new functionalities

[hurielreichel]

• 1. pass .p and .con as function parameters as in Edzer's commit.
• 2. group_by + summarise behaviour {if data cude is grouped, or not, different interactions} - add new subclass for that and therefore a (sub)method.
• 3. pass what slice does to filter.
• 4. year %in% (2000:2002) / group_by(year, month) [use list_connection()$extent as reference
• 5. check mandatory params -> is.null() stop()
• 6. "datacube" should be the first in the list of inherits(), not last
• 7. Maths and Ops Math.datacube = function(x, ...) { p = openeo::processes(); p[.Generic] }
• 8. Add examples for all functions + add them to the vignette
• 9. create slice interaction more similar dplyr slice
• 10. remove default con from all function... speed up code.

[edzer]

Yes; see also Open-EO/openeo-r-client#141

[hurielreichel]

Thanks a lot @edzer ! Working on it!

[hurielreichel]

@edzer for number 4, the filter using very dplyrish verbs...

the only way I see this working is to store somehow time related (maybe even space related) objects/columns in the class "datacube". I'd then add "year", "month", "day", "x", "y", create many filter_temporal and filter_spatial and finally use a merge_cubes() process.

What do you think of this idea?

[hurielreichel]

I'll discuss it with matthias, see if there's a way to do filter_temporal with length > 2 time_ext lists.

[hurielreichel]

@edzer could you have a look at https://github.com/hurielreichel/tidyopeneo/blob/main/R/Math.R
It's mainly your code, so it would help if you can check it.
Thanks :)