* chore(docs): rename benchmark CPU from 7950X3D to 9950X
The "Reproduced on" line in the umbrella + binding READMEs and the
benchmark page on the site listed the wrong AMD CPU. The benchmarks
were actually produced on a Ryzen 9 9950X, not a 7950X3D. Same
column for absolute µs values applies — the speedup ratios in the
tables are unchanged either way because they're relative across
libraries on the same machine.
The performance-regression issue template's CPU example also
updated for consistency (it was a generic placeholder, but matching
the canonical machine makes the example concrete).
* chore(npm): restore Windows ARM64 sub-package + napi matrix entry
npm Support unblocked the `wickra-win32-arm64-msvc` package name and
transferred write access to @kingchenc (placeholder 0.0.1-security
was published from their side; we ship our first real version on
top of that). This re-enables every change 8aa74cb temporarily
backed out for 0.2.1:
- bindings/node/package.json: re-add `aarch64-pc-windows-msvc` to
napi.triples.additional and `wickra-win32-arm64-msvc` to
optionalDependencies.
- bindings/node/npm/win32-arm64-msvc/package.json: restored — name,
cpu = arm64, os = win32, version pinned to the workspace.
- .github/workflows/release.yml: re-enable the
`windows-11-arm / aarch64-pc-windows-msvc` row in the node-build
matrix and drop the "temporarily skipped" comment block.
After the next tag-push this binding will be published alongside
the other five platforms and `npm install wickra` on Windows ARM64
will resolve to a native build instead of failing the loader's
optional-dep lookup.
* release: bump workspace + bindings to 0.2.7
Workspace, every binding (Python, Node, six platform stubs incl. the
restored win32-arm64-msvc), and the CHANGELOG all move together to
0.2.7. wickra-win32-arm64-msvc is now part of the standard publish
matrix and will land on npm alongside the other five binaries.
The 0.2.7 CHANGELOG entry consolidates the two changes this cycle:
- Windows ARM64 binding restored (npm Support unblocked the name).
- Benchmark CPU label corrected (Ryzen 9 9950X, not 7950X3D).
* fix(docs-rs): rename `doc_auto_cfg` to `doc_cfg` after Rust 1.92 merge
`doc_auto_cfg` was removed in Rust 1.92.0 and folded back into
`doc_cfg` (rust-lang/rust#138907). docs.rs builds with the latest
nightly and sets `--cfg docsrs`, so the previous
#![cfg_attr(docsrs, feature(doc_auto_cfg))]
aborts compilation with E0557 on every published 0.2.x. GitHub CI
never tripped this — stable rustc ignores the line because nothing
sets the `docsrs` cfg there.
Switch all three published library crates (`wickra`, `wickra-core`,
`wickra-data`) to the merged-into `doc_cfg` gate. Same intent, same
on-docs.rs output, builds again on nightly.
* docs(readme): float Wickra to the top of the comparison tables
Reorders the "Why Wickra exists" library-comparison table and the two
benchmark headers so Wickra is the first row (with a ★ marker) instead
of the last. The previous order placed Wickra at the bottom, which
buries the only row a reader landing on the README is here to compare
against. Same column data, same ★/winner annotations, just the row
order flipped and a ★ prefix on the Wickra label.
Mirrored across the umbrella README and every binding README so the
crates.io / PyPI / npm landing pages stay in sync.
* release: bump workspace + bindings to 0.2.6
Workspace, every binding (Python, Node, Node platform stubs), the
release.yml comment and the CHANGELOG all move together to 0.2.6 so
the next tagged release lines every artefact up.
0.2.6 carries two changes from the [0.2.6] CHANGELOG entry:
- fix(docs-rs): swap the now-removed `doc_auto_cfg` feature gate for
the merged-into `doc_cfg` so docs.rs nightly builds resume.
- docs(readme): float ★ Wickra to the top of every comparison table
across the umbrella + binding READMEs.
wickra-win32-arm64-msvc stays excluded for this release with the same
npm spam-filter rationale that held for 0.2.5.
Adds five new issue templates (bug_report_detailed, feature_request_detailed,
performance_regression, documentation, question) alongside the existing
short forms, plus an optional detailed PR template under
.github/PULL_REQUEST_TEMPLATE/ that contributors can opt into via the
?template=detailed.md URL. Existing templates keep their behavior; only
title prefixes and prose-paren wording were capitalized for consistency.
Workspace, every binding (Python, Node, Node platform stubs), and the
release.yml comment are all updated together so the next tagged release
on `v0.2.5` lines every artefact up.
Also adds a short README "Disclaimer" section pointing out that Wickra
is an indicator toolkit, not a trading system, and that production
use is at the caller's own risk. The legal terms in LICENSE (PolyForm
Noncommercial 1.0.0, "No Liability") already cover the warranty / as-is
language — the README section just makes the trading-specific framing
visible without burying it in a click-through.
CHANGELOG carries the new 0.2.5 entry with the API addition
(`BinanceConfig` + `connect_with_config`) and the best-effort Pong
write change in `BinanceKlineStream::next_event`. wickra-win32-arm64-msvc
stays excluded for this release with the same npm-spam-filter rationale
that held for 0.2.1.
* refactor(binance): introduce BinanceConfig for endpoint + timing knobs
Replaces the file-private READ_TIMEOUT / MAX_RECONNECT_ATTEMPTS / size
limit constants with a Default-equipped BinanceConfig the caller can hand
to a new connect_with_config(). connect() forwards to it with the
defaults, so the public API stays backwards-compatible.
Behaviour-preserving: every default matches the value of the constant it
replaces, and the WebSocketConfig is built the same way. The change
unlocks two real use-cases — pointing at Binance Testnet
(wss://testnet.binance.vision) and pointing at a local mock server with
millisecond-scale reconnect timing in tests.
* test(binance): cover the Interval table and the empty-symbol guard
Three quick wins that don't need a live or mock socket:
- interval_as_str_covers_every_variant pins every wire-format mapping in
one table so a typo on any of the 14 variants is caught.
- binance_config_default_matches_production_endpoint guards the default
base URL and timing knobs against an accidental drift.
- connect_rejects_an_empty_symbol_list exercises the guard before the
WebSocket handshake — the one async path we can hit without a server.
* test(binance): cover the async / reconnect / control-frame paths
Adds a small mock-WebSocket scaffold built on a `127.0.0.1:0` listener
and tokio-tungstenite's `accept_async`, plus nine integration tests that
drive `BinanceKlineStream::next_event` through every branch:
- text + binary kline frames decode to a KlineEvent
- inbound Ping is answered with a Pong, then the kline arrives
- inbound Pong / Frame variants are silently skipped
- a server-side Close triggers a transparent reconnect that then
serves the kline
- a stalled connection trips read_timeout and reconnects on its own
- close() flips the closed flag and next_event() yields None forever
- when every reconnect attempt is refused, next_event surfaces an Err
- a "kline" envelope whose numbers are unparseable bubbles up as
Error::Malformed rather than being silently skipped
`one_shot_server` drops the listener as soon as the first accept is
done, so a follow-up reconnect lands on a refused port — that is what
lets the exhaustion test hit the final `last_err.expect(...)`.
