tulip_rs 0.1.13

use crate::common::{min_process, validate_inputs, validate_options};
pub use crate::indicator_types::TIndicatorState;
use crate::indicators::ef::calc as calc_ef;
use crate::indicators::ema::multiplier as ema_multiplier;
use crate::types::{
    DisplayGroup, DisplayType, IndicatorError, IndicatorInfoOrInteger, IndicatorType, Info,
};
use serde::{Deserialize, Serialize};

/// Number of input price series required by this indicator.
pub const INPUTS_WIDTH: usize = 1;

/// Number of option parameters required by this indicator.
pub const OPTIONS_WIDTH: usize = 1;

/// SIMD-parallel variant that processes `N` assets with identical options simultaneously.
/// Requires the `simd_assets` Cargo feature. See [`by_assets`] for the module form.
#[cfg(feature = "simd_assets")]
pub use crate::indicators::simd_indicators::kama_simd::indicator_by_assets;

/// SIMD-parallel variant that processes a single asset with `N` different option
/// sets simultaneously. Requires the `simd_options` Cargo feature. See [`by_options`].
#[cfg(feature = "simd_options")]
pub use crate::indicators::simd_indicators::kama_simd::indicator_by_options;

/// Convenience module that re-exports [`indicator_by_assets`] as `indicator`,
/// allowing SIMD multi-asset computation to be used as a drop-in replacement
/// for the standard single-asset [`indicator`] function.
/// Requires the `simd_assets` Cargo feature.
#[cfg(feature = "simd_assets")]
pub mod by_assets {
    /// Processes `N` assets in parallel with shared options.
    /// See the parent module's [`super::indicator_by_assets`] for full documentation.
    pub use crate::indicators::simd_indicators::kama_simd::indicator_by_assets as indicator;
}

/// Convenience module that re-exports [`indicator_by_options`] as `indicator`,
/// allowing SIMD multi-option computation to be used as a drop-in replacement
/// for the standard single-asset [`indicator`] function.
/// Requires the `simd_options` Cargo feature.
#[cfg(feature = "simd_options")]
pub mod by_options {
    /// Processes a single asset with `N` different option sets in parallel.
    /// See the parent module's [`super::indicator_by_options`] for full documentation.
    pub use crate::indicators::simd_indicators::kama_simd::indicator_by_options as indicator;
}
/// Returns information about the Kaufman's Adaptive Moving Average (KAMA) indicator.
///
/// # Returns
///
/// An `Info` struct containing metadata about the KAMA indicator.
pub const INFO: Info = Info {
    name: "kama",
    indicator_type: IndicatorType::Trend,
    full_name: "Kaufman's Adaptive Moving Average",
    inputs: &["real"],
    options: &["period"],
    outputs: &["kama"],
    optional_outputs: &["ef"],
    display_groups: &[
        DisplayGroup {
            id: "kama",
            label: "KAMA",
            display_type: DisplayType::Overlay,
            outputs: &["kama"],
        },
        DisplayGroup {
            id: "ef",
            label: "Efficiency Ratio",
            display_type: DisplayType::Indicator,
            outputs: &["ef"],
        },
    ],
};
#[derive(Serialize, Deserialize)]
pub struct IndicatorState {
    real: Vec<f64>,
    period: usize,
    multipliers: (f64, f64),
    state: State,
}
impl IndicatorState {
    pub fn new(real: &[f64], period: usize, multipliers: (f64, f64), state: State) -> Self {
        Self {
            period,
            multipliers,
            state,
            real: real[real.len() - period - 1..].to_vec(),
        }
    }
}
impl TIndicatorState<1> for IndicatorState {
    fn batch_indicator(
        &mut self,
        inputs: &[&[f64]; INPUTS_WIDTH],
        optional_outputs: Option<&[bool]>,
    ) -> Result<Vec<Vec<f64>>, IndicatorError> {
        validate_inputs(inputs, 1)?;
        self.real.extend_from_slice(inputs[0]);

        let (mut kama_line, mut ef_line) = {
            let capacity = inputs[0].len();
            (
                crate::uninit_vec!(f64, capacity),
                crate::init_optional_outputs!(
                    optional_outputs, &[false],
                    ef_line: capacity
                ),
            )
        };

        cycle_kama(
            &self.real,
            &mut self.state,
            self.period,
            self.multipliers,
            &mut kama_line,
            &mut ef_line,
        );
        self.real.drain(..self.real.len() - self.period - 1);

