メニュー
Use when converting or porting code between MultiCharts PowerLanguage and TradingView Pine Script in either direction — concept mapping tables, semantic differences, conversion checklists.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use when converting or porting code between MultiCharts PowerLanguage and C++ in either direction — Strategy base-class scaffold, concept mapping tables, semantic differences, TA-Lib indicators, conversion checklists.
Use when converting or porting code between MultiCharts PowerLanguage and Python in either direction — Strategy ABC scaffold, concept mapping tables, semantic differences, pandas-ta/TA-Lib indicators, conversion checklists.
Use when converting or porting code between MultiCharts PowerLanguage and Rust in either direction — Strategy trait scaffold, concept mapping tables, semantic differences, ta-rs indicators, conversion checklists.
Use when orienting on the MultiCharts platform — choosing Indicator vs Signal vs Function, bar-close vs intra-bar-order-generation execution, multi-data references (data1, data2), Buy/Sell/SellShort/BuyToCover orders, and the unique-signal-name compile rule.
Use when writing Pine Script logic with built-in namespaces — ta.* (averages, oscillators, crossovers), strategy.* (orders, position info), request.* (multi-timeframe), math.*, str.*, array.*, color.*, bar state, symbol info.
Use when writing or reading Pine Script language fundamentals — @version, script types (indicator/strategy/library), type system (series/simple/const), var/varip/input.*, control flow, user-defined functions/types, gotchas (repainting, na handling, max_bars_back).
| name | powerlanguage-pinescript-conversion |
| description | Use when converting or porting code between MultiCharts PowerLanguage and TradingView Pine Script in either direction — concept mapping tables, semantic differences, conversion checklists. |
This skill covers the structural and semantic differences between MultiCharts PowerLanguage and TradingView Pine Script. It does not duplicate the syntax rules of either language — for PowerLanguage declarations, bar references, control flow, and built-in function signatures use the powerlanguage-syntax skill; for Pine Script fundamentals use pinescript-core, for built-in namespaces use pinescript-builtins, and for plotting/drawing use pinescript-visual. When performing a conversion in either direction, work through Part 1 (concept mapping) first, then review Part 2 (semantic differences and gotchas), and finally run through the relevant checklist in Part 3 before calling the output complete.
| PowerLanguage | Pine Script | Notes |
|---|---|---|
Inputs: Length(20) | length = input.int(20, "Length") | Pine requires an explicit title string; PL labels come from the Format window |
Variables: myVar(0) | var float myVar = 0.0 | var retains value across bars, matching PL default behavior |
Value1 .. Value99 | named variable (e.g., val1) | PL numbered built-ins have no Pine equivalent; must rename to meaningful identifiers |
Condition1 .. Condition99 | named bool (e.g., cond1) | Same: rename to typed bool or expression variable |
Arrays: buf[10](0) | buf = array.new_float(11, 0.0) | PL [10] means last index is 10 (11 elements); Pine size is element count |
IntraBarPersist myVar(0) | varip float myVar = 0.0 | Both update intrabar on live ticks; varip is the closest Pine equivalent |
| PowerLanguage | Pine Script | Notes |
|---|---|---|
Open, High, Low, Close, Volume | open, high, low, close, volume | Pine is all-lowercase; PL is case-insensitive but conventionally capitalized |
Close[1] | close[1] | Bracket indexing works the same; both are 0-based offset (0 = current bar) |
Close of Data2 | request.security(symbol, timeframe.period, close) | PL Data2 is a second chart feed; Pine requires explicit symbol and timeframe strings |
Date, Time | year(time_close), month(time_close), dayofmonth(time_close), hour(time_close), minute(time_close) | PL Date is a YYYMMDD integer with a 1900 offset (1260610 = 2026-06-10); reconstruct as (year(time_close) - 1900) * 10000 + .... PL Date/Time label the bar CLOSE — use time_close, not time (Pine time is the bar-OPEN timestamp) |
BarNumber | bar_index + 1 | PL BarNumber is 1-based (first bar = 1); Pine bar_index is 0-based (first bar = 0) |
