refactor: add agg_ops.TimeSeriesDiffOp

Author: chelsea-lin

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

@@ -293,3 +293,20 @@ def _(
     # Will be null if all inputs are null. Pandas defaults to zero sum though.
     zero = pd.to_timedelta(0) if column.dtype == dtypes.TIMEDELTA_DTYPE else 0
     return sge.func("IFNULL", expr, ir._literal(zero, column.dtype))
+
+
+@UNARY_OP_REGISTRATION.register(agg_ops.TimeSeriesDiffOp)
+def _(
+    op: agg_ops.TimeSeriesDiffOp,
+    column: typed_expr.TypedExpr,
+    window: typing.Optional[window_spec.WindowSpec] = None,
+) -> sge.Expression:
+    if column.dtype != dtypes.TIMESTAMP_DTYPE:
+        raise TypeError(f"Cannot perform time 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)
+    return sge.TimestampDiff(
+            this=column.expr,
+            expression=shifted,
+            unit=sge.Identifier(this="MICROSECOND"),
+        )

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

@@ -0,0 +1,17 @@
+WITH `bfcte_0` AS (
+  SELECT
+    `timestamp_col` AS `bfcol_0`
+  FROM `bigframes-dev`.`sqlglot_test`.`scalar_types`
+), `bfcte_1` AS (
+  SELECT
+    *,
+    TIMESTAMP_DIFF(
+      `bfcol_0`,
+      LAG(`bfcol_0`, 1) OVER (ORDER BY `bfcol_0` IS NULL ASC NULLS LAST, `bfcol_0` ASC NULLS LAST),
+      MICROSECOND
+    ) AS `bfcol_1`
+  FROM `bfcte_0`
+)
+SELECT
+  `bfcol_1` AS `diff_time`
+FROM `bfcte_1`

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

@@ -331,3 +331,14 @@ def test_sum(scalar_types_df: bpd.DataFrame, snapshot):
     )
 
     snapshot.assert_match(sql, "out.sql")
+
+
+def test_time_series_diff(scalar_types_df: bpd.DataFrame, snapshot):
+    col_name = "timestamp_col"
+    bf_df = scalar_types_df[[col_name]]
+    window = window_spec.WindowSpec(ordering=(ordering.ascending_over(col_name),))
+    op = agg_exprs.UnaryAggregation(
+        agg_ops.TimeSeriesDiffOp(periods=1), expression.deref(col_name)
+    )
+    sql = _apply_unary_window_op(bf_df, op, window, "diff_time")
+    snapshot.assert_match(sql, "out.sql")