DuckScript API Reference

Complete reference for all DuckScript functions, columns, and syntax

Built-in Functions

View in Guide →

Moving Averages

`sma(series, period)`

series<float>

Simple Moving Average

Parameters:

series: series<float> - Price series (e.g., close)
period: int - Period length

Returns:

series<float>

ma20 = sma(close, 20)

`ema(series, period)`

series<float>

Exponential Moving Average

Parameters:

series: series<float> - Price series
period: int - Period length

Returns:

series<float>

ema50 = ema(close, 50)

Technical Indicators

`rsi(series, period)`

series<float>

Relative Strength Index (simplified approximation)

Parameters:

series: series<float> - Price series (typically close)
period: int - Period length (typically 14)

Returns:

series<float>

rsi14 = rsi(close, 14)

Volatility

`atr(period)`

series<float>

Average True Range (simplified approximation)

Parameters:

period: int - Period length

Returns:

series<float>

volatility = atr(14)

Volume

`vwap()`

series<float>

Volume Weighted Average Price

Returns:

series<float>

vwap_value = vwap()

Math Functions

`abs(value)`

series<float>

Absolute value

Parameters:

value: series<float>

Returns:

series<float>

abs_change = abs(close - open)

`na(value)`

series<bool>

True if value is null/missing; use not na(x) for non-null check (Pine Script compatible)

Parameters:

value: series<float> | int | string - Expression to check for null

Returns:

series<bool>

valid = not na(close)

`highest(series, period)`

series<float>

Maximum value in a rolling window

Parameters:

series: series<float> - Price series (e.g., high)
period: int - Period length

Returns:

series<float>

highest_20 = highest(high, 20)

`lowest(series, period)`

series<float>

Minimum value in a rolling window

Parameters:

series: series<float> - Price series (e.g., low)
period: int - Period length

Returns:

series<float>

lowest_20 = lowest(low, 20)

Pattern Detection

`crossover(series1, series2)`

series<bool>

Detect when series1 crosses above series2

Parameters:

series1: series<float>
series2: series<float>

Returns:

series<bool>

golden_cross = crossover(ma_fast, ma_slow)

`crossunder(series1, series2)`

series<bool>

Detect when series1 crosses below series2

Parameters:

series1: series<float>
series2: series<float>

Returns:

series<bool>

death_cross = crossunder(ma_fast, ma_slow)

`pivothigh(source, leftBars, rightBars)`

series<float>

Swing high: returns source value when higher than leftBars bars left AND rightBars bars right; na otherwise (Pine Script compatible)

Parameters:

source: series<float> - Price series (e.g., high)
leftBars: int - Bars to the left (integer literal)
rightBars: int - Bars to the right (integer literal)

Returns:

series<float>

ph = pivothigh(high, 2, 2)

`pivotlow(source, leftBars, rightBars)`

series<float>

Swing low: returns source value when lower than leftBars bars left AND rightBars bars right; na otherwise

Parameters:

source: series<float> - Price series (e.g., low)
leftBars: int - Bars to the left (integer literal)
rightBars: int - Bars to the right (integer literal)

Returns:

series<float>

pl = pivotlow(low, 2, 2)

Pattern Functions

`count_consecutive(condition)`

series<float>

Count consecutive bars where condition is true

Parameters:

condition: series<bool>

Returns:

series<float>

count_consecutive(close > open)

`bars_since(condition)`

series<float>

Count bars since last time condition was true

Parameters:

condition: series<bool>

Returns:

series<float>

bars_since(close > high[1])

`high_since(condition)`

series<float>

Maximum high since last time condition was true

Parameters:

condition: series<bool>

Returns:

series<float>

high_since(close < open)

`low_since(condition)`

series<float>

Minimum low since last time condition was true

Parameters:

condition: series<bool>

Returns:

series<float>

low_since(close > open)

`max_since(condition, series)`

series<float>

Maximum value of series since last time condition was true

Parameters:

condition: series<bool>
series: series<float>

Returns:

series<float>

max_since(close < open, high)

`min_since(condition, series)`

series<float>

Minimum value of series since last time condition was true

Parameters:

condition: series<bool>
series: series<float>

Returns:

series<float>

min_since(close > open, low)

`accumulate_while(condition, value)`

series<float>

Accumulate value while condition is true, reset when false

Parameters:

condition: series<bool>
value: series<float>

Returns:

series<float>

accumulate_while(close > open, volume)

`carry_while(condition, value)`

series<float>

First value when condition became true, held during segment; na when false

Parameters:

condition: series<bool>
value: series<float>

Returns:

series<float>

dq3_open = carry_while(session == "DQ3", open)

`carry_forward(condition, value)`

series<float>

Last value when condition was true, propagated forward when false (LOCF)

Parameters:

condition: series<bool>
value: series<float>

Returns:

series<float>

last_dq3_close = carry_forward(session == "DQ3", close)