CurrentBar | bar_index + 1 | Same mapping as BarNumber — but CurrentBar counts from the first executed bar after MaxBarsBack, so bar_index + 1 is exact only when MaxBarsBack = 0 |
| PowerLanguage | Pine Script | Notes |
|---|---|---|
Average(Close, Length) | ta.sma(close, length) | Direct equivalent |
XAverage(Close, Length) | ta.ema(close, length) | Direct equivalent |
RSI(Close, Length) | ta.rsi(close, length) | Direct equivalent |
Stochastic(High, Low, Close, Length, 1, 3, 0, 0, 0, 0, 0) | ta.stoch(close, high, low, length) | PL arg order starts High, Low, Close; PL takes 11 params including smoothing; Pine returns raw %K only — %D must be smoothed manually |
ADX(Length) | [diplus, diminus, adx] = ta.dmi(length, length) | Pine ta.dmi returns a 3-tuple; destructure before use |
CCI(Length) | ta.cci(hlc3, length) | PL CCI takes only length (uses typical price internally); Pine computes on whatever source you pass — pass hlc3 to match |
AvgTrueRange(Length) | ta.sma(ta.tr(true), length) | PL AvgTrueRange is a SIMPLE average of TrueRange. Pine ta.atr is Wilder/RMA — use it only if the PL side used SmoothedAverage(TrueRange, Len), which is the matching Wilder pair |
BollingerBand(Close, Length, 2) | [mid, upper, lower] = ta.bb(close, length, 2) | Pine returns a 3-tuple; PL returns the upper or lower band depending on the final argument |
Close Crosses Over MA | ta.crossover(close, ma) | PL keyword phrase; Pine is a function returning bool |
Close Crosses Under MA | ta.crossunder(close, ma) | Same pattern as above |
Highest(Close, Length) | ta.highest(close, length) | Direct equivalent |
Lowest(Close, Length) | ta.lowest(close, length) | Direct equivalent |
Momentum(Close, Length) | ta.mom(close, length) | Direct equivalent |
TSI(Close, LongLen, ShortLen) | ta.tsi(close, ShortLen, LongLen) * 100 | Pine signature is ta.tsi(source, short_length, long_length) and returns −1..+1 — swap the length order and multiply by 100 |
AverageFC(Close, Length) | ta.sma(close, length) | FC = "fast calculation": AverageFC = SummationFC/Len (running-sum SMA), numerically identical to Average; Pine ta.sma is equivalent |
WAverage(Close, Length) | ta.wma(close, length) | Direct equivalent |
AdaptiveMovAvg(Close, Length) | manual KAMA formula | No Pine built-in; implement Kaufman AMA manually |
MidPoint(Close, Length) | (ta.highest(close, length) + ta.lowest(close, length)) / 2 | No built-in; average of highest and lowest |
MACD(Close, FastLen, SlowLen) | [macdLine, signal, hist] = ta.macd(close, FastLen, SlowLen, sigLen) | Pine returns a 3-tuple; PL returns a single value |
KeltnerChannel(Close, Length, NumATRs) | [mid, upper, lower] = ta.kc(close, length, NumATRs) | Pine returns a 3-tuple; select the band needed |
DMIPlus(Length) | [diplus, diminus, adx] = ta.dmi(Length, Length) | Destructure tuple; use diplus |
DMIMinus(Length) | [diplus, diminus, adx] = ta.dmi(Length, Length) | Destructure tuple; use diminus |
RateOfChange(Close, Length) | ta.roc(close, length) | Direct equivalent |
PercentR(Length) | ta.wpr(length) + 100 or 100 * (close − ta.lowest(low, Length)) / (ta.highest(high, Length) − ta.lowest(low, Length)) | PL PercentR takes length only and is POSITIVE 0..100 (= Williams %R + 100); overbought/oversold thresholds become 80/20 on the positive scale |
MoneyFlow(Length) | ta.mfi(hlc3, length) | Pine ta.mfi is Money Flow Index; PL uses typical price internally — pass hlc3, not close |
Parabolic(AFStep, AFLimit) | manual Parabolic SAR formula | No Pine built-in; implement SAR loop manually |
Volatility(Length) | ta.ema(ta.tr(true), length) (or ta.rma) | PL Volatility is a smoothed average of TrueRange weighted toward the most recent bar — NOT a stdev-based measure (StandardDev is stdev of closes; VolatilityStdDev is annualized stdev of log returns) |
UltimateOscillator(Len1, Len2, Len3) | manual formula using ta.atr and buying pressure sums | No Pine built-in; compute from three-period ratios |
ChaikinOsc(FastLen, SlowLen) | manual: ta.ema(ad, FastLen) − ta.ema(ad, SlowLen) | No Pine built-in; compute A/D line first, then EMA difference |
PriceOscillator(Close, FastLen, SlowLen) | manual: ta.ema(close, FastLen) − ta.ema(close, SlowLen) | No Pine built-in; difference of two EMAs |
DirMovement(Length, oADX, oDMIPlus, oDMIMinus) | [diplus, diminus, adx] = ta.dmi(Length, Length) | PL writes to reference params; Pine returns a tuple |
