refactor(exprs): temporals (#4446)

## Changes Made

Note: I did not move over the kernels because they are used for various
partitioning functions throughout core and other crates. To move the
kernels would be a bit bigger of a refactor that's out of scope for
getting everything converted to the new `ScalarUDF` methods.



## Related Issues

<!-- Link to related GitHub issues, e.g., "Closes #123" -->

## Checklist

- [ ] Documented in API Docs (if applicable)
- [ ] Documented in User Guide (if applicable)
- [ ] If adding a new documentation page, doc is added to
`docs/mkdocs.yml` navigation
- [ ] Documentation builds and is formatted properly (tag @/ccmao1130
for docs review)

Author: universalmind303

Cargo.lock

@@ -176,6 +176,7 @@ members = [
   "src/daft-functions-json",
   "src/daft-functions-list",
   "src/daft-functions-utf8",
+  "src/daft-functions-temporal",
   "src/daft-hash",
   "src/daft-image",
   "src/daft-io",
@@ -199,8 +200,7 @@ members = [
   "src/hyperloglog",
   "src/parquet2",
   "src/generated/spark-connect",
-  "src/daft-cli",
-  "src/daft-functions-temporal"
+  "src/daft-cli"
 ]
 
 [workspace.dependencies]
@@ -230,6 +230,7 @@ daft-functions = {path = "src/daft-functions"}
 daft-functions-binary = {path = "src/daft-functions-binary"}
 daft-functions-json = {path = "src/daft-functions-json"}
 daft-functions-list = {path = "src/daft-functions-list"}
+daft-functions-temporal = {path = "src/daft-functions-temporal"}
 daft-functions-utf8 = {path = "src/daft-functions-utf8"}
 daft-hash = {path = "src/daft-hash"}
 daft-local-execution = {path = "src/daft-local-execution"}

Cargo.toml

@@ -1229,30 +1229,6 @@ class ConnectionHandle:
     def shutdown(self) -> None: ...
     def port(self) -> int: ...
 
-# ---
-# expr.dt namespace
-# ---
-def dt_date(expr: PyExpr) -> PyExpr: ...
-def dt_day(expr: PyExpr) -> PyExpr: ...
-def dt_hour(expr: PyExpr) -> PyExpr: ...
-def dt_minute(expr: PyExpr) -> PyExpr: ...
-def dt_second(expr: PyExpr) -> PyExpr: ...
-def dt_millisecond(expr: PyExpr) -> PyExpr: ...
-def dt_microsecond(expr: PyExpr) -> PyExpr: ...
-def dt_nanosecond(expr: PyExpr) -> PyExpr: ...
-def dt_unix_date(expr: PyExpr) -> PyExpr: ...
-def dt_time(expr: PyExpr) -> PyExpr: ...
-def dt_month(expr: PyExpr) -> PyExpr: ...
-def dt_quarter(expr: PyExpr) -> PyExpr: ...
-def dt_year(expr: PyExpr) -> PyExpr: ...
-def dt_day_of_week(expr: PyExpr) -> PyExpr: ...
-def dt_day_of_month(expr: PyExpr) -> PyExpr: ...
-def dt_day_of_year(expr: PyExpr) -> PyExpr: ...
-def dt_week_of_year(expr: PyExpr) -> PyExpr: ...
-def dt_truncate(expr: PyExpr, interval: str, relative_to: PyExpr) -> PyExpr: ...
-def dt_to_unix_epoch(expr: PyExpr, timeunit: PyTimeUnit) -> PyExpr: ...
-def dt_strftime(expr: PyExpr, format: str | None = None) -> PyExpr: ...
-
 # ---
 # expr.list namespace
 # ---
@@ -1339,26 +1315,6 @@ class PySeries:
     def name(self) -> str: ...
     def rename(self, name: str) -> PySeries: ...
     def data_type(self) -> PyDataType: ...
-    def dt_date(self) -> PySeries: ...
-    def dt_day(self) -> PySeries: ...
-    def dt_hour(self) -> PySeries: ...
-    def dt_minute(self) -> PySeries: ...
-    def dt_second(self) -> PySeries: ...
-    def dt_millisecond(self) -> PySeries: ...
-    def dt_microsecond(self) -> PySeries: ...
-    def dt_nanosecond(self) -> PySeries: ...
-    def dt_unix_date(self) -> PySeries: ...
-    def dt_time(self) -> PySeries: ...
-    def dt_month(self) -> PySeries: ...
-    def dt_quarter(self) -> PySeries: ...
-    def dt_year(self) -> PySeries: ...
-    def dt_day_of_week(self) -> PySeries: ...
-    def dt_day_of_month(self) -> PySeries: ...
-    def dt_day_of_year(self) -> PySeries: ...
-    def dt_week_of_year(self) -> PySeries: ...
-    def dt_truncate(self, interval: str, relative_to: PySeries) -> PySeries: ...
-    def dt_strftime(self, format: str | None = None) -> PySeries: ...
-    def dt_to_unix_epoch(self, timeunit: PyTimeUnit) -> PySeries: ...
     def partitioning_days(self) -> PySeries: ...
     def partitioning_hours(self) -> PySeries: ...
     def partitioning_months(self) -> PySeries: ...