The whole suite runs in ~4 s with millisecond-scale reconnect timings
supplied via the new test-only [`test_config`].
* test(binance): drop defensive cold-paths in the mock-WS scaffolding
Codecov's patch report on PR #34 flagged seven uncovered lines, all of
them in the test scaffolding rather than in production code:
- the `let Ok((stream, _)) = … else { return }` shortcut and the
`if let Ok(ws) = accept_async(stream).await { … }` branch in the
mock-server helpers — both error arms never fire on a passing test
- the closing braces of the spawned-task bodies in the close-frame and
read-timeout reconnect tests — the spawned async blocks were getting
killed mid-drain when the test asserted and returned
Refactor the helpers to `.unwrap()` every Result (a failure here is a
bug in the scaffold, not in production) and have `multi_shot_server`
accept a fixed `n_accepts`, await every spawned inner task, and hand
the outer JoinHandle back to the caller.
Refactor the two affected tests to capture that JoinHandle, collapse
the per-index `if/else` so both arms reach the same trailing
expression, swap the read-timeout drain for a bounded sleep, and
await `server_done` at the end. Every handler now reaches its closing
brace before the runtime is torn down, so coverage on the patch should
collapse from 97.89 % to 100 %.
* test(binance): cover the non-kline-skip path and simplify the Ping arm
After the scaffolding fix landed three lines on binance.rs were still
uncovered:
- L305 / L313: the Text- and Binary-arm "frame was not a kline, keep
reading" fall-throughs. No existing test drove the loop through a
non-kline frame followed by a kline; the new
`next_event_skips_non_kline_frames_and_returns_the_next_kline` does
exactly that (Text ack, Binary id frame, then a real kline).
- L317: the Ping-Err defensive arm that forced a reconnect when the
Pong reply itself failed to write. A failed Pong reply means the
socket is already dead, so the very next read will surface the error
and reconnect through the existing timeout/err branch — one tokio
scheduling iteration later. Drop the defensive arm and write the
Pong reply best-effort. Same observable behaviour, no test back
door, no dead-line guard.
Repo coverage on `cov/binance-mock-ws` now sits at 100 %.
After PR #27 brought psar.rs to 99.03 %, Codecov still flagged the
'violation found' tuple arms in the trend tests (line 256 in
pure_uptrend_sar_below_lows, line 285 in pure_downtrend_sar_above_highs)
as missed: both tests are designed to NEVER find a violation, so the
filter_map branch that constructs the (index, sar, bound) tuple is dead
by design.
Restructure both tests to use `.all(|(i, sar)| sar.is_none_or(|s|
<bound>))` instead of collecting violations into a Vec. The closure
runs on every emitted Some, asserts the SAR-vs-extreme bound directly,
and the iterator short-circuits on the first false — no cold tuple
construction left to count as uncovered. Semantics are identical (still
asserts every SAR sits on the correct side of every candle's extreme);
the diagnostic message loses the violating index list, which the tests
never printed in any green run anyway.
psar.rs is now at 207/207 lines, no behavioural change.
* test(balance_of_power): cover name metadata
Codecov flagged 3 lines (file at 96.25%): Indicator-impl name body (73-75).
* test(median_price): cover name metadata
Codecov flagged 3 lines (file at 94.44%): Indicator-impl name body (62-64).
* test(true_range): cover name metadata
Codecov flagged 3 lines (file at 95.94%): Indicator-impl name body (73-75).
* test(typical_price): cover name metadata
Codecov flagged 3 lines (file at 94.44%): Indicator-impl name body (62-64).
* test(weighted_close): cover name metadata
Codecov flagged 3 lines (file at 94.44%): Indicator-impl name body (61-63).
* test(ema): cover period accessor + warmup/name metadata
Codecov flagged 9 lines in crates/wickra-core/src/indicators/ema.rs
(file at 94.03%): const accessor period (74-77), Indicator-impl
warmup_period (123-125), name (131-133). ema.rs now at 151/151.
* test(historical_volatility): cover periods/value accessors + name metadata
Codecov flagged 9 lines in crates/wickra-core/src/indicators/historical_volatility.rs
(file at 93.87%): const accessors periods (80-83), value (85-88) and
Indicator-impl name (153-155). historical_volatility.rs now at 147/147.
* test(kama): cover periods accessor + warmup/name metadata
Codecov flagged 9 lines in crates/wickra-core/src/indicators/kama.rs
(file at 91.26%): accessor periods (65-67), Indicator-impl
warmup_period (115-117), name (123-125). kama.rs now at 103/103.
* test(linreg_angle): cover period accessor + warmup/name metadata
Codecov flagged 9 lines in crates/wickra-core/src/indicators/linreg_angle.rs
(file at 88.15%): const accessor period (50-52), Indicator-impl
warmup_period (67-69), name (75-77). linreg_angle.rs now at 76/76.
* test(mass_index): cover periods/value accessors + name metadata
Codecov flagged 9 lines in crates/wickra-core/src/indicators/mass_index.rs
(file at 91.42%): const accessors periods (80-82), value (85-87) and
Indicator-impl name (134-136). mass_index.rs now at 105/105.
* test(vertical_horizontal_filter): cover period accessor + name metadata
Codecov flagged 6 lines (file at 94.44%): period (61-63) + name (119-121).
* test(z_score): cover period accessor + name metadata
Codecov flagged 6 lines (file at 93.75%): period (59-61) + name (106-108).
* test(vpt): cover value() Some branch, name, zero-prev fallback
Codecov flagged 5 lines (file at 94.38%): value() Some branch (57),
prev==0.0 ROC fallback (77), and Indicator-impl name (100-102).
Add accessors_and_metadata covering value()/name and zero_previous_
close_contributes_zero — feeding a 0.0 baseline + non-zero candle
proves the divide-by-zero guard yields a 0 contribution rather than NaN.
* test(csv): cover from_csv_reader + kill rejects_header dead panic arm
Codecov flagged 5 lines in csv.rs (file at 96.98%): from_csv_reader
(201-204) — never called by existing tests which use from_reader /
open — and the cold arm in
rejects_header_missing_a_column (279). Add from_csv_reader_accepts_a_
prebuilt_reader (demonstrates the API by building a custom-delimited
csv::Reader and passing it in), and refactor the header-missing test
to use a single matches!() assertion so the panic arm is gone.
* test(adl): cover name metadata
Codecov flagged 3 lines (file at 96.84%): Indicator-impl name body (94-96).
* test(keltner): cover periods accessor + name metadata
Codecov flagged 6 lines (file at 95.23%): periods (68-70) + name (106-108).