Extremes(Price, Length, oHighest, oLowest) | h = ta.highest(Price, Length) + l = ta.lowest(Price, Length) | PL writes to reference params; use two Pine calls |
TrueRange | ta.tr | Direct equivalent; both return current bar true range |
StandardDev(Close, Length) | ta.stdev(close, length) | Direct equivalent |
TrueHigh | math.max(high, close[1]) | No Pine built-in; max of current high and prior close |
TrueLow | math.min(low, close[1]) | No Pine built-in; min of current low and prior close |
Range | high - low | No Pine built-in; simple bar range |
HighestBar(Close, Length) | ta.highestbars(close, length) | Pine returns negative offset (e.g., −3); PL returns positive |
LowestBar(Close, Length) | ta.lowestbars(close, length) | Pine returns negative offset; PL returns positive |
NthHighest(Close, Length, N) | manual: sort last Length bars, pick Nth | No Pine built-in; use a loop or array.sort |
NthLowest(Close, Length, N) | manual: sort last Length bars, pick Nth | No Pine built-in; use a loop or array.sort |
NthHighestBar(Close, Length, N) | manual: find bar offset of Nth highest | No Pine built-in; combine loop with comparison |
NthLowestBar(Close, Length, N) | manual: find bar offset of Nth lowest | No Pine built-in; combine loop with comparison |
SwingHigh(Occur, Price, Strength, Length) | ta.pivothigh(Price, Strength, Strength) | Occurrence is the FIRST PL arg; PL returns −1 when no swing found (Pine returns na); Length must exceed Strength |
SwingLow(Occur, Price, Strength, Length) | ta.pivotlow(Price, Strength, Strength) | Same: Occurrence first, −1 when none found, Length > Strength |
SwingHighBar(Occur, Price, Strength, Length) | manual: track bar offset of ta.pivothigh | No direct built-in; record bar_index when pivot appears |
SwingLowBar(Occur, Price, Strength, Length) | manual: track bar offset of ta.pivotlow | No direct built-in; record bar_index when pivot appears |
Summation(Close, Length) | math.sum(close, length) | Direct equivalent — math.sum exists in v5 |
Cum(Close) | ta.cum(close) | Direct equivalent; cumulative sum from bar 0 |
LinearRegValue(Close, Length, Offset) | ta.linreg(close, length, offset) | Direct equivalent |
LinearRegAngle(Close, Length, Offset) | manual: math.todegrees(math.atan(ta.linreg(close, length, offset) − ta.linreg(close, length, offset + 1))) | No Pine built-in; derive from regression slope |
LinearRegSlope(Close, Length) | manual: ta.linreg(close, length, 0) − ta.linreg(close, length, 1) | No Pine built-in; approximate from two offsets |
Correlation(X, Y, Length) | ta.correlation(X, Y, length) | Direct equivalent |
RSquared(Close, Length) | math.pow(ta.correlation(close, bar_index, length), 2) | R-squared of linear regression; square the correlation |
StdError(Close, Length) | manual: math.sqrt(sumSqResid / (Length - 2)) from residuals against ta.linreg(close, Length, 0) | Standard error of the linear-regression residuals — NOT SE of the mean; no Pine built-in |
Median(Close, Length) | ta.median(close, length) | Direct equivalent |
ELDate | (year(time) - 1900) * 10000 + month(time) * 100 + dayofmonth(time) | PL YYYMMDD format (1900-based century); reconstruct in Pine |
MinutesToTime(mins) | manual: int(mins / 60) * 100 + (mins % 60) | PL returns HHMM integer; replicate arithmetic |
TimeToMinutes(t) | manual: int(t / 100) * 60 + (t % 100) | Inverse of MinutesToTime; convert HHMM to minutes |
AvgPrice | ohlc4 | Pine built-in ohlc4 = (O+H+L+C)/4 |
MedianPrice | hl2 | Pine built-in hl2 = (H+L)/2 |
TypicalPrice | hlc3 | Pine built-in hlc3 = (H+L+C)/3 |
WeightedClose | (high + low + close * 2) / 4 | No Pine built-in; compute manually |
CountIF(Cond, Length) | manual: ta.cum(Cond ? 1 : 0) − ta.cum(Cond ? 1 : 0)[length] | No Pine built-in; rolling sum of boolean-as-int |
MRO(Cond, Length, N) | manual: loop back through Length bars | No Pine built-in; most recent occurrence of condition |
AccumDist | manual: ta.cum(((close − low) − (high − close)) / (high − low) * volume) | No Pine built-in; cumulative A/D line formula |
IFF(Cond, TrueVal, FalseVal) | Cond ? TrueVal : FalseVal | Pine ternary operator is the direct equivalent |
TriAverage(Close, Length) | halfLen = math.ceil((length + 1) * 0.5) then ta.sma(ta.sma(close, halfLen), halfLen) | Triangular MA = SMA of SMA with HALVED length ceil((Length+1)*0.5) — not the full length applied twice |
FastK(StochLength) | ta.stoch(close, high, low, StochLength) | Raw %K; Pine ta.stoch returns raw %K |
