Skip to content

chore: implement comparison_ops for sqlglot compiler #1999

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 6 commits into from
Closed

chore: implement comparison_ops for sqlglot compiler #1999

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
26f8a77
chore: implement comparison_ops for sqlglot compiler
chelsea-lin Aug 15, 2025
6f83b54
implement eq_null_match_op
chelsea-lin Aug 19, 2025
7ad273a
implement ne_op
chelsea-lin Aug 19, 2025
68a77eb
implement gt_op
chelsea-lin Aug 19, 2025
f485d26
implement lt_op
chelsea-lin Aug 19, 2025
4b2a651
implement le
chelsea-lin Aug 20, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
83 changes: 82 additions & 1 deletion bigframes/core/compile/sqlglot/expressions/binary_compiler.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -73,6 +73,37 @@ def _(op, left: TypedExpr, right: TypedExpr) -> sge.Expression:
)


@BINARY_OP_REGISTRATION.register(ops.eq_op)
def _(op, left: TypedExpr, right: TypedExpr) -> sge.Expression:
left_expr = left.expr
if left.dtype == dtypes.BOOL_DTYPE:
left_expr = sge.Cast(this=left_expr, to="INT64")
right_expr = right.expr
if right.dtype == dtypes.BOOL_DTYPE:
right_expr = sge.Cast(this=right_expr, to="INT64")
return sge.EQ(this=left_expr, expression=right_expr)


@BINARY_OP_REGISTRATION.register(ops.eq_null_match_op)
def _(op, left: TypedExpr, right: TypedExpr) -> sge.Expression:
left_expr = left.expr
if left.dtype == dtypes.BOOL_DTYPE and right.dtype != dtypes.BOOL_DTYPE:
left_expr = sge.Cast(this=left_expr, to="INT64")

right_expr = right.expr
if right.dtype == dtypes.BOOL_DTYPE and left.dtype != dtypes.BOOL_DTYPE:
right_expr = sge.Cast(this=right_expr, to="INT64")

sentinel = sge.convert("$NULL_SENTINEL$")
left_coalesce = sge.Coalesce(
this=sge.Cast(this=left_expr, to="STRING"), expressions=[sentinel]
)
right_coalesce = sge.Coalesce(
this=sge.Cast(this=right_expr, to="STRING"), expressions=[sentinel]
)
return sge.EQ(this=left_coalesce, expression=right_coalesce)


@BINARY_OP_REGISTRATION.register(ops.div_op)
def _(op, left: TypedExpr, right: TypedExpr) -> sge.Expression:
left_expr = left.expr
Expand All @@ -91,14 +122,53 @@ def _(op, left: TypedExpr, right: TypedExpr) -> sge.Expression:

@BINARY_OP_REGISTRATION.register(ops.ge_op)
def _(op, left: TypedExpr, right: TypedExpr) -> sge.Expression:
return sge.GTE(this=left.expr, expression=right.expr)
left_expr = left.expr
if left.dtype == dtypes.BOOL_DTYPE:
left_expr = sge.Cast(this=left_expr, to="INT64")
right_expr = right.expr
if right.dtype == dtypes.BOOL_DTYPE:
right_expr = sge.Cast(this=right_expr, to="INT64")
return sge.GTE(this=left_expr, expression=right_expr)


@BINARY_OP_REGISTRATION.register(ops.gt_op)
def _(op, left: TypedExpr, right: TypedExpr) -> sge.Expression:
left_expr = left.expr
if left.dtype == dtypes.BOOL_DTYPE:
left_expr = sge.Cast(this=left_expr, to="INT64")
right_expr = right.expr
if right.dtype == dtypes.BOOL_DTYPE:
right_expr = sge.Cast(this=right_expr, to="INT64")
return sge.GT(this=left_expr, expression=right_expr)


@BINARY_OP_REGISTRATION.register(ops.JSONSet)
def _(op, left: TypedExpr, right: TypedExpr) -> sge.Expression:
return sge.func("JSON_SET", left.expr, sge.convert(op.json_path), right.expr)


