Did Alpaca change the order of the OHLC in the dataframe when retrieving historical bars? I’m using the alpaca_trade_api to retrieve historical bars at the start of the day and then subscribing to minute bars and updating the dataframe as new bars come in as I’m sure a lot of people do.
I happened to notice today that open prices on all symbols was way off. After a little troubleshooting I found that the order of the columns was different in the historical dataframe. In the past the order of the columns in the returned dataframe was “timestamp, open, high, low, close, volume, trade_count, vwap”. Today, and I don’t how long this has been going on, the order of the historical dataframe is “timestamp, close, high, low, trade_count, open, volume, vwap”. Because of this, the subscribed bars were updating the dataframe with open prices on the close column, close prices on trade_count column, volume on the open column and trade_count on the volume column. This obviously is causing some issues.
Is this just an error for today or is this permanently going to be in this order in the historical dataframe? I just tried retrieving the same historical data with the new API and the columns are in the correct order or at least the order I am expecting to see. Was this done to get people to stop using the old API?