FastD(StochLength) | ta.sma(ta.stoch(close, high, low, StochLength), 3) | Smoothed %K; apply SMA(3) to FastK |
SlowK(StochLength) | ta.sma(ta.stoch(close, high, low, StochLength), 3) | Same as FastD |
SlowD(StochLength) | ta.sma(ta.sma(ta.stoch(close, high, low, StochLength), 3), 3) | Double-smoothed; SMA of SlowK |
FastKCustom(H, L, C, Length) | ta.stoch(C, H, L, Length) | Custom prices; Pine signature is (close, high, low, length) |
FastDCustom(H, L, C, Length) | ta.sma(ta.stoch(C, H, L, Length), 3) | Smooth FastKCustom with SMA(3) |
SlowKCustom(H, L, C, Length) | ta.sma(ta.stoch(C, H, L, Length), 3) | Same as FastDCustom |
SlowDCustom(H, L, C, Length) | ta.sma(ta.sma(ta.stoch(C, H, L, Length), 3), 3) | Slow %D with custom prices |
StochasticExp(H, L, C, Length, S1, S2, ...) | ta.ema(ta.stoch(C, H, L, Length), S1) | Uses EMA smoothing instead of SMA; compute %D from %K manually |
ADXR(Length) | [diPlus, diMinus, adxV] = ta.dmi(Length, Length) at global scope, then (adxV + adxV[Length]) / 2 | Average of current ADX and ADX N bars ago; [] indexing on a function call (or chained [2][Length]) is invalid — destructure into a variable first |
ADXCustom(H, L, C, Length) | Manual: replicate ADX using custom H/L/C | No Pine built-in; implement Wilder smoothing with custom prices |
DMI(Length) | ta.dmi(Length, Length) | Same as ADX; wrapper that returns DMI value |
DMIPlusCustom(H, L, C, Length) | Manual: +DI with custom prices | No Pine built-in; compute +DM from custom highs |
DMIMinusCustom(H, L, C, Length) | Manual: −DI with custom prices | No Pine built-in; compute −DM from custom lows |
ParabolicCustom(AfStep, AfLimit) | Manual Parabolic SAR with limit | No Pine built-in; implement SAR loop with AfLimit cap |
TRIX(Close, Length) | Manual: ta.change(ta.ema(ta.ema(ta.ema(close, Length), Length), Length)) | Triple EMA ROC; no Pine built-in |
MassIndex(SmoothLen, SumLen) | Manual: math.sum(ta.ema(high - low, SmoothLen) / ta.ema(ta.ema(high - low, SmoothLen), SmoothLen), SumLen) | No Pine built-in; ratio of single/double EMA of range |
EaseOfMovement | Manual: ((high + low) / 2 - (high[1] + low[1]) / 2) / (volume / (high - low)) | No Pine built-in; distance moved / box ratio |
SwingIndex | Manual: Wilder swing index formula | No Pine built-in; complex formula using O/H/L/C of current and previous bars |
AccumSwingIndex | Manual: cumulative sum of SwingIndex | No Pine built-in; ta.cum(swingIndex) |
Detrend(Close, Length) | close - ta.sma(close, Length)[int(Length / 2 + 1)] | No Pine built-in; offset SMA detrending |
PercentChange(Close, Length) | ta.change(close, Length) / close[Length] * 100 | Manual computation; or ta.roc(close, Length) if ROC matches |
UlcerIndex(Close, Length) | Manual: math.sqrt(ta.sma(math.pow((close - ta.highest(close, Length)) / ta.highest(close, Length) * 100, 2), Length)) | No Pine built-in; RMS of drawdown percentage |
ParabolicSAR(AfStep, AfLimit, ...) | Manual SAR with position/transition outputs | Multi-output version; implement SAR loop tracking direction changes |
LinearReg(Close, Length, TgtBar, ...) | ta.linreg(close, Length, TgtBar) + manual slope/angle/intercept | Multi-output; Pine ta.linreg returns value only |
TrueRangeCustom(H, L, C) | math.max(H - L, math.abs(H - C[1]), math.abs(L - C[1])) | Custom prices; same formula as ta.tr |
VolatilityStdDev(NumDays) | ta.stdev(math.log(close / close[1]), NumDays) * math.sqrt(252) | Historical vol; annualized stdev of log returns |
StandardDevAnnual(Close, Length, DataType) | ta.stdev(close, Length) * math.sqrt(252) | Annualized standard deviation |
HighestFC(Close, Length) | ta.highest(close, Length) | Fast calc variant; same as Highest in Pine |
LowestFC(Close, Length) | ta.lowest(close, Length) | Fast calc variant; same as Lowest in Pine |
PivotHighVS(Inst, Price, LStr, RStr, Length) | ta.pivothigh(Price, LStr, RStr) | Pine ta.pivothigh supports asymmetric left/right; returns na when no pivot |
PivotLowVS(Inst, Price, LStr, RStr, Length) | ta.pivotlow(Price, LStr, RStr) | Same as above for lows |
PivotHighVSBar(Inst, Price, LStr, RStr, Length) | Manual: track bar_index when ta.pivothigh fires | Record bar offset when pivot appears |
PivotLowVSBar(Inst, Price, LStr, RStr, Length) | Manual: track bar_index when ta.pivotlow fires | Record bar offset when pivot appears |
