DateTime.before?(a, b) is much nicer than DateTime.compare(a, b) == :lt. It doesn't completely remove the argument order issue but I reckon it would resolve it for me. I run DateTime.compare(a, b) in iex every time I use the function because I'm terribly forgetful and paranoid. I would prefer DateTime.eq?/lt?/le?/gt?/ge? instead of before?/after?/on_or_before?/on_or_after? which is shorter, matches compare/2 and might allow the le/ge equivalents to sneak through. I think it would be a shame to leave out le and ge.
DateTime.is?/compare?(a, :lt, b) is a whole lot less ambiguous to me. It reads how you would write it in maths or spoken language. On Monday, 31 October 2022 at 5:08:35 pm UTC+10 zachary....@gmail.com wrote: > I wonder how much of the issue is the Api and how much of the issue is > just the docs? I.e its not a given that all arguments in every position > always make sense, but we typically rely on things like elixir_ls to help > us when the answer isn't obvious. > > Could we perhaps just improve the docs in some way? i.e update the specs > to say `datetime :: Calendar.datetime(), compares_to :: > Calendar.datetime()`, and have the args say `compare(datetime, > compares_to)` and have part of the first line of text say something a bit > more informative? > > > On Mon, Oct 31, 2022 at 3:02 AM, Jon Rowe <ma...@jonrowe.co.uk> wrote: > >> I'm not sure the name is right, but I like >> >> DateTime.is?(a <http://datetime.is/?(a>, operator, b), when operator :lt >> | :le | :eq | :ge | :gt, which would capture the :le and :ge options. >> >> >> As a usage api, we could actually have `compare?/3` especially as the >> name doesn't overlap with `compare/2` which would hopefully alleviate >> anyones concerns about the return type changing >> >> On Mon, 31 Oct 2022, at 6:23 AM, José Valim wrote: >> > My thought process is that a simple to use API should be the focus, >> because we already have a complete API in Date.compare/2 >> <http://date.compare/2> and friends. >> >> On Mon, Oct 31, 2022 at 02:16 Simon McConnell <simonmcconn...@gmail.com> >> wrote: >> >> would we want on_or_after? and on_or_before? as well then? Or something >> like DateTime.is?(a <http://datetime.is/?(a>, operator, b), when >> operator :lt | :le | :eq | :ge | :gt, which would capture the :le and :ge >> options. >> >> On Monday, 31 October 2022 at 7:26:42 am UTC+10 José Valim wrote: >> >> Thank you! >> >> A PR that adds before?/after? to Time, Date, NaiveDateTime, and DateTime >> is welcome! >> >> >> On Sun, Oct 30, 2022 at 6:46 PM Cliff <notcliff...@gmail.com> wrote: >> >> I did a bit of research. Many other languages use some form of operator >> overloading to do datetime comparison. The ones that do something different: >> >> - Java has LocalDateTime.compareTo(other) >> >> <https://docs.oracle.com/en/java/javase/11/docs/api/java.base/java/time/LocalDateTime.html#compareTo(java.time.chrono.ChronoLocalDateTime)>, >> >> returning an integer representing gt/lt/eq. There is also >> LocalDateTime.isBefore(other) >> >> <https://docs.oracle.com/en/java/javase/11/docs/api/java.base/java/time/LocalDateTime.html#isBefore(java.time.chrono.ChronoLocalDateTime)>, >> >> LocalDateTime.isAfter(other), and LocalDateTime.isEqual(other). The >> LocalDateTime.is <http://localdatetime.is/>{Before, After} methods >> are non-inclusive (<, >) comparisons. They are instance methods, so usage >> is like `myTime1.isBefore(myTime2)` >> - OCaml's "calendar" library provides a Date.compare >> >> <https://ocaml.org/p/calendar/3.0.0/doc/CalendarLib/Date/index.html#val-compare> >> >> function that returns an integer representing gt/lt/eq (for use in >> OCaml's >> List.sort function, which sorts a list according to the provided >> comparison >> function). It also provides Date.> >> >> <https://ocaml.org/p/calendar/3.0.0/doc/CalendarLib/Date/index.html#val-(%3E)>, >> >> and Date.