Preprocessor extract_season does not return complete seasons

[k-a-webb]

When the dataset time vector starts or ends mid-season, extract_season will return incomplete seasons for the first and/or last years.
For example, if a data set spans 2000-01-01 to 2010-12-31, the season DJF will have just JF in 2000 and just D in 2010.

In the related pull request, I propose to add an additional argument to extract_season, full (default False), which will drop incomplete seasons.

[valeriupredoi]

technically correct though 🤣 I suspect there's some string slicing going on in there, it'd be great if you could have a closer look, and many thanks 🍺

[valeriupredoi]

@schlunma is there even such a thing as an "incomplete season" in the modelling nomenclature ie as Kristi mentions, is JF a valid season? (I thought not, but am a mere engineer)

[schlunma]

Well, if it say's "DJF", all of the seasons should be there, not just "JF". So in my opinion it's reasonable to add this here.

[valeriupredoi]

cheers, Manu! What I thought too, unless you live in Wales, where all seasons are called MR (for mild and rainy) 🏴󠁧󠁢󠁷󠁬󠁳󠁿