Divergence(Price1, Price2, Str, Len, HiLo) | Manual: compare pivots of two series | No Pine built-in; detect when price makes new high/low but indicator doesn't |
TimeSeriesForecast(Close, Length) | ta.linreg(close, Length, 0) | Direct equivalent |
SummationFC(Close, Length) | math.sum(close, Length) | Fast calc Summation; same as Summation in Pine |
OpenD(N) | request.security(syminfo.tickerid, "D", open[N]) | Daily open; requires request.security for MTF |
HighD(N) | request.security(syminfo.tickerid, "D", high[N]) | Daily high |
LowD(N) | request.security(syminfo.tickerid, "D", low[N]) | Daily low |
CloseD(N) | request.security(syminfo.tickerid, "D", close[N]) | Daily close |
OpenW(N) | request.security(syminfo.tickerid, "W", open[N]) | Weekly open |
HighW(N) | request.security(syminfo.tickerid, "W", high[N]) | Weekly high |
LowW(N) | request.security(syminfo.tickerid, "W", low[N]) | Weekly low |
CloseW(N) | request.security(syminfo.tickerid, "W", close[N]) | Weekly close |
OpenM(N) | request.security(syminfo.tickerid, "M", open[N]) | Monthly open |
HighM(N) | request.security(syminfo.tickerid, "M", high[N]) | Monthly high |
LowM(N) | request.security(syminfo.tickerid, "M", low[N]) | Monthly low |
CloseM(N) | request.security(syminfo.tickerid, "M", close[N]) | Monthly close |
OpenY(N) | request.security(syminfo.tickerid, "12M", open[N]) | Yearly open; Pine uses "12M" for yearly |
HighY(N) | request.security(syminfo.tickerid, "12M", high[N]) | Yearly high |
LowY(N) | request.security(syminfo.tickerid, "12M", low[N]) | Yearly low |
CloseY(N) | request.security(syminfo.tickerid, "12M", close[N]) | Yearly close |
LRO(Cond, Length, N) | Manual: find Nth-oldest True in lookback | No Pine built-in; scan from Length bars ago forward |
SummationIf(Cond, Price, Length) | Manual: ta.cum(Cond ? Price : 0) - nz(ta.cum(Cond ? Price : 0)[Length]) | Conditional rolling sum |
IFFString(Cond, TrueStr, FalseStr) | Cond ? TrueStr : FalseStr | Pine ternary works with strings |
OBV | ta.obv | Direct equivalent; Pine built-in |
VolumeROC(Length) | ta.roc(volume, Length) | Rate of change applied to volume |
VolumeOsc(ShortLen, LongLen) | ta.sma(volume, ShortLen) - ta.sma(volume, LongLen) | Difference of two volume SMAs |
PriceVolTrend | Manual: ta.cum(ta.change(close) / close[1] * volume) | Cumulative price-volume trend |
LWAccDis | Manual: ta.cum((close - open) / (high - low) * volume) | Larry Williams A/D |
Fisher(Price) | Manual: 0.5 * math.log((1 + norm) / (1 - norm)) | Normalize price to −0.999..+0.999 first |
FisherINV(Price) | Manual: (math.exp(2 * Price) - 1) / (math.exp(2 * Price) + 1) | Inverse Fisher transformation |
C_Doji(Percent) | Manual: math.abs(close - open) <= (high - low) * Percent / 100 | Detect doji pattern |
C_Hammer_HangingMan(Len, Factor, ...) | Manual: check body/shadow ratios | Lower shadow ≥ 2× body, small upper shadow |
C_BullEng_BearEng(Len, ...) | Manual: current body engulfs previous body | Bullish: down bar followed by larger up bar |
C_BullHar_BearHar(Len, ...) | Manual: current body inside previous body | Opposite of engulfing; small body inside large |
C_MornDoji_EveDoji(Len, Pct, ...) | Manual: 3-bar pattern with middle doji | Morning/Evening star variant with doji middle |
C_MornStar_EveStar(Len, ...) | Manual: 3-bar reversal pattern | Down-small-up (morning) or up-small-down (evening) |
C_PierceLine_DkCloud(Len, ...) | Manual: 2-bar pattern, close pierces midpoint | Gap down then close above midpoint of prior bar |
C_ShootingStar(Len, Factor) | Manual: small body at low, long upper shadow | Upper shadow ≥ 2× body, small lower shadow |
C_3WhSolds_3BlkCrows(Len, Factor, ...) | Manual: 3 consecutive bars in same direction | Three ascending closes (soldiers) or three descending (crows) |
| Statistical extended | ||
AvgDeviation(Close, N) | Manual: math.sum(math.abs(close - ta.sma(close, N)), N) / N | Mean absolute deviation |
Variance(Close, N) | ta.variance(close, N) | Population variance |
Kurtosis(Close, N) | Manual: 4th moment calculation | Excess kurtosis |
Skew(Close, N) | Manual: 3rd moment calculation | Skewness |
PercentRank(ValToRank, Price, N) | ta.percentrank(close, N) | Percent rank within lookback |
Covariance(P1, P2, N) | Manual: ta.sma(P1*P2, N) - ta.sma(P1, N)*ta.sma(P2, N) | Covariance — note this sma(xy) − sma(x)·sma(y) identity gives the POPULATION (ddof=0) covariance |
Quartile(Close, N, Q) | ta.percentile_nearest_rank(close, N, Q*25) | Quartile value |