* test(linreg): cover period accessor + name metadata
Codecov flagged 6 lines (file at 96.10%): period (92-94) + name (142-144).
* test(linreg_slope): cover period accessor + name metadata
Codecov flagged 6 lines (file at 95.91%): period (80-82) + name (125-127).
* test(macd): cover periods/value accessors + name metadata
Codecov flagged 6 lines (file at 95.45%): periods (81-83) + name (135-137).
* test(super_trend): cover params accessor + name metadata
Codecov flagged 6 lines (file at 96.36%): params (99-101) + name (176-178).
* test(chaikin_volatility): cover periods accessor + name metadata
Codecov flagged 6 lines (file at 94.91%): periods (69-71) + name (99-101).
* test(chande_kroll_stop): cover params accessor + name metadata
Codecov flagged 6 lines (file at 95.45%): params (97-99) + name (164-166).
* test(chandelier_exit): cover params accessor + name metadata
Codecov flagged 6 lines (file at 95.12%): params (83-85) + name (128-130).
* test(choppiness_index): cover period accessor + name metadata
Codecov flagged 6 lines (file at 95.04%): period (73-75) + name (125-127).
* test(force_index): cover period accessor + name metadata
Codecov flagged 6 lines (file at 93.33%): period (58-60) + name (93-95).
* test(rsi): cover period/value accessors, name, naive flat-series branch
Codecov flagged 7 lines in indicators/rsi.rs (file at 96.42%): const
accessors period (60-62), value (65-67), Indicator-impl name (145-147),
and line 167 in the test-helper rsi_naive's ag==0 fallback. The
proptest reference never lands on a fully flat series so the helper's
50.0 branch was dead.
Add accessors_and_metadata covering period/value/name and
naive_helper_flat_series_yields_50 driving rsi_naive on [42.0; 20] —
both avg_gain and avg_loss converge to 0, hitting the 50.0 branch.
rsi.rs now at 196/196.
* test(accelerator_oscillator): cover params accessor + name metadata
Codecov flagged 6 lines in indicators/accelerator_oscillator.rs (file
at 93.68%): const accessor params (69-71) and Indicator-impl name
(99-101). ac.rs now at 95/95.
* test(aroon): cover period accessor + name metadata
Codecov flagged 6 lines in indicators/aroon.rs (file at 94.28%): const
accessor period (56-58) and Indicator-impl name (104-106). aroon.rs
now at 105/105.
* test(atr_trailing_stop): cover params accessor + name metadata
Codecov flagged 6 lines in indicators/atr_trailing_stop.rs (file at
95.91%): const accessor params (77-79) and Indicator-impl name
(130-132). atr_trailing_stop.rs now at 147/147.
* test(chaikin_oscillator): cover periods accessor + name metadata
Codecov flagged 6 lines in indicators/chaikin_oscillator.rs (file at
95.27%): const accessor periods (76-78) and Indicator-impl name
(109-111). chaikin_oscillator.rs now at 127/127.
* test(mfi): cover period accessor, name, flat-TP fallback
Codecov flagged 8 lines in indicators/mfi.rs (file at 93.10%): const
accessor period (58-60), (0.0, 0.0) arm when tp==prev (85), the
Some(50.0) flat-flow fallback (105), and Indicator-impl name body
(132-134). Add accessors_and_metadata and flat_typical_prices_default_to_50.
mfi.rs now at 116/116.
* test(psar): cover warmup/name, drop cold format-arg + panic-only asserts
Codecov flagged 8 lines in indicators/psar.rs (file at 95.69%):
warmup_period (206-208), name (220-222), the cold format-arg line
254 in pure_uptrend_sar_below_lows, and the in-loop assert! at line
275 in pure_downtrend_sar_above_highs (its panic body is dead).
Add accessors_and_metadata for warmup/name. Refactor both trend
tests to collect violations into a Vec and assert once outside the
loop — the single assert can now legitimately reach its panic body
in a regression, while removing the dead cold-path lines from the
happy-path coverage.
* test(cmf): cover period accessor, name, zero-range branch
Codecov flagged 7 lines in indicators/cmf.rs (file at 95.03%): const
accessor period (71-73), the range==0.0 zero-MFV branch (84), and
Indicator-impl name body (124-126). Add accessors_and_metadata and
zero_range_candle_contributes_zero_mfv (flat H=L=close candles).
cmf.rs now at 141/141.
* test(hma): cover period accessor + name, kill dead naive panic arm
Codecov flagged 7 lines in indicators/hma.rs (file at 92.22%): const
accessor period (51-53), Indicator-impl name body (87-89), and the
unreachable arm at line 167 in matches_independent_wmas.
Refactor that test to assert the warmup-shape invariant via
assert_eq!(got.is_some(), want.is_some()) + if let, removing the
dead panic arm. Add accessors_and_metadata covering period/name.
hma.rs now at 90/90.
* test(obv): cover value() Some branch + warmup/name metadata
Codecov flagged 7 lines in indicators/obv.rs (file at 92.92%): the
Some(self.total) branch of value() (47) — only the None branch was
hit by reset_clears_state — plus Indicator-impl warmup_period
(79-81), name (87-89). Add accessors_and_metadata covering all four.
obv.rs now at 99/99.
* test(tsi): cover periods/value accessors + name metadata
Codecov flagged 9 lines in indicators/tsi.rs (file at 92.30%): const
accessors periods (70-72), value (75-77) and Indicator-impl name
(137-139). tsi.rs now at 117/117.
* test(ultimate_oscillator): cover periods/value accessors + name metadata
Codecov flagged 9 lines in indicators/ultimate_oscillator.rs
(file at 94.76%): const accessors periods (96-98), value (101-103)
and Indicator-impl name (193-195). uo.rs now at 172/172.
* test(vortex): cover period/value accessors + name metadata
Codecov flagged 9 lines in indicators/vortex.rs (file at 93.18%): const
accessors period (84-86), value (89-91) and Indicator-impl name
(157-159). vortex.rs now at 132/132.
* test(vwma): cover period/value accessors + name metadata
Codecov flagged 9 lines in indicators/vwma.rs (file at 92.56%): const
accessors period (72-74), value (77-79) and Indicator-impl name
(129-131). vwma.rs now at 121/121.
* test(zlema): cover period/value accessors + name metadata
Codecov flagged 9 lines in indicators/zlema.rs (file at 90.62%): const
accessors period (62-64), value (72-74) and Indicator-impl name
(111-113). zlema.rs now at 96/96.
* test(mom): cover period/value accessors + name metadata
Codecov flagged 9 lines in indicators/mom.rs (file at 89.53%): const
accessors period (56-58), value (61-63) and Indicator-impl name
(101-103). mom.rs now at 86/86.
* test(sma): cover period accessor + warmup/name metadata
Codecov flagged 9 lines in indicators/sma.rs (file at 93.12%): const
accessor period (70-72), Indicator-impl warmup_period (115-117),
name (123-125). sma.rs now at 131/131.
