Repository navigation
1.25: "every nth day of month" calculating wrong dates with certain forecast start dates #2032
Description
I found a fairly obscure regression with forecast dates in 1.25:
~ every 31st day of month
(a) 1
Here, 1.24 correctly approximates the 31st day of each month:
$ hledger-1.24 -f a.j reg --forecast=202303..
2023-03-31 (a) 1 1
2023-04-30 (a) 1 2
2023-05-31 (a) 1 3
2023-06-30 (a) 1 4
2023-07-31 (a) 1 5
2023-08-31 (a) 1 6
2023-09-30 (a) 1 7
But 1.25 gets confused by the forecast start date and calculates wrong dates:
$ hledger-1.25 -f a.j reg --forecast=202303..
2023-03-28 (a) 1 1
2023-04-28 (a) 1 2
2023-05-28 (a) 1 3
2023-06-28 (a) 1 4
2023-07-28 (a) 1 5
2023-08-28 (a) 1 6
2023-09-28 (a) 1 7
2023-10-28 (a) 1 8
Activity
[-]"every nth day of month" calculating wrong forecast dates near end of month (since 1.25)[/-][+]"every nth day of month" calculating wrong dates with a forecast start date (since 1.25)[/+][-]"every nth day of month" calculating wrong dates with a forecast start date (since 1.25)[/-][+]"every nth day of month" calculating wrong dates with certain forecast start dates (since 1.25)[/+]fix: "every 29th/30th/31st day of month" dates with a start date (fix #…
dev: fix a date-sensitive test (#2032)
[-]"every nth day of month" calculating wrong dates with certain forecast start dates (since 1.25)[/-][+]1.25: "every nth day of month" calculating wrong dates with certain forecast start dates[/+]