`flag_once(condition)`

series<bool>

Flag that becomes true once condition is true and stays true

Parameters:

condition: series<bool>

Returns:

series<bool>

flag_once(close > high[1])

`latch(condition, reset?)`

series<bool>

Flag that holds when condition is true. With two args: latch(condition, reset) — holds until reset is true, then can trigger again

Parameters:

condition: series<bool> - Condition that sets the flag
reset?: series<bool> - Optional: when true, clears the latch so it can trigger again

Returns:

series<bool>

lat = latch(signal, new_bar)

Date / Time Functions

`weekday(date)`

series<float>

Day of week (0=Sunday, 1=Monday, ..., 6=Saturday)

Parameters:

date: string - Date column or string (e.g., date)

Returns:

series<float>

weekday(date)

`hour(timestamp)`

int

Extract hour from timestamp (0-23)

Parameters:

timestamp: int - Timestamp (e.g., timestamp column)

Returns:

int

hour(timestamp)

`minute(timestamp)`

int

Extract minute from timestamp (0-59)

Parameters:

timestamp: int - Timestamp

Returns:

int

minute(timestamp)

`month(date)`

series<float>

Extract month from date or timestamp (1-12)

Parameters:

date: string - Date column or timestamp

Returns:

series<float>

month(date)

`day_of_month(date)`

series<float>

Extract day of month from date or timestamp (1-31)

Parameters:

date: string - Date column or timestamp

Returns:

series<float>

day_of_month(date)

`year(date)`

series<float>

Extract year from date or timestamp

Parameters:

date: string - Date column or timestamp

Returns:

series<float>

year(date)

`bar_size(timestamp?)`

int

Bar size (time delta between current and previous bar)

Parameters:

timestamp?: int - Optional override timestamp (defaults to timestamp column)

Returns:

int

bar = bar_size()

`session_end_timestamp(session, timestamp?)`

int

Epoch timestamp of the current row’s session end (based on config.sessions and config.timezone)

Parameters:

session: series<string> - Session name (often the session column)
timestamp?: int - Optional override timestamp (defaults to timestamp column)

Returns:

int

end_ts = session_end_timestamp(session)

`session_start_timestamp(session, timestamp?)`

int

Epoch timestamp of the current row’s session start (based on config.sessions and config.timezone)

Parameters:

session: series<string> - Session name (often the session column)
timestamp?: int - Optional override timestamp (defaults to timestamp column)

Returns:

int

start_ts = session_start_timestamp(session)

`date_end_timestamp(date?)`

int

Epoch timestamp for end of day (based on config.timezone)

Parameters:

date?: series<string> - Optional date override (defaults to date column)

Returns:

int

eod = date_end_timestamp()

`date_start_timestamp(date?)`

int

Epoch timestamp for start of day (based on config.timezone)

Parameters:

date?: series<string> - Optional date override (defaults to date column)

Returns:

int

sod = date_start_timestamp()

`time_to_utc(timestamp)`

int

Canonical UTC epoch in seconds (for timezone adjustments; timestamp should be in seconds)

Parameters:

timestamp: int - Epoch timestamp (seconds)

Returns:

int

utc = time_to_utc(timestamp)

`date_from_epoch(epoch)`

series<string>

Date string YYYY-MM-DD in UTC for the given epoch (seconds)

Parameters:

epoch: int - Unix epoch in seconds

Returns:

series<string>

trading_date = date_from_epoch(utc + 6 * 3600)

Built-in Columns

View in Guide →

OHLCV Prices

`open`

series<float>

Opening price of the bar

`high`

series<float>

Highest price of the bar

`low`

series<float>

Lowest price of the bar

`close`

series<float>

Closing price of the bar

`volume`

series<float>

Volume of the bar

Metadata

`timestamp`

series<int>

Timestamp of the bar

`symbol`

series<string>

Symbol identifier

Structural Columns

These columns are automatically generated based on your dataset metadata and can be used in aggregations.

`session`

series<string>

Session identifier (e.g., "premarket", "regular", "afterhours")

`date`

string

Date as string (e.g., "2025-01-15")

`interval_id`

string

Interval identifier

`interval_start`

series<int>

Interval start timestamp

`interval_index`

int

Global sequential index that continues summing across dates and sessions. Does NOT reset when date or session changes. Depends on interval_duration (e.g., with 90min intervals: 1010, 1011, 1012...).

`interval_number`

series<int>

Follows the sequence of interval_index but resets at the start of each session. Allows identifying position within each session (1, 2, 3, 4, 5... per session). Depends on interval_duration.

Keywords

View in Guide →

input

func

signal

output

stats

summary

by

count

percent

avg

max

min

stddev

median

first

last

first_when

prev

lead

if

then

else

as

int

float

bool

string

true

false

Operators

View in Guide →

Arithmetic Operators

+

-

*

/

Comparison Operators

>

<

>=

<=

==

!=

Logical Operators

and

or

not

← Back to DSL Guide (DuckScript)