refactor: add agg_ops.DateSeriesDiffOp

Author: chelsea-lin

bigframes/core/compile/sqlglot/aggregations/unary_compiler.py

@@ -98,6 +98,27 @@ def _(
     return apply_window_if_present(sge.func("COUNT", column.expr), window)
 
 
+@UNARY_OP_REGISTRATION.register(agg_ops.DateSeriesDiffOp)
+def _(
+    op: agg_ops.DateSeriesDiffOp,
+    column: typed_expr.TypedExpr,
+    window: typing.Optional[window_spec.WindowSpec] = None,
+) -> sge.Expression:
+    if column.dtype != dtypes.DATE_DTYPE:
+        raise TypeError(f"Cannot perform date series diff on type {column.dtype}")
+    shift_op_impl = UNARY_OP_REGISTRATION[agg_ops.ShiftOp(0)]
+    shifted = shift_op_impl(agg_ops.ShiftOp(op.periods), column, window)
+    # Conversion factor from days to microseconds
+    conversion_factor = 24 * 60 * 60 * 1_000_000
+    return sge.Cast(
+        this=sge.DateDiff(
+            this=column.expr, expression=shifted, unit=sge.Identifier(this="DAY")
+        )
+        * sge.convert(conversion_factor),
+        to="INT64",
+    )
+
+
 @UNARY_OP_REGISTRATION.register(agg_ops.DenseRankOp)
 def _(
     op: agg_ops.DenseRankOp,
@@ -306,7 +327,7 @@ def _(
     shift_op_impl = UNARY_OP_REGISTRATION[agg_ops.ShiftOp(0)]
     shifted = shift_op_impl(agg_ops.ShiftOp(op.periods), column, window)
     return sge.TimestampDiff(
-            this=column.expr,
-            expression=shifted,
-            unit=sge.Identifier(this="MICROSECOND"),
-        )
+        this=column.expr,
+        expression=shifted,
+        unit=sge.Identifier(this="MICROSECOND"),
+    )

tests/unit/core/compile/sqlglot/aggregations/snapshots/test_unary_compiler/test_date_series_diff/out.sql

@@ -0,0 +1,17 @@
+WITH `bfcte_0` AS (
+  SELECT
+    `date_col` AS `bfcol_0`
+  FROM `bigframes-dev`.`sqlglot_test`.`scalar_types`
+), `bfcte_1` AS (
+  SELECT
+    *,
+    CAST(DATE_DIFF(
+      `bfcol_0`,
+      LAG(`bfcol_0`, 1) OVER (ORDER BY `bfcol_0` IS NULL ASC NULLS LAST, `bfcol_0` ASC NULLS LAST),
+      day
+    ) * 86400000000 AS INT64) AS `bfcol_1`
+  FROM `bfcte_0`
+)
+SELECT
+  `bfcol_1` AS `diff_date`
+FROM `bfcte_1`

tests/unit/core/compile/sqlglot/aggregations/test_unary_compiler.py

@@ -127,6 +127,17 @@ def test_dense_rank(scalar_types_df: bpd.DataFrame, snapshot):
     snapshot.assert_match(sql, "out.sql")
 
 
+def test_date_series_diff(scalar_types_df: bpd.DataFrame, snapshot):
+    col_name = "date_col"
+    bf_df = scalar_types_df[[col_name]]
+    window = window_spec.WindowSpec(ordering=(ordering.ascending_over(col_name),))
+    op = agg_exprs.UnaryAggregation(
+        agg_ops.DateSeriesDiffOp(periods=1), expression.deref(col_name)
+    )
+    sql = _apply_unary_window_op(bf_df, op, window, "diff_date")
+    snapshot.assert_match(sql, "out.sql")
+
+
 def test_diff(scalar_types_df: bpd.DataFrame, snapshot):
     # Test integer
     int_col = "int64_col"