* test(stoch_rsi): cover periods/value accessors + name metadata
Codecov flagged 9 lines in indicators/stoch_rsi.rs (file at 92.37%):
const accessors periods (69-71), value (74-76) and Indicator-impl
name (131-133). stoch_rsi.rs now at 118/118.
* test(tema): cover period accessor + warmup/name metadata
Codecov flagged 9 lines in indicators/tema.rs (file at 83.63%): const
accessor period (45-47), Indicator-impl warmup_period (67-69), name
(75-77). tema.rs now at 55/55.
* test(trima): cover period/value accessors + name metadata
Codecov flagged 9 lines in indicators/trima.rs (file at 89.53%): const
accessors period (59-61), value (64-66) and Indicator-impl name
(99-101). trima.rs now at 86/86.
* test(cmo): cover period/value accessors + name metadata
Codecov flagged 9 lines in crates/wickra-core/src/indicators/cmo.rs
(file at 92.30%): const accessors period (66-68), value (71-73) and
Indicator-impl name (134-136). cmo.rs now at 117/117.
* test(dema): cover period accessor + warmup/name metadata
Codecov flagged 9 lines in crates/wickra-core/src/indicators/dema.rs
(file at 85.00%): const accessor period (43-45), Indicator-impl
warmup_period (63,65,66) and name (72-74). dema.rs now at 60/60.
* test(donchian): cover period accessor + warmup/name metadata
Codecov flagged 9 lines in crates/wickra-core/src/indicators/donchian.rs
(file at 90.21%): const accessor period (57-59), Indicator-impl
warmup_period (95-97), name (103-105). donchian.rs now at 92/92.
* test(dpo): cover period/value accessors + name metadata
Codecov flagged 9 lines in crates/wickra-core/src/indicators/dpo.rs
(file at 91.74%): const accessors period (73-75), value (83-85) and
Indicator-impl name (132-134). dpo.rs now at 109/109.
* test(ease_of_movement): cover period/divisor accessors + name metadata
Codecov flagged 9 lines in crates/wickra-core/src/indicators/ease_of_movement.rs
(file at 94.15%): const accessors period (83-85), divisor (88-90) and
Indicator-impl name (141-143). ease_of_movement.rs now at 154/154.
* test(wma): cover period/warmup/name + kill dead naive panic arm
Codecov flagged 10 lines in crates/wickra-core/src/indicators/wma.rs
(file at 92.48%): const accessor period (56-58), Indicator-impl
warmup_period (111-113), name (119-121), and line 186 — the
`_ => panic!("warmup mismatch")` arm in matches_naive_over_random_
inputs, an invariant guard that never fires when both streams share
a warmup period.
Add accessors_and_metadata covering the three metadata methods.
Refactor matches_naive_over_random_inputs to assert the warmup-shape
invariant via assert_eq!(g.is_some(), w.is_some()) + if let,
removing the dead panic arm.
wma.rs is now at 133/133 lines, no behavioural change.
* test(aroon_oscillator): cover period/value accessors + name metadata
Codecov flagged 9 lines in crates/wickra-core/src/indicators/aroon_
oscillator.rs (file at 90.42%): const accessors period (57-59),
value (62-64) and Indicator-impl name (90-92). warmup_period is
already covered by warmup_period_matches_aroon.
Add accessors_and_metadata asserting period == 7, name ==
"AroonOscillator", and value() across the None (pre-warmup) and
Some (post-warmup) branches.
aroon_oscillator.rs is now at 94/94 lines, no behavioural change.
* test(atr): cover period/value accessors + name metadata
Codecov flagged 9 lines in crates/wickra-core/src/indicators/atr.rs
(file at 93.70%): const accessors period (54-57), value (59-62) and
Indicator-impl name body (103-105). warmup_period is exercised
indirectly via downstream indicators; the metadata getters were
never queried directly.
Add accessors_and_metadata asserting period == 14, name == "ATR",
and value() across the None (pre-warmup) and Some (post-warmup)
branches.
atr.rs is now at 143/143 lines, no behavioural change.
* test(awesome_oscillator): cover periods accessor + warmup/name metadata
Codecov flagged 9 lines in crates/wickra-core/src/indicators/awesome_
oscillator.rs (file at 88.15%): const accessor periods (59-61),
Indicator-impl warmup_period (83-85), name (91-93). The classic()
constructor is covered indirectly through the existing tests; only
the metadata methods were dead.
Add accessors_and_metadata asserting periods == (5, 34),
warmup_period == 34 (= slow_period), name == "AwesomeOscillator".
awesome_oscillator.rs is now at 76/76 lines, no behavioural change.
* test(cci): cover period accessor + warmup/name metadata
Codecov flagged 9 lines in crates/wickra-core/src/indicators/cci.rs
(file at 89.65%): const accessor period (68-70), Indicator-impl
warmup_period (102-104), name (110-112). Existing tests never
inspected the metadata surface.
Add accessors_and_metadata asserting period == 20, warmup_period ==
20, name == "CCI".
cci.rs is now at 87/87 lines, no behavioural change.
* test(pmo): cover periods/value accessors, name, zero-prev fallback
Codecov flagged 10 lines in crates/wickra-core/src/indicators/pmo.rs
(file at 90.56%):
- const accessors periods (76-78), value (81-83) — never queried
- line 103 (`0.0` in the prev == 0.0 ROC fallback) — every existing
test used prices > 0, so the divide-by-zero guard never fired
- Indicator-impl name body (130-132) — never queried
Add accessors_and_metadata covering periods/value/name. Add
zero_previous_price_treats_roc_as_flat seeding prev_price = 0 then
pushing a non-zero price — the ROC must take the flat-momentum
fallback (0.0) and the doubly-smoothed PMO emits exactly 0.0
rather than NaN.
pmo.rs is now at 106/106 lines, no behavioural change.
* test(ppo): cover periods/value accessors, name, zero-slow-EMA fallback
Codecov flagged 10 lines in crates/wickra-core/src/indicators/ppo.rs
(file at 90.29%):
- const accessors periods (71-73), value (76-78) — never queried
- line 96 (`0.0` in the s == 0.0 PPO fallback) — every existing test
used prices ≈ 100, so the slow EMA was never 0 and the
divide-by-zero guard never fired
- Indicator-impl name body (122-124) — never queried
Add accessors_and_metadata covering periods/value/name. Add
zero_slow_ema_yields_zero_ppo feeding a stream of zeros — both EMAs
converge to 0.0 and the indicator must emit exactly 0.0 (flat
momentum) rather than NaN.
ppo.rs is now at 103/103 lines, no behavioural change.
