Bug fixing

[jmaslek]

Fixes #3670 (Also changes the sdk function slightly)

Fixes #3674

[montezdesousa]

I'm getting this warning when inputting a date that is not available, in this case 2022-12-01 onwards. In this case we should fallback to the last available as before imo

[jmaslek]

I'm getting this warning when inputting a date that is not available, in this case 2022-12-01 onwards. In this case we should fallback to the last available as before imo

I silenced that warning by changing the fred model:

However I am noticing some weird things. If I specify a date, I always get the most recent it looks like?

    if date not in df.index:
        date_of_yield = df.index[-1].strftime("%Y-%m-%d")
        rates = pd.DataFrame(df.iloc[-1, :].values, columns=["Rate"])

This does what you suggest by getting the closest date, and there is the following block in the model which I think is causing the problems of not getting the most recent:

    if date is None:
        date = (datetime.now() - timedelta(days=7)).strftime("%Y-%m-%d")

Will push some better logic later today. Also open to thoughts

[jmaslek]

@montezdesousa I have the following. Without a date argument, it passes the last date data is available:

If I enter a weekend, it gets the next day:

And if I enter the future, new message:

2022 Dec 02, 09:52 (🦋) /economy/ $ ycrv -d 2040-10-10

Date cannot be today or in the future
Yield data not found for 2040-10-10.

[montezdesousa]

@montezdesousa I have the following. Without a date argument, it passes the last date data is available:

If I enter a weekend, it gets the next day:

And if I enter the future, new message:

2022 Dec 02, 09:52 (🦋) /economy/ $ ycrv -d 2040-10-10

Date cannot be today or in the future
Yield data not found for 2040-10-10.

oof, sorry I must be looking to wrong branch, this ok in yours

[montezdesousa]

Seems to work fine for me