TrimMean(Close, N, Pct) | Manual: sort window, trim, average | Trimmed mean |
Mode(Close, N, Type) | Manual: frequency count over window | Modal value |
HarmonicMean(Close, N) | Manual: N / math.sum(1/close, N) | Harmonic mean |
| Moving averages extended | ||
SmoothedAverage(Close, N) | ta.rma(close, N) | Wilder smoothing = RMA |
| Miscellaneous | ||
BarAnnualization | Manual: compute from timeframe.period | Bars-per-year factor |
LastBarOnChart | barstate.islast | True on last bar |
| Custom functions | ||
StochRSI(Close, N, M) | rsiVal = ta.rsi(close, N) → (rsiVal - ta.lowest(rsiVal, M)) / (ta.highest(rsiVal, M) - ta.lowest(rsiVal, M)) | Stochastic of RSI (0–1) |
supertrend(N, Mult) | [st, dir] = ta.supertrend(Mult, N) | Pine order is (factor, atrPeriod) — reversed |
NVI(Start) | Manual: accumulate prev*(1+(close-close[1])/close[1]) when volume < volume[1] | Negative Volume Index |
PVI(Start) | Manual: accumulate prev*(1+(close-close[1])/close[1]) when volume > volume[1] | Positive Volume Index |
Coppo(N1, N2, N3) | ta.wma(ta.roc(close, N1) + ta.roc(close, N2), N3) | Coppock Curve |
LWTI(Close, P, N) | Manual: ta.sma(close-close[P], N) / ta.sma(high-low, N) * 50 + 50 | Larry Williams Trading Index |
TVI(Close, Vol, Tick) | Manual: cumulative directional volume with tick threshold | Trade Volume Index |
SharpeRatio(Period, Rate, Calc, Cap) | Manual: portfolio-level (avgReturn - riskFree) / stdReturn | No Pine equivalent; portfolio only |
WRSI(N, Close) | ta.rsi(close, N) | Pine RSI already uses Wilder smoothing |
NewMA(Close, N) | Manual: Heikin-Ashi + triple EMA hybrid | No direct Pine equivalent |
| PowerLanguage | Pine Script | Notes |
|---|---|---|
Buy("label") next bar market | strategy.entry("label", strategy.long) | PL keyword Buy always means enter long |
SellShort("label") next bar market | strategy.entry("label", strategy.short) | PL keyword SellShort always means enter short |
Sell("label") next bar market | strategy.close("label") | WARNING: Sell in PL exits an existing long — use strategy.close, NOT strategy.entry(strategy.short) |
BuyToCover("label") next bar market | strategy.close("label") | PL BuyToCover exits an existing short |
Buy(qty, "label") shares next bar market | strategy.entry("label", strategy.long, qty=qty) | Pass contract/share count via qty parameter |
Buy("label") next bar at price limit | strategy.entry("label", strategy.long, limit=price) | Limit orders use the limit param |
Buy("label") next bar at price stop | strategy.entry("label", strategy.long, stop=price) | Stop orders use the stop param |
SetStopLoss(dollars) | strategy.exit("id", stop=strategy.position_avg_price - dollars / (qty * syminfo.pointvalue)) | PL takes a dollar amount; Pine takes an absolute price level — convert via point value |
SetProfitTarget(dollars) | strategy.exit("id", limit=strategy.position_avg_price + dollars / (qty * syminfo.pointvalue)) | Same dollar-to-price conversion required |
SetDollarTrailing(dollars) | strategy.exit("id", trail_points=activationTicks, trail_offset=dollars / (qty * syminfo.pointvalue) / syminfo.mintick) | Pine trailing needs an activation level (trail_price or trail_points) that PL does not have, and trail_points/trail_offset are in TICKS — convert dollars → price distance → ticks |
SetPercentTrailing(dollars, pct) | strategy.exit("id", trail_price=activationPrice, trail_offset=offsetTicks) | Same tick-unit and activation-level warnings; PL arms after dollars of open profit then gives back pct% of the peak profit — compute the activation price for trail_price manually |
[IntrabarOrderGeneration = true] | strategy(..., calc_on_every_tick=true) | Closest mapping, but semantics differ: PL IOG evaluates intrabar on real ticks both live and in backtest; Pine calc_on_every_tick affects real-time only — historical bars still fill on OHLC-modeled prices unless bar magnifier is enabled |
| PowerLanguage | Pine Script | Notes |
|---|---|---|
Plot1(value, "label") | plot(value, title="label") | Pine plot() is a statement, not an indexed output slot |
SetPlotColor(1, Red) | plot(value, color=color.red) | PL sets color separately by plot index; Pine sets it inline as a parameter |
SetPlotWidth(1, 2) | plot(value, linewidth=2) | Same: inline in Pine, separate call in PL |
NoPlot(1) | plot(na) | Passing na suppresses the plot for that bar |