* test(roc): cover period accessor, warmup/name, zero-prev fallback
Codecov flagged 10 lines in crates/wickra-core/src/indicators/roc.rs
(file at 87.80%):
- const accessor period (47-49) — never queried
- line 70 (`0.0` in the prev == 0.0 ROC fallback) — every test used
prices ≥ 1.0, so the divide-by-zero guard never fired
- Indicator-impl warmup_period (83-85), name (91-93) — never queried
Add accessors_and_metadata covering period == 5, warmup_period == 6
(= period + 1), name == "ROC". Add zero_previous_price_yields_zero_roc
feeding a leading zero followed by `period` more values so the front
of the window is exactly 0.0; the next emission must be the
flat-momentum fallback 0.0 (not NaN).
roc.rs is now at 82/82 lines, no behavioural change.
* test(ulcer_index): cover period/value accessors, name, zero-max fallback
Codecov flagged 10 lines in crates/wickra-core/src/indicators/ulcer_index.rs
(file at 93.86%):
- const accessors period (77-80), value (82-85) — never queried
- line 123 (`0.0` in the max_price == 0.0 drawdown fallback) — every
test used prices > 0, so the trailing-max divisor was always positive
- Indicator-impl name body (162-164) — never queried
Add accessors_and_metadata covering period/value/name. Add
zero_max_price_yields_zero_drawdown feeding a stream of zeros — the
trailing max is exactly 0.0 and the drawdown computation would
otherwise hit 0/0 NaN; the indicator must emit exactly 0.0
(drawdown is 0% by convention).
ulcer_index.rs is now at 163/163 lines, no behavioural change.
* test(williams_r): cover period accessor, warmup/name, zero-range branch
Codecov flagged 10 lines in crates/wickra-core/src/indicators/williams_r.rs
(file at 89.79%):
- const accessor period (49-51) — never queried
- line 78 (`Some(-50.0)` in the range == 0.0 fallback) — every test
used H != L candles, so the lookback range was always positive
- Indicator-impl warmup_period (87-89), name (95-97) — never queried
Add accessors_and_metadata covering period == 14, warmup_period == 14,
name == "WilliamsR". Add zero_range_yields_minus_fifty feeding flat
candles (H == L == close) — the lookback hi/lo coincide and the
divide-by-zero guard fires, returning the neutral mid-range value
-50.0.
williams_r.rs is now at 98/98 lines, no behavioural change.
* test(t3): cover period/volume_factor/value accessors + name metadata
Codecov flagged 12 lines in crates/wickra-core/src/indicators/t3.rs
(file at 91.48%): const accessors period (95-97), volume_factor
(100-102), value (105-107) and Indicator-impl name (148-150). The
warmup_period method is already covered by first_emission_at_warmup_
period; the other four metadata methods were never queried.
Add accessors_and_metadata asserting period == 5, volume_factor == 0.7,
name == "T3", and value() across both the None (pre-warmup) and Some
(post-warmup) branches.
t3.rs is now at 141/141 lines, no behavioural change.
* test(adx): cover period accessor, warmup/name metadata, zero-TR branch
Codecov flagged 11 lines in crates/wickra-core/src/indicators/adx.rs
(file at 94.17%): the const accessor period (89-91), the tr_v == 0.0
defensive branches inside update (142, 147), and the Indicator-impl
warmup_period (199-201) and name (207-209) bodies.
Add accessors_and_metadata asserting period == 14, warmup_period == 28,
name == "ADX". Add zero_true_range_yields_zero_di_and_zero_adx feeding
flat all-zero candles (H == L == close == 0) — every TR is 0, so the
smoothed tr_smooth stays at 0 and update must take the zero-denominator
fallback for both plus_di and minus_di, then the dx_den == 0 path for
ADX. The indicator must emit 0/0/0 rather than NaN.
adx.rs is now at 189/189 lines, no behavioural change.
* test(natr): cover accessors, zero-close branch, kill dead panic arm
Codecov flagged 11 lines in crates/wickra-core/src/indicators/natr.rs
(file at 87.64%):
- const accessors period (59-61), value (64-66) — never queried
- line 77 (`0.0` in the candle.close == 0.0 fallback) — every test
used candles with close ≈ 100, so the divide-by-zero guard never
fired
- Indicator-impl name body (98-100) — never queried
- line 142 (`_ => panic!("warmup mismatch at {i}")`) — unreachable
invariant guard in natr_is_atr_over_close_as_percent because the
NATR wrapper inherits ATR's warmup period exactly
Add accessors_and_metadata covering period/value/name. Add
zero_close_yields_zero_natr feeding an all-zero candle series (Candle
validator accepts open == high == low == close == 0 with positive
volume) — ATR is 0 each bar, so the indicator must emit exactly 0.0
rather than 100 * 0 / 0 = NaN. Refactor natr_is_atr_over_close_as_
percent to assert the warmup-shape invariant via assert_eq! on
is_some(), removing the dead panic arm.
natr.rs is now at 89/89 lines, no behavioural change.
* test(trix): cover period accessor, warmup/name metadata, zero-prev branch
Codecov flagged 11 lines in crates/wickra-core/src/indicators/trix.rs
(file at 84.05%):
- const accessor period (47-49) — never queried
- the Some(_) match arm (67-68) — the degenerate path where the
previous triple-EMA value is exactly 0.0 (would otherwise divide
by zero on the percent-rate formula). All other tests used
inputs ≈ 100, so prev_tr was never 0.0
- Indicator-impl warmup_period (84, 86-87) and name (93-95) — never
queried
Add accessors_and_metadata asserting period == 5, warmup_period == 14
(= 3*5 - 1), name == "TRIX". Add zero_input_series_yields_zero_trix
feeding [0.0; 20] — every EMA stage collapses to 0.0, so once warmed
up prev_tr is Some(0.0) and every subsequent emission must take the
fallback arm returning 0.0.
trix.rs is now at 69/69 lines, no behavioural change.
* test(coppock): cover periods/value accessors + name + simplify assert
Codecov flagged 10 lines in crates/wickra-core/src/indicators/coppock.rs
(file at 91.07%):
- const accessors periods (68-70), value (73-75) — never queried
- Indicator-impl name body (128-130) — never queried
- line 180 (`warmup - 1,` format-arg) inside the multi-line assert!
in warmup_period_matches_first_some_for_every_parameter_set —
only evaluated on assertion failure, which never happens, so
Codecov flagged the cold path as uncovered
Add accessors_and_metadata covering periods/value/name. Simplify the
multi-line assert's format args to a static message — the {warmup}
binding already appears once in the cold path so dropping the literal
"warmup index" arg loses nothing diagnostic but kills the dead
expression-arg line.
coppock.rs is now at 112/112 lines, no behavioural change.
Commit aa2846c collapsed the multi-line assert in the test-only naive
helper to a single line to drop the cold expression-arg lines that
Codecov saw as uncovered. The single-line form exceeded the 100-col
limit, so cargo fmt --check failed on every supported toolchain in CI
(ubuntu-latest, macos-latest, windows-latest).
