Closed nielsdevleeschouwer closed 2 years ago
Certainly a valid point, as we rely on the Pandas date conversion, this is interpreted as month-day-year
, see https://github.com/fluves/pywaterinfo/blob/master/src/pywaterinfo/waterinfo.py#L358 This provides more flexibility compared to requiring an actual datetime.datetime
(or pd.Timestamp)
object with a trade off for 'wrong' date representations....
I would propose to keep the functionality as such - we can not know what the intention is of the user month-day or day-month. If as a user/client want to be sure, using ISO 8601
, https://nl.wikipedia.org/wiki/ISO_8601 or converting to a datetime.datetime
object is both a valid input which won't run into this issue.
Closing this issue with the yyyy-mm-dd as preferred input which will ensure to not encounter this issue.
Description
Downloading a time series based on timeseries id, start and end date of an event. Date string '01-03-2018' is interpreted as month-day-year instead of day-month-year Question: not standard date formatting in Flanders --> changing date format? / documenting alternative date format? / only allowing year-month-day format?
What I Did