@vtorralba noticed a bug in the subseasonal case. When providing multiple start dates in this way the recipe is not divided correctly:
Time:
sdate:
- 20240823 #%Y%m%d
- 20240830
This is because of a bug in the code of tools/divide_recipe.R that considers only the current start_date in the loop, and generates a vector of only one start date even when there are multiple, resulting in only one recipe.
This MR fixes this bug.
Victòria