        Ok(vec![kama_line, ef_line])
    }
}

#[derive(Serialize, Deserialize)]
pub struct State {
    pub kama: f64,
    pub sum: f64,
}
impl State {
    pub fn new(kama: f64, sum: f64) -> Self {
        Self { kama, sum }
    }
    pub fn init_state(
        real: &[f64],
        period: usize,
        kama_line: &mut [f64],
        ef_line: &mut [f64],
    ) -> Self {
        let mut state = Self::new(
            real[period - 1],
            (1..period).map(|i| (real[i] - real[i - 1]).abs()).sum(),
        );
        let multipliers = multiplier();
        let values = unsafe {
            (
                real.get_unchecked(period),
                real.get_unchecked(period - 1),
                real.get_unchecked(0),
                &0.0,
            )
        };
        let (kama, efficiency_ratio) = state.calc(values, multipliers, period, period);
        kama_line[0] = kama;
        let (_, want_ef) = crate::calc_want_flags!(ef_line);
        crate::store_optional_outputs!(0,
            want_ef, ef_line => efficiency_ratio
        );

        state
    }
    /// Kaufman's Efficiency Ratio (ER) is defined as:
    ///
    /// ```text
    /// ER = |price[t] - price[t-n]| / sum(|Δprice[i]|)
    /// ```
    ///
    /// If the denominator (the sum of absolute price changes) is zero, the ER is
    /// defined as **0.0**.  This condition occurs when price has not moved at all
    /// over the lookback window, or when every up‑move is exactly cancelled by a
    /// down‑move.  In either case, the market exhibits **maximum noise and zero
    /// trend efficiency**, so ER = 0.0 is the correct interpretation.
    ///
    /// ER must **not** be forced to 1.0 in this case.  Doing so would imply a
    /// perfectly efficient trend despite *no net movement*, which in turn would
    /// cause KAMA to switch into its fastest smoothing regime—opposite of the
    /// intended behaviour. (yet another bug in c tulip! and many other indicator libraries)
    ///
    /// In this implementation, an ER of 0.0 is completely safe: the KAMA smoothing
    /// constant is computed as:
    ///
    /// ```text
    /// SC = (ER * (fast - slow) + slow).powi(2)
    /// ```
    ///
    /// Because `slow` is always greater than zero, the smoothing constant can never
    /// collapse to 0.0.  When ER = 0.0, KAMA simply falls back to its slowest
    /// possible smoothing, exactly as Kaufman designed.
    #[inline(always)]
    pub fn calc(
        &mut self,
        values: (&f64, &f64, &f64, &f64),
        multipliers: (f64, f64),
        period: usize,
        i: usize,
    ) -> (f64, f64) {
        let (fast_ema, slow_ema) = multipliers;
        let efficiency_ratio = calc_ef(&mut self.sum, values, period, i);
        let (value, _, _, _) = values;
        //let smoothing_constant = (efficiency_ratio * (fast_ema - slow_ema) + slow_ema).powi(2);
        let smoothing_constant = (fast_ema - slow_ema)
            .mul_add(efficiency_ratio, slow_ema)
            .powi(2);

        // Optimized calculation using C-style EMA pattern
        let per1 = 1.0 - smoothing_constant;
        //self.kama = self.kama * per1 + value * smoothing_constant;
        self.kama = self.kama.mul_add(per1, value * smoothing_constant);
        (self.kama, efficiency_ratio)
    }
}
/// Returns the minimum number of input bars required to produce results
/// accurate to `decimals` decimal places.
///
/// For indicators with exponential smoothing the seed value's influence
/// must decay below the requested precision, so this value grows with
/// `decimals`. Internally uses `min_process` with the smoothing
/// multiplier to calculate the required lookback.
///
/// # Arguments
///
/// * `options` - A slice containing the indicator options (e.g. period).
/// * `decimals` - The number of decimal places of accuracy required.
///
/// # Returns
///
/// The minimum number of input bars needed for the requested accuracy.
pub fn min_data_accuracy(options: &[f64], decimals: usize) -> usize {
    if options[0] > 12.0 {
        let (short_multiplier, long_multiplier) = multiplier();
        let alpha = short_multiplier - long_multiplier;
        return min_process(
            options,
            Some((decimals, 0)),
            &[ema_multiplier(options[0] as usize).0, alpha],
            IndicatorInfoOrInteger::Info(INFO),
            min_data,
        );
    }
    min_process(
        options,
        Some((decimals, 0)),
        &[ema_multiplier(options[0] as usize).0],
        IndicatorInfoOrInteger::Info(INFO),
        min_data,
    )
}
/// Returns the minimum amount of data required for the KAMA indicator.
///
/// # Arguments
///
/// * `options` - A slice containing the options for the KAMA calculation.
///
/// # Returns
///
/// The minimum amount of data required.
pub fn min_data(options: &[f64]) -> usize {
    options[0] as usize + 1
}