@BINARY_OP_REGISTRATION.register(ops.lt_op)
def _(op, left: TypedExpr, right: TypedExpr) -> sge.Expression:
left_expr = left.expr
if left.dtype == dtypes.BOOL_DTYPE:
left_expr = sge.Cast(this=left_expr, to="INT64")
right_expr = right.expr
if right.dtype == dtypes.BOOL_DTYPE:
right_expr = sge.Cast(this=right_expr, to="INT64")
return sge.LT(this=left_expr, expression=right_expr)


@BINARY_OP_REGISTRATION.register(ops.le_op)
def _(op, left: TypedExpr, right: TypedExpr) -> sge.Expression:
left_expr = left.expr
if left.dtype == dtypes.BOOL_DTYPE:
left_expr = sge.Cast(this=left_expr, to="INT64")
right_expr = right.expr
if right.dtype == dtypes.BOOL_DTYPE:
right_expr = sge.Cast(this=right_expr, to="INT64")
return sge.LTE(this=left_expr, expression=right_expr)


@BINARY_OP_REGISTRATION.register(ops.mul_op)
def _(op, left: TypedExpr, right: TypedExpr) -> sge.Expression:
left_expr = left.expr
Expand All @@ -118,6 +188,17 @@ def _(op, left: TypedExpr, right: TypedExpr) -> sge.Expression:
return result


@BINARY_OP_REGISTRATION.register(ops.ne_op)
def _(op, left: TypedExpr, right: TypedExpr) -> sge.Expression:
left_expr = left.expr
if left.dtype == dtypes.BOOL_DTYPE:
left_expr = sge.Cast(this=left_expr, to="INT64")
right_expr = right.expr
if right.dtype == dtypes.BOOL_DTYPE:
right_expr = sge.Cast(this=right_expr, to="INT64")
return sge.NEQ(this=left_expr, expression=right_expr)


@BINARY_OP_REGISTRATION.register(ops.sub_op)
def _(op, left: TypedExpr, right: TypedExpr) -> sge.Expression:
if dtypes.is_numeric(left.dtype) and dtypes.is_numeric(right.dtype):
Expand Down
3 changes: 2 additions & 1 deletion tests/system/small/engines/test_comparison_ops.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,7 +48,8 @@ def apply_op_pairwise(
return new_arr


