polars-df 0.15.0 → 0.26.0

data/lib/polars/date_time_name_space.rb

@@ -18,6 +18,71 @@ module Polars
       s[item]
     end
 
+    # Offset by `n` business days.
+    #
+    # @note
+    #   This functionality is considered **unstable**. It may be changed
+    #   at any point without it being considered a breaking change.
+    #
+    # @param n [Object]
+    #   Number of business days to offset by. Can be a single number of an
+    #   expression.
+    # @param week_mask [Array]
+    #   Which days of the week to count. The default is Monday to Friday.
+    #   If you wanted to count only Monday to Thursday, you would pass
+    #   `[true, true, true, true, false, false, false]`.
+    # @param holidays [Object]
+    #   Holidays to exclude from the count.
+    # roll
+    #   What to do when the start date lands on a non-business day. Options are:
+    #
+    #   - `'raise'`: raise an error
+    #   - `'forward'`: move to the next business day
+    #   - `'backward'`: move to the previous business day
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new("start", [Date.new(2020, 1, 1), Date.new(2020, 1, 2)])
+    #   s.dt.add_business_days(5)
+    #   # =>
+    #   # shape: (2,)
+    #   # Series: 'start' [date]
+    #   # [
+    #   #         2020-01-08
+    #   #         2020-01-09
+    #   # ]
+    #
+    # @example You can pass a custom weekend - for example, if you only take Sunday off:
+    #   week_mask = [true, true, true, true, true, true, false]
+    #   s.dt.add_business_days(5, week_mask: week_mask)
+    #   # =>
+    #   # shape: (2,)
+    #   # Series: 'start' [date]
+    #   # [
+    #   #         2020-01-07
+    #   #         2020-01-08
+    #   # ]
+    #
+    # @example Roll all dates forwards to the next business day:
+    #   s = Polars::Series.new("start", [Date.new(2020, 1, 5), Date.new(2020, 1, 6)])
+    #   s.dt.add_business_days(0, roll: "forward")
+    #   # =>
+    #   # shape: (2,)
+    #   # Series: 'start' [date]
+    #   # [
+    #   #         2020-01-06
+    #   #         2020-01-06
+    #   # ]
+    def add_business_days(
+      n,
+      week_mask: [true, true, true, true, true, false, false],
+      holidays: [],
+      roll: "raise"
+    )
+      super
+    end
+
     # Return minimum as Ruby object.
     #
     # @return [Object]
@@ -42,49 +107,6 @@ module Polars
       Utils.wrap_s(_s).max
     end
 
-    # Return median as Ruby object.
-    #
-    # @return [Object]
-    #
-    # @example
-    #   date = Polars.datetime_range(
-    #     DateTime.new(2001, 1, 1), DateTime.new(2001, 1, 3), "1d", eager: true
-    #   ).alias("datetime")
-    #   # =>
-    #   # shape: (3,)
-    #   # Series: 'datetime' [datetime[ns]]
-    #   # [
-    #   #         2001-01-01 00:00:00
-    #   #         2001-01-02 00:00:00
-    #   #         2001-01-03 00:00:00
-    #   # ]
-    #
-    # @example
-    #   date.dt.median
-    #   # => 2001-01-02 00:00:00 UTC
-    def median
-      _s.median
-    end
-
-    # Return mean as Ruby object.
-    #
-    # @return [Object]
-    #
-    # @example
-    #   s = Polars::Series.new([Date.new(2001, 1, 1), Date.new(2001, 1, 2)])
-    #   s.dt.mean
-    #   # => 2001-01-01 12:00:00 UTC
-    #
-    # @example
-    #   s = Polars::Series.new(
-    #     [DateTime.new(2001, 1, 1), DateTime.new(2001, 1, 2), DateTime.new(2001, 1, 3)]
-    #   )
-    #   s.dt.mean
-    #   # => 2001-01-02 00:00:00 UTC
-    def mean
-      _s.mean
-    end
-
     # Convert a Date/Time/Datetime column into a String column with the given format.
     #
     # Similar to `cast(Polars::String)`, but this method allows you to customize the
@@ -111,13 +133,18 @@ module Polars
     #   #         "2020/04/01"
     #   #         "2020/05/01"
     #   # ]
-    def to_string(format)
+    def to_string(format = nil)
       super
     end
 