| PowerLanguage | Pine Script | Notes |
|---|---|---|
If cond Then Begin ... End; | if cond\n ... | Pine uses indentation instead of Begin/End delimiters |
If cond Then Begin ... End Else Begin ... End; | if cond\n ...\nelse\n ... | Same indentation pattern for else blocks |
For idx = 1 to n Begin ... End; | for i = 1 to n\n ... | Loop body is indented; no Begin/End needed. PL side uses idx because i is reserved (alias for OpenInterest) |
While cond Begin ... End; | while cond\n ... | Same indentation pattern |
Switch (expr) Begin Case 1: ...; End; | switch expr\n 1 => ... | Pine switch uses => arrows and indentation; PL empty case body is a compile error — use Value1 = Value1; as no-op |
Once Begin ... End; | if barstate.isfirst\n ... | PL Once runs code on the first bar only; Pine uses the barstate.isfirst built-in |
| PowerLanguage | Pine Script | Notes |
|---|---|---|
MarketPosition | strategy.position_size | PL returns -1/0/1; Pine returns the actual signed position size — check sign rather than comparing to 1 or -1 |
EntryPrice | strategy.position_avg_price | Pine gives the average entry price of the current position |
CurrentContracts | math.abs(strategy.position_size) | PL gives absolute contract count; Pine position size is signed |
BarsSinceEntry | bar_index - strategy.opentrades.entry_bar_index(strategy.opentrades - 1) | Built-in since v5; gives bars since the most recent open trade's entry — no manual tracker needed |
Print("text") | label.new(bar_index, high, "text") or log.info("text") | label.new creates on-chart labels; log.info writes to the Pine console |
#BeginCmtry ... #EndCmtry | no equivalent | Pine has no commentary output; use label.new for visible annotations |
Alert("msg", AlertType) | alert("msg", alert.freq_once_per_bar) or alertcondition(cond, "title") | Pine alertcondition registers a condition for the Alerts dialog; alert() fires immediately |
Buy next bar at market fills at the NEXT bar's OPEN. Buy next bar at X Stop / Limit fills intrabar on the next bar at the stop/limit price (or at the open if the bar gaps through it). Any conversion that flips position on the signal bar at that bar's close is one bar early and uses the wrong price.strategy.entry with the default process_orders_on_close=false fills at the next bar's open — same as PL next bar at market. Setting process_orders_on_close=true fills at the SAME bar's close, which does NOT match PL — never set it when converting from PL.XAverage seeds recursively from the first bar's price; Pine ta.ema is na during warmup. Early-history values (and therefore early signals) differ between the two platforms — discard roughly the first 3×Length bars before comparing backtests.Sell does not mean "go short". In PowerLanguage, Sell exits an existing long position. The Pine equivalent is strategy.close(). If you write strategy.entry("label", strategy.short) as a translation of Sell, you will create a new short position rather than closing the long — this is one of the most common and costly conversion mistakes.
Multi-data feed architecture is fundamentally different. PowerLanguage allows a second (or third) price feed to be attached directly to the chart as Data2/Data3, and accesses its bars with Close of Data2. Pine Script has no concept of a second chart feed; instead you must call request.security(syminfo.tickerid, "D", close) with an explicit symbol string and timeframe. There is no direct bar-by-bar alignment guarantee across different timeframes in Pine the way there is when you add a second data series in MultiCharts.
Dollar amounts versus price levels. SetStopLoss and SetProfitTarget in PowerLanguage accept a dollar (currency) amount, which MultiCharts converts internally to a price level using the instrument's point value and position size. Pine Script's strategy.exit() stop and limit parameters require an absolute price level. You must perform the conversion explicitly: stop_price = strategy.position_avg_price - stop_dollars / (qty * syminfo.pointvalue).
Stochastic parameter mismatch. PowerLanguage's Stochastic function accepts 11 parameters covering %K and %D periods, smoothing type, and detrending options. Pine's ta.stoch() accepts 4 parameters and returns raw %K only. If your PL strategy relies on the smoothed %D line or any of the advanced smoothing options, you must replicate that smoothing manually in Pine after calling ta.stoch().