daft/daft/__init__.pyi

@@ -2168,7 +2168,7 @@ def date(self) -> Expression:
             (Showing first 3 of 3 rows)
 
         """
-        return Expression._from_pyexpr(native.dt_date(self._expr))
+        return self._eval_expressions("date")
 
     def day(self) -> Expression:
         """Retrieves the day for a datetime column.
@@ -2204,7 +2204,7 @@ def day(self) -> Expression:
             (Showing first 3 of 3 rows)
 
         """
-        return Expression._from_pyexpr(native.dt_day(self._expr))
+        return self._eval_expressions("day")
 
     def hour(self) -> Expression:
         """Retrieves the day for a datetime column.
@@ -2240,7 +2240,7 @@ def hour(self) -> Expression:
             (Showing first 3 of 3 rows)
 
         """
-        return Expression._from_pyexpr(native.dt_hour(self._expr))
+        return self._eval_expressions("hour")
 
     def minute(self) -> Expression:
         """Retrieves the minute for a datetime column.
@@ -2276,7 +2276,7 @@ def minute(self) -> Expression:
             (Showing first 3 of 3 rows)
 
         """
-        return Expression._from_pyexpr(native.dt_minute(self._expr))
+        return self._eval_expressions("minute")
 
     def second(self) -> Expression:
         """Retrieves the second for a datetime column.
@@ -2312,7 +2312,7 @@ def second(self) -> Expression:
             (Showing first 3 of 3 rows)
 
         """
-        return Expression._from_pyexpr(native.dt_second(self._expr))
+        return self._eval_expressions("second")
 
     def millisecond(self) -> Expression:
         """Retrieves the millisecond for a datetime column.
@@ -2345,7 +2345,7 @@ def millisecond(self) -> Expression:
             <BLANKLINE>
             (Showing first 3 of 3 rows)
         """
-        return Expression._from_pyexpr(native.dt_millisecond(self._expr))
+        return self._eval_expressions("millisecond")
 
     def microsecond(self) -> Expression:
         """Retrieves the microsecond for a datetime column.
@@ -2378,7 +2378,7 @@ def microsecond(self) -> Expression:
             (Showing first 3 of 3 rows)
 
         """
-        return Expression._from_pyexpr(native.dt_microsecond(self._expr))
+        return self._eval_expressions("microsecond")
 
     def nanosecond(self) -> Expression:
         """Retrieves the nanosecond for a datetime column.
@@ -2412,7 +2412,7 @@ def nanosecond(self) -> Expression:
             (Showing first 3 of 3 rows)
 
         """
-        return Expression._from_pyexpr(native.dt_nanosecond(self._expr))
+        return self._eval_expressions("nanosecond")
 
     def unix_date(self) -> Expression:
         """Retrieves the number of days since 1970-01-01 00:00:00 UTC.
@@ -2449,7 +2449,7 @@ def unix_date(self) -> Expression:
             (Showing first 3 of 3 rows)
 