-    # Format Date/datetime with a formatting rule.
+    # Convert a Date/Time/Datetime column into a String column with the given format.
     #
-    # See [chrono strftime/strptime](https://docs.rs/chrono/0.4.19/chrono/format/strftime/index.html).
+    # Similar to `cast(Polars::String)`, but this method allows you to customize the
+    # formatting of the resulting string.
+    #
+    # @param format [String]
+    #   Format to use, refer to the [chrono strftime documentation](https://docs.rs/chrono/latest/chrono/format/strftime/index.html)
+    #   for specification. Example: `"%y-%m-%d"`.
     #
     # @return [Series]
     #
@@ -135,7 +162,75 @@ module Polars
     #   #         "2020/04/01"
     #   #         "2020/05/01"
     #   # ]
-    def strftime(fmt)
+    def strftime(format)
+      super
+    end
+
+    # Extract the millennium from underlying representation.
+    #
+    # Applies to Date and Datetime columns.
+    #
+    # Returns the millennium number in the calendar date.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new(
+    #     "dt",
+    #     [
+    #       Date.new(999, 12, 31),
+    #       Date.new(1897, 5, 7),
+    #       Date.new(2000, 1, 1),
+    #       Date.new(2001, 7, 5),
+    #       Date.new(3002, 10, 20)
+    #     ]
+    #   )
+    #   s.dt.millennium
+    #   # =>
+    #   # shape: (5,)
+    #   # Series: 'dt' [i32]
+    #   # [
+    #   #         1
+    #   #         2
+    #   #         2
+    #   #         3
+    #   #         4
+    #   # ]
+    def millennium
+      super
+    end
+
+    # Extract the century from underlying representation.
+    #
+    # Applies to Date and Datetime columns.
+    #
+    # Returns the century number in the calendar date.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new(
+    #     "dt",
+    #     [
+    #       Date.new(999, 12, 31),
+    #       Date.new(1897, 5, 7),
+    #       Date.new(2000, 1, 1),
+    #       Date.new(2001, 7, 5),
+    #       Date.new(3002, 10, 20)
+    #     ]
+    #   )
+    #   s.dt.century
+    #   # =>
+    #   # shape: (5,)
+    #   # Series: 'dt' [i32]
+    #   # [
+    #   #         10
+    #   #         19
+    #   #         20
+    #   #         21
+    #   #         31
+    #   # ]
+    def century
       super
     end
 
@@ -161,6 +256,72 @@ module Polars
       super
     end
 
+    # Determine whether each day lands on a business day.
+    #
+    # @note
+    #   This functionality is considered **unstable**. It may be changed
+    #   at any point without it being considered a breaking change.
+    #
+    # @param week_mask [Array]
+    #   Which days of the week to count. The default is Monday to Friday.
+    #   If you wanted to count only Monday to Thursday, you would pass
+    #   `[true, true, true, true, false, false, false]`.
+    # @param holidays [Object]
+    #   Holidays to exclude from the count.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new([Date.new(2020, 1, 3), Date.new(2020, 1, 5)])
+    #   s.dt.is_business_day
+    #   # =>
+    #   # shape: (2,)
+    #   # Series: '' [bool]
+    #   # [
+    #   #         true
+    #   #         false
+    #   # ]
+    #
+    # @example You can pass a custom weekend - for example, if you only take Sunday off:
+    #   week_mask = [true, true, true, true, true, true, false]
+    #   s.dt.is_business_day(week_mask: week_mask)
+    #   # =>
+    #   # shape: (2,)
+    #   # Series: '' [bool]
+    #   # [
+    #   #         true
+    #   #         false
+    #   # ]
+    def is_business_day(
+      week_mask: [true, true, true, true, true, false, false],
+      holidays: []
+    )
+      super
+    end
+
+    # Determine whether the year of the underlying date representation is a leap year.
+    #
+    # Applies to Date and Datetime columns.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new(
+    #     "date", [Date.new(2000, 1, 1), Date.new(2001, 1, 1), Date.new(2002, 1, 1)]
+    #   )
+    #   s.dt.is_leap_year
+    #   # =>
+    #   # shape: (3,)
+    #   # Series: 'date' [bool]
+    #   # [
+    #   #         true
+    #   #         false
+    #   #         false
+    #   # ]
+    def is_leap_year
+      super
+    end
+
     # Extract ISO year from underlying Date representation.
     #
     # Applies to Date and Datetime columns.
@@ -236,6 +397,32 @@ module Polars
       super
     end
 
+    # Extract the number of days in the month from the underlying date representation.
+    #
+    # Applies to Date and Datetime columns.
+    #
+    # Returns the number of days in the month.
+    # The return value ranges from 28 to 31.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new(
+    #     "date", [Date.new(2001, 1, 1), Date.new(2001, 2, 1), Date.new(2000, 2, 1)]
+    #   )
+    #   s.dt.days_in_month
+    #   # =>
+    #   # shape: (3,)
+    #   # Series: 'date' [i8]
+    #   # [
+    #   #         31
+    #   #         28
+    #   #         29
+    #   # ]
+    def days_in_month
+      super
+    end
+
     # Extract the week from the underlying date representation.
     #
     # Applies to Date and Datetime columns.
@@ -346,6 +533,48 @@ module Polars
       super
     end
 
