Candlestick Patterns

tulip-rs detects 77+ Japanese candlestick patterns. In tulip-rs-lwc these are rendered as series markers directly on the candlestick chart — each detected pattern is annotated with an arrow and the pattern name at the bar it was found on.

This uses addCandlestickPatterns(), a separate function from addIndicator(), because candlestick patterns return labelled objects (not numeric series) and are rendered using the LWC v5 createSeriesMarkers() plugin rather than as line or histogram series.

---

Quick Start

import { init, addCandlestickPatterns } from 'tulip-rs-lwc';

await init();

const chart   = createChart(container);
const candles = chart.addSeries(CandlestickSeries);
candles.setData(ohlcv);

// All patterns — annotated with pattern name
const patterns = addCandlestickPatterns(candles, ohlcv);

// Streaming — O(1) per bar
candles.update(newBar);
patterns.appendBar(newBar);

// Clean up
patterns.remove();

---

How It Works

addCandlestickPatterns() calls tulip-rs-wasm's candlestick indicator internally and maps each detected pattern to an LWC v5 SeriesMarkerBar:

Forecast	Shape	Position	Default Colour
BullishReversal	arrowUp	below bar	#4CAF50 (green)
BearishReversal	arrowDown	above bar	#ef5350 (red)
Continuation	circle	above bar	#FF9800 (orange)
Unknown	square	above bar	#9E9E9E (grey)

The pattern short-name (e.g. "hammer", "doji", "engulfing") is shown as an inline text annotation on the marker by default (showText: true).

Markers are managed via the LWC v5 createSeriesMarkers() plugin, which is created and attached to the series automatically. Calling remove() detaches the plugin cleanly.

---

Forecast Filtering

Pass a filter option to show only patterns with a specific forecast type. The filter maps directly to the Rust ForecastType enum and is passed on every batchIndicator() call during streaming — the candlestick state does not capture it internally, so it is always explicitly forwarded.

// Bullish reversals only
const bullish = addCandlestickPatterns(candles, ohlcv, [5, 1, 1], {
  filter: 'BullishReversal',
});

// Bearish reversals only
const bearish = addCandlestickPatterns(candles, ohlcv, [5, 1, 1], {
  filter: 'BearishReversal',
});

// Bullish continuation patterns
const bullCont = addCandlestickPatterns(candles, ohlcv, [5, 1, 1], {
  filter: 'BullishContinuation',
});

// Any bearish pattern — reversal or continuation
const anyBear = addCandlestickPatterns(candles, ohlcv, [5, 1, 1], {
  filter: 'BearishReversalOrContinuation',
});

When filter is omitted, all detected patterns across all forecast types are shown.

The full set of accepted filter values (matching the Rust ForecastType enum exactly):

Filter value	Description
'BullishReversal'	Bullish reversal patterns
'BearishReversal'	Bearish reversal patterns
'BullishContinuation'	Bullish continuation patterns
'BearishContinuation'	Bearish continuation patterns
'BullishReversalOrContinuation'	Any bullish pattern
'BearishReversalOrContinuation'	Any bearish pattern

---

Options

The second positional argument is [candle_period, trend_period, trend_signal]:

Option	Description	Default
candle_period	Period for OHLC body/shadow size classification	5
trend_period	Period for the prior trend calculation	1
trend_signal	Trend signal threshold	1

// Default [5, 1, 1]
addCandlestickPatterns(candles, ohlcv);

// Longer candle classification period
addCandlestickPatterns(candles, ohlcv, [10, 3, 1]);

The candle_period controls how the library classifies candle body and shadow sizes relative to recent bars. A longer period produces more conservative pattern detection on noisy data; a shorter period is more sensitive.

---

Colour Customisation

Override the default colours per forecast type:

const patterns = addCandlestickPatterns(candles, ohlcv, [5, 1, 1], {
  bullishColor:      '#00E676',  // bright green
  bearishColor:      '#FF1744',  // bright red
  continuationColor: '#2979FF',  // blue
  unknownColor:      '#757575',  // grey
});

---

Hiding the Text Annotation

Pattern names are shown by default. Set showText: false to render bare arrows without labels — useful if you have high pattern density and the annotations become crowded:

const patterns = addCandlestickPatterns(candles, ohlcv, [5, 1, 1], {
  showText: false,
});

---

Streaming

appendBar() uses state.batchIndicator() internally — O(1) per bar, no history reprocessing. The filter (if set) is applied automatically.

const patterns = addCandlestickPatterns(candles, ohlcv, [5, 1, 1], {
  filter: 'BullishReversal',
});

ws.onmessage = (event) => {
  const bar = JSON.parse(event.data);
  candles.update(bar);
  patterns.appendBar(bar);  // O(1), filter automatically applied
};

---

Combined with Indicators

addCandlestickPatterns() and addIndicator() are fully independent — you can use both on the same series simultaneously:

await init();

// Numeric indicators
const sma = addIndicator(chart, candles, 'sma', ohlcv, [20]);
const rsi = addIndicator(chart, candles, 'rsi', ohlcv, [14]);

// Candlestick patterns alongside
const patterns = addCandlestickPatterns(candles, ohlcv, [5, 1, 1], {
  filter: 'BullishReversal',
});

// Stream all together
function onNewBar(bar) {
  candles.update(bar);
  sma.appendBar(bar);
  rsi.appendBar(bar);
  patterns.appendBar(bar);
}

// Remove independently
patterns.remove(); // markers gone, SMA and RSI unaffected

---

Multiple Pattern Sets

Because each addCandlestickPatterns() call creates its own independent ISeriesMarkersPluginApi instance, you can add multiple pattern sets with different filters to the same series. Each set manages its own marker array:

const bullish = addCandlestickPatterns(candles, ohlcv, [5, 1, 1], {
  filter: 'BullishReversal',
  bullishColor: '#00E676',
});

const bearish = addCandlestickPatterns(candles, ohlcv, [5, 1, 1], {
  filter: 'BearishReversal',
  bearishColor: '#FF1744',
});

// Both sets of markers are visible simultaneously
// Remove them independently when needed
bullish.remove();
bearish.remove();

---

API

addCandlestickPatterns(sourceSeries, data, options?, addOptions?)

function addCandlestickPatterns(
  sourceSeries: ISeriesApi<...>,
  data:         OhlcvBar[],
  options?:     [number, number, number],   // [candle_period, trend_period, trend_signal]
  addOptions?:  AddCandlestickPatternOptions,
): IndicatorHandle

Returns an IndicatorHandle — the same type as addIndicator() — with remove(), setData(), and appendBar().

AddCandlestickPatternOptions

type AddCandlestickPatternOptions = {
  filter?:            'BullishReversal' | 'BearishReversal'
                    | 'BullishContinuation' | 'BearishContinuation'
                    | 'BullishReversalOrContinuation' | 'BearishReversalOrContinuation';
  bullishColor?:      string;   // default '#4CAF50'
  bearishColor?:      string;   // default '#ef5350'
  continuationColor?: string;   // default '#FF9800'
  unknownColor?:      string;   // default '#9E9E9E'
  showText?:          boolean;  // default true
};