         """
-        return Expression._from_pyexpr(native.dt_unix_date(self._expr))
+        return self._eval_expressions("unix_date")
 
     def time(self) -> Expression:
         """Retrieves the time for a datetime column.
@@ -2485,7 +2485,7 @@ def time(self) -> Expression:
             (Showing first 3 of 3 rows)
 
         """
-        return Expression._from_pyexpr(native.dt_time(self._expr))
+        return self._eval_expressions("time")
 
     def month(self) -> Expression:
         """Retrieves the month for a datetime column.
@@ -2520,7 +2520,7 @@ def month(self) -> Expression:
             (Showing first 3 of 3 rows)
 
         """
-        return Expression._from_pyexpr(native.dt_month(self._expr))
+        return self._eval_expressions("month")
 
     def quarter(self) -> Expression:
         """Retrieves the quarter for a datetime column.
@@ -2555,7 +2555,7 @@ def quarter(self) -> Expression:
             (Showing first 3 of 3 rows)
 
         """
-        return Expression._from_pyexpr(native.dt_quarter(self._expr))
+        return self._eval_expressions("quarter")
 
     def year(self) -> Expression:
         """Retrieves the year for a datetime column.
@@ -2590,7 +2590,7 @@ def year(self) -> Expression:
             (Showing first 3 of 3 rows)
 
         """
-        return Expression._from_pyexpr(native.dt_year(self._expr))
+        return self._eval_expressions("year")
 
     def day_of_week(self) -> Expression:
         """Retrieves the day of the week for a datetime column, starting at 0 for Monday and ending at 6 for Sunday.
@@ -2625,7 +2625,7 @@ def day_of_week(self) -> Expression:
             (Showing first 3 of 3 rows)
 
         """
-        return Expression._from_pyexpr(native.dt_day_of_week(self._expr))
+        return self._eval_expressions("day_of_week")
 
     def day_of_month(self) -> Expression:
         """Retrieves the day of the month for a datetime column.
@@ -2663,7 +2663,7 @@ def day_of_month(self) -> Expression:
             <BLANKLINE>
             (Showing first 4 of 4 rows)
         """
-        return Expression._from_pyexpr(native.dt_day_of_month(self._expr))
+        return self._eval_expressions("day_of_month")
 
     def day_of_year(self) -> Expression:
         """Retrieves the ordinal day for a datetime column. Starting at 1 for January 1st and ending at 365 or 366 for December 31st.
@@ -2701,7 +2701,7 @@ def day_of_year(self) -> Expression:
             <BLANKLINE>
             (Showing first 4 of 4 rows)
         """
-        return Expression._from_pyexpr(native.dt_day_of_year(self._expr))
+        return self._eval_expressions("day_of_year")
 
     def week_of_year(self) -> Expression:
         """Retrieves the week of the year for a datetime column.
@@ -2739,7 +2739,7 @@ def week_of_year(self) -> Expression:
             <BLANKLINE>
             (Showing first 4 of 4 rows)
         """
-        return Expression._from_pyexpr(native.dt_week_of_year(self._expr))
+        return self._eval_expressions("week_of_year")
 
     def truncate(self, interval: str, relative_to: Expression | None = None) -> Expression:
         """Truncates the datetime column to the specified interval.
@@ -2778,8 +2778,7 @@ def truncate(self, interval: str, relative_to: Expression | None = None) -> Expr
             (Showing first 3 of 3 rows)
 
