As you can see in the following example, you may have a mix of chart, market and exchange usage within the same Formula or Condition.

The Dependency Filter is a mechanism that analyzes the code of all conditions, formulas and Study Bots in order to get from the code which data dependencies need to be loaded at the bot once ran. The procedure scans every line of code in search for the exchange, market or chart objects and once it finds them, it then breaks it down to understand which product will be needed for the bot. These filters are needed in order to prevent the bot to load all dependency files at all time frames when at the code many of these files are evidently not used.

A Dependency Filter is a list of Indicators and Studies a Strategy depends on, that is later used to filter out all the other indicators and studies the Trading Bot depends on.

The function will scan a node branch, most likely a Trading System, looking into the code property of each node. It will analyze it's content and try to make a list of all indicators and studies mentioned at the code text and at which time frames they are mentioned.

It is important to note that all nodes that are not of the type Procedure Javascript, Javascript Code or Formula are going to be ignored.

These are instructions that the Dependency Filter will recognize without any problems:

market.BTC.USDT.chart.at01hs.popularSMA.sma200 - market.ETC.USDT.chart.at01hs.popularSMA.sma100 < 10

If you removed spaces between instructions then the Dependency Filter might get confused and you will get an undefined value for the indicator you are trying to use.

Inside a commented area of the code you can write something like this:

market.anyBaseAsset.BTC.chart.atAnyTimeFrame.candle

The valid keyworks are:

anyBaseAsset

anyQuotedAsset

atAnyTimeFrame