/// Returns the number of output values produced by the KAMA indicator given input data length and options.
///
/// # Arguments
///
/// * `data_len` - The length of the input data.
/// * `options` - A slice containing the options for the KAMA calculation.
///
/// # Returns
///
/// The number of output values.
pub fn output_length(data_len: usize, options: &[f64]) -> usize {
    data_len - min_data(options) + 1
}

/// Calculates the Kaufman's Adaptive Moving Average (KAMA) indicator for an entire dataset.
///
/// # Inputs
///
/// * `inputs[0]` — real (close) prices
///
/// # Options
///
/// * `options[0]` — period
///
/// # Outputs
///
/// * `outputs[0]` — `kama` line
///
/// # Arguments
///
/// * `inputs` - Array of input price slices (see Inputs above).
/// * `options` - Array of indicator options (see Options above).
/// * `_optional_outputs` - Unused; KAMA has no optional outputs.
///
/// # Returns
///
/// `Ok((outputs, state))` where `outputs[0]` is the `kama` line and
/// `state` can be passed to `IndicatorState::batch_indicator` for streaming.
/// Returns `Err(IndicatorError)` if inputs are too short or options are invalid.
pub fn indicator(
    inputs: &[&[f64]; INPUTS_WIDTH],
    options: &[f64; OPTIONS_WIDTH],
    optional_outputs: Option<&[bool]>,
) -> Result<(Vec<Vec<f64>>, IndicatorState), IndicatorError> {
    validate_options(options)?;
    let period = options[0] as usize;
    let multipliers = multiplier();

    validate_inputs(inputs, min_data(options))?;
    let real = inputs[0];

    let (mut kama_line, mut ef_line) = {
        let capacity = output_length(real.len(), options);
        (
            crate::uninit_vec!(f64, capacity),
            crate::init_optional_outputs!(
                optional_outputs, &[false],
                ef_line: capacity
            ),
        )
    };
    let mut state = State::init_state(real, period, &mut kama_line, &mut ef_line);
    let ef = {
        let (_, want_ef) = crate::calc_want_flags!(ef_line);

        if want_ef {
            &mut ef_line[1..]
        } else {
            &mut ef_line
        }
    };
    // Perform the main KAMA calculation
    cycle_kama(
        real,
        &mut state,
        period,
        multipliers,
        &mut kama_line[1..],
        ef,
    );

    Ok((
        vec![kama_line, ef_line],
        IndicatorState::new(real, period, multipliers, state),
    ))
}

/// Performs the main calculation loop for the KAMA indicator.
///
/// # Arguments
///
/// * `real` - A slice of input data.
/// * `state` - A mutable reference to the indicator state.
/// * `period` - The period for the KAMA calculation.
/// * `multipliers` - A tuple of `(fast_ema, slow_ema)` smoothing constants.
/// * `kama_line` - A mutable slice for storing the KAMA output values.
fn cycle_kama(
    real: &[f64],
    state: &mut State,
    period: usize,
    multipliers: (f64, f64),
    kama_line: &mut [f64],
    ef_line: &mut [f64],
) {
    let (_, want_ef) = crate::calc_want_flags!(ef_line);
    for (j, i) in (period + 1..real.len()).enumerate() {
        let values = unsafe {
            (
                real.get_unchecked(i),
                real.get_unchecked(i - 1),
                real.get_unchecked(j + 1),
                real.get_unchecked(j),
            )
        };
        let (kama, efficiency_ratio) = state.calc(values, multipliers, period, i);

        unsafe { *kama_line.get_unchecked_mut(j) = kama };

        crate::store_optional_outputs!(j,
            want_ef, ef_line => efficiency_ratio
        );
    }
}

/// Calculates the KAMA value for a single bar.
///
/// # Arguments
///
/// * `state` - A mutable reference to the indicator state.
/// * `values` - A tuple of price references: `(value, prev_value, last_value, old_value)`.
/// * `multipliers` - A tuple of `(fast_ema, slow_ema)` smoothing constants.
/// * `period` - The period for the KAMA calculation.
/// * `i` - The current index in the full data slice (used to gate the rolling-sum subtraction).
///
/// # Returns
///
/// The calculated KAMA value.
#[inline(always)]
pub fn calc(
    state: &mut State,
    values: (&f64, &f64, &f64, &f64),
    multipliers: (f64, f64),
    period: usize,
    i: usize,
) -> (f64, f64) {
    state.calc(values, multipliers, period, i)
}

#[inline(always)]
pub fn multiplier() -> (f64, f64) {
    (ema_multiplier(2).0, ema_multiplier(30).0)
}