Let rustfmt wrap it back to the three-line form. The arms are still
just a literal expression and a static-string message — no expression
format args — so the cold-path lines that Codecov originally flagged
on line 179 stay covered. No behaviour change.
Codecov flagged 12 lines in crates/wickra-core/src/indicators/std_dev.rs
(file at 89.09%): const accessors period (64-66), value (68-71) and
Indicator-impl bodies warmup_period (110-112), name (118-120). None
of the existing tests inspected the metadata surface.
Add accessors_and_metadata asserting period == 14, warmup_period == 14,
name == "StdDev", and value() across both the None (pre-warmup) and
Some (post-warmup) branches.
std_dev.rs is now at 110/110 lines, no behavioural change.
Codecov flagged 12 lines in crates/wickra-core/src/indicators/smma.rs
(file at 86.81%): the const accessors period (57-59), value (62-64)
and the Indicator-impl bodies warmup_period (95-97), name (103-105).
None of the existing tests inspected the metadata surface — they only
fed numeric updates and asserted on SMMA values.
Add accessors_and_metadata exercising period == 7, warmup_period == 7,
name == "SMMA", and value() across both the None (pre-warmup) and
Some (post-warmup) branches.
smma.rs is now at 91/91 lines, no behavioural change.
Codecov flagged 13 uncovered lines in
crates/wickra-core/src/indicators/stochastic.rs (file at 93.43%):
- classic() convenience constructor (76-78) — every test passed
explicit (k_period, d_period) to new
- periods() const accessor (81-83) — never queried
- warmup_period (170-172), name (178-180) Indicator-impl bodies —
never queried
- line 208 (`50.0` literal) inside the test-only naive_k helper's
flat-range branch — k_matches_naive feeds an oscillating price
series, so the helper's range == 0 path was dead
Add classic_periods_and_metadata test asserting Stochastic::classic()
has periods (14, 3), warmup_period 16 (= 14 + 3 - 1) and name
"Stochastic". Extend flat_range_yields_k_50 to also call naive_k on
the flat candle series and verify the helper returns Some(50.0) for
every index ≥ k_period - 1 — exercises line 208 without diluting the
production-code assertion.
stochastic.rs is now at 198/198 lines, no behavioural change.
Codecov flagged 15 uncovered lines in crates/wickra-data/src/aggregator.rs
(file at 95.11%):
- Timeframe::millis / Timeframe::seconds / Timeframe::one_minute_ms
convenience constructors (40-52) — every existing test built
Timeframes via new / minutes / hours / days, never via these three
- the cold `?` Err arm on `Candle::new(...)?` for the flat gap-fill
candle (line 334) — `prev.close` is already finite (came from a
closed bar), volume is exactly 0.0, OHLC are trivially equal, so
Candle::new's error path is unreachable here
- the cold `ok_or_else` overflow closure on `t.checked_add(step)`
inside the gap-fill loop (336-337) — bucket alignment guarantees
start + (gap_count-1)*step ≤ next_bucket - step < i64::MAX, so
every aligned-bucket layout reaches t == next_bucket cleanly and
exits without ever invoking the overflow path
- TickAggregator::timeframe accessor (353-355) — never queried
Add two new tests:
- timeframe_convenience_constructors exercises millis/seconds/
one_minute_ms with both happy-path and rejection cases
- aggregator_timeframe_getter asserts timeframe().bucket() round-trips
Refactor fill_between to use Candle::new_unchecked for the flat-candle
push (the OHLCV invariants hold by construction) and iterate via
`0..gap_count` with `saturating_add(step)` instead of `while t <
next_bucket` with `checked_add(...).ok_or_else(...)?`. gap_count
already controls iteration count and saturating_add cannot panic,
preserving observable behaviour on every reachable input while
removing the unreachable overflow-error branch.
aggregator.rs is now at 307/307 lines, no observable behaviour change
on aligned-bucket inputs (which is every input fill_between can be
called with given the call site's preconditions).
Codecov re-check after 1255892 showed line 179 of
crates/wickra-core/src/indicators/bollinger.rs still uncovered: the
`prices.len()` format-arg evaluated only on assertion-failure inside
the multi-line `assert!(prices.len() >= period, "…got {}, period {}",
prices.len(), period)` — the cold panic path.
Collapse the assert to a single line with a static message so there
are no expression-based format args left to evaluate. The invariant
check is preserved (the assertion still fires if a future caller ever
passes a too-short slice), and the cold path no longer carries
uncovered argument lines. bollinger.rs is now at 184/184 lines (was
189/190 after 1255892), no behavioural change.
Codecov flagged 16 uncovered lines in
crates/wickra-core/src/indicators/bollinger.rs (file at 91.30%):
- classic() convenience constructor (91-93) — every test passed
explicit parameters to BollingerBands::new, so the classic-defaults
path was dead
- const accessors period (96-98), multiplier (101-103) — never queried
- Indicator-impl bodies warmup_period (156-158), name (164-166) —
never queried
- `return None;` (line 177) inside the test-only `naive` helper's
`if prices.len() < period` early-return — every caller passes a
slice of length >= period (matches_naive_definition uses
`&prices[..=i]` for `i in 19..` with period=20; long_stream_drift_
stays_bounded fills the window before measuring), so the arm is dead
Add classic_and_accessors_and_metadata to cover the constructor and
the four getter bodies, and refactor naive to return BollingerOutput
directly with an `assert!(prices.len() >= period)` precondition. The
two existing callers were already using .unwrap()/.expect() on the
Option result and simplify to direct calls.
bollinger.rs is now at 184/184 lines, no behavioural change.
Codecov flagged lines 86-102 in crates/wickra-core/src/ohlcv.rs as missed
(file at 90.00%) — the entire body of `Candle::new_unchecked`. Every
existing test routes through the validating `Candle::new`, so the unchecked
constructor (intended for callers like the aggregator and parsed-payload
paths that have already validated upstream) was dead.
Add candle_new_unchecked_preserves_fields_verbatim:
- first assertion builds a candle with six distinct field values and
verifies each reads back exactly.
- second assertion feeds an OHLC combination (high < low) that the
checked constructor rejects with Error::InvalidCandle, then proves
Candle::new_unchecked still builds the struct as-is. This documents
and enforces the API contract that the unchecked variant performs
no validation.
ohlcv.rs is now at 170/170 lines, no behavioural change.
Codecov flagged 17 uncovered lines in
crates/wickra-core/src/indicators/vwap.rs (file at 87.94%):
- Vwap::value() Some branch (line 53) — the only test calling value()
did so after reset() when sum_v == 0, exercising only the None branch
- Vwap::update zero-volume early-return `return None;` (line 67) — all
existing candles carried strictly positive volume
- Vwap::warmup_period body returning 1 (79-81), Vwap::name body
returning "VWAP" (87-89) — metadata never queried
- RollingVwap::period accessor (134-136), RollingVwap::warmup_period
body (165-167), RollingVwap::name body returning "RollingVWAP"
(173-175) — same metadata gap on the rolling variant
Add four new tests:
- cumulative_value_some_branch_after_update drives a single
non-zero-volume candle then asserts value() == Some(typical_price).