@pytest.mark.parametrize("engine", ["polars", "bq"], indirect=True)
# @pytest.mark.parametrize("engine", ["polars", "bq", "bq-sqlglot"], indirect=True)
@pytest.mark.parametrize("engine", ["bq-sqlglot"], indirect=True)
@pytest.mark.parametrize(
"op",
[
Expand Down
14 changes: 14 additions & 0 deletions ...ore/compile/sqlglot/expressions/snapshots/test_binary_compiler/test_eq_null_match/out.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,14 @@
WITH `bfcte_0` AS (
SELECT
`bool_col` AS `bfcol_0`,
`int64_col` AS `bfcol_1`
FROM `bigframes-dev`.`sqlglot_test`.`scalar_types`
), `bfcte_1` AS (
SELECT
*,
COALESCE(CAST(`bfcol_1` AS STRING), '$NULL_SENTINEL$') = COALESCE(CAST(CAST(`bfcol_0` AS INT64) AS STRING), '$NULL_SENTINEL$') AS `bfcol_4`
FROM `bfcte_0`
)
SELECT
`bfcol_4` AS `int64_col`
FROM `bfcte_1`
54 changes: 54 additions & 0 deletions ...t/core/compile/sqlglot/expressions/snapshots/test_binary_compiler/test_ge_numeric/out.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,54 @@
WITH `bfcte_0` AS (
SELECT
`bool_col` AS `bfcol_0`,
`int64_col` AS `bfcol_1`,
`rowindex` AS `bfcol_2`
FROM `bigframes-dev`.`sqlglot_test`.`scalar_types`
), `bfcte_1` AS (
SELECT
*,
`bfcol_2` AS `bfcol_6`,
`bfcol_1` AS `bfcol_7`,
`bfcol_0` AS `bfcol_8`,
`bfcol_1` >= `bfcol_1` AS `bfcol_9`
FROM `bfcte_0`
), `bfcte_2` AS (
SELECT
*,
`bfcol_6` AS `bfcol_14`,
`bfcol_7` AS `bfcol_15`,
`bfcol_8` AS `bfcol_16`,
`bfcol_9` AS `bfcol_17`,
`bfcol_7` >= 1 AS `bfcol_18`
FROM `bfcte_1`
), `bfcte_3` AS (
SELECT
*,
`bfcol_14` AS `bfcol_24`,
`bfcol_15` AS `bfcol_25`,
`bfcol_16` AS `bfcol_26`,
`bfcol_17` AS `bfcol_27`,
`bfcol_18` AS `bfcol_28`,
`bfcol_15` >= CAST(`bfcol_16` AS INT64) AS `bfcol_29`
FROM `bfcte_2`
), `bfcte_4` AS (
SELECT
*,
`bfcol_24` AS `bfcol_36`,
`bfcol_25` AS `bfcol_37`,
`bfcol_26` AS `bfcol_38`,
`bfcol_27` AS `bfcol_39`,
`bfcol_28` AS `bfcol_40`,
`bfcol_29` AS `bfcol_41`,
CAST(`bfcol_26` AS INT64) >= `bfcol_25` AS `bfcol_42`
FROM `bfcte_3`
)
SELECT
`bfcol_36` AS `rowindex`,
`bfcol_37` AS `int64_col`,
`bfcol_38` AS `bool_col`,
`bfcol_39` AS `int_ge_int`,
`bfcol_40` AS `int_ge_1`,
`bfcol_41` AS `int_ge_bool`,
`bfcol_42` AS `bool_ge_int`
FROM `bfcte_4`
54 changes: 54 additions & 0 deletions ...t/core/compile/sqlglot/expressions/snapshots/test_binary_compiler/test_gt_numeric/out.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,54 @@
WITH `bfcte_0` AS (
SELECT
`bool_col` AS `bfcol_0`,
`int64_col` AS `bfcol_1`,
`rowindex` AS `bfcol_2`
FROM `bigframes-dev`.`sqlglot_test`.`scalar_types`
), `bfcte_1` AS (
SELECT
*,
`bfcol_2` AS `bfcol_6`,
`bfcol_1` AS `bfcol_7`,
`bfcol_0` AS `bfcol_8`,
`bfcol_1` > `bfcol_1` AS `bfcol_9`
FROM `bfcte_0`
), `bfcte_2` AS (
SELECT
*,
`bfcol_6` AS `bfcol_14`,
`bfcol_7` AS `bfcol_15`,
`bfcol_8` AS `bfcol_16`,
`bfcol_9` AS `bfcol_17`,
`bfcol_7` > 1 AS `bfcol_18`
FROM `bfcte_1`
), `bfcte_3` AS (
SELECT
*,
`bfcol_14` AS `bfcol_24`,
`bfcol_15` AS `bfcol_25`,
`bfcol_16` AS `bfcol_26`,
`bfcol_17` AS `bfcol_27`,
`bfcol_18` AS `bfcol_28`,
`bfcol_15` > CAST(`bfcol_16` AS INT64) AS `bfcol_29`
FROM `bfcte_2`
), `bfcte_4` AS (
SELECT
*,
`bfcol_24` AS `bfcol_36`,