Bar numbering is offset by one. PowerLanguage's BarNumber and CurrentBar start at 1 for the first historical bar. Pine's bar_index starts at 0. Any logic that compares bar counts, computes bar offsets, or initializes arrays based on BarNumber must subtract or add 1 when converting. Caveat: CurrentBar counts from the first EXECUTED bar after MaxBarsBack is satisfied, so bar_index + 1 is exact only when MaxBarsBack = 0.
Position size semantics differ. MarketPosition in PowerLanguage returns exactly -1 (short), 0 (flat), or 1 (long) regardless of how many contracts are held. Pine's strategy.position_size returns the signed number of contracts — for example, +3 if three long contracts are open. Code that tests if MarketPosition = 1 should become if strategy.position_size > 0 in Pine, not if strategy.position_size = 1.
Execution model: confirmed close versus real-time tick. PowerLanguage strategies execute on bar close by default (the bar is confirmed before orders are evaluated). Pine Script strategies also default to executing on bar close, but the behavior of calc_on_every_tick and calc_on_order_fills can alter this. When porting a PL strategy, verify that the Pine strategy() declaration does not add intrabar recalculation that was absent in the original.
Portfolio Money Management (PMM) has no Pine equivalent. MultiCharts PMM allows a single money-management script to govern position sizing across a portfolio of instruments. Pine Script has no portfolio-level concept; each script runs independently on a single symbol. PMM logic must be reimplemented as per-symbol position sizing inside the Pine strategy itself.
PowerLanguage Functions (.elf files) must become Pine user-defined functions or libraries. A PL .elf (EasyLanguage Function) is a reusable calculation unit compiled separately and called from indicators or strategies. In Pine, the equivalent is either a user-defined function defined with f_name(params) => syntax within the same script, or a published Pine library imported with import. There is no separate compilation step in Pine.
History-referencing functions must be called at global scope in Pine. In PL, Crosses Over / Crosses Under can appear inside any If block without issue. In Pine, ta.crossover(), ta.crossunder(), ta.change(), ta.pivothigh(), ta.pivotlow(), and similar functions must be called on every bar at the global scope to maintain correct internal state. Calling them inside a conditional block (e.g. if strategy.position_size > 0) produces a compiler warning and potentially incorrect results. Extract the call to a variable at the top level, then use that variable inside conditions.
Value1..Value99 and Condition1..Condition99 must be renamed. These PL numbered built-ins have no Pine counterpart. Every occurrence must be replaced with a descriptive typed variable (var float, var bool, or a plain expression). Do not carry over numbers as Pine variable names — names like val1 are legal but Value1 in a Pine context is just an undefined identifier that will cause a compile error.
//@version=5 for all conversions — v5 is the proven baseline; v6 may require additional syntax adjustmentsInputs: declaration; note the type (numeric / string / bool) and default value so each can be mapped to the correct input.*() functionData2 / Data3 references; confirm what symbol and timeframe each feed represents so request.security() calls can be written correctlySetStopLoss, SetProfitTarget, and dollar-based order parameters; note the instrument's point value for the price-level conversionIntraBarPersist variables are used; if so, decide whether Pine varip behavior is acceptable or whether the strategy must run on confirmed bars only.elf function calls; locate the function source code and plan whether to inline it or create a Pine librarySell order maps to strategy.close(), not strategy.entry(strategy.short) — audit all order keywordsBuyToCover order maps to strategy.close(), not strategy.entry(strategy.long)BarNumber / CurrentBar comparisons adjusted for 0-based bar_indexMarketPosition comparisons converted from = 1 / = -1 to > 0 / < 0CurrentContracts replaced with math.abs(strategy.position_size)BarsSinceEntry replaced with bar_index - strategy.opentrades.entry_bar_index(strategy.opentrades - 1)strategy.exit() callsValue1..Value99 and Condition1..Condition99 renamed to typed Pine variablesta.crossover(), ta.crossunder(), and other history-referencing functions called at global scope, not inside conditional blocksrequest.security() call; determine whether MultiCharts can supply the required symbol and timeframe as a secondary data feedimport statements); locate equivalent PL functions or plan to reimplement the library logic inlinestrategy.position_size comparisons; note the exact values tested so the sign-based logic can be converted to PL's -1/0/1 modelvarip variables are used and whether the equivalent PL IntraBarPersist behavior is needed or can be droppedstrategy.entry(strategy.short) maps to SellShort, not Sellstrategy.close() on a long position maps to Sell, not SellShortbar_index comparisons incremented by 1 for PL's 1-based BarNumberstrategy.position_size > 0 / < 0 converted to MarketPosition = 1 / = -1strategy.exit() price levels converted back to dollar amounts for SetStopLoss / SetProfitTargetta.dmi, ta.bb) unpacked before the equivalent PL function calls — PL functions return scalar valuesta.stoch() %K result wrapped with appropriate smoothing if the PL target uses Stochastic's built-in %D