input_modifier does not apply to sub_indicators

[powellnorma]

Is this on purpose? Let's say I want to pass OHLCV to MACD.add, which I initialized with input_modifier=ValueExtractor.extract_close.

However, here the value gets added to the sub_indicators before the input_modifier got applied:

talipp/talipp/indicators/Indicator.py

Lines 103 to 114 in ed5e0e1

	else:
	for sub_indicator in self.sub_indicators:
	sub_indicator.add(value)
	if not isinstance(value, list):
	value = [value]
	for input_value in value:
	if input_value is not None and self.input_modifier is not None:
	input_value = self.input_modifier(input_value)
	self.input_values.append(input_value)

My goal is to use the resampling logic for MACD, for this to work I think I have to pass OHLCV to Indicator.add.

[nardew]

Hi, input modifiers are intended only for indicator chaining and if an indicator accepts only floats, auto-sampling is not supported.

However, you seem to be right that chaining does not work for composite indicators. I will look into this.

[nardew]

Delivered in 2.4.1.