>= >> >> <https://ocaml.org/p/calendar/3.0.0/doc/CalendarLib/Date/index.html#val-(%3E=)>, >> >> etc. Worth noting is that OCaml allows you to do expression-level module >> imports, like *Date.(my_t1 > my_t2)* to use Date's *>* function in >> the parenthesized expression without needing to *open Date* in the >> entire scope ("open" is OCaml's "import") - this could potentially be >> possible in Elixir using a macro? >> - Golang: t1.After(t2) <https://pkg.go.dev/time#Time.After>, >> t1.Before(t2), t1.Equal(t2). Non-inclusive (> and <). >> - Clojure clj-time library: (after? t1 t2) >> >> <https://clj-time.github.io/clj-time/doc/clj-time.core.html#var-after.3F>, >> (before? t1 t2) >> >> <https://clj-time.github.io/clj-time/doc/clj-time.core.html#var-before.3F>, >> and (equal? t1 t2) >> >> <https://clj-time.github.io/clj-time/doc/clj-time.core.html#var-equal.3F>. >> IMO the argument order is still confusing in these. >> >> >> >> >> On Sunday, October 30, 2022 at 3:15:14 AM UTC-4 José Valim wrote: >> >> I am definitely in favor of clearer APIs. >> >> However, it would probably be best to explore how different libraries in >> different languages tackle this. Can you please explore this? In >> particular, I am curious to know if before/after mean "<" and ">" >> respectively or if they mean "<=" and "=>" (I assume the former). And also >> if some libraries feel compelled to expose functions such as >> "after_or_equal" or if users would have to write Date.equal?(date1, date2) >> or Date.earlier?(date1, date2), which would end-up doing the double of >> conversions. >> >> >> >> -- >> You received this message because you are subscribed to the Google Groups >> "elixir-lang-core" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to elixir-lang-co...@googlegroups.com. >> >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/elixir-lang-core/fcd07389-c6a0-497d-9c09-7f1eacf620c6n%40googlegroups.com >> >> <https://groups.google.com/d/msgid/elixir-lang-core/fcd07389-c6a0-497d-9c09-7f1eacf620c6n%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> >> >> -- >> You received this message because you are subscribed to the Google Groups >> "elixir-lang-core" group. >> >> To unsubscribe from this group and stop receiving emails from it, send an >> email to elixir-lang-core+unsubscr...@googlegroups.com. >> >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/elixir-lang-core/e6c55604-c3ea-464c-908c-5a6092f4d8edn%40googlegroups.com >> >> <https://groups.google.com/d/msgid/elixir-lang-core/e6c55604-c3ea-464c-908c-5a6092f4d8edn%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> >> >> -- >> You received this message because you are subscribed to the Google Groups >> "elixir-lang-core" group. >> >> To unsubscribe from this group and stop receiving emails from it, send an >> email to elixir-lang-core+unsubscr...@googlegroups.com. >> >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/elixir-lang-core/CAGnRm4%2ByT9jA7uqGX0Cyapgfx0AjW%2BU_d4Ai-NQ6vD9UsEb2uQ%40mail.gmail.com >> >> <https://groups.google.com/d/msgid/elixir-lang-core/CAGnRm4%2ByT9jA7uqGX0Cyapgfx0AjW%2BU_d4Ai-NQ6vD9UsEb2uQ%40mail.gmail.com?utm_medium=email&utm_source=footer> >> . >> >> -- >> You received this message because you are subscribed to the Google Groups >> "elixir-lang-core" group. >> > To unsubscribe from this group and stop receiving emails from it, send an >> email to elixir-lang-core+unsubscr...@googlegroups.com. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/elixir-lang-core/2e821e87-6ee0-4702-b69f-e2616b61b1dd%40app.fastmail.com >> >> <https://groups.google.com/d/msgid/elixir-lang-core/2e821e87-6ee0-4702-b69f-e2616b61b1dd%40app.fastmail.com?utm_medium=email&utm_source=footer> >> . >> > > -- You received this message because you are subscribed to the Google Groups "elixir-lang-core" group. To unsubscribe from this group and stop receiving emails from it, send an email to elixir-lang-core+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/elixir-lang-core/f2f2ee39-ea91-4e05-8f83-62525defe251n%40googlegroups.com.
