Today's project is relatively straightforward. I've started going through the tropical weather outlooks that cyclocane carries and am parsing the dates that the outlooks were transmitted.
The biggest reason I'm finally doing this is for things like the National Hurricane Season and the Fiji Meteorological Service who have been guilty of having out-of-date tropical outlooks that actually have a probability area.
In the case of the NHC, two years in a row, their last Atlantic tropical weather outlook of the hurricane season has mentioned a super low chance (near 0%), but since this is still a chance, it ends up showing up on cyclocane as a probability until the next season starts.
Fiji has been guilty of doing the same thing, but on the scale of, I don't know, 3 or 4 days, instead of 6 flipping months like the NHC.
So I'm taking matters into my own hands now in determining whether an outlook is out of date and still matters.
This also means I'll be able to sort the tropical storm risk page by which outlooks are most recent.
At the moment, I've probably done about half of the parsing work. So tomorrow, I'll work on completing the parsing side of things, and then the work will begin on actually doing something useful with that raw data.