I added an explicit test with a bit more context to explain why the normalise_ranges() is checking for the end of the year (or at least my understanding of it).
The problem
Example:
$ curl "https://api.carbonintensity.org.uk/regional/intensity/2022-12-31T00:01Z/2023-01-01T05:00Z/regionid/13" | jq '.data.data[].from'
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 17235 100 17235 0 0 71119 0 --:--:-- --:--:-- --:--:-- 71219
"2022-12-31T00:00Z"
"2022-12-31T00:30Z"
[...]
"2022-12-31T22:30Z"
"2022-12-31T23:00Z"
What happened to the values for the 1st January?
This is why currently the logic checks if a range is spanning multiple
years, e.g. 2023-12-30..2024-01-10, and it splits it into 2 ranges,
e.g. 2023-12-30..2024-01-01 and 2024-01-01..2024-01-10.
I added an explicit test with a bit more context to explain why the
normalise_ranges()is checking for the end of the year (or at least my understanding of it).The problem
Example:
What happened to the values for the 1st January?
This is why currently the logic checks if a range is spanning multiple years, e.g. 2023-12-30..2024-01-10, and it splits it into 2 ranges, e.g. 2023-12-30..2024-01-01 and 2024-01-01..2024-01-10.