         """
-        relative_to = Expression._to_expression(relative_to)
-        return Expression._from_pyexpr(native.dt_truncate(self._expr, interval, relative_to._expr))
+        return self._eval_expressions("truncate", relative_to, interval=interval)
 
     def to_unix_epoch(self, time_unit: str | TimeUnit | None = None) -> Expression:
         """Converts a datetime column to a Unix timestamp. with the specified time unit. (default: seconds).
@@ -2816,12 +2815,7 @@ def to_unix_epoch(self, time_unit: str | TimeUnit | None = None) -> Expression:
             <BLANKLINE>
             (Showing first 4 of 4 rows)
         """
-        if time_unit is None:
-            time_unit = TimeUnit.s()
-        if isinstance(time_unit, str):
-            time_unit = TimeUnit.from_str(time_unit)
-
-        return Expression._from_pyexpr(native.dt_to_unix_epoch(self._expr, time_unit._timeunit))
+        return self._eval_expressions("to_unix_epoch", time_unit=time_unit)
 
     def strftime(self, format: str | None = None) -> Expression:
         """Converts a datetime/date column to a string column.
@@ -2869,7 +2863,7 @@ def strftime(self, format: str | None = None) -> Expression:
             <BLANKLINE>
             (Showing first 3 of 3 rows)
         """
-        return Expression._from_pyexpr(native.dt_strftime(self._expr, format))
+        return self._eval_expressions("strftime", format=format)
 
     def total_seconds(self) -> Expression:
         """Calculates the total number of seconds for a duration column.
@@ -2913,8 +2907,7 @@ def total_seconds(self) -> Expression:
             <BLANKLINE>
             (Showing first 6 of 6 rows)
         """
-        f = native.get_function_from_registry("total_seconds")
-        return Expression._from_pyexpr(f(self._expr))
+        return self._eval_expressions("total_seconds")
 
     def total_milliseconds(self) -> Expression:
         """Calculates the total number of milliseconds for a duration column.
@@ -2958,8 +2951,7 @@ def total_milliseconds(self) -> Expression:
             <BLANKLINE>
             (Showing first 6 of 6 rows)
         """
-        f = native.get_function_from_registry("total_milliseconds")
-        return Expression._from_pyexpr(f(self._expr))
+        return self._eval_expressions("total_milliseconds")
 
     def total_microseconds(self) -> Expression:
         """Calculates the total number of microseconds for a duration column.
@@ -3003,8 +2995,7 @@ def total_microseconds(self) -> Expression:
             <BLANKLINE>
             (Showing first 6 of 6 rows)
         """
-        f = native.get_function_from_registry("total_microseconds")
-        return Expression._from_pyexpr(f(self._expr))
+        return self._eval_expressions("total_microseconds")
 
     def total_nanoseconds(self) -> Expression:
         """Calculates the total number of nanoseconds for a duration column.
@@ -3048,8 +3039,7 @@ def total_nanoseconds(self) -> Expression:
             <BLANKLINE>
             (Showing first 6 of 6 rows)
         """
-        f = native.get_function_from_registry("total_nanoseconds")
-        return Expression._from_pyexpr(f(self._expr))
+        return self._eval_expressions("total_nanoseconds")
 
     def total_minutes(self) -> Expression:
         """Calculates the total number of minutes for a duration column.
@@ -3093,8 +3083,7 @@ def total_minutes(self) -> Expression:
             <BLANKLINE>
             (Showing first 6 of 6 rows)
         """
-        f = native.get_function_from_registry("total_minutes")
-        return Expression._from_pyexpr(f(self._expr))
+        return self._eval_expressions("total_minutes")
 
     def total_hours(self) -> Expression:
         """Calculates the total number of hours for a duration column.
@@ -3138,8 +3127,7 @@ def total_hours(self) -> Expression:
             <BLANKLINE>
             (Showing first 6 of 6 rows)
         """
-        f = native.get_function_from_registry("total_hours")
-        return Expression._from_pyexpr(f(self._expr))
+        return self._eval_expressions("total_hours")
 
     def total_days(self) -> Expression:
         """Calculates the total number of days for a duration column.
@@ -3183,8 +3171,7 @@ def total_days(self) -> Expression:
             <BLANKLINE>
             (Showing first 6 of 6 rows)
         """
-        f = native.get_function_from_registry("total_days")
-        return Expression._from_pyexpr(f(self._expr))
+        return self._eval_expressions("total_days")
 
 
 class ExpressionStringNamespace(ExpressionNamespace):