

Added — Type stubs & typing

• python/ferro_ta/py.typed marker added (PEP 561 compliance).
• pyproject.toml [tool.mypy] section added for IDE / CI use.
• Typing :: Typed PyPI classifier present in pyproject.toml.

Added — Error model & validation

• ferro_ta.exceptions module: FerroTAError, FerroTAValueError, FerroTAInputError.
• Validation helpers: check_timeperiod, check_equal_length, check_finite.
• All three exception classes exported from ferro_ta top-level namespace.

Added — Backtesting utilities

• ferro_ta.backtest module: backtest() entry point, BacktestResult container.
• Built-in strategies: rsi_strategy (RSI 30/70) and sma_crossover_strategy.
• Clear scope note: "minimal harness for testing strategies."

Added — CI/CD & quality expansion

• pytest-cov coverage reporting added to CI (tests job); coverage XML uploaded.
• CHANGELOG.md (this file).
• VERSIONING.md — semantic versioning policy and release playbook.

Added — Plugin / extension system

• ferro_ta.registry module: register, unregister, get, run, list_indicators.
• All built-in indicators auto-registered at import time.
• FerroTARegistryError raised for unknown indicator names.

---