+    # Extract (local) time.
+    #
+    # Applies to Date/Datetime/Time columns.
+    #
+    # @return
+    #
+    # @example
+    #   ser = Polars::Series.new([DateTime.new(2021, 1, 2, 5)]).dt.replace_time_zone(
+    #     "Asia/Kathmandu"
+    #   )
+    #   ser.dt.time
+    #   # =>
+    #   # shape: (1,)
+    #   # Series: '' [time]
+    #   # [
+    #   #         05:00:00
+    #   # ]
+    def time
+      super
+    end
+
+    # Extract (local) date.
+    #
+    # Applies to Date/Datetime columns.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   ser = Polars::Series.new([DateTime.new(2021, 1, 2, 5)]).dt.replace_time_zone(
+    #     "Asia/Kathmandu"
+    #   )
+    #   ser.dt.date
+    #   # =>
+    #   # shape: (1,)
+    #   # Series: '' [date]
+    #   # [
+    #   #         2021-01-02
+    #   # ]
+    def date
+      super
+    end
+
     # Extract the hour from the underlying DateTime representation.
     #
     # Applies to Datetime columns.
@@ -624,43 +853,6 @@ module Polars
       super
     end
 
-    # Set time unit a Series of dtype Datetime or Duration.
-    #
-    # This does not modify underlying data, and should be used to fix an incorrect
-    # time unit.
-    #
-    # @param time_unit ["ns", "us", "ms"]
-    #   Time unit for the `Datetime` Series.
-    #
-    # @return [Series]
-    #
-    # @example
-    #   start = DateTime.new(2001, 1, 1)
-    #   stop = DateTime.new(2001, 1, 3)
-    #   date = Polars.datetime_range(start, stop, "1d", time_unit: "ns", eager: true).alias("datetime")
-    #   # =>
-    #   # shape: (3,)
-    #   # Series: 'datetime' [datetime[ns]]
-    #   # [
-    #   #         2001-01-01 00:00:00
-    #   #         2001-01-02 00:00:00
-    #   #         2001-01-03 00:00:00
-    #   # ]
-    #
-    # @example
-    #   date.dt.with_time_unit("us").alias("tu_us")
-    #   # =>
-    #   # shape: (3,)
-    #   # Series: 'tu_us' [datetime[μs]]
-    #   # [
-    #   #         +32971-04-28 00:00:00
-    #   #         +32974-01-22 00:00:00
-    #   #         +32976-10-18 00:00:00
-    #   # ]
-    def with_time_unit(time_unit)
-      super
-    end
-
     # Cast the underlying data to another time unit. This may lose precision.
     #
     # @param time_unit ["ns", "us", "ms"]
@@ -825,21 +1017,11 @@ module Polars
       super
     end
 
-    # Localize tz-naive Datetime Series to tz-aware Datetime Series.
-    #
-    # This method takes a naive Datetime Series and makes this time zone aware.
-    # It does not move the time to another time zone.
-    #
-    # @param tz [String]
-    #   Time zone for the `Datetime` Series.
-    #
-    # @return [Series]
-    def tz_localize(tz)
-      super
-    end
-
     # Extract the days from a Duration type.
     #
+    # @param fractional [Boolean]
+    #   Whether to include the fractional component of the day.
+    #
     # @return [Series]
     #
     # @example
@@ -855,13 +1037,15 @@ module Polars
     #   #         31
     #   #         30
     #   # ]
-    def total_days
+    def total_days(fractional: false)
       super
     end
-    alias_method :days, :total_days
 
     # Extract the hours from a Duration type.
     #
+    # @param fractional [Boolean]
+    #   Whether to include the fractional component of the hour.
+    #
     # @return [Series]
     #
     # @example
@@ -887,13 +1071,15 @@ module Polars
     #   #         24
     #   #         24
     #   # ]
-    def total_hours
+    def total_hours(fractional: false)
       super
     end
-    alias_method :hours, :total_hours
 
     # Extract the minutes from a Duration type.
     #
+    # @param fractional [Boolean]
+    #   Whether to include the fractional component of the minute.
+    #
     # @return [Series]
     #
     # @example
@@ -919,13 +1105,15 @@ module Polars
     #   #         1440
     #   #         1440
     #   # ]
-    def total_minutes
+    def total_minutes(fractional: false)
       super
     end
-    alias_method :minutes, :total_minutes
 
