distinguishing actual (historical) measured values from forecasted / timing of meteostat pulling actual historical data

[lefkothea1]

Hi, thank you so much for you efforts on this very useful and effective repo!

I am aiming to distinguish actual measured values from forecasted ones, and store both.

I would assume that when pulling data today (D), all rows until yesterday (D-1) would be historical, and all rows where time>= today(D) would be forecasted.

However I notice that calling Hourly(station, start, end, timezone=timezone, model=False).fetch() on day (D), returns data up to and including D-2.

For example, running the above line of code on 15/11/2024 at 14.00, I get data up to and including 14/11/2024 00.00 (so complete until 13/11/2024 --> D-2)

I wanted to ask if I can expect this behavior to be consistent? (ie data up to D-2 are actual measured values, rows for D-1 consitute forecasted values) Or do the historical and forecasted rows returned, depend on the timing of my API call? (depending on when the meteostat backend updates from its data sources?)

Ultimately, I just want to be sure about which values are actually measured or forecasted. Would it be sufficient to treat the last returned datetime from the Hourly(station, start, end, timezone=timezone, model=False).fetch() as the last currenty available actually measured value?

Thanks again for your efforts and support