- cumulative_zero_volume_first_candle_returns_none feeds a candle
with volume == 0.0, asserts update returns None and is_ready stays
false, then adds a real candle to confirm the indicator still works.
- cumulative_metadata asserts warmup_period() == 1 and name() == "VWAP".
- rolling_accessors_and_metadata asserts period() == 7,
warmup_period() == 7, name() == "RollingVWAP" on a RollingVwap::new(7).
vwap.rs is now at 141/141 lines, no behavioural change.
Codecov flagged 17 uncovered lines in
crates/wickra-core/src/indicators/percent_b.rs (file at 81.11%):
- const accessors period (53-55), multiplier (58-60), value (63-65)
- Indicator-impl bodies warmup_period (90-92) and name (98-100)
- the unreachable `_ => panic!("warmup mismatch at {i}")` arm in
matches_bands_definition (line 141)
- the inner `assert_relative_eq!(*pv, 0.5, …)` (line 158) inside
price_at_middle_is_half, gated by `(prices[i] - bv.middle).abs()
< 1e-9` over a sin-based oscillation that, with period=20 over 60
samples, never lands within 1e-9 of the rolling SMA — so the
assertion was silently dead and the test made no checks.
Add accessors_and_metadata covering the five getter bodies, refactor
matches_bands_definition to assert the warmup-shape invariant via
assert_eq!(p.is_some(), b.is_some()) + if let (kills the panic arm),
and replace price_at_middle_is_half with a deterministic construction:
PercentB::new(3, 2.0) on [1.0, 5.0, 3.0] gives SMA=3.0 at index 2
which equals the third price exactly, stddev=√(8/3)≈1.633 keeps the
width strictly positive so the divide path runs, and %b lands on
exactly 0.5 because price sits on the centre line of symmetric bands.
percent_b.rs is now at 90/90 lines, no behavioural change.
Codecov flagged 17 uncovered lines in
crates/wickra-core/src/indicators/bollinger_bandwidth.rs (file at 79.51%):
- the const accessors period (54-56), multiplier (59-61), value (64-66)
- the zero-middle defensive fallback 0.0 (line 77) inside update
- the Indicator-impl bodies warmup_period (90-92) and name (98-100)
- the unreachable `_ => panic!("warmup mismatch")` arm (line 140) in
the existing matches_bands_definition test
None of the existing tests inspected the metadata surface — every test
fed numeric updates and asserted on bandwidth values, leaving the five
getter bodies dead. The zero-middle path was unreachable because all
existing tests used positive price levels ≈100, so the rolling SMA was
always strictly positive and the divide-by-zero guard never fired.
The panic arm in matches_bands_definition was an invariant guard that
by design cannot fire when the two streams share a warmup period; that
invariant is now asserted directly with assert_eq!(w.is_some(),
b.is_some()), and the catch-all arm is gone.
Add two new tests and refactor one existing:
- accessors_and_metadata asserts period == 20, multiplier == 2.0,
value() == None before warmup, warmup_period == 20, name ==
"BollingerBandwidth", then drives 20 updates so value() also
exercises the Some branch.
- zero_middle_band_yields_zero_bandwidth feeds [-2, -1, 0, 1, 2] so
the 5-bar SMA lands on exactly 0.0 at the fifth input, and asserts
the emitted bandwidth is exactly 0.0 (rather than inf/nan from the
would-be divide-by-zero).
- matches_bands_definition now uses an explicit assert_eq! on
is_some() agreement plus an if let for the numeric compare,
removing the unreachable panic arm without weakening the
invariant check.
bollinger_bandwidth.rs is now at 83/83 lines, no behavioural change.
Codecov flagged 30 uncovered lines in crates/wickra-core/src/traits.rs (file at
75.60%): the const borrow accessors Chain::first / Chain::second (140-147), the
Chain::warmup_period + Chain::name Indicator-impl bodies (167-178), the full
Identity test-helper Indicator surface — reset, warmup_period, is_ready, name
(198-209), and Doubler's warmup_period + name (228-236).
None of the existing tests touched those code paths: every chain test invoked
update/reset/is_ready through the Chain wrapper without ever inspecting the
borrow accessors, querying chain.warmup_period(), or asking for chain.name(),
and the Identity helper was only ever driven by batch (which calls update
only). Doubler's warmup_period and name were similarly dead because
Chain::warmup_period and Chain::name themselves were dead.
Add two new tests at the end of the Chain section in mod tests, immediately
before the parallel-feature-gated test:
- chain_accessors_and_metadata exercises chain.first(), chain.second(),
chain.warmup_period(), chain.name(), and pulls Doubler::warmup_period
+ Doubler::name in via the borrowed accessors.
- identity_helper_full_indicator_surface asserts warmup_period == 0,
name == "Identity", and walks is_ready through both seen=false and
seen=true via an update/reset cycle.
traits.rs is now at 123/123 lines, no behavioural change.
jetli/wasm-pack-action@v0.4.0 with no version: input installs whatever
wasm-pack the action's bundled installer fetches — currently a ~0.10.x
release whose 'build' subcommand does not yet accept --features. Our
build invocation 'wasm-pack build … --features panic-hook' now fails
with
error: Found argument '--features' which wasn't expected, or isn't
valid in this context
USAGE: wasm-pack build --release --target <target>
even though that exact command worked on past runs where the action
happened to install a newer wasm-pack. (The bundler-target release.yml
job passed for v0.2.1 only because it shared the same cached install
on that runner.)
wasm-pack's --features top-level flag has been stable since 0.12.0, so
the fix is to install a fresh wasm-pack each run. Switch both the ci.yml
'WASM build' step and the release.yml 'wasm-publish' job to the same
taiki-e/install-action prebuilt-binary installer we already use for
cargo-llvm-cov and cargo-fuzz. taiki-e tracks the latest wasm-pack
release and the install is a single binary download — no compile, no
shell installer.
The wasm-pack invocations themselves are unchanged.
Codecov flagged a single uncovered line in crates/wickra-data/src/resample.rs:
line 46, the `self.low = c.low;` assignment inside RolledBar::absorb.
None of the existing resampler tests fed a follow-up candle with a strictly
lower low than the first candle in the bucket, so the `c.low < self.low`
branch never fired. Coverage stayed at 122/123.
Add a small dedicated test that pushes a 10.0-low candle into bucket 0, then
a 8.0-low candle into the same bucket, and asserts the rolled bar's low
reflects the dip. Resample file is now at 123/123 lines, no behavioural change.