`bfcol_25` AS `bfcol_37`,
`bfcol_26` AS `bfcol_38`,
`bfcol_27` AS `bfcol_39`,
`bfcol_28` AS `bfcol_40`,
`bfcol_29` AS `bfcol_41`,
CAST(`bfcol_26` AS INT64) > `bfcol_25` AS `bfcol_42`
FROM `bfcte_3`
)
SELECT
`bfcol_36` AS `rowindex`,
`bfcol_37` AS `int64_col`,
`bfcol_38` AS `bool_col`,
`bfcol_39` AS `int_gt_int`,
`bfcol_40` AS `int_gt_1`,
`bfcol_41` AS `int_gt_bool`,
`bfcol_42` AS `bool_gt_int`
FROM `bfcte_4`
54 changes: 54 additions & 0 deletions ...t/core/compile/sqlglot/expressions/snapshots/test_binary_compiler/test_le_numeric/out.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,54 @@
WITH `bfcte_0` AS (
SELECT
`bool_col` AS `bfcol_0`,
`int64_col` AS `bfcol_1`,
`rowindex` AS `bfcol_2`
FROM `bigframes-dev`.`sqlglot_test`.`scalar_types`
), `bfcte_1` AS (
SELECT
*,
`bfcol_2` AS `bfcol_6`,
`bfcol_1` AS `bfcol_7`,
`bfcol_0` AS `bfcol_8`,
`bfcol_1` <= `bfcol_1` AS `bfcol_9`
FROM `bfcte_0`
), `bfcte_2` AS (
SELECT
*,
`bfcol_6` AS `bfcol_14`,
`bfcol_7` AS `bfcol_15`,
`bfcol_8` AS `bfcol_16`,
`bfcol_9` AS `bfcol_17`,
`bfcol_7` <= 1 AS `bfcol_18`
FROM `bfcte_1`
), `bfcte_3` AS (
SELECT
*,
`bfcol_14` AS `bfcol_24`,
`bfcol_15` AS `bfcol_25`,
`bfcol_16` AS `bfcol_26`,
`bfcol_17` AS `bfcol_27`,
`bfcol_18` AS `bfcol_28`,
`bfcol_15` <= CAST(`bfcol_16` AS INT64) AS `bfcol_29`
FROM `bfcte_2`
), `bfcte_4` AS (
SELECT
*,
`bfcol_24` AS `bfcol_36`,
`bfcol_25` AS `bfcol_37`,
`bfcol_26` AS `bfcol_38`,
`bfcol_27` AS `bfcol_39`,
`bfcol_28` AS `bfcol_40`,
`bfcol_29` AS `bfcol_41`,
CAST(`bfcol_26` AS INT64) <= `bfcol_25` AS `bfcol_42`
FROM `bfcte_3`
)
SELECT
`bfcol_36` AS `rowindex`,
`bfcol_37` AS `int64_col`,
`bfcol_38` AS `bool_col`,
`bfcol_39` AS `int_le_int`,
`bfcol_40` AS `int_le_1`,
`bfcol_41` AS `int_le_bool`,
`bfcol_42` AS `bool_le_int`
FROM `bfcte_4`
54 changes: 54 additions & 0 deletions ...t/core/compile/sqlglot/expressions/snapshots/test_binary_compiler/test_lt_numeric/out.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,54 @@
WITH `bfcte_0` AS (
SELECT
`bool_col` AS `bfcol_0`,
`int64_col` AS `bfcol_1`,
`rowindex` AS `bfcol_2`
FROM `bigframes-dev`.`sqlglot_test`.`scalar_types`
), `bfcte_1` AS (
SELECT
*,
`bfcol_2` AS `bfcol_6`,
`bfcol_1` AS `bfcol_7`,
`bfcol_0` AS `bfcol_8`,
`bfcol_1` < `bfcol_1` AS `bfcol_9`
FROM `bfcte_0`
), `bfcte_2` AS (
SELECT
*,
`bfcol_6` AS `bfcol_14`,
`bfcol_7` AS `bfcol_15`,
`bfcol_8` AS `bfcol_16`,
`bfcol_9` AS `bfcol_17`,
`bfcol_7` < 1 AS `bfcol_18`
FROM `bfcte_1`
), `bfcte_3` AS (
SELECT
*,
`bfcol_14` AS `bfcol_24`,
`bfcol_15` AS `bfcol_25`,
`bfcol_16` AS `bfcol_26`,
`bfcol_17` AS `bfcol_27`,
`bfcol_18` AS `bfcol_28`,
`bfcol_15` < CAST(`bfcol_16` AS INT64) AS `bfcol_29`
FROM `bfcte_2`
), `bfcte_4` AS (
SELECT
*,
`bfcol_24` AS `bfcol_36`,
`bfcol_25` AS `bfcol_37`,
`bfcol_26` AS `bfcol_38`,
`bfcol_27` AS `bfcol_39`,
`bfcol_28` AS `bfcol_40`,
`bfcol_29` AS `bfcol_41`,
CAST(`bfcol_26` AS INT64) < `bfcol_25` AS `bfcol_42`
FROM `bfcte_3`
)
SELECT
`bfcol_36` AS `rowindex`,
`bfcol_37` AS `int64_col`,
`bfcol_38` AS `bool_col`,
`bfcol_39` AS `int_lt_int`,
`bfcol_40` AS `int_lt_1`,
`bfcol_41` AS `int_lt_bool`,
`bfcol_42` AS `bool_lt_int`
FROM `bfcte_4`
Loading