Reminder to use something like the date-fns library for immutable operations on dates. The native JS date library can be painful to work on when doing operations like adding or subtracting time.
A function that takes a simple struct (containing each component) can also handle the conundrum. The function could of course even take a string representation.
e.g. If today is the 30th of January 2024. What is the date in 1 months time?
Should it be the 29th of February? (as the 30th doesn't exist). The 28th? (the penultimate day of the month, like the 30th of Jan). The 1st of March (a 'standard' month should be considered 30 days, so 30 days from now).
Date-fns is fine for simpler use cases but Luxon is a lot more complete, especially where it comes to time zones.
This is not the kind of thing you just want to blindly hack your way through without a really thorough understanding of how different cultures/locales/governments handle dates and times.
If you have to do math across daylight savings time boundaries that change over time (because governments and regulations change), converted to two or more time zones (which again have their own separate DST rules), and possibly span some 30 or 45 minute time zones (they exist!) this will quickly get out of hand and unreadable. Not just unreadable, but incredibly difficult to reason about.
It gets even worse when the backend stores only the offset (as in the ISO time, like T23:00:00+08:00) because then you lose the original time zone and can't be sure whether it came from a DST zone or another country in the same zone (but only during part of the year).
Some random examples...
When you cross DST switchovers, for example, you may magically gain or lose an hour. A naive milliseconds calculation will miss that. It gets worse because DST in a country isn't a static thing either, but will change with the laws over time. So over decades, you need to have several lookup tables.
And people are ambiguous. Is January 31 plus 1 month the end of February or the beginning of March? What about during a leap year?
And a "day" isn't necessarily 24 hours (because of DST, again). Humans doing day math across those boundaries will just ignore (or really, never think about) the missing or gained hours, but computers have to explicitly account for it.
Then once you throw in different time zones it gets even crazier, especially for the half hour and 45 minute zones.
It's okay to store datetimes in epoch milliseconds (ideally with a separate field for the time zone, not just offset, which holds less information). But you can't easily/correctly do math on that without more specific instructions and cultural/locale adjustments.
Use an integer type denoting days since the start date. See Modified Julian date.
Thus tomorrow is always 1 more than today - if you have times you need leap seconds.
Slightly sad to have been downvoted, though, just for making a tentative suggestion (note use of question mark) :'(
targetEndDate = new Date();
targetEndDate.setFullYear(endDate.getFullYear());
targetEndDate.setMonth(endDate.getMonth());
targetEndDate.setDate(endDate.getDate());
If you just did:
targetEndDate = createDate(endDate.getFullYear(), endDate.getMonth(), endDate.getDate());
// createDate is an intentional placeholder, I didn't want to double-check the actual syntax
In this case, using mutation on the individual fields makes you have to transition through an invalid state to get back to a valid one, and the JS date object does something unexpected (though I think there's nothing good to do here, an exception is probably the best you could do). So mutation really is at issue here. In general, mutation creates room for these kind of counter-intuitive state transitions to arise.
"Change the code completely so you supply all 3 components at the same time" obviously fixes the bug, but that doesn't require an immutable type.
.NET's (immutable) DateTime, on the other hand, has AddMonths(Int32), which sets the day to Min(original day number, last day of result month), which, while arguably reasonable, isn't obvious without reading the documentation.
My favorite counterintutive DateTime "mutator," however is AddMilliseconds(Double): prior to .NET 7, its floating-point argument is rounded to the nearest integer (!?!).
If you want to do this mutation-style for some reason, you should use the builder pattern (or whatever it's called). So in this case you'd instantiate a DateBuilder instance instead of a Date instance, and finally call .date() to get the actual Date instance from the builder instance.
For a simple date/time class, the builder pattern reeks of pointless overengineering.
But I've seen the mutating pattern in other cases causing the weird state bug which could have avoided using a builder.