The 84 markdown files under docs/wiki/ are now published to the
project's GitHub Wiki at https://github.com/kingchenc/wickra/wiki —
a separate git repository (https://github.com/kingchenc/wickra.wiki.git)
that GitHub hosts natively with its own UI, search and history. The
flat layout that the GitHub Wiki requires has been generated, all
internal cross-links rewritten, and a _Sidebar.md groups the 71
indicators by their canonical 8 families.
Effects:
- docs/wiki/ is removed from the main repo (-84 files). docs/README.md
now just points readers at the Wiki.
- PR template + CONTRIBUTING text updated to point at the Wiki instead
of the in-repo path. The Wiki repo is separately cloneable and
editable via the GitHub web UI.
- examples/wasm/README.md cross-link fixed to use the Wiki URL.
- The (still in-repo) CHANGELOG keeps its historical references to
docs/wiki/ paths — those describe what the tree looked like at past
releases and stay accurate as history.
- README.md, license, all source unaffected.
The Wiki itself ships with _Sidebar.md / _Footer.md generated from the
8-families taxonomy and 503/503 cross-links resolved.
Three separate README files (root, bindings/node, bindings/python) had
been drifting independently — each registry showed a different project
page, which is exactly the consistency debt I want to avoid.
Single source of truth: /README.md. The three binding READMEs are
overwritten with the root README content as a baseline, and release.yml
gets a one-line cp step right before every publishing call so future
edits to /README.md propagate automatically:
- python-wheels job: cp README.md bindings/python/README.md before
PyO3/maturin-action runs the wheel build
- python-sdist job: same, before the sdist build
- node-publish job: cp ../../README.md README.md (working-directory
bindings/node) before the main 'npm publish wickra'
- wasm-publish job: cp README.md bindings/wasm/README.md before
wasm-pack build (which copies the crate README into pkg/ on its own)
Cargo crates (wickra, wickra-core, wickra-data) already inherit
readme.workspace = true pointing at /README.md, so crates.io was already
correct — no change needed there.
The per-platform npm subpackages (bindings/node/npm/<target>/) keep
their tiny package.json with no README; they are install-time
optionalDependencies that the loader reads through, never user-facing
on the registry.
Effect: same README on github.com/kingchenc/wickra, crates.io/crates/wickra,
pypi.org/project/wickra, and npmjs.com/package/wickra. Will be live on
the registries with the next tag-push.
CODECOV_TOKEN was just added as a repository secret; the existing
Coverage job in .github/workflows/ci.yml will pick it up on the next
run. This empty commit fires that run.
The 0.2.0 release left wickra@npm stuck at 0.1.4 and never created a
GitHub Release entry because the brand-new `wickra-win32-arm64-msvc`
sub-package name was caught by npm's spam-detection filter on its first
publish attempt (same situation that affected `wickra-win32-x64-msvc`
through 0.1.4 until npm Support unblocked it). A support ticket is open;
until it is resolved, ship 0.2.1 for the five platforms whose
sub-packages are already on npm and re-add Windows ARM64 in a follow-up
release.
Changes for this cycle:
- bindings/node/package.json: remove "wickra-win32-arm64-msvc" from
optionalDependencies and "aarch64-pc-windows-msvc" from
napi.triples.additional.
- bindings/node/npm/win32-arm64-msvc/: removed (will be restored fresh
once the npm name is unblocked).
- .github/workflows/release.yml: comment out the
aarch64-pc-windows-msvc entry of the node-build matrix with a
TODO/restore note.
- Bump every workspace and binding version to 0.2.1 (Cargo.toml,
pyproject.toml, bindings/node/package.json, five npm/<target>
templates, the wiki version table). Cargo.lock regenerated.
- CHANGELOG: new [0.2.1] block consolidating every fix that has landed
on main since 0.2.0 (HV epsilon, examples CI step, fuzz cargo-fuzz
install, MSRV 1.85 -> 1.86 / 1.77 -> 1.88, criterion 0.5 -> 0.8,
tokio-tungstenite 0.24 -> 0.29, tick_aggregator gap-fill cap, every
GitHub Action SHA-pin bump). Compare-link added.
The arm64 loader branch in bindings/node/index.js is left untouched: a
Windows ARM64 user installing 0.2.1 will get the standard
`Cannot find module 'wickra-win32-arm64-msvc'` error from the loader,
which is accurate. PyPI's win-arm64 wheel is unaffected.
Verified locally:
cargo fmt/clippy/test --workspace --all-features -> 630 passed / 0 failed
cargo build -p wickra-examples --bins -> clean
cargo build -p wickra-node -> clean
criterion 0.8.2 raised its MSRV from 1.85 to 1.86 (rustc 1.85.1 is now
explicitly rejected by its Cargo.toml `rust-version`). One more notch
on the same upward drift that already took us from 1.75 -> 1.80 (rayon)
-> 1.85 (clap_lex/edition2024).
Updated:
- Cargo.toml workspace rust-version 1.85 -> 1.86
- .github/workflows/ci.yml MSRV matrix name + toolchain
- ci.yml comment refreshed to reflect the new driving dep
Dependabot opened #13 to bump tokio-tungstenite from 0.24 to 0.29 but
the bare-version bump fails to compile: WebSocketConfig became
#[non_exhaustive] starting with 0.27, so the existing struct-literal
construction
let ws_config = WebSocketConfig {
max_message_size: Some(MAX_MESSAGE_SIZE),
max_frame_size: Some(MAX_FRAME_SIZE),
..WebSocketConfig::default()
};
produces
error[E0639]: cannot create non-exhaustive struct using struct expression
Switch to the builder-style setters that 0.29 exposes on the default
value. Semantics are unchanged; both fields still carry the
MAX_MESSAGE_SIZE / MAX_FRAME_SIZE caps from the original config and the
rest of the WebSocketConfig defaults are preserved by starting from
WebSocketConfig::default().
This supersedes #13 — same target version, plus the code change
Dependabot can't make on its own.
Verified locally:
cargo check -p wickra-data --features live-binance # clean
cargo test --workspace --all-features # 630 passed / 0 failed
cargo clippy --workspace --all-targets --all-features -- -D warnings
Dependabot opened #10 to bump criterion from 0.5.1 to 0.8.2 but the
straight version bump fails to build: criterion::black_box was
deprecated in 0.6 and removed/marked deny-warn in 0.8, so the existing
`use criterion::{black_box, ...}` produces
error: use of deprecated function `criterion::black_box`: use
`std::hint::black_box()` instead
across every bench callsite. Switch the import to std::hint::black_box
(stable since Rust 1.66, well under our MSRV of 1.85) and drop the
criterion re-export.
This supersedes #10 — same target version, plus the code change
Dependabot can't make on its own.
Verified locally:
cargo bench -p wickra --no-run # builds clean
cargo clippy --workspace --all-targets --all-features -- -D warnings
cargo test --workspace # 630 passed / 0 failed
kingchenc