Closed ivelascog closed 5 months ago
I think it makes sense to return the selected ids, and then provide a way of also accessing the individual rows including the time component. But given that the data is in tidy format
id | x | y |
---|---|---|
a | 0 | 10 |
a | 1 | 20 |
a | 2 | 10 |
b | 0 | 30 |
b | 1 | 20 |
b | 2 | 10 |
It is more useful to return that a and b as selected, rather than which specific combination of id, x, y are
Ok, so I can think of two options:
Similar to what we have now
Map of timeGroupName => Map itemID => Actual input data selected
e.g.
{
"Group_Spanish_Speaking" :
{
"Colombia":
[ { Country: "Colombia", temp: 27, year: 2020},
{ Country: "Colombia", temp: 26, year: 2021},
{ Country: "Colombia", temp: 28, year: 2022}
],
"España":
[ { Country: "España", temp: 17, year: 2020},
{ Country: "España", temp: 16, year: 2021},
{ Country: "España", temp: 18, year: 2022}
],
....
},
....
}
itemID | timeGroupName | pointsSelected in Original data |
---|---|---|
Colombia | Group_Spanish_Speaking | [ { Country: "Colombia", temp: 27, year: 2020}, { Country: "Colombia", temp: 26, year: 2021}, { Country: "Colombia", temp: 28, year: 2022} ] |
España | Group_Spanish_Speaking | [ { Country: "España", temp: 17, year: 2020}, { Country: "España", temp: 16, year: 2021}, { Country: "España", temp: 18, year: 2022} ] |
I think that option 1 gives more options to obtain in a simple way both the Ids per group and the selected points in the original data. I will make the change, although we will probably break some notebooks and examples.
Let's add a selectedIds attribute in the returned value that contains the ids of the currently selected group.
Also return a selectedGroup attribute that contains the name of the currently selected group
TimeSearcher+ receives all data without grouping, however, it returns the data grouped by the "id" parameter. I don't know if it would be better to return the selected data in the same format as the input data to avoid confusion.