     # Extract the seconds from a Duration type.
     #
+    # @param fractional [Boolean]
+    #   Whether to include the fractional component of the second.
+    #
     # @return [Series]
     #
     # @example
@@ -955,13 +1143,15 @@ module Polars
     #   #         60
     #   #         60
     #   # ]
-    def total_seconds
+    def total_seconds(fractional: false)
       super
     end
-    alias_method :seconds, :total_seconds
 
     # Extract the milliseconds from a Duration type.
     #
+    # @param fractional [Boolean]
+    #   Whether to include the fractional component of the millisecond.
+    #
     # @return [Series]
     #
     # @example
@@ -987,13 +1177,15 @@ module Polars
     #   #         1
     #   #         1
     #   # ]
-    def total_milliseconds
+    def total_milliseconds(fractional: false)
       super
     end
-    alias_method :milliseconds, :total_milliseconds
 
     # Extract the microseconds from a Duration type.
     #
+    # @param fractional [Boolean]
+    #   Whether to include the fractional component of the microsecond.
+    #
     # @return [Series]
     #
     # @example
@@ -1019,13 +1211,17 @@ module Polars
     #   #         1000
     #   #         1000
     #   # ]
-    def total_microseconds
+    def total_microseconds(fractional: false)
       super
     end
-    alias_method :microseconds, :total_microseconds
 
     # Extract the nanoseconds from a Duration type.
     #
+    # @param fractional [Boolean]
+    #   Whether to include return the result as a `Float64`.
+    #   Because the smallest `TimeUnit` is `'ns'`, the
+    #   fractional component will always be zero.
+    #
     # @return [Series]
     #
     # @example
@@ -1051,10 +1247,9 @@ module Polars
     #   #         1000000
     #   #         1000000
     #   # ]
-    def total_nanoseconds
+    def total_nanoseconds(fractional: false)
       super
     end
-    alias_method :nanoseconds, :total_nanoseconds
 
     # Offset this date by a relative time offset.
     #
@@ -1276,6 +1471,35 @@ module Polars
       super
     end
 
+    # Create a naive Datetime from an existing Date/Datetime expression and a Time.
+    #
+    # If the underlying expression is a Datetime then its time component is replaced,
+    # and if it is a Date then a new Datetime is created by combining the two values.
+    #
+    # @param time [Object]
+    #   A Ruby time literal or Series of the same length as this Series.
+    # @param time_unit ['ns', 'us', 'ms']
+    #   Unit of time.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new(
+    #     "dtm",
+    #     [DateTime.new(2022, 12, 31, 10, 30, 45), DateTime.new(2023, 7, 5, 23, 59, 59)]
+    #   )
+    #   s.dt.combine(Time.new(2000, 1, 1, 1, 2, 3.456))
+    #   # =>
+    #   # shape: (2,)
+    #   # Series: 'dtm' [datetime[μs]]
+    #   # [
+    #   #         2022-12-31 01:02:03.456
+    #   #         2023-07-05 01:02:03.456
+    #   # ]
+    def combine(time, time_unit: "us")
+      super
+    end
+
     # Roll backward to the first day of the month.
     #
     # @return [Series]
@@ -1370,5 +1594,54 @@ module Polars
     def dst_offset
       super
     end
+
+    # Replace time unit.
+    #
+    # @param year [Object]
+    #   Literal or Series.
+    # @param month [Object]
+    #   Literal or Series, ranging from 1-12.
+    # @param day [Object]
+    #   Literal or Series, ranging from 1-31.
+    # @param hour [Object]
+    #   Literal or Series, ranging from 0-23.
+    # @param minute [Object]
+    #   Literal or Series, ranging from 0-59.
+    # @param second [Object]
+    #   Literal or Series, ranging from 0-59.
+    # @param microsecond [Object]
+    #   Literal or Series, ranging from 0-999999.
+    # @param ambiguous [String]
+    #   Determine how to deal with ambiguous datetimes:
+    #
+    #   - `'raise'` (default): raise
+    #   - `'earliest'`: use the earliest datetime
+    #   - `'latest'`: use the latest datetime
+    #   - `'null'`: set to null
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new("date", [Date.new(2013, 1, 1), Date.new(2024, 1, 2)])
+    #   s.dt.replace(year: 1800)
+    #   # =>
+    #   # shape: (2,)
+    #   # Series: 'date' [date]
+    #   # [
+    #   #         1800-01-01
+    #   #         1800-01-02
+    #   # ]
+    def replace(
+      year: nil,
+      month: nil,
+      day: nil,
+      hour: nil,
+      minute: nil,
+      second: nil,
+      microsecond: nil,
+      ambiguous: "raise"
+    )
+      super
+    end
   end
 end